Compress-and-forward strategy for the relay channel with causal state information

In this paper, we study a state-dependent relay channel where perfect channel state information (CSI) is causally known to transmitters in both symmetric and asymmetric manner. So, three different setups are investigated in which causal CSI is available: 1) at the relay only, 2) at the source only and 3) both at the source and the relay nodes. In each situation, we obtain the lower bound on the capacity (achievable rate) for the general discrete memoryless case. The lower bounds are derived based on Shannon's strategy where CSI is known, and compress-and-forward (CF) strategy at the relay.