User pairing and network performance optimization in cooperative wireless network coding

Abstract

In today's wireless networks, diversity is regarded as an efficient and established means to combat multipath fading. Moreover, user cooperation has emerged lately as an elegant technique to achieve spatial diversity over wireless channels, where the installation of multiple antennas on handheld, battery-powered, mobile terminals is often impractical. Recently, the application of network coding in cooperative wireless networks has gained increasing interest with its potential to further boost the network performance, such as in terms of the achievable throughput. With network coding, the relaying nodes are allowed to linearly combine packets from multiple source nodes, and then forward the combined packets for better resource utilization. -- We propose mutual user pairing in amulti-user infrastructure-based network-coded cooperative wireless network to realize network coding, in the absence of dedicated relay nodes. We propose an optimal user pairing algorithm, and tailor it to maximize the network capacity. Next, we develop heuristic pairing algorithms which approach the optimal performance at a reduced complexity. Performance analysis is conducted in terms of the average capacity per user, average outage probability per user, and user-fairness. -- For energy-constrained network-coded cooperative networks, we subsequently address the problem of transmission power minimization. A joint optimization problem is formulated and solved to find the pairing which maximizes the network capacity, and minimizes the transmission power, such that certain performance constraints in terms of the average capacity per user or average outage probability per user are satisfied.