Modern society is increasingly relying on wireless networks. Spectrum is the most valuable resource in a wireless network. How to share this limited resource among different users is one of the main challenges. When different users share the same spectrum, their signals cause interference to each other. Therefore, the most distinctive feature of wireless networks is the phenomenon of interference. In this research, we place special emphasis on practical considerations to ultimately find new ? realistic ? methods to deal with interference.
A recent development is the idea of interference alignment which has shown that the capacity of wireless networks may be significantly higher than previously believed. Since higher rates invariably come at the cost of lower reliability, the emerging capacity results present only half the picture. This research is motivated by the need to complete this picture by evaluating the benefits of interference alignment schemes on the performance of wireless networks when both rate and reliability are of concern. The research follows three main thrusts. First, we examine the rate-reliability tradeoff of interference alignment schemes from the traditional coding perspective which places emphasis on low decoding complexity, usually at the cost of a restricted notion of optimality. Second, we examine the rate-reliability tradeoff of interference alignment schemes in the Shannon framework which allows strong definitions of optimality, usually at the cost of unbounded delay and complexity. These two thrusts are the stepping stones to the final thrust to reconcile the findings from the two distinct perspectives and use the collective insights to develop new physical layer schemes that can operate at the frontier of the rate-reliability tradeoff.
