Data-assisted channel estimation and performance analysis for two-way amplify-and-forward relay networks with M-ary frequency shift keying modulation

In this study, a new method is proposed for channel estimation and symbol detection in amplify-and-forward two-way relay networks with frequency shift keying signalling, over block fading as well as time-varying Rayleigh flat fading channels. It relies solely on information symbols to estimate reciprocal and non-reciprocal channels without pilot symbols assistance. To achieve this goal, users apply known periodic phase shifts to their transmit symbols and adopt maximum likelihood or linear minimum mean square error criteria to discriminate and estimate the two composite channels. Optimal and suboptimal phase shifts are designed to minimise the mean square error (MSE). Minimum distance receiver is then used for symbol detection. An approximate upper bound for the resulting bit error rate (BER) is derived and the effect of channel variation rate on the MSE and BER is studied. Finally, simulation results are provided to verify the effectiveness of the suggested method.