Cognitive radio with causal primary message and partial channel state information at the transmitter

This paper investigates the achievable rate of cognitive radio network when one primary user (PU) and one secondary user (SU) are present, where there is partial channel state information at the transmitter (CSIT). On one hand, SU learns primary message causally in the first phase and then relays it with decode-and-forward strategy in the second phase. On the other hand, SU employs linear assignment Gel´fand-Pinsker coding (LA-GPC) on its own message to cancel the interference from the primary transmitter. We formulate an optimization problem to maximize SU´s achievable rate while keeping PU´s rate unaffected. An algorithm is proposed to jointly determine the fraction of channel uses in the first phase, the percentage of SU´s transmit power used for relaying, and the filter factors in LA-GPC method. The performance of the proposed algorithm is evaluated through numerical simulations.