Relay Channel with Causal Channel State Information
Dajin Wang, Liang-Yan Gui
Abstract
Dajin Wang, Liang-Yan Gui
Abstract
In this paper, state dependent relay channel (SD- RC) with causal channel state information (CSI) is considered. Two different cases are investigated in which causal CSI is available: 1) only at the relay node, 2) at all the nodes. The second situation is specialized to cases where causal CSI is known: at both source and relay; at both destination and relay. We established the lower bounds for these situations. Our bound contains all the bounds in previous work, and it is strictly better under some cases. In our scheme, the CSI at relay is compressed by Wyner-Ziv compression and transmitted with the message information by exploiting Shannon strategy. The transmission of CSI has twofold effect. On the one hand, it may reduce the message rate to be relayed. On the other hand, the destination can use the received CSI to help decode message from both the source and relay. Note that if the channel between the relay and destination is good enough, the relay can transmit all the message information it got and use the extra capacity to transmit compressed CSI. In this situation the transmission of CSI doesn't reduce the message rate to be relayed, and our scheme is better than compress- forward (CF) relaying with Shannon strategy.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper, state dependent relay channel (SD- RC) with causal channel state information (CSI) is considered. Two different cases are investigated in which causal CSI is available: 1) only at the relay node, 2) at all the nodes. The second situation is specialized to cases where causal CSI is known: at both source and relay; at both destination and relay. We established the lower bounds for these situations. Our bound contains all the bounds in previous work, and it is strictly better under some cases. In our scheme, the CSI at relay is compressed by Wyner-Ziv compression and transmitted with the message information by exploiting Shannon strategy. The transmission of CSI has twofold effect. On the one hand, it may reduce the message rate to be relayed. On the other hand, the destination can use the received CSI to help decode message from both the source and relay. Note that if the channel between the relay and destination is good enough, the relay can transmit all the message information it got and use the extra capacity to transmit compressed CSI. In this situation the transmission of CSI doesn't reduce the message rate to be relayed, and our scheme is better than compress- forward (CF) relaying with Shannon strategy.
Key concepts: Relay, Relay channel, Channel state information, Computer science, Channel (broadcasting), Transmission (telecommunications), Computer network, Node (physics)