Cooperative spectrum sensing with Slepian-Wolf coded cooperations

A new Slepian-Wolf coded cooperation scheme is proposed for a cognitive radio network with two secondary users (SUs) performing cooperative spectrum sensing through a fusion center (FC). The SUs sense the spectrum by measuring the energy statistics of the received signals. The measured energy statistics are quantized with a Lloyd-Max quantizer at the SUs and forwarded to the FC, which then performs soft combining over the quantized information. This is different from most previous works that forward hard decisions based on local sensing results at the SUs. Due to the wireless nature of the channel, signals transmitted by one SU to the FC will also be observed by the other SU, which can cooperate with the transmitting SU by relaying the observed signals to the FC. In recognition of the strong correlation between the signals observed at the FC and the relay SU, a new asymmetric Slepian-Wolf code is employed at the relay SU to reduce the amount of cooperation information, thus to improve the cooperation efficiency. In addition, we propose to unequally allocate energy among the coded bits to compensate the energy loss due to the redundancy introduced by the coded cooperation, and this yields better performance compared to conventional equal energy coding schemes. Simulation results demonstrate that the proposed cooperative spectrum sensing scheme operating in practical fading channels can achieve a performance that is almost identical to the ideal case with soft combining performed over unquantized energy statistics transmitted through distortion-free channels.