Distributed convolutional-based coding for system non-idealities

This paper proposes a coding scheme for cooperative networks in the non-ideal case with system imperfections where the source and relays share their antennas to create a virtual transmit array to transmit towards their destination. We focus on the problem of coding for the relay channels. While the relays may use several forwarding strategies, including amplify-and-forward (AF) and decode-and-forward (DF), we focus on coded DF relaying. In particular, we consider the case when all the nodes estimate the channel state information in the transmission process. We derive upper bounded expressions for the bit error rate (BER) assuming M-ary phase shift keying (M-PSK) transmission. It is shown that the performance is degraded due to the presence of channel estimation error. However, the observations made in ideal scenarios still hold for the non-ideal case. Also our analytical results have shown that a performance close to perfect channel knowledge can be obtained when the number of pilot symbols k_p increases. Also, with k_p pilot symbols, the performance was shown to approach the perfect channel knowledge case at high pilot to noise ratio (PNR) (E_b/N₀). Finally, at low PNR, the proposed scheme cannot provide significant performance improvements.