A novel joint and iterative scheme for synchronization and channel estimation in MC-CDMA power line communications

The paper is dedicated to synchronization and channel estimation for multicarrier code division multiple access (MC-CDMA) communications over outdoor power lines (PLC). A new data-aided algorithm is presented to cope with the horrible PLC channel and consists of a joint and iterative estimation of sampling clock offset, FFT window misalignment, and frequency response of the channel. The estimation process is derived from studying the synchronization errors that affect received training symbols. The basis of the algorithm is the tracking of the phase gradient of such known symbols from frame to frame. Synchronization convergence is shown to be reachable in few iterations. A frequency filtering and a subcarrier selection within the training sequence enable high Gaussian noise levels to be coped with. Simulations have been run on measured channel responses of an outdoor PLC network. The algorithm exhibits very good results since highly accurate frequency and symbol synchronization is obtained in less than five iterations and low mean square error channel estimation is shown after ten iterations.