Case Studies:General advice, Crossing the divide with a simple bridge in Timbuktu, Networking Mérida State

<< Economic Sustainability:Create a Mission Statement, Evaluate the Demand for Potential Offerings

Appendix A: Resources:Antennas and antenna design, Security >>

Case Studies

No matter how much planning goes into building a link or node location, you

will inevitably have to jump in and actually install something. This is the mo-

ment of truth that demonstrates just how accurate your estimates and predic-

tions prove to be.

It is a rare day when everything goes precisely as planned. Even after you

install your 1st, 10th, or 100th node, you will still find that things do not al-

ways work out as you might have intended. This chapter describes some of

our more memorable network projects. Whether you are about to embark on

your first wireless project or you are an old hand at this, it is reassuring to

remember that there is always more to learn.

General advice

The economies of developing countries are very different from the devel-

oped world, and thus a process or solution designed for a more developed

country may not be suitable in West Africa, or Southern Asia. Specifically,

the cost of locally produced materials and the cost of labour will be negligi-

ble, whereas imported goods can be much more expensive when com-

pared to its cost in the developed world. For example, one can manufacture

and install a tower for a tenth of the cost of a tower in the United States, but

the price of an antenna might be double. Solutions that capitalize on local

competitive advantages, namely cheap labour and locally found materials,

will be the easiest to replicate.

Finding the right equipment is one of the most difficult tasks in developing

markets. Because transportation, communication and economic systems are

not developed, the right materials or equipment can be difficult and often im-

299

300

Chapter 11: Case Studies

possible to find. A fuse, for example, is difficult to find, thus finding wire that

has a burn-up at a certain amperage and can substitute is a great advantage.

Finding local substitutes for materials also encourages local entrepreneur-

ship, ownership, and can save money.

Equipment enclosures

Cheap plastics are everywhere in the developing world, but they are made of

poor materials and are thin, thus mostly unsuitable for enclosing equipment.

PVC tubing is far more resilient and is made to be waterproof. In West Africa,

the most common PVC is found in plumbing, sized from 90mm to 220mm.

Access points such as the Routerboard 500 and 200 can fit into such tubing,

and with end-caps that are torched-on, they can make very robust waterproof

enclosures. They also have the added benefit of being aerodynamic and un-

interesting to passers-by. The resulting space left around the equipment as-

sures adequate air circulation. Also, it is often best to leave an exhaust hole

at the bottom of the PVC enclosure. The author did find that leaving open

holes can become a problem. In one instance ants decided to nest 25 meters

above ground inside the PVC holding the access point. Using a wire mesh

cover made from locally available screen material is advised to secure the

exhaust hole from infestations.

Antenna masts

Recovering used materials has become an important industry for the poorest

countries. From old cars to televisions, any material that has value will be

stripped, sold, or re-used. For example, you will see vehicles torn apart piece

by piece and day by day. The resulting metal is sorted and then tossed into a

truck to be sold. Local metal workers will already be familiar with how to

make television masts from scrap metal. A few quick adaptations and these

same masts can be re-purposed for wireless networks.

The typical mast is the 5 meter pole, comprised of a single 30mm diameter

pipe which is then planted into cement. It's best to construct the mast in two

parts, with a removable mast that fits into a base which is slightly larger in

diameter. Alternately, the mast may be made with arms that can be securely

cemented into a wall. This project is easy, but requires the use of a ladder to

complete and therefore some caution is suggested.

This type of mast can be augmented by several meters with the use of guy

lines. To sturdy the pole, plant three lines 120 degrees apart, forming an

angle of at least 33 degrees with the tower.

Chapter 11: Case Studies

301

Above all: involve the local community

Community involvement is imperative in assuring the success and sustain-

ability of a project. Involving the community in a project can be the greatest

challenge, but if the community is not involved the technology will not serve

their needs, nor will it be accepted. Moreover, a community might be afraid

and could subvert an initiative. Regardless of the complexity of the undertak-

ing, a successful project needs support and buy-in from those it will serve.

An effective strategy in gaining support is to find a respected champion

whose motives are palatable. Find the person, or persons whom are most

likely to be interested in the project. Often, you will need to involve such

champions as advisors, or as members of a steering committee. These peo-

ple will already have the trust of the community, will know who to approach,

and can speak the language of the community. Take your time and be selec-

tive in finding the right people for your project. No other decision will affect

your project more than having effective, trusted local people on your team.

In addition, take note of key players in an institution, or community. Identify

those people whom are likely to be opponents and proponents of your pro-

ject. As early as possible, attempt to earn the support of the potential propo-

nents and to diffuse the opponents. This is a difficult task and one that re-

quires intimate knowledge of the institution or community. If the project does

not have a local ally, the project must take time to acquire this knowledge

and trust from the community.

Be careful in choosing your allies. A "town-hall" meeting is often useful to see

local politics, alliances, and feuds in play. Thereafter, it is easier to decide on

whom to ally, champion and whom to diffuse. Try to not build unwarranted

enthusiasm. It is important to be honest, frank, and not to make promises

that you cannot keep.

In largely illiterate communities, focus on digital to analog services such as

Internet for radio stations, printing on-line articles and photos, and other non-

textual applications. Do not try to introduce a technology to a community

without understanding which applications will truly will serve the community.

Often the community will have little idea how new technologies will help their

problems. Simply providing new features is useless without an understand-

ing of how the community will benefit.

When gathering information, verify the facts that you are given. If you want to

know the financial status of a company/organization, ask to see an electricity bill,

or phone bill. Have they been paying their bills? At times, potential beneficiaries

will compromise their own values in hopes of winning funds or equipment.

Most often, local partners who trust you will be very frank, honest, and helpful.

302

Chapter 11: Case Studies

Another common pitfall is what I call "divorced parents" syndrome, where

NGOs, donors, and partners are not told of each others involvement with the

beneficiary. Savvy beneficiaries can earn handsome rewards by letting

NGOs and donors lavish them with equipment, training and funds. It is impor-

tant to know which other organizations are involved so you can understand

how their activities might impact your own. For example, I once designed a

project for a rural school in Mali. My team installed an open source system

with used computers and spent several days training people how to use it.

The project was deemed a success, but shortly after the installation, another

donor arrived with brand-new Pentium 4 computers running Windows XP.

The students quickly abandoned the older computers and lined-up to use the

new computers. It would have been better to negotiate with the school in ad-

vance, to know their commitment to the project. If they had been frank, the

computers that are now sitting unused could have been deployed to another

school where they would be used.

In many rural communities in under-developed economies, law and policies

are weak, and contracts can be effectively meaningless. Often, other assur-

ances must be found. This is where pre-paid services are ideal, as they do

not require a legal contract. Commitment is assured by the investment of

funds before service is given.

Buy-in also requires that those involved invest in the project themselves. A

project should ask for reciprocal involvement from the community.

Above all, the "no-go" option should always be evaluated. If a local ally and

community buy-in cannot be had, the project should consider choosing a dif-

ferent community or beneficiary. There must be a negotiation; equipment,

money, and training cannot be gifts. The community must be involved and

they too must contribute.

Ian Howard

Case study: Crossing the divide with a

simple bridge in Timbuktu

Networks ultimately connect people together, and therefore always involve a

political component. The cost of Internet in less developed economies is high

and the ability to pay is low, which adds to the political challenges. Attempt-

ing to superimpose a network where human networks are not fully function-

ing is nearly impossible in the long term. Trying to do so can leave a project

on unstable social ground, threatening its existence. This is where the low

cost and mobility of a wireless network can be advantageous.

Chapter 11: Case Studies

303

The author's team was asked by funders to determine how to connect a rural

radio station with a very small (2 computer) telecentre to the Internet in Tim-

buktu, the desert capital of Mali. Timbuktu is widely known as an outpost in

the most remote area of the world. At this site, the team decided to imple-

ment a model which has been called the parasitic wireless model. This

model takes a wireless "feed" that is spliced from an existing network, and

extends that network to a client site using a simple bridged network. This

model was chosen because it requires no significant investment by the sup-

porting organization. While it added a source of revenue for the telecentre, it

did not add a significant operational cost. This solution meant that the client

site could get cheap Internet, albeit not as fast or as reliable as a dedicated

solution. Because of opposed usage patterns between an office and a tele-

centre there was no perceptible slowing of the network for either party.

Though in an ideal situation it would be best to encourage more development

of the small telecentre into an ISP, neither the telecentre nor the market were

deemed ready. As is often the case, there were serious concerns about

whether this telecentre could become self-sustaining once its funders de-

parted. Thus, this solution minimized the initial investment while achieving

two goals: first, it extended the Internet to the target beneficiary, a radio sta-

tion, at an affordable cost. Second, it added a small additional revenue

source for the telecentre while not increasing its operational costs, or adding

complexity to the system.

The people

Timbuktu is remote, though having a world renowned name. Being a symbol

of remoteness, many projects have wanted to "stake a flag" in the sands of

this desert city. Thus, there are a number of information and communications

technologies (ICT) activities in the area. At last count there were 8 satellite

connections into Timbuktu, most of which service special interests except for

the two carriers, SOTELMA and Ikatel. They currently use VSAT to link their

telephone networks to the rest of the country. This telecentre used an X.25

connection to one of these telcos, which then relayed the connection back to

Bamako. Relative to other remote cities in the country, Timbuktu has a fair

number of trained IT staff, three existing telecentres, plus the newly installed

telecentre at the radio station. The city is to some degree over saturated with

Internet, precluding any private, commercial interests from being sustainable.

Design Choices

In this installation the client site is only 1 km away directly by line of sight.

Two modified Linksys access points, flashed with OpenWRT and set to

bridge mode, were installed. One was installed on the wall of the telecentre,

and the other was installed 5 meters up the radio station's mast. The only

configuration parameters required on both devices were the ssid and the

channel. Simple 14 dBi panel antennas (from http://hyperlinktech.com/)

304

Chapter 11: Case Studies

were used. At the Internet side, the access point and antenna were fastened

using cement plugs and screws onto the side of the building, facing the client

site. At the client site, an existing antenna mast was used. The access point

and antenna were mounted using pipe rings.

To disconnect the client, the telecentre simply unplugs the bridge on their

side. An additional site will eventually be installed, and it too will have its own

bridge at the telecentre so that staff can physically disconnect the client if

they have not paid. Though crude, this solution is effective and reduces risk

that the staff would make a mistake while making changes to the configura-

tion of the system. Having a bridge dedicated to one connection also simpli-

fied installation at the central site. as the installation team was able to choose

the best spot for connecting the client sites. Though it is not optimal to

bridge a network (rather than route network traffic), when technology knowl-

edge is low and one wants to install a very simple system this can be a rea-

sonable solution for small networks. The bridge makes systems installed at

the remote site (the radio station) appear as though they are simply con-

nected to the local network.

Financial model

The financial model here is simple. The telecentre charges a monthly fee,

about $30 per connected computer to the radio station. This was many times

cheaper than the alternative. The telecentre is located in the court of the

Mayor's office, so the principle client of the telecentre is the Mayor s staff.

This was important because the radio station did not want to compete for

clientele with the telecentre and the radio station's systems were primarily

intended for the radio station staff. This quick bridge reduced costs, meaning

that this selective client base could support the cost of the Internet without

competing with the telecentre, its supplier. The telecentre also has the ability

to easily disconnect the radio station should they not pay. This model also

allowed sharing of network resources. For example, the radio station has a

new laser printer, while the telecentre has a color printer. Because the client

systems are on the same network, clients can print at either site.

Training

To support this network, very little training was required. The telecentre staff

were shown how to install the equipment and basic trouble shooting, such as

rebooting (power cycling) the access points, and how to replace the unit

should one fail. This allows the author's team to simply ship a replacement

and avoid the two day trek to Timbuktu.

Chapter 11: Case Studies

305

Summary

The installation was considered an interim measure. It was meant to serve

as a stop-gap measure while moving forward with a more complete solution.

While it can be considered a success, it has not yet led to building more

physical infrastructure. It has brought ICTs closer to a radio solution, and re-

enforced local client/supplier relationships.

As it stands, Internet access is still an expensive undertaking in Timbuktu.

Local politics and competing subsidized initiatives are underway, but this

simple solution has proven to be an ideal use case. It took the team several

months of analysis and critical thought to arrive here, but it seems the sim-

plest solution provided the most benefit.

Ian Howard

Case study: Finding solid ground in Gao

One day's drive east from Timbuktu, in Eastern Mali, is Gao. This rural city,

which seems more more like a big village, sits up the the river Niger just be-

fore it dips South crossing into Niger and onto Nigeria. The city slopes into

the river gently, and has few buildings taller than two stories. In 2004, a tele-

centre was installed in Gao. The project's goal was to provide information to

the community in the hope that a better informed community would yield a

healthier and more educated citizenry.

The centre provides information via CD-ROMs, films and radio, but the cor-

nucopic source of information for the centre is the Internet. It is a standard

telecentre, with 8 computers, an all-in-one printer, scanner, fax, a telephone

and a digital camera. A small two room building was built to house the tele-

centre. It is located a bit outside of downtown, which is not an ideal location

for attracting customers, but the site was chosen because of its sympathetic

host. The site received funding for all construction needed, and equipment

and initial training was supplied as well. The telecentre was expected to be

self-sustaining after one year.

Several months after its opening, the telecentre was attracting few custom-

ers. It used a modem to dial-up to connect to an Internet provider in the capi-

tal. This connection was too slow and unreliable, and so the funder spon-

sored the installation of a VSAT system. There are a number of VSAT sys-

tems now available to the region; most of these services have just recently

become available. Previously only C-band (which cover a larger area than

Ku-band) systems were available. Recently, fiber has been laid in almost

every subway tunnel and canal throughout Europe, and thus it has sup-

planted the more expensive satellite services. As a result, providers are now

306

Chapter 11: Case Studies

redirecting their VSAT systems to new markets, including middle and West-

ern Africa, and South Asia. This has led to a number of projects which use

satellite systems for an Internet connection.

After the VSAT was installed, the connection provided 128 kbps down and

64 kbps up, and cost about $400 per month. The site was having trouble

earning enough revenue to pay for this high monthly cost, so the telecentre

asked for help. A private contractor was hired, who had been trained by the

author to install a wireless system. This system would split the connection

between three clients: a second beneficiary, a radio station, and the telecen-

tre, each paying $140. This collectively covered the costs of the VSAT, and

the extra revenue from the telecentre and the radio station would cover sup-

port and administration of the system.

The people

Though capable and willing, the author s team did not do the actual installa-

tion. Instead, we encouraged the telecentre to hire the local contractor to do

it. We were able to reassure the client by agreeing to train and support the

contractor in the fulfillment of this installation. The premise of this decision

was to discourage a reliance on a short-term NGO, and rather to build trust

and relationships between domestic service providers and their clients. This

design proved to be fruitful. This approach took much more time from the

author's team, perhaps twice as much, but this investment has already be-

gun to pay-off. Networks are still being installed and the author and his team

are now home in Europe and North America.

Design choices

Initially, it was conceived that a backbone connection would be made to the

radio station, which already had a 25 meter tower. That tower would be used

to relay to the other clients, avoiding the need to install towers at the client

sites, as this tower was well above any obstacles in the city. To do this, three

approaches were discussed: installing an access point in repeater mode,

using the WDS protocol, or using a mesh routing protocol. A repeater was not

desirable as it would introduce latency (due to the one-armed repeater prob-

lem) to an already slow connection. VSAT connections need to send packets

up to the satellite and back down, often introducing up to 3000 ms in delay

for a round trip. To avoid this problem, it was decided to use one radio to

connect to clients, and a second radio for to the dedicated backbone connec-

tion. For simplicity it was decided to make that link a simple bridge, so that

the access point at the radio station would appear to be on the same physical

LAN as the telecentre.

In testing this approach functioned, though in the real world, its performance

was dismal. After many different changes, including replacing the access

Chapter 11: Case Studies

307

points, the technician decided that there must be a software or hardware bug

affecting this design. The installer then decided to place the access point at

the telecentre directly using a small 3 meter mast, and to not use a relay site

at the radio station. The client sites also required small masts in this design.

All sites were able to connect, though the connections were at times too fee-

ble, and introduced massive packet loss.

Later, during the dust season, these connections became more erratic and

even less stable. The client sites were 2 to 5 km away, using 802.11b. The

team theorized that the towers on either side were too short, cutting off too

much of the Fresnel zone. After discussing many theories, the team also re-

alized the problem with the performance at the radio station: the radio fre-

quency 90.0 MHz was about the same as the frequency of the high-speed

(100BT) Ethernet connection. While transmitting, the FM signal (at 500 watts)

was completely consuming the signal on the Ethernet cable. Thus, shielded

cable would be required, or the frequency of the Ethernet link would need to

be changed. The masts were then raised, and at the radio station the speed

of the Ethernet was changed to 10 Mbps. This changed the frequency on the

wire to 20 MHz, and so avoided interference from the FM transmission.

These changes resolved both problems, increasing the strength and reliabil-

ity of the network. The advantage of using mesh or WDS here would be that

client sites could connect to either access point, either directly to the telecen-

tre to the radio station. Eventually, removing the reliance on the radio station

as a repeater likely made the installation more stable in the longer-term.

Financial model

The satellite system used at this site cost approximately $400 per month. For

many IT for Development projects this expensive monthly cost is difficult to

manage. Typically these projects can purchase equipment and pay for the

establishment of a wireless network, but most are not able to pay for the cost

of the network after a short period of time (including the recurring Internet

costs and operational costs). It is necessary to find a model where the

monthly costs for a network can be met by those who use. For most commu-

nity telecenters or radio stations, this is simply too expensive. Often, the only

feasible plan is to share the costs with other users. To make the Internet

more affordable, this site used wireless to share the Internet to the commu-

nity, allowing a greater number of organizations to access the Internet while

reducing the cost per client.

Typically in Mali, a rural community has only a few organizations or companies

that could afford an Internet connection. Where there are few clients, and the

Internet connection cost is high, the model developed by his team included

anchor clients: clients whom are solid and are low-risk. For this region, for-

eign NGOs (Non Governmental Organizations), the United Nations Agencies

and large commercial enterprises are among the very few whom qualify.

308

Chapter 11: Case Studies

Among the clients selected for this project were three anchor clients, who

collectively paid the entire monthly cost of the satellite connection. A second

beneficiary, a community radio station, was also connected. Any revenue

earned from the beneficiaries contributed to a windfall, or deposit for future

costs, but was not counted upon due to the small margins that both of these

community services operated on. Those clients could be disconnected and

could resume their service once they can afford it again.

Training needed: who, what, for how long

The contractor taught the telecentre technician the basics of supporting the

network, which was fairly rudimentary. Any non-routine work, such as adding

a new client, was contracted out. Therefore it was not imperative to teach

the telecentre staff how to support the system in its entirety.

Lessons learned

By sharing the connection, the telecentre is now self-sustaining, and in addi-

tion, three other sites have Internet access. Though it takes more time and

perhaps more money, it is valuable to find the right local talent and to en-

courage them to build relationships with clients. A local implementor will be

able to provide the follow-up support needed to maintain and expand a net-

work. This activity is building local expertise, and demand, which will allow

subsequent ICT projects to build on this base.

Ian Howard

Case Study: Fantsuam Foundation's

Community Wireless Network

Kafanchan is a community of 83,000 people located 200 km northeast of

Abuja, in central Nigeria. Kafanchan used to be known as a busy and thriving

town as it was the host of one of the main junctions of the national railway.

When the railway industry was booming, almost 80% of Kafanchan's popula-

tions relied on it in one way or another. Following the complete breakdown of

the Nigerian railway system, the population of Kafanchan has been forced to

go back to its original source of income, which is agriculture.

Kafanchan is a poorly connected area in terms of fixed telephony and Inter-

net connectivity. Today, no fixed telephony (PSTN) is available in the area

and GSM only just arrived in 2005. However, the GSM coverage is just as

poor as the quality of the service. At the moment, SMS services are the most

reliable communication service because voice conversations tend to cut off in

the middle and suffer heavy noise.

Chapter 11: Case Studies

309

Poor access to electricity brings further challenges to the people of Kafan-

chan. The national electric power company of Nigeria, generally known as

NEPA (National Electric Power Authority), is more commonly known to Nige-

rians as "Never Expect Power Always". In 2005, NEPA changed its name to

Power Holding Company of Nigeria (PHCN).

Kafanchan is receiving power from NEPA on an average of 3 hours per day.

For the remaining 21 hours, the population relies on expensive diesel gen-

erators or kerosene for illumination and cooking. When NEPA is available on

the grid, it provides an unregulated voltage in the range of 100-120 V in a

system designed for 240 V. This voltage must be regulated to 240 V before

most loads can be connected. Only light bulbs can be fed straight to the grid

power since they can handle the low voltage without damage.

Project participants

Given the challenging background of Kafanchan, how could anyone come up

with the idea of establishing the first rural Wireless ISP in Nigeria there?

Fantsuam Foundation did and they made it happen.

Fantsuam Foundation is a local, non-governmental organization that has

been working together with the community of Kafanchan since 1996 to fight

poverty and disadvantage through integrated development programs. Fant-

suam's focus lies on micro finance, ICT services and social development in

rural communities of Nigeria. Becoming the first rural wireless ISP in Nigeria

was part of their mission to be a recognized leader in the provision of rural

development initiatives, as well as the foremost rural knowledge economy

driver in Nigeria.

The Wireless ISP of Fantsuam Foundation, also know as Zittnet, is funded

by IDRC, the International Development Research Centre of Canada. IT +46,

a Swedish based consultancy company focusing on ICTs for development,

has worked together with the Zittnet team to provide technical support for

wireless communications, bandwidth management, solar energy, power

backup systems and VoIP deployments.

Objectives

The main objective of Zittnet is to improve access to communications in the

area of Kafanchan by implementing a community wireless network. The net-

work provides intranet and Internet access to local partners in the commu-

nity. The community network is formed by community-based organizations

such as educational institutions, faith-based institutions, health services,

small enterprises and individuals.

310

Chapter 11: Case Studies

Power Backup System

In order to provide a reliable service to the community, Zittnet needed to be

equipped with a stable power backup system that would make the network

run independently of the NEPA.

A hybrid power system was designed for Fantsuam, consisting of a deep-

cycle battery bank and 2 kW (peak) solar panels. The system can charge

from three different sources: a diesel generator, a solar array, and from NEPA

when electricity is available. The network operation center (NOC) of the or-

ganization runs completely from solar energy. The rest of the Fantsuam's

premises runs from NEPA or the generator via the battery bank, which pro-

vides uninterrupted voltage stability. The NOC load has been separated from

the rest of the load of Fantsuam to ensure a reliable power source to the

critical infrastructure in the NOC, even when the battery bank is running low

on power.

Figure 11.1: 24 solar panels with a nominal power of 80 W have been mounted to the

roof of the NOC to provide power to the system 24/7.

Simulations with the best existing solar data reveal that Kaduna State, where

Kafanchan is located, receives at least 4 sun peak hours during its worst

months which stretch from June to August (the rainy season).

Each of the solar panels (Suntech 80 W peak) provides a maximum current

of 5 A (when the solar radiation is highest during the day). In the worst

months of the year, the system is expected to produce not less than 6 KWh/

day.

Chapter 11: Case Studies

311

The solar system has been designed to provide 12 and 24 V DC output in

order to match the input voltage of all low power servers and workstations for

NOC infrastructure and training classrooms.

The solar panels used are Suntech STP080S-12/Bb-1 with the following

specifications:

· Open-circuit Voltage (VOC): 21.6 V

· Optimum operating voltage (VMP): 17.2 V

· Short-circuit current (ISC): 5 A

· Optimum operating current (IMP): 4.65 A

· Maximum power at STC (PMAX): 80 W (Peak)

The minimum 6 KWh/day that feeds the NOC is used to power the following

equipment:

Device

Hours/Day

Units

Power (W)

Access points

1080

Low power servers

960

LCD screens

160

Laptops

1500

Lamps

480

VSAT modem

1440

Total

5620

The power consumption for servers and LCD screens is based on Inveneo's

Low Power Computing Station, http://www.inveneo.org/?q=Computingstation.

The total estimated power consumption of the NOC is 5.6 kWh/day which is

less than the daily power generated from the solar panels in the worst month.

312

Chapter 11: Case Studies

Figure 11.2: The NOC is built by locally made laterite brick stones, produced and laid

by youths in Kafanchan.

Network Operating Center (NOC)

A new Network Operating Center was established to host the power backup sys-

tem and server room facilities. The NOC was designed to provide a place safe

from dust, with good cooling capabilities for the batteries and the inverters. The

NOC uses natural methods and is made from locally available materials.

The building is comprised of four rooms: a battery storage room, a server

room, a working space and a room for equipment storage.

The battery storage room hosts seventy 200 Ah deep cycle batteries, as well

as five inverters (one of them pure sine wave), two solar regulators, power

stabilizers and DC and AC disconnects. The batteries are stacked vertically

on a metal shelf structure for better cooling.

The server space accommodates a rack unit for servers and a fan. The room

has no regular windows, to avoid dust and overheating. The server room and

battery room face south to improve natural cooling and to help keep the room

at an appropriate temperature.

Chapter 11: Case Studies

313

The server room and the battery space require effective low cost/low energy

cooling as they need to operate 24x7. To achieve this goal, natural cooling

techniques have been introduced in the NOC design: small fans and extrac-

tors and thick walls of bricks (double width) in the direction of the sunset.

The south side of the building hosts 24 solar panels in a shadow-free area on

its metal roof. The roof was designed with an inclination of 20 degrees to

host the panels and limit corrosion and dust. Extra efforts have been made to

keep the panels easily reachable for cleaning and maintenance. The roof has

also been strengthened in order to carry the extra load of 150-200 kg.

The NOC building is constructed of locally produced laterite mud bricks. The

material is cheap since it is frequently used and comes from the top layer of

soil. The bricks are produced locally by hand using a low-tech pressing tech-

nique. The NOC is unique for its kind in Kaduna State.

Figure 11.3: Omolayo Samuel, one of the staff of Zittnet, does not fear the height of

the 45m tall tower as she is aligning the antennas hosted in the top of the tower.

Physical infrastructure: A communication mast

Most potential clients to Zittnet are located between 1 km and 10 km from the

premises of Fantsuam. In order to reach these clients, Fantsuam established a

communication mast on their premises. In October 2006, a 45m (150 foot) tall self-

314

Chapter 11: Case Studies

standing mast was installed at Fantsuam Foundation. The mast was equipped

with grounding and lighting protection as well as a mandatory signal light.

A metal ring was buried at the base of the tower at a depth of 4 feet. All three

legs of the mast were then connected to the grounding circuit. A lightning rod

was mounted at the highest point of the mast to protect the equipment

against lighting strikes. The rod is made of pure copper and is connected to

the earth ring at the base of the mast using copper tape.

The signal light mounted at the top of the mast is a requirement from the Civil

Aviation Authorities. The light is equipped with a photocell which enables

automated switching based on the level of ambient light. In this way, the light

comes on at night and goes off during the day.

Wireless backbone infrastructure

The wireless backbone infrastructure is built using SmartBridges multi-band

access points and client units from the Nexus PROTM TOTAL series. The

units are designed for service providers and enterprises to establish high

performance point-to-multipoint outdoor wireless links. They come with an

integrated multi-band sectoral antenna that can operate both in 2.4 GHz and

5.1-5.8 GHz frequencies. The Nexus PROTM TOTAL series offers QoS for

traffic prioritization and bandwidth management per client using the IEEE

802.11e compliant WMM (WiFi Multimedia) extensions.

Figure 11.4: The network topology of Zittnet in October 2007.

Chapter 11: Case Studies

315

Currently, the topology of the network is a star topology with two access

points in the communication mast at Fantsuam's premises. One access point

hosts a 90 degree sectoral antenna (blue dotted lines) and the other access

point provides omnidirectional coverage to the surroundings (red dotted

rings). Clients that are located within the area between the dotted lines are

connected to the sectoral antenna, while the remaining clients are connected

to the omnidirectional antenna.

Plans are underway to expand the wireless backbone by setting up two wire-

less repeaters. One repeater will be located in Kafanchan city using an exist-

ing NITEL tower to enhance the wireless coverage in the city center. The

second repeater will be established in the Kagoro Hills, a small mountain

group with a relative altitude to Kafanchan of about 500m, which is located

about 7 km from Kafanchan. This repeater will provide coverage to many

surrounding towns and may even enable a long-distance link to Abuja.

Zittnet connected its first client in early August 2007. Two months later, no

less than eight clients are connected to Zittnet. These clients include:

· The general hospital

· New Era Hospital

· Jagindi Street Clinic (health clinic)

· Zenith Bank (for private use)

· Isaiah Balat (Internet café)

· New World Hotel

· Throne Room GuestHouse

· Fulke

Problems encountered

A few problem areas that have been constantly present throughout the pro-

ject are as follows.

Low buildings

Most client premises are single-story buildings with a height of no more than

3 meters. Many houses have very weak roof structures which makes it im-

possible to mount equipment on the roof, as physical access is not possible.

The low buildings force us to mount the equipment at a fairly low height, as

clients can not afford to invest in small (10 m) masts to host the equipment.

Most installations make use of water tanks or a simple 3 meter metal pole

attached to the wall of the premise.

316

Chapter 11: Case Studies

When the equipment is mounted low, the first Fresnel zone is not cleared and

lower throughput is experienced. Although the landscape in Kafanchan is very

flat, vegetation in the form of thick mango trees easily block the line-of-sight.

Lightning strikes

Heavy thunder storms are frequent during the rainy season in Kafanchan. In

September 2007, a nearby lightning strike damaged equipment mounted on

a mast, as well as its power supply. At the moment, the access point and its

PoE injector are grounded to the tower itself. Further means need to be in-

vestigated to prevent damage to equipment caused by nearby lightning. The

Zittnet team is currently working on improving the surge protection by adding

extra coaxial surge arrestors. Furthermore, the shield of the UTP cable con-

necting the access point with the NOC will be grounded using grounding

blocks and fasteners.

Low Quality Equipment

Unfortunately, a lack of quality products on the market is a widespread prob-

lem across the whole African continent. As most sub-Sahara countries lack

policies for quality assurance of imported goods, the market is flooded by

"cheap" and very low quality articles. Since quality products are hard to find,

you often find yourself buying locally available merchandise that breaks even

before it is put into operation. As no sort of warranty exists for these minor

purchases, this ends up being very expensive. This problem is almost always

present in common accessories such as power sockets, power bars, RJ45

connectors, CAT5 cabling, and other low-tech equipment.

Business Model

The only alternative for Internet access in Kafanchan is via satellite. During

2006, Fantsuam had a subscription of 128/64 kbps dedicated bandwidth at a

cost of $1800 USD/month. This huge monthly cost of connectivity has been a

big burden for Fantsuam and a constant stress of being unable to meet the

monthly bill.

As an alternative to the high risk "flat fee" model, Fantsuam has implemented

a system called HookMeUP provided by Koochi Communications. The sys-

tem offers flexible Pay-As-You-Go charges over broadband VSAT Internet

connections to countries across sub-Sahara Africa.

This kind of access model is typically found in airports, hotels or large shop-

ping malls in western countries where end-users buy vouchers online and log

in using an access code.

Chapter 11: Case Studies

317

The HookMeUP system offers a 512/256 kbps dedicated VSAT connection to

Fantsuam (from their ground station in the UK). Fantsuam buys vouchers

from Koochi Communications and resells them to its local clients in Kafan-

chan. In this way, Fantsuam is no longer stuck with a fixed monthly cost but

has only to pay Koochi for the bandwidth they actually have consumed. The

risk of buying expensive international bandwidth has now been transferred to

the Internet provider instead of the end user, at a cost of a higher price for

the end user.

Fantsuam foundation now acts as a reseller of vouchers from Koochi and a

supplier of wireless infrastructure to the end users. The Wireless Community

Network now provides the Fantsuam Foundation with five sources of income:

1. Installation of client premises equipment (one occasion per client)

2. Leasing of wireless equipment (monthly cost per client)

3. Reselling wireless equipment (one occasion per client)

4. Installation of wireless hotspot at client's premise (one occasion per client)

5. Reselling of vouchers (continuously)

The voucher system is based on three parameters: access time, data limit

and validity time. Whichever parameter runs out first will consume the

voucher.

Access

Data limit

Validity

Price

USD / h

USD /

time

(MB)

time

(USD)

700 MB

30 min

1 day

0.80

1.60

112.00

60 min

5 days

1.28

89.60

12 hours

14 days

10.40

0.87

121.33

24 hours

150

30 days

26.00

1.08

121.33

1 month

500

1 month

71.50

0.10

100.10

3 months

1600

3 months

208.00

0.10

91.00

6 months

3500

6 months

416.00

0.10

83.20

12 months

7500

12 months

728.00

0.08

67.95

318

Chapter 11: Case Studies

The greatest advantage of this system is that Fantsuam Foundation no

longer has the burden of a huge monthly bill for international bandwidth. Hav-

ing a flat-fee model means that you are forced to sell a certain amount of

bandwidth every month. With the Pay-As-You-Go (PAYG) model, Fantsuam's

income from reselling vouchers depends on how much bandwidth their cli-

ents consume. The client pays in advance (pre-paid model) with the result

that Fantsuam will never end up in huge debt with the provider.

The pre-paid model works well in Africa since people are familiar with this

model from mobile operators. It is even used by electricity companies in

some counties. The pre-paid model is appreciated by many as it helps them

to keep track of their expenditures. One of the main limitations of the PAYG

model is the lack of flexibility and transparency. The current PAYG system

provides very little feedback to the user about consumed time or volume.

Only when the user logs off will he/she be informed about how many minutes

are left to spend.

However, the business model seems to fit the local reality of Kafanchan and

many other rural communities in Africa quite well. Although there is room for

improvement, the advantage of avoiding debts is far greater than the disad-

vantages. With time, when the number of clients have increased and they

can rely on a substantial monthly income from the wireless network, it might

be beneficial to go back to the flat-fee model again.

Clients

The clients are free to use the Internet access for any purpose. For example,

Isaiah Balat is reselling vouchers (that he bought from Fantsuam) to his cli-

ents. His Internet café hosts 10 computers that all are connected to Zittnet.

The clients purchase vouchers from the owner with a margin of 25% over the

price offered by Fantsuam. In return, clients that do not have access to a

computer connected to Zittnet can access the network though the PC's at

Isaiah Balat's café.

The New World Hotel is another client that aims to create a similar business

model but on a larger scale. They will provide wireless Internet access to all

of their rooms and offer access to Zittnet's uplink by reselling vouchers.

Other clients, like the General Hospital and the Jagindi Street Clinic, are us-

ing the Internet access for professional and private use without reselling ac-

cess to its clients.

--Louise Berthilson

Chapter 11: Case Studies

319

Case study: The quest for affordable

Internet in rural Mali

For several years the international development community has promoted

the idea of closing the digital divide. This invisible chasm that has formed

separating access to the wealth of information and communications tech-

nologies (ICT) between the developed and the developing world. Access to

information and communications tools has been shown to have a dramatic

impact on quality of life. For many donors fatigued by decades of supporting

traditional development activities, the installation of a telecentre in the devel-

oping world seems like a realizable and worthwhile effort. Because the infra-

structure does not exist, this is much more expensive and difficult to do in the

developing world than it is in the West. Moreover, few models have been

shown to sustain these activities. To help mitigate some of the cost of bring-

ing the Internet to rural areas of the developed world, the author's team has

promoted the use of wireless systems to share the cost of an Internet con-

nection. In November of 2004, an affiliated project asked the author's team to

pilot such a wireless system at a recently installed telecentre in rural Mali, 8

hours South-West by four-by-four from Bamako, the capital.

This rural city, located on the margin of a man-made reservoir, holds water

for the Manitali dam that powers a third of the country. This location is fortu-

nate as hydroelectric power is much more stable and available than diesel

generated power. While diesel generated power is far less stable, some rural

communities are lucky to have any electricity at all.

The city is also endowed to be in one of the most fertile regions of the coun-

try, in its cotton belt, Mali's main cash crop. It was believed that this site

would be the least difficult of the rural areas in Mali to make a self-sustaining

telecentre. Like many experiments, this pilot was fraught with challenges.

Technologically it was a simple task. In 24 hours the team installed an

802.11b wireless network that shares the telecenter's VSAT Internet connec-

tion with 5 other local services: the Mayor, the Governor, the health service,

the district's Mayor's council (CC) and the community advisory service

(CCC).

These clients had been selected during a reconnaissance two months prior.

During that visit the team had interviewed potential clients and determined

which clients could be connected without complicated or expensive installa-

tions. The telecentre itself is housed at the community radio station. Radio

stations tend to be great sites to host wireless networks in rural Mali as they

are often well placed, have electricity, security and people who understand at

least the basics of radio transmissions. They are also natural hubs for a vil-

320

Chapter 11: Case Studies

lage. Providing Internet to a radio station provides better information to its

listeners. And for a culture which is principally oral, radio happens to be the

most effect means to provide information.

From the list of clients above, you will note that the clients were all govern-

ment or para-governmental. This proved to be a difficult mix, as there is con-

siderable animosity and resentment between the various levels of govern-

ment, and there were continuing disputes regarding taxes and other fiscal

matters. Fortunately the director of the radio station, the network's champion,

was very dynamic and was able to wade through most of these politics,

though not all.

Design choices

The technical team determined that the access point would be installed at 20

meters up the radio station tower, just below the FM radio dipoles, and not so

high as to interfere with coverage to client sites below in the bowl-like de-

pression where most were found. The team then focused on how to connect

each client site to this site. An 8 dBi omni (from Hyperlinktech,

http://hyperlinktech.com/) would suffice, providing coverage to all client

sites. The 8 dBi antenna that was chosen has a 15 degree vertical beam-

width, assuring that the two clients less than a kilometer away could still re-

ceive a strong signal. Some antennae have very narrow beam width and thus

"overshoot" sites that are close. Panel antennae were considered, though at

least two would be required and either a second radio or a channel splitter. It

was deemed unnecessary for this installation. The following calculation

shows how to calculate the angle between the client site's antenna and the

base station's antenna, using standard trigonometry.

tan(x) =

difference in elevation

height of base station antenna

height of CPE antenna

distance between the sites

tan(x) = 5m + 20m - 3m / 400m

x = tan-1 (22m / 400m)

x =~ 3 degrees

In addition to the equipment in the telecentre (4 computers, a laser printer, 16

port switch), the radio station itself has one Linux workstation installed by the

author's project for audio editing. A small switch was installed in the radio

station, an Ethernet cable was run through plastic tubing buried at 5 cm

across to the telecentre, across the yard.

From the main switch, two cables run up to a Mikrotik RB220, access point.

The RB220 has two Ethernet ports, one that connects to the VSAT through a

cross-over cable, and the second that connects to the radio station's central

Chapter 11: Case Studies

321

switch. The RB 220 is housed in a D-I-Y PVC enclosure and an 8 dBi omni

(Hyperlink Technologies) is mounted directly to the top of the PVC cap.

The RB220 runs a derivative of Linux, Mikrotik version 2.8.27. It controls the

network, providing DHCP, firewall, and DNS-caching services, while routing

traffic to the VSAT using NAT. The Mikrotik comes with a powerful command

line and a relatively friendly and comprehensive graphical interface. It is a

small x86 based computer, designed for use as an access point or embed-

ded computer. These access points are POE capable, have two Ethernet

ports, a mini-pci port, two PCMCIA slots, a CF reader (which is used for its

NVRAM), are temperature tolerant and support a variety of x86 operating

systems. Despite that the Mikrotik software requires licensing, there was al-

ready a substantial user base in Mali. The system has a powerful and friendly

graphical interface that was superior to other products. Due to the above fac-

tors the team agreed to use these systems, including the Mikrotik software to

control these networks. The total cost of the RB220, with License Level 5,

Atheros mini-pci a/b/g and POE was $461. You can find these parts at Mik-

rotik online at http://www.mikrotik.com/routers.php#linx1part0.

The network was designed to accommodate expansion by segregating the

various sub-networks of each client; 24 bit private subnets were alloted. The

AP has a virtual interface on each subnet and does all routing between, also

allowing fire-walling at the IP layer. Note: this does not provide a firewall at

the network layer, thus, using a network sniffer like tcpdump one can see all

traffic on the wireless link.

To limit access to subscribers, the network uses MAC level access control.

There was little perceived security risk to the network. For this first phase, a

more thorough security system was left to be implemented in the future, when

time could be found to find an easier interface for controlling access. Users

were encouraged to use secure protocols, such as https, pops, imaps etc.

The affiliate project had installed a C-band VSAT (DVB-S) system. These

satellite systems are normally very reliable and are often used by ISPs. It is a

large unit, in this case the dish was 2.2 meters in diameter and expensive,

costing approximately $12,000 including installation. It is also expensive to

operate. A 128 kbps down and 64 kbps up Internet connection costs ap-

proximately $700 per month. This system has several advantages compared

to a Ku system though, including: greater resilience to bad weather, lower

contention rates (number of competing users on the same service) and it is

more efficient at transferring data.

The installation of this VSAT was not ideal. Since the system ran Windows,

users were able to quickly change a few settings, including adding a pass-

word to the default account. The system had no UPS or battery back up, so

once a power outage occurred the system would reboot and sit waiting for a

322

Chapter 11: Case Studies

password, which had since been forgotten. To make this situation worse,

because the VSAT software was not configured as an automatic background

service it did not automatically launch and establish the link. Though the C-

band systems are typically reliable, this installation caused needless outages

which could have been resolved with the use of a UPS, proper configuration

of the VSAT software as a service, and by limiting physical access to the

modem. Like all owners of new equipment, the radio station wanted to dis-

play it, hence it was not hidden from view. Preferably a space with glass

doors would have kept the unit secure while keeping it visible.

The wireless system was fairly simple. All of the client sites selected were

within 2 km of the radio station. Each site had a part of the building that could

physically see the radio station. At the client site, the team chose to use

commercial, client grade CPEs: Based on price, the Powernoc 802.11b CPE

bridge, small SuperPass 7 dBi patch antennas and home-made Power Over

Ethernet (POE) adaptors. To facilitate installation, the CPE and the patch

antenna were mounted on a small piece of wood that could be installed on

the outside wall of the building facing the radio station.

In some cases the piece of wood was an angled block to optimize the posi-

tion of the antenna. Inside, a POE made from a repurposed television signal

amplifier (12V) was used to power the units. At the client sites there were not

local networks, so the team also had to install cable and hubs to provide

Internet for each computer. In some cases it was necessary to install Ether-

net adapters and their drivers (this was not determined during the assess-

ment). It was decided that because the client's networks were simple, that it

would be easiest to bridge their networks. Should it be required, the IP archi-

tecture could allow future partitioning and the CPE equipment supported STA

mode. We used a PowerNOC CPE bridge that cost $249.

Local staff were involved during the installation of the wireless network.

They learned everything from wiring to antenna placement. An intensive

training program followed the installation. It lasted several weeks, and was

meant to teach the staff the day to day tasks, as well as basic network

troubleshooting.

A young university graduate who had returned to the community was chosen

to support the system, except for the cable installation, which the radio sta-

tion technician quickly learned. Wiring Ethernet networks is very similar to

coaxial cable repairs and installations which the radio technician already per-

formed regularly. The young graduate also required little training. The team

spent most of its time helping him learn how to support the basics of the sys-

tem and the telecentre. Soon after the telecentre opened, students were

lined up for the computer training, which offered 20 hours of training and

Internet use per month for only $40, a bargain compared to the $2 an hour

Chapter 11: Case Studies

323

for Internet access. Providing this training was a significant revenue and was

a task that the young computer savvy graduate was well suited for.

Unfortunately, and somewhat unsurprisingly, the young graduate left for the

capital, Bamako, after receiving an offer for a government job. This left the

telecentre effectively marooned. Their most technically savvy member, and

the only one who was trained in how to support the system, had left. Most of

the knowledge needed to operate the telecentre and network left with him.

After much deliberation, the team determined that it was best not to train an-

other tech savvy youth, but rather to focus on the permanent local staff, de-

spite their limited technical experience. This took much more time. Our train-

ers have had to return for a total of 150 hours of training. Several people

were taught each function, and the telecentre support tasks were divided

among the staff.

Training did not stop there. Once the community services were connected,

they too needed access. It seemed that although they were participating,

the principals, including the mayor, were not using the systems themselves.

The team realized the importance of assuring that the decision makers

used the system, and provided training for them and their staff. This did

remove some of the mystique of the network and got the city's decision

makers involved.

Following training, the program monitored the site and began to provide in-

put, evaluating ways that this model could be improved. Lessons learned

here were applied to other sites.

Financial Model

The community telecentre was already established as a non-profit, and was

mandated to be self-sustaining through the sale of its services. The wireless

system was included as a supplementary source of revenue because early

financial projections for the telecentre indicated that they would fall short of

paying for the VSAT connection.

Based on the survey, and in consultation with the radio station that manages

the telecentre, several clients were selected. The radio station negotiated

contracts with some support from its funding partner. For this first phase, cli-

ents were selected based on ease of installation and expressed ability to pay.

Clients were asked to pay a subscription fee, as described later.

Deciding how much to charge was a major activity which required consultation

and expertise that the community did not have in financial projections. The

equipment was paid for by the grant, to help offset the costs to the community,

but clients were still required to pay a subscription fee, which served to as-

sure their commitment. This was equivalent to one month of the service fee.

324

Chapter 11: Case Studies

To determine the monthly cost for an equal slice of bandwidth we started with

the following formula:

VSAT + salaries + expenses (electricity, supplies) =

telecentre revenue + wireless client revenue

We had estimated that the telecentre should earn about $200 to $300 per

month in revenue. Total expenses were estimated to be $1050 per month,

and were broken down as: $700 for the VSAT, $100 for salaries, $150 for

electricity, and about $100 for supplies. About $750 in revenue from the wire-

less clients was required to balance this equation. This amounted to roughly

$150 from each client. This was just tolerable by the clients, and looked fea-

sible, but required fair weather, and had no room for complications.

Because this was becoming complicated, we brought in business geeks, who

modified the formula as such:

Monthly expenses + amortization + safety funds = total

revenue

The business experts were quick to point out the need of amortization of the

equipment, or one could say "re-investment funds" as well as safety funds, to

assure that the network can continue if a client defaults, or if some equip-

ment breaks. This added about $150 per month for amortization (equipment

valued at about $3,000, amortized over 24 months) and the value of one cli-

ent for default payments, at $100. Add another 10% to account for currency

devaluation ($80), and that equals an expense of $1380 per month. In trying

to implement this model, it was finally determined that amortization is a con-

cept that was too difficult to convey to the community, and that they would

not consider that clients might default on payment. Thus, both formulae were

used, the first by the telecentre and the second for our internal analysis.

As was soon discovered, regular payments are not part of the culture in rural

Mali. In an agrarian society everything is seasonal, and so too is income.

This means that the community's income fluctuates wildly. Moreover, as

many public institutions were involved, they had long budget cycles with little

flexibility. Although they theoretically had the budget to pay for their service, it

would take many months for the payments to be made. Other fiscal compli-

cations arose as well. For example, the mayor signed on and used the back-

taxes owed by the radio to pay for its subscription. This of course did not

contribute to cash flow. Unfortunately, the VSAT providers have little flexibil-

ity or patience, as they have limited bandwidth and only have room for those

that can pay.

Cash flow management became a primary concern. First, the revenue fore-

seen in financial projections showed that even with an optimistic outlook,

they would not only have trouble earning enough revenue on time to pay the

Chapter 11: Case Studies

325

fee, but getting the money to the Bamako-based bank also presented a prob-

lem. Roads near the village can be dangerous, due to the number of smug-

glers from Guinea and wayward rebels from the Ivory Coast. As projected,

the telecentre was not able to pay for its service and its service was sus-

pended, thereby suspending payment from their clients as well.

Before the project was able to find solutions to these problems, the cost of

the VSAT already began to dig the telecentre into debt. After several months,

due to technical problems, as well as concerns raised in this analysis, the

large C-band VSAT was replaced with a cheaper Ku band system. Although

cheaper, it still sufficed for the size of the network. This system was only

$450, which by ignoring amortization and safety margins is affordable by the

network. Unfortunately, due to default payments, the network was not able to

pay for the VSAT connection after the initial subsidized period.

Conclusions

Building a wireless network is relatively easy, but making it work is much

more of a business problem than a technical problem. A payment model that

considers re-investment and risk is a necessity, or eventually the network will

fail. In this case, the payment model was not appropriate as it did not con-

form to fiscal cycles of the clients, nor did it conform to social expectations. A

proper risk analysis would have concluded that a $700 (or even a $450)

monthly payment left too narrow a margin between revenue and expenses to

compensate for fiscal shortcomings. High demand and education needs lim-

ited the expansion of the network.

Following training the network operated for 8 months without significant techni-

cal problems. Then, a major power surge caused by a lightning strike de-

stroyed much of the equipment at the station, including the access point and

VSAT. As a result, the telecentre was still off-line at the time that this book was

written. By that time this formula was finally deemed an unsuitable solution.

Ian Howard

Case study: Commercial deployments in

East Africa

Describing commercial wireless deployments in Tanzania and Kenya, this

chapter highlights technical solutions providing solid, 99.5% availability Inter-

net and data connectivity in developing countries. In contrast to projects de-

voted to ubiquitous access, we focused on delivering services to organiza-

tions, typically those with critical international communications needs. I will

326

Chapter 11: Case Studies

describe two radically different commercial approaches to wireless data con-

nectivity, summarizing key lessons learned over ten years in East Africa.

Tanzania

In 1995, with Bill Sangiwa, I founded CyberTwiga, one of the first ISPs in Af-

rica. Commercial services, limited to dialup email traffic carried over a

9.6 kbps SITA link (costing over $4000/month!), began in mid-1996. Frus-

trated by erratic PSTN services, and buoyed by a successful deployment of a

3-node point-multipoint (PMP) network for the Tanzania Harbours authority,

we negotiated with a local cellular company to place a PMP base station on

their central mast. Connecting a handful of corporations to this WiLan pro-

prietary 2.4 GHz system in late 1998, we validated the market and our tech-

nical capacity to provide wireless services.

As competitors haphazardly deployed 2.4 GHz networks, two facts emerged:

a healthy market for wireless services existed, but a rising RF noise floor in

2.4 GHz would diminish network quality. Our merger with the cellular carrier,

in mid-2000, included plans for a nationwide wireless network built on the

existing cellular infrastructure (towers and transmission links) and proprietary

RF spectrum allocations.

Infrastructure was in place (cellular towers, transmission links, etc.) so wire-

less data network design and deployment were straightforward. Dar es Sa-

laam is very flat, and because the cellular partner operated an analog net-

work, towers were very tall. A sister company in the UK, Tele2, had com-

menced operations with Breezecom (now Alvarion) equipment in

3.8/3.9 GHz, so we followed their lead.

By late 2000, we had established coverage in several cities, using fractional

E1 transmission circuits for backhaul. In most cases the small size of the

cities connected justified the use of a single omnidirectional PMP base sta-

tion; only in the commercial capital, Dar es Salaam, were 3-sector base sta-

tions installed. Bandwidth limits were configured directly on the customer

radio; clients were normally issued a single public IP address. Leaf routers

at each base station sent traffic to static IP addresses at client locations, and

prevented broadcast traffic from suffocating the network. Market pressures

kept prices down to about $100/month for 64 kbps, but at that time (mid/late

2000) ISPs could operate with impressive, very profitable, contention ratios.

Hungry applications such as peer-peer file sharing, voice, and ERPs simply

did not exist in East Africa. With grossly high PSTN international charges,

organizations rapidly shifted from fax to email traffic, even though their wire-

less equipment purchase costs ranged from $2000-3000.

Technical capabilities were developed in-house, requiring staff training over-

seas in subjects such as SNMP and UNIX. Beyond enhancing the company

Chapter 11: Case Studies

327

skills set, these training opportunities generated staff loyalty. We had to com-

pete in a very limited IT labor market with international gold mining compa-

nies, the UN, and other international agencies.

To insure quality at customer sites, a top local radio and telecoms contractor

executed installations, tightly tracking progress with job cards. High tempera-

tures, harsh equatorial sunlight, drenching rain, and lightning were among

the environmental insults tossed at outside plant components; RF cabling

integrity was vital.

Customers often lacked competent IT staff, burdening our employees with

the task of configuring many species of network hardware and topology.

Infrastructure and regulatory obstacles often impeded operations. The cellu-

lar company tightly controlled towers, so that if there was a technical issue at

a base station hours or days could pass before we gained access. Despite

backup generators and UPS systems at every site, electrical power was al-

ways problematic. For the cellular company, electrical mains supplies at

base stations were less critical. Cellular subscribers simply associated with a

different base station; our fixed wireless data subscribers went offline.

On the regulatory side, a major disruption occurred when the telecoms

authority decided that our operation was responsible for disrupting C-band

satellite operations for the entire country and ordered us to shut down our

network.

Despite hard data demonstrating that we were not at fault, the regulator con-

ducted a highly publicized seizure of our equipment. Of course the interfer-

ence persisted, and later was determined to emanate from a Russian radar

ship, involved in tracking space activities. We quietly negotiated with the

regulator, and ultimately were rewarded with 2 x 42 MHz of proprietary spec-

trum in the 3.4/3.5 GHz bands. Customers were switched over to dialup in

the month or so it took to reconfigure base stations and install new CPE.

Ultimately the network grew to about 100 nodes providing good, although not

great, connectivity to 7 cities over 3000+km of transmission links. Only the

merger with the cellular operator made this network feasible--the scale of

the Internet/data business alone would not have justified building a data net-

work of these dimensions and making the investments needed for proprietary

frequencies. Unfortunately, the cellular operator took the decision to close

the Internet business in mid-2002.

Nairobi

In early 2003 I was approached by a Kenyan company, AccessKenya, with

strong UK business and technical backup to design and deploy a wireless

328

Chapter 11: Case Studies

network in Nairobi and environs. Benefiting from superb networking and

business professionals, improved wireless hardware, progress in internet-

working, and bigger market we designed a high availability network in line

with regulatory constraints.

Two regulatory factors drove our network design. At the time in Kenya, Inter-

net services were licensed separately from public data network operators,

and a single company could not hold both licenses. Carrying traffic of multi-

ple, competing ISPs or corporate users, the network had to operate with total

neutrality. Also, "proprietary" frequencies, namely 3.4/3.5 GHz, were not ex-

clusively licensed to a single provider, and we were concerned about inter-

ference and the technical ability/political will of the regulator to enforce. Also,

spectrum in 3.4/3.5 GHz was expensive, costing about USD1000 per MHz

per year per base station. Restated, a base station using 2 x 12 MHz at-

tracted license fees of over $10,000 year. Since Nairobi is a hilly place with

lots of tall trees and valleys, wireless broadband networks demanded many

base stations. The licensing overheads simply were not sensible. In con-

trast, 5.7/5.8 GHz frequencies were subject only to an annual fee, about USD

120, per deployed radio.

To meet the first regulatory requirement we chose to provide services using

point-point VPN tunnels, not via a network of static IP routes. An ISP would

deliver a public IP address to our network at their NOC. Our network con-

ducted a public-private IP conversion, and traffic transited our network in pri-

vate IP space. At the customer site, a private-public IP conversion delivered

the globally routable address (or range) to the customer network.

Security and encryption added to network neutrality, and flexibility, as unique

sales properties of our network. Bandwidth was limited at the VPN tunnel

level. Based on the operating experience of our sister UK company, Virtu-

alIT, we selected Netscreen (now subsumed under Juniper Networks) as the

vendor for VPN firewall routers.

Our criteria for wireless broadband equipment eliminated big pipes and

feature-rich, high performance gear. Form factor, reliability, and ease of in-

stallation and management were more important than throughput. All inter-

national Internet connections to Kenya in 2003, and at this writing, are car-

ried by satellite. With costs 100X greater than global fiber, satellite connec-

tivity put a financial ceiling on the amount of bandwidth purchased by end-

users. We judged that the bulk of our user population required capacity on

the order of 128 to 256 kbps. We selected Motorola s recently introduced

Canopy platform in line with our business and network model.

Broadband Access, Ltd., went live in July 2003, launching the "Blue" network.

We started small, with a single base station. We wanted demand to drive our

Chapter 11: Case Studies

329

network expansion, rather than relying on a strategy of building big pipes and

hoping we could fill them.

Canopy, and third-party enhancements such as omnidirectional base sta-

tions, permitted us to grow our network as traffic grew, softening initial capi-

tal expenditures. We knew the tradeoff was that as the network expanded,

we would have to sectorize traffic and realign client radios. The gentle learn-

ing curve of a small network paid big dividends later. Technical staff became

comfortable with customer support issues in a simple network environment,

rather than have to deal with them on top of a complex RF and logical

framework. Technical staff attended two-day Motorola training sessions.

A typical PMP design, with base stations linked to a central facility via a Can-

opy high-speed microwave backbone, the network was deployed on building

rooftops, not antenna towers. All leases stipulated 24x7 access for staff,

mains power and, critically, protected the exclusivity of our radio frequencies.

We did not want to restrict landlords from offering roof space to competitors,

rather to simply guarantee that our own services would not be interrupted.

Rooftop deployments provided many advantages. Unlimited physical ac-

cess, unconstrained by night or rain, helped meet the goal of 99.5% network

availability. Big buildings also housed many big clients, and it was possible

to connect them directly into our core microwave network. Rooftop sites did

have the downside of more human traffic--workers maintaining equipment

(a/c) or patching leaks would occasionally damage cabling. As a result all

base stations were set up with two sets of cabling for all network elements, a

primary and a spare.

Site surveys confirmed radio path availability and client requirements. Sur-

vey staff logged GPS positions for each client, and carried a laser range-

finder to determine height of obstacles. Following receipt of payment for

hardware, contractors under the supervision of a technical staffer performed

installations. Canopy has the advantage that the CPE and base station ele-

ments are light, so that most installations do not need extensive civil works or

guying. Cabling Canopy units was also simple, with outdoor UTP connecting

radios directly to customer networks. Proper planning enabled completion of

many installations in less than an hour, and contractor crews did not need

any advanced training or tools.

As we compiled hundreds of customer GPS positions we began to work

closely with a local survey company to overlay these sites on topographical

maps. These became a key planning tool for base station placement.

Note that the point-point VPN tunnel architecture, with its separate physical

and logical layers, required clients to purchase both wireless broadband and

VPN hardware. In order to tightly control quality, we categorically refused to

330

Chapter 11: Case Studies

permit clients to supply their own hardware--they had to buy from us in order

to have service and hardware guarantees. Every client had the same hard-

ware package. Typical installations cost on the order of USD 2500, but that

compares to the $500-600 monthly charges for 64 to 128 kbps of bandwidth.

A benefit of the VPN tunnel approach was that we could prevent a client s

traffic from passing over the logical network (i.e. if their network was hit by a

worm or if they didn t pay a bill) while the radio layer remained intact and

manageable.

As it grew from one base station to ten, and service was expanded to Mom-

basa, the network RF design evolved and wherever possible network ele-

ments (routers) were configured with fallover or hot swap redundancy. Major

investments in inverters and dual conversion UPS equipment at each base

station were required to keep the network stable in the face of an erratic

power grid. After a number of customer issues (dropped VPN connections)

were ascribed to power blackouts, we simply included a small UPS as part of

the equipment package.

Adding a portable spectrum analyzer to our initial capital investment was

costly, but hugely justified as we operated the network. Tracing rogue opera-

tors, confirming the operating characteristics of equipment, and verifying RF

coverage enhanced our performance.

Fanatical attention to monitoring permitted us to uptweak network perform-

ance, and gather valuable historical data. Graphed via MRTG or Cacti (as

described in chapter six), parameters such as jitter, RSSI, and traffic warned

of rogue operators, potential deterioration of cable/connectors, and presence

of worms in client networks. It was not uncommon for clients to claim that

service to their site had been interrupted for hours/days and demand a credit.

Historical monitoring verified or invalidated these claims.

The Blue network combined a number of lessons from Tanzania with im-

proved RF and networking technologies.

Lessons learned

For the next few years satellite circuits will provide all international Internet

connectivity in East Africa. Several groups have floated proposals for sub-

marine fiber connectivity, which will energize telecommunications when it

happens. Compared to regions with fiber connectivity, bandwidth costs in

East Africa will remain very high.

Wireless broadband networks for delivery of Internet services therefore do

not need to focus on throughput. Instead, emphasis should be placed on

reliability, redundancy, and flexibility.

Chapter 11: Case Studies

331

Reliability for our wireless networks was our key selling point. On the net-

work side this translated into sizable investments in infrastructure substitu-

tion, such as backup power, and attention to details such as crimping and

cabling. The most ordinary reasons for a single customer to lose connectivity

were cabling or crimping issues. Radio failures were essentially unheard of.

A key competitive advantage of our customer installation process is that we

pushed contractors to adhere to tight specifications. It was common for well-

managed customer sites to remain connected for hundreds of days with zero

unscheduled downtime. We controlled as much of our infrastructure as pos-

sible (i.e building rooftops).

As attractive as potential alliances with cellular providers seem, in our experi-

ence they raise more problems than they solve. In East Africa, Internet busi-

nesses generate a fraction of the revenue of mobile telephony, and so are mar-

ginal to the cellular companies. Trying to run a network on top of infrastructure

that doesn t belong to you and is, from the point of view of the cellular provider,

a goodwill gesture, will make it impossible to meet service commitments.

Implementing fully redundant networks, with fail-over or hotswap capability is

an expensive proposition in Africa. Nonetheless the core routers and VPN

hardware at our central point of presence were fully redundant, configured for

seamless fail-over, and routinely tested. For base stations we took the deci-

sion not to install dual routers, but kept spare routers in stock. We judged

that the 2-3 hours of downtime in the worst case (failure at 1AM Sunday

morning in the rain) would be acceptable to clients. Similarly weekend staff

members had access to an emergency cupboard containing spare customer

premises equipment, such as radios and power supplies.

Flexibility was engineered into both the logical and RF designs of the net-

work. The point-to-point VPN tunnel architecture rolled out in Nairobi was

extraordinarily flexible in service of client or network needs. Client connec-

tions could be set to burst during off-peak hours to enable offsite backup, as

a single example. We could also sell multiple links to separate destinations,

increasing the return on our network investments while opening up new serv-

ices (such remote monitoring of CCTV cameras) to clients.

On the RF side we had enough spectrum to plan for expansion, as well as

cook up an alternative radio network design in case of interference. With the

growing number of base stations, probably 80% of our customer sites had

two possible base station radios in sight so that if a base station were de-

stroyed we could restore service rapidly.

Separating the logical and RF layers of the Blue network introduced an addi-

tional level of complexity and cost. Consider the long-term reality that radio

technologies will advance more rapidly than internetworking techniques.

Separating the networks, in theory, gives us the flexibility to replace the exist-

332

Chapter 11: Case Studies

ing RF network without upsetting the logical network. Or we may install a

different radio network in line with evolving technologies (Wimax) or client

needs, while maintaining the logical network.

Finally, one must surrender to the obvious point that the exquisite networks

we deployed would be utterly useless without unrelenting commitment to

customer service. That is, after all, what we got paid for.

More information

· Broadband Access, Ltd.: http://www.blue.co.ke/

· AccessKenya, Ltd.: http://www.accesskenya.com/

· VirtualIT: http://www.virtualit.biz/

--Adam Messer, Ph.D

Case study: Dharamsala Community

Wireless Mesh Network

The Dharamsala Wireless-Mesh Community Network came to life in Febru-

ary 2005, following the deregulation of WiFi for outdoor use in India. By the

end of February 2005, the mesh had already connected 8 campuses.

Extensive testing during February of 2005 showed that the hard mountainous

terrain is most suitable for mesh networking, as conventional point-to-

multipoint networks, cannot overcome the line-of-sight limitations presented

by the mountains. mesh topology also offered much larger area coverage,

while the "self healing" nature of mesh routing, proved to be essential in

places where electricity supply is very erratic at best.

The mesh backbone includes over 30 nodes, all sharing a single radio channel.

Broadband Internet services are provided to all mesh members. The total up-

stream Internet bandwidth available is 6 Mbps. There are over 2,000 computers

connected to the mesh, The broadband internet connection is putting the mesh

under great load. At present, the system seems to handle the load without any

increase in latency or packet-loss. It is clear that scalability will become an issue

if we continue to use a single radio channel. To solve this problem, new mesh

routers with multiple radio channel support are being developed and tested in

Dharamsala, with an emphasis on products that meet our technical require-

ments and our economically viable. The initial results are very promising.

The mesh network is based on recurring deployments of a hardware device,

which is designed and built locally known as the Himalayan-Mesh-Router

Chapter 11: Case Studies

333

(http://drupal.airjaldi.com/node/9). The same mesh-routers are installed at

every location, with only different antennas, depending on the geographical

locations and needs. We use a wide range of antennas, from 8 - 11 dBi om-

nidirectional, to 12 - 24 dBi directional antennas and occasionally some high-

gain (and cost) sector antennas.

The mesh is primarily used for:

· Internet access

· File-sharing applications

· Off-site backups

· Playback of high quality video from remote archives.

A central VoIP, software-based PBX is installed (Asterisk) and it provides ad-

vanced telephony services to members. The Asterisk PBX is also interfacing

the PSTN telephone network. However, due to legal issues it is presently

used only for incoming calls into the mesh. Subscribers use a large variety of

software-phones, as well as numerous ATAs (Analog Telephone Adaptors)

and full-featured IP phones.

Figure 11.5: Dharamsala installer working on a tower

The encrypted mesh back-bone does not allow access to roaming mobile

devices (notebooks and PDAs), so we have placed multiple 802.11b access-

points at many of the same locations where mesh-routers are installed. The

mesh provides the backbone infrastructure while these APs provide access

to mobile roaming devices, where needed.

Access to the mesh back-bone is only possible by mesh-routers. Simple

wireless clients lack the intelligence needed to "speak" the mesh routing pro-

tocols and strict access policies. The mesh channel is therefore encrypted

(WPA), and also "hidden" to prevent mobile devices from finding it or attempt-

ing to access it. Allowing access to the mesh only by mesh-routers allows for

strict access control policies and limitations to be enforced at the CPE (Client

Premises Equipment) which is a crucial element needed to achieve end-to-

end security, traffic-shaping, and quality-of-service.

334

Chapter 11: Case Studies

Power consumption of the mesh-Router is less than 4 Watts. This makes

them ideal for using with solar panels. Many of the Dharamsala Mesh routers

are powered solely by small solar panels. The use of solar power in combina-

tion with small antennas and low power routers is ideally suitable for disaster

areas, as it very likely to survive when all other communication infrastructure

is damaged.

--AirJaldi, http://airjaldi.com/

Case study: Networking Mérida State

The city of Mérida lies at the foot of the highest mountain in Venezuela, on

a plateau at about 1600 m. It is the capital of the state of Mérida, and home

to a two- centuries-old university, with some 35,000 students. The Univer-

sity of Los Andes (ULA) deployed the first academic computer network in

1989 which, despite economic limitations, has grown to encompass 26 km

of fiber optic cable over which both a TDM and an ATM (asynchronous

transfer mode) network are overlaid. In 2006, over the same fiber optic

cable, a 50 km Gigabit Ethernet network has been deployed.

Figure 11.6: Mérida is one of the three mountainous states of Venezuela, where the

Andes reach 5000 m.

Nevertheless, many places in the city and the surrounding villages are out of

reach of the fiber optic ring. The university operates a communication server

with telephone lines to provide remote access to its network, but local calls

are charged by the minute and many villages lack phone lines altogether.

Chapter 11: Case Studies

335

For these reasons, efforts to develop wireless access to the university's

network, called RedULA, were undertaken from the very beginning. The

first attempts took advantage of the existing packet network operated by

radio amateurs. As early as 1987, amateurs had a gateway with an HF

(High Frequency) station working at 300 bps for contacts overseas, as

well as several VHF (Very High Frequency) stations linked at 1200 bps

that crisscrossed the country.

While the rugged mountains of the region are a big obstacle for laying cables

and building roads, they can be helpful in deploying a radio network. This

task is aided by the existence of a cable car system, reputedly the highest in

the world, which links the city to a 4765 m peak.

Figure 11.7: On its way to the peak, the cable car passes by an intermediate station

called La Aguada, which is 3450 m high and has an astounding view of the city of

Mérida and other villages at distances up to 50 km.

Packet radio

Local amateurs operate a packet radio network. Initially it worked at 1200

bps, using VHF amateur FM voice radios connected to a personal computer

by means of a terminal node controller (TNC). The TNC is the interface

between the analog radio and the digital signals handled by the PC.

The TNC keys the Push To Talk circuits in the radio to change from transmit

to receive, performs modulation/demodulation and the assembly/disassembly

of packets using a variation of the X.25 protocol known as AX.25. Gateways

between VHF and HF radios were built by attaching two modems to the

same TNC and computer. Typically, a gateway would connect the local VHF

packet network to stations overseas by means of HF stations that could span

thousands of kilometers, albeit at a speed of only 300 bps. A national packet

336

Chapter 11: Case Studies

radio network was also built, which relayed on digipeaters (digital repeaters,

essentially a TNC connected to two radios with antennas pointing in different

directions), to extend the network from Mérida to Caracas by means of just

two such repeater stations. The digipeaters operated at 1200 bps and al-

lowed for the sharing of programs and some text files among amateurs.

Phil Karn, a radio amateur with a strong background in computer networks,

wrote the KA9Q program that implements TCP/IP over AX.25. Using this

program, named after the call sign of its developer, amateurs all over the

world were soon able to connect to the Internet using different kinds of ra-

dios. KA9Q keeps the functions of the TNC to a bare minimum, harnessing

the power of the attached PC for most processing functions. This approach

allows for much greater flexibility and easy upgrades. In Mérida, we were

soon able to upgrade our network to 9600 bps by use of more advanced mo-

dems, and several radio amateurs were now able to access the Internet

through the RedULA wired network. The limit on the radio bandwidth avail-

able on the VHF band puts a cap on the highest attainable speed. To in-

crease that speed, one must move to higher frequency carriers.

Amateurs are allowed to use 100 kHz wide channels using UHF (Ultra-High

Frequency) signals. Digital radios coupled with 19.2 kbps modems doubled

the transmission bandwidth. A project was developed using this technology to

link the House of Science in the city of El Vigia, to Mérida and the Internet.

UHF antennas were built at LabCom, the communications laboratory of ULA.

Figure 11.8: A UHF antenna for packet radio developed at ULA, LabCom.

Although El Vigia is only 100 km from Mérida by road, the mountainous ter-

rain called for the use of two repeaters. One is located at La Aguada, at 3600

m altitude, and the other at Tusta, at 2000 m. The project was financed by

FUNDACITE MERIDA, a government institution that promotes science and

technology in the state. FUNDACITE also operates a pool of 56 kbps tele-

phone modems to provide Internet access for institutions and individuals.

The need for two repeater stations underscores the limitations imposed by

using higher frequency carriers, which require line of sight to establish a reli-

Chapter 11: Case Studies

337

able transmission. In the much lower VHF band, signals are easily reflected

and can reach beyond hills.

Sometimes it is possible to reflect signals using a passive repeater, which is

made by connecting two directional antennas back to back with a coaxial

cable, without any radio. This scheme was tested to connect my residence to

LabCom. The distance is only 11 km, but there is a hill in between that blocks

radio signals. A connection was made by using a passive repeater to reflect

off La Aguada, with the two antennas of the repeater pointing 40 degrees

apart. While this was very exciting and certainly much cheaper than access

through the telephone modems, a faster medium would obviously be needed

for a wireless backbone to connect remote villages.

We therefore explored the use of 56 kbps modems developed by Dale

Heatherington. These modems are housed in a PI2 card built by Ottawa

amateurs, and connected directly to a PC using Linux as the network operat-

ing system. While this system functions very well, the emergence of the

World Wide Web with its plethora of images and other bandwidth-hogging

files made it clear that if we were to satisfy the needs of schools and hospi-

tals we had to deploy a higher bandwidth solution, at least on the backbone.

This meant the use of even higher carrier frequencies in the microwave

range, which entailed high costs.

Fortunately, an alternative technology widely used in military applications

was becoming available for civilian uses at affordable prices. Called spread

spectrum, it first found a use in civilian applications as a short-reach wire-

less local area network, but soon proved to be very useful in places where

the electromagnetic spectrum is not overcrowded, allowing the bridging of

distances of several kilometers.

Spread spectrum

Spread spectrum uses low power signals with its spectrum expanded on

purpose to span all the allocated bandwidth, while at the same time allowing

a number of users to share the medium by using different codes for each

subscriber.

There are two ways to accomplish this: Direct Sequence Spread Spectrum

(DSSS) and Frequency Hopping Spread Spectrum (FHSS).

· In DSSS the information to be transmitted is digitally multiplied by a higher

frequency sequence, thereby augmenting the transmission bandwidth. Al-

though this might seem to be a waste of bandwidth, the recovery system is

so efficient that it can decode very weak signals, allowing for the simulta-

neous use of the same spectrum by several stations.

338

Chapter 11: Case Studies

· In FHSS, the transmitter is constantly changing its carrier frequency inside

the allotted bandwidth according to a specified code. The receiver must

know this code in order to track the carrier frequency.

Both techniques exchange transmission power for bandwidth, allowing many

stations to share a certain portion of the spectrum. During the First Latin

American Networking School (EsLaRed '92), held in Mérida in 1992, we were

able to demonstrate this technique. We established some trial networks mak-

ing use of external antennas built at the LabCom, allowing transmission at

several kilometers. In 1993, the Venezuelan Ministry of Telecommunications

opened up four bands for use with DSSS:

· 400 - 512 MHz

· 806 - 960 MHz

· 2.4 - 2.4835 GHz

· 5.725 - 5.850 GHz

In any of the above bands, maximum transmitter power was restricted to 1 Watt

and the maximum antenna gain to 6 dBi, for a total EIRP (effective isotropic radi-

ated power) of 36 dBm. This ruling paved the way for the deployment of a DSSS

network with a nominal bandwidth of 2 Mbps in the 900 MHz band. This tech-

nology satisfied the needs caused by the surge in World Wide Web activity.

The network started at LabCom, where the connection to RedULA was avail-

able. LabCom housed an inhouse-built Yagi antenna pointed towards a corner

reflector at Aguada. This provided a 90 degree beamwidth, illuminating most of

the city of Mérida. Several subscriber sites, all sharing the nominal 2 Mbps

bandwidth, were soon exchanging files, including images and video clips. Some

subscriber sites that required longer cables between the antenna and the spread

spectrum radio were accommodated by the use of bidirectional amplifiers.

These encouraging results were reported to a group set up at the Interna-

tional Centre for Theoretical Physics (ICTP) in Trieste, Italy, in 1995. This

group was aimed at providing connectivity between the Computer Center,

Physical Sciences Building, and the Technology Building at the University of

Ile-Ife in Nigeria. Later that year, the network was set up by ICTP staff with

funding from the United Nations University and has been running satisfacto-

rily ever since, proving to be a much more cost-effective solution than the

fiber optic network originally planned would have been.

Back in Mérida, as the number of sites increased, the observed throughput per

user declined. We started looking at the 2.4 GHz band to provide additional

capacity. This band can carry three simultaneously independent 2 Mbps

streams, but the effective range is lower than what can be achieved in the 900

MHz band. We were very busy planning the extension of the backbone using

Chapter 11: Case Studies

339

2.4 GHz when we found out about a start-up company that was offering a new

solution that promised longer distances, dramatically higher throughput, and

the possibility of frequency reuse with narrowband microwaves.

Broadband delivery system

After visiting the Nashua, New Hampshire, facilities of Spike Technologies,

we were convinced that their proprietary antenna and radio system was the

best solution for the requirements of our state network, for the following

reasons:

Their broadband delivery system employs a special sectored antenna (Figure

11.9), with 20 dBi gain on each of up to 22 independent sectors. Each sector

transmits and receives on independent channels at 10 Mbps full duplex, for an

aggregate throughput of 440 Mbps. Frequency reuse on interleaved sectors

makes for a spectrally efficient system.

Figure 11.9: Spike Technologies' full duplex, high density sectoral system.

The narrowband digital radios can operate anywhere from 1 to 10 GHz, with

a coverage of up to 50 km. The radios work with a variety of cable TV mo-

dems, delivering a standard 10Base-T LAN connection to the subscriber. At

the base station, the sectors are interconnected with a high-speed switch that

has a very small latency (see Figure 11.10), allowing applications such as

streaming video at up to 30 frames per second. Each sector acts as an inde-

pendent Ethernet LAN.

340

Chapter 11: Case Studies

Cabling to

Network

transceivers

Management

Software

Internet

Switch

Video Servers

File Servers

Telephony

· Conferencing

· Data

Servers

· VOD

· Mail

· Broadcast

· Web pages

Figure 11.10: Spike Technologies' system interconnections.

At the subscriber site, a similar radio and modem provide a 10BaseT connec-

tion to the local Ethernet.

Outside

Equipment

RF Cable

Transceiver

Intermediate

Frequency Cable

Modem

10baseT

connection

Figure 11.11: The subscriber end of the link.

With funding from Fundacite, a trial system was soon installed in Mérida, with

the base station located just above the cable car station of La Aguada at an

altitude of 3600 m.

Chapter 11: Case Studies

341

Figure 11.12: Installation above Mérida at La Aguada, at 3600 meters.

Initially only 5 sectors were installed, with a beamwidth of 16 degrees each.

The first subscriber site was at Fundacite´s premises, where a satellite system

provides Internet access. Sector two served the Governor's Palace. Sector

three served FUNDEM, a relief organization of the local government. Sector

four served a penitentiary near the town of Lagunillas, about 35 km from Mé-

rida. The fifth sector transmitted to a mountaintop repeater close to the village

of La Trampa, 40 km from La Aguada. From La Trampa, another 41 km link

extended the network to the House of Science in the town of Tovar.

On January 31, 1998, a videoconference between the penitentiary and the Jus-

tice Palace in Mérida proved that, aside from Internet access, the system could

also support streaming video. In this case it was used for the arraignment of

prisoners, thus avoiding the inconveniences and risks of their transportation.

The success of the trial prompted the state government to allocate the funding

for a complete system to give high-speed Internet access to the state health

system, educational system, libraries, community centers, and several gov-

ernmental agencies. In January 1999 we had 3 hospitals, 6 educational institu-

tions, 4 research institutions, 2 newspapers, 1 TV station, 1 public library, and

20 social and governmental institutions sharing information and accessing the

Internet. Plans call for 400 sites to be connected within this year at full duplex

10 Mbps speed, and funding has already been allocated for this purpose.

Figure 11.13 shows a map of the state of Mérida. The dark lines show the

initial backbone, while the light lines show the extension.

342

Chapter 11: Case Studies

Figure 11.13: The Mérida State network

Among the many activities supported by the network, it is worthwhile to men-

tion the following:

· Educational: Schools have found an endless supply of material of the

highest quality for pupils and teachers, especially in the areas of geogra-

phy, languages, and sciences, and as a tool to communicate with other

groups that share common interests. Libraries have rooms with computers

accessible to the general public with full Internet capabilities. Newspaper

and TV stations have an amazing source of information to make available

to their audience.

· Health: The university hospital has a direct link to the intensive care unit,

where a staff of specialist physicians is always on duty. These doctors are

available to be queried by their colleagues in remote villages to discuss

specific cases. A group of researchers at the university is developing sev-

eral telemedicine applications based on the network.

Chapter 11: Case Studies

343

· Research: The astronomic observatory of Llano del Hato, located on a

mountain at 3600 m and 8 degrees off the equator will soon be linked, al-

lowing astronomers from all over the world access to the images collected

there. Field researchers in many villages will enjoy Internet access.

· Government: Most government agencies are already connected and start-

ing to put information online for the citizens. We expect this to have a pro-

found impact on the relationship of citizens with the government. Relief

agencies and law enforcement agencies make heavy use of the network.

· Entertainment and Productivity: For people living outside the city, the

opportunities offered by the Net have a significant impact on the quality of

their lives. We hope that this will help to reverse the trend of migrating out

of the countryside, alleviating the overcrowding of the urban areas. Farm-

ers have access to information about the commanding prices of their crops

and supplies, as well as improved agricultural practices.

SUPERCOMM '98, held in Atlanta in June, cited the Mérida broadband deliv-

ery network as winner of the SUPERQuest award in category 8-Remote Ac-

cess as the best in that particular field of nominees.

Training

Since our earliest efforts to establish a computer network, we realized that

training was of paramount importance for the people involved in the network

construction, management, and maintenance. Given our very limited budget,

we decided that we had to pool our resources with those of other people who

also required training. In 1990 the ICTP organized the First International

School on computer network analysis and management, which was attended

by Professor Jose Silva and Professor Luis Nunez from our university. Upon

returning to Mérida, they proposed that we should somehow emulate this

activity in our university. To this end, taking advantage of my sabbatical, I

spent three months at Bellcore in Morristown, New Jersey, and three more

months at the ICTP helping in the preparation of the Second Networking

School in 1992, where I was joined by my colleague Professor Edmundo Vi-

tale. I spent the rest of my sabbatical at SURANET in College Park, Mary-

land, under the guidance of Dr. Glenn Ricart, who introduced me to Dr. Saul

Hahn of the Organization of American States, who offered financial support

for a training activity in Latin America. These experiences allowed us to

launch the First Latin American Networking School (EsLaRed'92) in Mérida,

attended by 45 participants from 8 countries in the region, with instructors

from Europe, the United States, and Latin America. This hands-on training

lasted three weeks, and wireless technologies were emphasized.

EsLaRed'95 gathered again in Mérida with 110 participants and 20 instruc-

tors. EsLaRed'97 had 120 participants, and it was endorsed by the Internet

Society, which also sponsored a Spanish and Portuguese first Networking

344

Chapter 11: Case Studies

Workshop for Latin America and the Caribbean, held in Rio de Janeiro in 1998

with EsLaRed responsible for the training content. Now ten years later, Es-

LaRed continues to expand its training efforts throughout South America.

Concluding remarks

The Internet has an even more profound impact in developing countries

than elsewhere, owing to the high cost of international phone calls, faxes,

magazines, and books. This is obviously exacerbated by the lower average

income of people. Some dwellers in remote villages that do not have tele-

phones are experiencing a transition from the 19th to the 21st century

thanks to wireless networking. It is hoped that this will contribute to the im-

provement of lifestyles in the fields of health, education, entertainment, and

productivity, as well as create a more equitable relationship between citi-

zens and government.

References

· Karn, Phil, "The KA9Q Internet (TCP/IP) Package: A Progress Report,"

Sixth ARRL Computer Networking Conference, Redondo Beach, CA, 29

August 1987.

· Heatherington, D., "A 56 kilobaud RF modem," Sixth ARRL Computer Net-

working Conference, Redondo Beach, CA, 29 August 1987.

· Conatel, Comision Nacional de Comunicaciones, Ministerio de Transporte

y Comunicaciones, "NORMAS PARA LA OPERACION DE SISTEMAS DE

TELECOMUNICACIONES CON TECNOLOGIA DE BANDA ESPARCIDA

(SPREAD SPECTRUM)," Caracas, 17 November 1993.

· International Centre For Theoretical Physics, "Programme of Training and

System Development on Networking and Radiocommunications," Trieste,

Italy, 1996, http://www.ictp.trieste.it/

· Escuela Latinoamericana de Redes, http://www.eslared.org.ve/

--Ermanno Pietrosemoli

Chapter 11: Case Studies

345

Case study: Chilesincables.org

Recent wireless data transmission technologies allow the creation of high

speed, geographically separated networks at a relatively low cost. If these

networks are built around the idea of removing restrictions to data access,

we call them free networks. Such networks can bring great benefits to every

user, independent of the their political, economic, or social conditions. This

kind of network is a direct response to the often restrictive commercial model

ruling over much of our modern western society.

In order for free networks to flourish, wireless technologies must be adapted

and put to the best possible use. This is carried out by groups of hackers

who do the research, investigation, development and implementation of pro-

jects, as well as permit free access to the knowledge gained.

Chilesincables.org endeavors to promote and organize wireless free net-

works in Chile in a professional way. We do this by providing education

about the related legal and technical aspects of wireless networking; en-

couraging the adaptation of new technologies through adequate research;

and stimulating the adaptation of these technologies to meet the specific

needs of Chilean communities and society.

Description of technology

We employ a variety of wireless technologies, including IEEE 802.11a/b/g.

We are also investigating recent innovations in the field, such as WiMAX. In

most cases, the equipment has been modified in order to be accept external

locally built antennas which meet local telecommunications regulations.

Even though a majority of wireless hardware available on the market will suit

our goals, we encourage utilization and exploration of a few vendors that al-

low for better control and adaptation to our needs (without necessarily in-

creasing the prices). These include Wi-Fi cards with chipsets offered by Athe-

ros, Prism, Orinoco, and Ralink, as well as some models of access points

manufactured by Linksys, Netgear, and Motorola. The hacker community has

developed firmware that provides new functionality on this equipment.

For the network backbone itself, we employ Open Source operating systems,

including GNU/Linux, FreeBSD, OpenBSD, and Minix. This fits our needs in

the areas of routing as well as implementation of services such as proxies,

web and FTP servers, etc. In addition, they share our project s philosophy of

being free technology with open source code.

346

Chapter 11: Case Studies

Uses and applications

The networks implemented so far allow the following tasks:

· Transfer of data via FTP or web servers

· VoIP services

· Audio and video streaming

· Instant messaging

· Exploration and implementation of new services such as LDAP, name reso-

lution, new security methods, etc.

· Services provided by the clients. The users are free to use the net s infra-

structure in order to create their own services.

Administration and maintenance

The operational unit of the network is the node. Each node allows clients to

associate to the network and obtain basic network services. In addition,

each node must be associated to at least another node, by convention.

This allows the network to grow and to make more services available to

every client.

A node is maintained by an administrator who is a member of the community

committed to the following tasks:

· Maintenance of an adequate uptime (over 90%).

· Providing basic services (typically web access).

· Keeping the clients updated about the node s services (for example, how to

get access to the network). This is generally provided by a captive portal.

The general administration of the network (specifically, tasks related to de-

ployment of new nodes, selection of sites, network s topology, etc.) is carried

out by the board of the community, or by technicians trained for this purpose.

Chilesincables.org is currently in the process of acquiring legal organization

status, a step that will allow the regulation of its internal administrative proce-

dures and the formalization of the community in our society.

Chapter 11: Case Studies

347

Training and capacity building

Chilesincables.org considers training of its members and clients to be of vital

importance for the following reasons:

· The radio spectrum must be kept as clear as possible in order to guarantee

adequate quality of wireless connections. Therefore, training in radio com-

munications techniques is essential.

· The employment of materials and methods approved by the current regula-

tions is a requirement for the normal development of the activities.

· In order to comply with Internet standards, all of our network administrators

are trained in TCP/IP networking.

· To ensure continuity in network operations, knowledge of networking tech-

nology must be transferred to the users.

To support these principles, Chilesincables.org undertakes the following

activities:

· Antenna Workshop. Attendees are trained in the construction of anten-

nas, and introduced to basic concepts of radio communication.

· Operating Systems Workshop. Training on the implementation of routers

and other devices based on GNU/Linux or other software such as

m0n0wall or pfsense. Basic networking concepts are also taught.

· Promotion and Advertising. Events for different communities that pursue

our same goals are promoted. These include college workshops, lectures,

free software gatherings, etc.

· Updating of Materials. Chilesincables.org maintains a number of free-

access documents and materials made available to people interested in a

specific activity.

The pictures on the following pages present a brief account of the activities in

our community.

348

Chapter 11: Case Studies

Figure 11.14: Omnidirectional slotted antenna workshop. In this session, attendants

learned about building antennas and related theory.

Figure 11.15: One of our staff members lecturing on the implementation of a

m0nowall-based router in the administration of a node.

Chapter 11: Case Studies

349

Figure 11.16: Detail of mini tower with samples of antennas, cables and pigtails.

350

Chapter 11: Case Studies

Figure 11.17: Wireless station and parabolic antenna used for the transmission of

Santiago-2006 FLISOL via streaming video.

Figure 11.18: Location of the other end of the link.

Chapter 11: Case Studies

351

Figure 11.19: Schematic representing Santiago-2006 FLISOL video streaming

transmission, using free software. The wireless transmission speed achieved was

36 Mbps at 1 km.

352

Chapter 11: Case Studies

Figure 11.20: Quiani node. This is one of the world's highest nodes. Its located at an

elevation of 4000 m, about 2000 km north of the country's capital.

Figure 11.21: Node in southern Santiago, consisting of a 15 m tower, a Trevor

Marshall 16+16 antenna, and 30 clients. The node is connected to a downtown node

more than 12 km away.

Chapter 11: Case Studies

353

Figure 11.22: Panoramic view of a node from the top of the tower.

Figure 11.23: Downtown node connected to the Santiago southern node. Note the

parabolic antenna for backhaul and the slotted antenna to connect the clients.

354

Chapter 11: Case Studies

Figure 11.24: Implementation of node over a water tower in Batuco, Metropolitan

Region, providing backhaul to Cabrati telecenter.

Figure 11.25: Workshop on Yagi antennas organized by our community. Participants

are building their own antennas.

Chapter 11: Case Studies

355

Credits

Our community is made up of a group of committed volunteer associates

among which are worthy of notice:

Felipe Cortez (Pulpo), Felipe Benavides (Colcad), Mario Wagenknecht

(Kaneda), Daniel Ortiz (Zaterio), Cesar Urquejo (Xeuron), Oscar Vasquez

(Machine), Jose San Martin (Packet), Carlos Campano (Campano), Christian

Vasquez (Crossfading), Andres Peralta (Cantenario), Ariel Orellana (Ariel),

Miguel Bizama (Picunche), Eric Azua (Mr. Floppy), David Paco (Dpaco),

Marcelo Jara (Alaska).

--Chilesincables.org

Case study: Long Distance 802.11

Thanks to a favorable topography, Venezuela already has some long range

WLAN links, like the 70 km long operated by Fundacite Mérida between Pico

Espejo and Canagua.

To test the limits of this technology, it is necessary to find a path with an unob-

structed line of sight and a clearance of at least 60% of the first Fresnel zone.

While looking at the terrain in Venezuela, in search of a stretch with high ele-

vation at the ends and low ground in between, I first focused in the Guayana

region. Although plenty of high grounds are to be found, in particular the fa-

mous "tepuys" (tall mesas with steep walls), there were always obstacles in

the middle ground.

My attention shifted to the Andes, whose steep slopes (rising abruptly from

the plains) proved adequate to the task. For several years, I have been trav-

eling through sparsely populated areas due to my passion for mountain bik-

ing. In the back of my head, I kept a record of the suitability of different spots

for long distance communications.

Pico del Aguila is a very favorable place. It has an altitude of 4200 m and is

about a two hour drive from my home town of Mérida. For the other end, I

finally located the town of El Baúl, in Cojedes State. Using the free software

Radio Mobile (available at http://www.cplus.org/rmw/english1.html), I found

that there was no obstruction of the first Fresnel zone (spanning 280 km)

between Pico del Aguila and El Baúl.

356

Chapter 11: Case Studies

Action Plan

Once satisfied with the existence of a suitable trajectory, we looked at the

equipment needed to achieve the goal. We have been using Orinoco cards

for a number of years. Sporting an output power of 15 dBm and receive

threshold of -84 dBm, they are robust and trustworthy. The free space loss at

282 km is 149 dB. So, we would need 30 dBi antennas at both ends and

even that would leave very little margin for other losses.

On the other hand, the popular Linksys WRT54G wireless router runs

Linux. The Open Source community has written several firmware versions

for it that allow for a complete customization of every transmission parame-

ter. In particular, OpenWRT firmware allows for the adjustment of the ac-

knowledgment time of the MAC layer, as well as the output power. Another

firmware, DD-WRT, has a GUI interface and a very convenient site survey

utility. Furthermore, the Linksys can be located closer to the antenna than a

laptop. So, we decided to go with a pair of these boxes. One was config-

ured as an AP (access point) and the other as a client. The WRT54G can

operate at 100 mW output power with good linearity, and can even be

pushed up to 200 mW. But at this value, non linearity is very severe and

spurious signals are generated, which should be avoided. Although this is

consumer grade equipment and quite inexpensive, after years of using it,

we felt confident that it could serve our purpose. Of course, we kept a

spare set handy just in case.

By setting the output power to 100 mW (20 dBm), we could obtain a 5dB

advantage compared with the Orinoco card. Therefore, we settled for a pair

of WRT54Gs.

Pico del Águila site survey

On January 15, 2006, I went to Pico Águila to check out the site that Radio

Mobile had reported as suitable. The azimuth towards El Baúl is 86°, but

since the magnetic declination is 8° 16 , our antenna should be pointed to a

magnetic bearing of 94°.

Unfortunately, when I looked towards 94°, I found the line of sight obstructed

by an obstacle that had not been shown by the software, due to the limited

resolution of the digital elevation maps that are freely available.

I rode my mountain bike for several hours examining the surrounding area

looking for a clear path towards the East. Several promising spots were iden-

tified, and for each of them I took photos and recorded the coordinates with a

GPS for later processing with the Radio Mobile software. This led me to re-

fine my path selection, resulting in the one depicted in Figure 11.26 using

Google Earth:

Chapter 11: Case Studies

357

Figure 11.26: View of the 280 km link. Maracaibo's Lake is to the West, and the

Peninsula of Paraguaná is to the North.

The radio profile obtained with Radio Mobile is shown in Figure 11.27:

Figure 11.27: Map and profile of the proposed path between Pico Aguila, and

Morrocoy hill, near the town of El Baúl.

358

Chapter 11: Case Studies

The details of the wireless link are displayed in Figure 11.28:

Figure 11.28: Propagation details of the 280 km link.

In order to achieve a reasonable margin of some 12 dB for the link, we

needed antennas with at least 30 dBi gain at each end.

Antennas

High gain antennas for the 2.4 GHz band are not available in Venezuela. The

importation costs are considerable, so we decided instead to recycle para-

bolic reflectors (formerly used for satellite service) and replaced the feed with

one designed for 2.4 GHz. We proved the concept with an 80 cm dish. The

gain was way too low, so we tried an offset fed 2.4 m reflector. This offered

ample gain, albeit with some difficulties in the aiming of the 3.5° beam. The

22.5° offset meant that the dish appeared to be pointing downwards when it

was horizontally aligned.

Several tests were performed using various cantennas and a 12 dBi Yagi as a

feed. We pointed the antenna at a base station of the university wireless net-

work that was located 11 km away on a 3500 m mountain. The test site sits at

2000 m and therefore the elevation angle is 8°. Because of the offset feed, we

pointed the dish 14° downward, as can be seen in the following picture:

Chapter 11: Case Studies

359

Figure 11.29: 2.4 m offset fed reflector with a 12 dBi antenna at its focus, looking

14° down. The actual elevation is 8° up.

We were able to establish a link with the base station at Aguada, but our efforts

to measure the gain of the setup using Netstumbler were not successful. There

was too much fluctuation on the received power values of live traffic.

For a meaningful measurement of the gain, we needed a signal generator

and spectrum analyzer. These instruments were also required for the field

trip in order to align the antennas properly.

While waiting for the required equipment, we looked for an antenna to be used at

the other end, and also a pointing system better suited to the narrow radio beam.

In February 2006, I traveled to Trieste to partake in the annual wireless train-

ing event that I have been attending since 1996. While there, I mentioned the

project to my colleague Carlo Fonda, who was immediately thrilled and eager

to participate.

The collaboration between the Latin American Networking School (Es-

LaRed) and the Abdus Salam International Centre for Theoretical Phys-

ics (ICTP) goes back to 1992, when the first Networking School was held in

Mérida with ICTP support. Since then, members of both institutions have col-

laborated in several activities. Some of these include an annual training

school on wireless networking (organized by ICTP) and another on computer

360

Chapter 11: Case Studies

networks (organized by EsLaRed) that are hosted in several countries

throughout Latin America. Accordingly, it was not difficult to persuade Dr.

Sandro Radicella, the head of the Aeronomy and Radio Propagation Labora-

tory at ICTP, to support Carlo Fonda s trip in early April to Venezuela in order

to participate in the experiment.

Back at home, I found a 2.75 m parabolic central fed mesh antenna at a

neighbors house. Mr. Ismael Santos graciously lent his antenna for the ex-

periment.

Figure 11.30 shows the disassembly of the mesh reflector.

Figure 11.30: Carlo and Ermanno disassembling the satellite dish supplied by

Mr. Ismael Santos.

We exchanged the feed for a 2.4 GHz one, and aimed the antenna at a sig-

nal generator that was located on top of a ladder some 30 m away. With a

spectrum analyzer, we measured the maximum of the signal and located the

focus. We also pinpointed the boresight for both the central fed and the offset

antennas. This is shown in Figure 11.31:

Chapter 11: Case Studies

361

Figure 11.31: Finding the focus of the antennas with the 2.4 GHz feed

We also compared the power of the received signal with the output of a

commercial 24 dBi antenna. This showed a difference of 8 dB, which led us

to believe that the overall gain of our antenna was about 32 dBi. Of course,

there is some uncertainty about this value. We were receiving reflected sig-

nals, but the value agreed with the calculation from the antenna dimension.

El Baúl Site Survey

Once we were satisfied with the proper functioning and aim of both anten-

nas, we decided to do a site survey at the other end of the El Baúl link.

Carlo Fonda, Gaya Fior and Ermanno Pietrosemoli reached the site on

April 8th. The following day, we found a hill (south of the town) with two

telecom towers from two cell phone operators and one belonging to the

mayor of El Baúl. The hill of Morrocoy is some 75 m above the surrounding

area, about 125 m above sea level. It provides an unobstructed view to-

wards El Aguila. There is a dirt road to the top, a must for our purpose,

given the weight of the antenna.

362

Chapter 11: Case Studies

Performing the experiment

On Wednesday April 12th, Javier Triviño and Ermanno Pietrosemoli traveled

towards Baúl with the offset antenna loaded on top of a four-wheel drive

truck. Early the morning of April 13th, we installed the antenna and pointed it

at a compass bearing of 276°, given that the declination is 8° and therefore

the true Azimuth is 268°.

At the same time, the other team (composed by Carlo Fonda and Gaya Fior

from ICTP, with assistance of Franco Bellarosa, Lourdes Pietrosemoli and

José Triviño) rode to the previously surveyed area at Pico del Aguila in a

Bronco truck that carried the 2.7 m mesh antenna.

Figure 11.32: Pico del Águila and surrounds map with Bronco truck.

Poor weather is common at altitudes of 4100 m above sea level. The Aguila team

was able to install and point the mesh antenna before the fog and sleet began.

Figure 11.33 shows the antenna and the rope used for aiming the 3° radio beam.

Power for the signal generator was supplied from the truck by means of a 12

VDC to 120 VAC inverter. At 11 A.M in El Baúl, we were able to observe a

-82 dBm signal at the agreed upon 2450 MHz frequency using the spectrum

analyzer. To be sure we had found the proper source, we asked Carlo to

switch off the signal. Indeed, the trace on the spectrum analyzer showed only

noise. This confirmed that we were really seeing the signal that originated

some 280 km away.

After turning the signal generator on again, we performed a fine tuning in

elevation and azimuth at both ends. Once we were satisfied that we had at-

tained the maximum received signal, Carlo removed the signal generator and

replaced it with a Linksys WRT54G wireless router configured as an access

point. Javier substituted the spectrum analyzer on our end for another

WRT54G configured as a client.

Chapter 11: Case Studies

363

Figure 11.33: Aiming the antenna at el Águila.

At once, we started receiving "beacons" but ping packets did not get through.

This was expected, since the propagation time of the radio wave over a 300 km

link is 1 ms. It takes at least 2 ms for an acknowledgment to reach the transmitter.

Fortunately, the OpenWRT firmware allows for adjusting the ACK timing. After Carlo

adjusted for the 3 orders of magnitude increase in delay above what the standard

Wi-Fi link expects, we began receiving packets with a delay of about 5 ms.

Figure 11.34: El Baúl antenna installation. Actual elevation was 1° upward, since the

antenna has an offset of 22.5°.

364

Chapter 11: Case Studies

We proceeded to transfer several PDF files between Carlo s and Javier s

laptops. The results are shown in Figure 11.35.

Figure 11.35: Screenshot of Javier's laptop showing details of PDF file transfer from

Carlo's laptop 280 km away, using two WRT54G wireless routers, no amplifiers.

Note the ping time of a few milliseconds.

Figure 11.36: Javier Triviño (right) and Ermanno Pietrosemoli beaming from the

El Baúl antenna

Chapter 11: Case Studies

365

Figure 11.37: Carlo Fonda at the Aguila Site

Mérida, Venezuela, 17 April 2006.

One year after performing this experiment, we found the time and resources

to repeat it. We used commercial 30 dBi antennas, and also a couple of wire-

less routers which had been modified by the TIER group led by Dr. Eric

Brewer of Berkeley University.

The purpose of the modification of the standard WiFi MAC is to make it suit-

able for long distance applications by replacing the CSMA Media Access

Control with TDMA. The latter is better suited for long distance point-to-point

links since it does not require the reception of ACKs. This eliminates the

need to wait for the 2 ms round trip propagation time on a 300 km path.

On April 28th, 2007, a team formed by Javier Triviño, José Torres and Fran-

cisco Torres installed one of the antennas at El Aguila site. The other team,

formed by Leonardo González V., Leonardo González G., Alejandro

González and Ermanno Pietrosemoli, installed the other antenna at El Baúl.

A solid link was quickly established using the Linksys WRT54G routers.

This allowed for video transmission at a measured throughput of 65 kbps.

With the TDMA routers, the measured throughput was 3 Mbps in each di-

rection. This produced the total of 6 Mbps as predicted by the simulations

done at Berkeley.

366

Chapter 11: Case Studies

Can we do better?

Thrilled by these results, which pave the way for really inexpensive long dis-

tance broadband links, the second team moved to another location previ-

ously identified at 382 km from El Aguila, in a place called Platillón. Platillón

is 1500 m above sea level and there is an unobstructed first Fresnel zone

towards El Aguila (located at 4200 m above sea level). The proposed path is

shown in Figure 11.38:

Figure 11.38: Map and profile of the 380 km path.

Again, the link was quickly established with the Linksys and the TIER sup-

plied routers. The Linksys link showed approximately 1% packet loss, with an

average round trip time of 12 ms. The TIER equipment showed no packet

loss, with propagation times below 1 ms. This allowed for video transmission,

but the link was not stable. We noticed considerable signal fluctuations that

often interrupted the communication.

However, when the received signal was about -78 dBm, the measured

throughput was a total of 6 Mbps bidirectional with the TIER routers imple-

menting TDMA.

Chapter 11: Case Studies

367

Figure 11.39: The team at el Aguila, José Torres (left), Javier Triviño (center) and Fran-

cisco Torres (right)

Although further tests must be conducted to ascertain the limits for stable

throughput, we are confident that Wi-Fi has a great potential for long distance

broadband communication. It is particularly well suited for rural areas were

the spectrum is still not crowded and interference is not a problem, provided

there is good radio line of sight.

Acknowledgments

We wish to express our gratitude to Mr. Ismael Santos for lending the mesh

antenna to be installed at El Aguila and to Eng. Andrés Pietrosemoli for sup-

plying the special scaffolding joints used for the installation and transporta-

tion of the antennas.

We'd also like to thank the Abdus Salam International Centre of Theoretical

Physics for supporting Carlo Fonda s trip from Italy to Venezuela.

368

Chapter 11: Case Studies

Figure 11.40: The team at Platillon. From left to right: Leonardo González V.,

Leonardo González G., Ermanno Pietrosemoli and Alejandro González .

The 2006 experiment was performed by Ermanno Pietrosemoli, Javier

Triviño from EsLaRed, Carlo Fonda, and Gaya Fior from ICTP. With the help

of Franco Bellarosa, Lourdes Pietrosemoli, and José Triviño.

For the 2007 experiments, Dr. Eric Brewer from Berkeley University provided

the wireless routers with the modified MAC for long distance, as well as en-

thusiastic support through his collaborator, Sonesh Surana. RedULA, CPTM,

Dirección de Servicios ULA Universidad de los Andes and Fundacite Mérida

contributed to this trial.

This work was funded by ICA-IDRC.

Chapter 11: Case Studies

369

References

· Fundación Escuela Latinoamericana de Redes, Latin American Networking

School, http://www.eslared.org.ve/

· Abdus Salam International Centre for Theoretical Physics,

http://wireless.ictp.it/

· OpenWRT Open Source firmware for Linksys, http://openwrt.org/

· Fundacite Mérida, http://www.funmrd.gov.ve/

--Ermanno Pietrosemoli

Table of Contents: