Table of Contents

Welcome

Chapter 1: Introduction
1.1: Introduction to Current and Upcoming Technologies
1.2: Why Traditional Models Won't Work in Rural Canada
1.3: Overview of the Structure of This Thesis

Chapter 2: Internet Use in Northern Ontario
2.1: Communities and Infrastructure
2.2: Community Based Networks and Their Efforts
2.3: Computer, Internet and Broadband Adoption Statistics
2.4: Outlook for Northern Ontario

Chapter 3: Wired Networks
3.1: Fibre Optic Networks
3.2: DSL
3.3: Cable
3.4: BPL

Chapter 4:Wireless Networks
4.1: WiFi
4.2: Broadband Cellular
4.3: MMDS and LMDS
4.4: Motorola Canopy
4.5: WiMAX

Chapter 5: Satellite Networks
5.1: KU-band and KA-band
5.2: Satellite as a Pipeline

Chapter 6: Broadband Funding Programs
6.1: Government Support for Broadband
6.2: Northern Ontario Heritage Fund Corporation (NOHFC)
6.3: Satellite Internet for Remote Areas (SIRA)
6.4: Federal Funding

Chapter 7: Infrastructure Feasibility Algorithms
7.1: The Need for Algorithms and the Definition of Feasibility
7.2: Reasoning Behind the Algorithms
7.3: Broadband Pipeline Physical Feasibility Algorithm
7.4: Broadband Last Mile Infrastructure Physical Feasibility Algorithm
7.5: Case Study: The Cottage
7.6: Case Study: Village of Hilton Beach
7.7: Afterthoughts on the Algorithms

Chapter 8: Conclusion

Appendix A: List of References
Appendix B: Glossary

Chapter 4: Wireless Networks

4.1: WiFi

WiFi, which is short for Wireless Fidelity, is a wireless version of Ethernet operating in the 2.4 GHz and 5 GHz frequency range. This technology exploded in 2005, and 802.11 (the name of this IEEE standard) devices are now very inexpensive and widely available everywhere. The heart of a WiFi network is the Wireless Access Point (WAP), which usually has a small “rubber ducky” antenna and often incorporates a wired switch or router as well. The WAP broadcasts its Service Set Identifier (SSID) to identify the network by name every 100ms in a packet called a beacon. A client computer or PDA equipped with a WiFi network device will hear these beacons from nearby access points, and will display a list of networks the client can connect to. Once a connection is established, the client can move around within the range of the WAP and utilize network services (i.e. The Internet) just as if it was connected by a regular Ethernet cat-5 cable. Because WiFi is a point-to-multipoint technology, several clients are able to connect to a WAP simultaneously.

The first 802.11 standard was released in 1997 but was never widely implemented because of its slow 1 Mbps throughput. The 802.11b amendment was introduced in 1999 and these devices appeared on the market very quickly, becoming popular thanks to their affordability [79]. With a bandwidth of 11 Mbps, these 802.11b WAPs have enough throughput for any Internet application and most other network services. The 802.11a amendment was released at about the same time as 802.11b, but the products didn't hit the market until 2001 because of the slow availability of 5 GHz components. Because of this amendment's use of the 5 GHz frequency band, the distance covered by an “a” access point is not as great, and a line of sight is needed, even indoors. The throughput of 802.11a is 54 Mbps, however, so this amendment did get more attention from consumers than the original 802.11 did. The next standard, released in 2003, brought togetherthe best of both worlds. Operating at 54 Mbps on the reliable 2.4 GHz frequency band, 802.11g is the current WiFi standard and is slowly replacing 802.11a and 802.11b networks. With an indoor range of about 100 feet and an outdoor range of about 500 feet, 802.11g can handle most any networking requirement in a home or small office. Consumer-grade 802.11g WAPs and network cards are priced under $100, while commercial-grade WAPs having greater security and user management features can be up to $2000 [58]. The range of a WAP can be extended somewhat with high-gain 2.4 GHz antennas and amplifiers [74]. Using highly directional parabolic or yagi antennas, a WiFi link can be established over about 100km [29], but a reliable, fast link would probably have a distance limitation of under 20km. When a high-gain omnidirectional antenna is used, however, the range can in practice be extended only up to a kilometre or two (based on this author's experience). A point-to-point link using directional antennas can be as inexpensive as $500, with the ability to cover a distance of under 5km. Greater distances could be covered using larger, more expensive antennas and amplifiers. A point-to-multipoint setup using an outdoor omnidirectional antenna would also start off in the same price range.

Because of WiFi's affordable hardware and simple installation, these small wireless networks have popped up across the world in coffee shops, airports and other public places. Some cities have even began rolling out community-wide wireless networks; some free, some by subscription. If you take this community-wide idea to the extreme, you get wireless mesh networks. Using a grid of 802.11 access points that also act as repeaters, an entire community can be blanketed with a WiFi signal quite inexpensively. Each WAP can hear its neighbouring WAP and they share information about their neighbours and client machines so that data can be routed properly by hopping across the grid [65]. To give the grid access to the Internet, one or more WAPs need to have a high speed connection to an Internet pipeline. The Northern Ontario community of Chapleau is playing host to the first wireless mesh network in Canada. “Project Chapleau” was launched in November 2005 by Bell Canada and Nortel to “research the impact of advanced technology on rural and dispersed communities” [61]. The network covers the core of this densely-populated community of 2800 people.

Outdoor WAPs specially suited for wireless mesh network applications, such as the BelAir 200 by Kanata, Ontario's BelAir Networks, have an average range of 450 metres depending on the environment. The cost for each WAP, which serves both as a backhaul and an access point, ranges from about $4,000 to $10,000, so the cost per kilometre (two WAPs) would be between $8,000 and $20,000 [85].

WiFi has certainly changed the way homes and offices are networked, but only recently has this technology moved outdoors. Specialized outdoor WiFi equipment can easily be deployed to provide a community with reliable last-mile Internet access. Because of the short range and line-of-sight requirements of a standalone WiFi access point, this kind of infrastructure is not likely to provide a solid broadband solution to a community or rural area. While WiFi was not intended to be anything more than a local area networking technology, it has been successfully demonstrated that it can work over several kilometres in a point-to-point link using directional antennas. This may be an option for connecting remote houses and buildings to a network, but the low throughput of under 10 Mbps means it's not viable for use as an Internet pipeline to serve a last-mile network.

Considering that a WiFi mesh network only requires a pipeline going into one mesh node to provide the entire wireless grid with Internet access, and that each node plays the dual-role of access point for clients and pipeline to neighbouring nodes, this solution is actually quite affordable. It is an attractive solution for small communities with a relatively high population density, as the entire population may be covered by five or six nodes. Provided that a pipeline is available to at least one of them, this is a far less expensive solution than, say, DSL or cable. The grid can easily be expanded by clamping another node onto a telephone pole within eye shot of another node; it is not dependent on preexisting wired infrastructure or 5.5km range issues like cable or DSL. Wireless mesh may not hold much promise for truly rural areas, however. It is unlikely that a network provider would be willing to spend eight to twenty thousand dollars to extend the grid another kilometre to allow one or two rural customers to connect to the network.

While unlikely to be feasible in rural areas, small communities where DSL and cable may not be economically feasible should explore and invest in WiFi mesh technology. It is affordable, expandable and easy to implement, and may provide a solution to the broadband problem.

<< Previous :: Next >>

© Jake Cormier, 2006 [jake (at) stormcloudstudios.com]
Completed as a partial requirement for the degree of Bachelor of Science (specialized)
Department of Computer Science :: Algoma University College :: Sault Ste. Marie, Ontario :: Spring 2006