Wireless Networks

WPAN

One of the key areas of research in this group is the concept of very short-hop wireless communications. These are known as wireless personal area networks (PAN) or body area networks (BAN). These networks consist of a single person and his immediate environs (i.e. within hands reach) in which there are many wireless devices communicating via a netwok. These devices would generally be small (i.e. wearable), low power, and portable. Such a network would be self configurable as the user moves from area to area.

While Bluetooth is the current network standard in this area, its functionality is limited and the 802.15 standards group has been expanded to look for better solutions. One solution that we're currently looking at is replacing the direct sequence physical layer with the chirp spread spectrum technology developed at TRLabs. In this system, the terminal power consumption is reduced by using analog signal processing elements (in this case SAW filters) and by reducing the component count by having the terminal reflecting a modified version of the transmitter's signal to return data rather than generating the signal internally.

WLAN

Interoperability between networks has recently become the subject of a new standardization group (IEEE 802.21). This will have particular interests to our telco sponsors as we will eventually have users handing off from their private LANs to public PCS systems and back as the move from place to place. Most of the work to date in this topic has concentrated on higher layer of the stack (networking, transport, and session), which is important, but it will also have huge implications on the lower layers. In the short term, interoperability will probably push from the top layers down resulting in bulky and expensive multi-mode terminals or software radios. For the longer term, TRLabs has been working on a concept (called "simple terminal") where the terminal is designed from the physical layer up to be interoperable with different networks, resulting in a much simpler hardware implementation (ideally single chip).

The goal of the physical layer research is to find the right mix of technologies that will best implement the diverse requirements of the network including: high data rate, low bit rate, good coverage, good spectral efficiency, interference robustness, low terminal power consumption, and interoperability between networks. Current standards are quite strict about the physical layer implementation, so it can be quite difficult to do research to improve systems within the standards.

Current work includes characterizing and modeling the frame loss process in 802.11 in order to: develop a frame/packet loss measurement system, collect frame-level (or packet-level) traces, perform statistical analysis of the collected data. The output should be an analytic discrete-time channel model for link layer and transport layers. With the understanding of the processes that go on within a wireless network, we should be able to develop improvements such as improved reliability and quality of service for home technologies, digital media, and e-Health. All of these applications areas aren't well served under the current standards.

WMAN

The group needs to become more familiar with the emerging Wi-Max standard (802.16) to understand what its strength and limitations are and look for opportunities to improve systems for our sponsors. One of the possible areas of research is the implementation of some of the adaptive antenna technology developed within TRLabs to the Wi-Max standard. As Wi-Max systems become more crowded, the requirements for higher capacity and reduced interference will demand some kind of a smart antenna solution. Another area of expertise that can be applied to the wireless MAN problem is our research in optical-radio hybrid systems. In this solution, the wireless signal is carried close to the potential users using a optical fibre, then the last hop using wireless. This would have increased capacity and coverage over a wireless only solution.