Symbol Timing Estimation for Physical-Layer Network Coding

Recently, physical-layer network coding (PNC) has attracted much attention due to its ability to double the throughput in relay-aided communications. However, most research in PNC assumes that the symbol timings of the end nodes are ideally aligned. In fact, they have to be recovered accurately at the relay. In this letter, we present a novel symbol timing error estimator for PNC. It is shown that accurate error estimates can be obtained by the interpolation algorithm based on a small-size Discrete Fourier Transformation (DFT). The Mean Square Error (MSE) performance is one order of magnitude better than that of the conventional maximum likelihood based optimum sample (OS) algorithm for signal-to-noise ratio (SNR) greater than 10 dB.