Capacity and Power Allocation for Transmitter and Receiver Cooperation in Fading Channels

Capacity gain from transmitter and receiver cooperation under channel fading are compared in a relay network where the cooperating nodes are close together. We assume a Rayleigh flat-fading environment in the high signal-to-noise ratio (SNR) regime where the transmitters only have channel distribution information (CDI) but not channel state information (CSI). When all nodes have equal average transmit power, we show that the decode-and-forward transmitter cooperation strategy is capacity-achieving and is superior to receiver cooperation. However, the compress-and-forward receiver cooperation strategy is shown to outperform transmitter cooperation when power is optimally allocated among the nodes. Furthermore, we show that cooperative systems provide resilience to channel fading. However, in a fading channel, capacity becomes more sensitive to power allocation, and the cooperating nodes need to be closer together. With respect to limits on cooperation, it is shown that in a large cluster of M cooperating nodes, transmitter cooperation without CSI at the transmitter (CSIT), or receiver cooperation under equal power allocation, provides no capacity gain in a static channel, and at most a constant capacity gain that fails to grow with M in a fading channel.