Wireless
From TIER
1 Research Overview
Many rural regions in developing and developed countries with low user densities do not have good connectivity solutions. To date, networking research has largely focused on urban areas of the industrialized world with high user densities.
WiFi-based Long Distance networks (WiLDNet) are emerging as a potential low-cost alternative to traditional connectivity solutions for rural regions. The primary cost gains arise from the use of low cost and low power single board computers, high-volume off-the-shelf 802.11 wireless cards originally intended for industrialized markets and low cost towers.
Unlike mesh networks, which use omni-directional antennas to cater to short ranges (less than 1-2 km at most), WiLD networks are comprised of point-to-point wireless links that use high-gain directional antennas (e.g. 24 dBi, 8 deg beam-width) with line of sight (LOS) over long distances (10-100 km).
We currently have several deployments of our low-cost WiFi-based Long Distance (WiLD) network in India, Ghana, Guinea Bissau and Philippines in addition to our local testbed in the Bay Area.
Here is a list of some of the long distance links that have been tried with our equipment and software (including the record-breaking 279km link in Venezuela).
For more details on the research agenda in long distance point-to-point wireless networks, read our paper, Rethinking Wireless in the Developing World in Hotnets V, 2006.
The full list of publications is here.
2 Projects
2.1 WiLDNet: Network protocol design for WiFi based long distance networks
The performance of WiLD networks in real-world deployments is abysmal. The poor performance characteristics results from the fact that the conventional 802.11 MAC protocol has several fundamental deficiencies when being adopted in multi-hop WiLD network settings.
In this project, we address the following question: What are the link- and MAC-layer modifications essential to achieve good transport performance in multi-hop WiLD networks? In addressing this problem, an important constraint is that any solution should continue to leverage existing 802.11 hardware to preserve the cost savings.
For more details, read our paper, WiLDNet: Design and Implementation of High Performance WiFi Based Long Distance Networks in NSDI, 2007.
2.2 Remote Monitoring, Diagnosis and Management of Wireless Networks
Generally WiFi-enabled Long Distance (WiLD) networks are managed by non-local users who cannot guarantee long-term support beyond the pilot. For long term operational sustainability, it is essential that maintenance duties be transferred to local administrators. This research is focussed on designing simplified diagnosis solutions as an enabler for locally managed WiLD networks.
For more details, read our paper, Simplifying Fault Diagnosis in Locally Managed Rural WiFi Networks that is going to appear in the Workshop on Networked Systems for Developing Regions (NSDR) as part of SIGCOMM, 2007 in Kyoto.
2.3 Wireless Channel Characterization
The aim is to perform a systematic study to investigate the commonly cited sources of packet loss induced by the wireless channel for long-distance links. The channel induced losses include external WiFi, non-WiFi and multipath interference. The protocol induced losses include protocol timeouts and the breakdown of CSMA over WiLD links.
Our measurements are performed on real-world WiLD deployments and a wireless channel emulator. The channel emulator allows us to study each source of packet loss in isolation in a controlled environment. Based on our experiments we observe that the presence of external WiFi interference leads to significant amount of packet loss in WiLD links. In addition to identifying the sources of packet loss, we analyze the loss variability across time. We also explore the solution space and propose a range of MAC and network layer adaptation algorithms to mitigate the channel and protocol induced losses.
For more details, read our paper, Packet Loss Characterization in WiFi-based Long Distance Networks which is to appear in INFOCOM, 2007.
2.4 Outdoor Measurement Testbed
We have set up a number of long distance links in Berkeley and the sorrounding areas. We are using high-power 802.11 radios and high gain directional antennas connected to single-board computer based routers.
Network Information: Wireless TestBed
Real time status: TestBed Map2
2.5 Electronically Steerable Antennas
2.6 Economic Analysis of Wireless Technologies
Providing network connectivity to rural regions in the de- veloping world is an economically challenging problem especially given the low income levels and low population densities in such regions. Many existing connectivity technologies incur a high deployment cost that lim- its their affordability. The case for economically viable networks in rural developing regions is presented in the paper, Economic Analysis of Networking Technologies for Rural Developing Regions which leverages several emerging wireless technologies.
We have also evaluated CDMA450 as a potential solution for rural data and voice connectivity. CDMA450 is a promising technology, competitive in both capacity-centric urban environments and in coverage-centric rural environments.
There are advantages in deploying CDMA450 using a fixed wireless (WLL) model and we have explored the the use of directional antennas and receive antenna diversity in a real-world commercial CDMA450 deployment in Romania.
This is presented in the paper, Potential of CDMA450 for Rural Network Connectivity in the Special Issue on New Directions In Networking Technologies In Emerging Economies of IEEE Communications Magazine, 2007.
2.7 Related Projects
2.7.1 Low cost Power
Local electricity generation is one of the key challenges of technology for developing regions. To address the issue of providing power to the wireless routers, we focus on solar power. Depending on the situation, solar power can be the only source of power or can complement existing power from the grid. The main components of a solar power setup (solar panel, battery and charge controller) are readily available, and can be used to supply power for wireless routers. However, current designs don't meet our goals in terms of cost, ease of use, and remote monitoring.
2.7.2 Low cost Towers
A major component of the cost of a long-distance wireless network is the cost of the towers.
3 Resources
A list of resources on wireless issues, the hardware choices and the customized software is available. We are also working on a Installation Guide for installation of a long distance wireless network on the field.
