Channel shortening techniques for differential encoded OFDM

Abstract

In OFDM systems, the Inter-Symbol Interference (ISI) and the Inter-Block Interference (IBI) are mitigated by using Null-Tones (NTs) and the Guard Interval (GI) redundancy of length higher than the channel size. To preserve a high bit rate, channel shortening is required to reduce the GI length. We propose here a new adaptive method for the blind estimation of the Time domain EQualizer (TEQ) for time varying communication channels. Our contribution is three-fold. First, we exploit the knowledge of the first emitted symbol in a differential encoded OFDM system to derive the initial estimate of the TEQ. Then, we update the TEQ coefficients according to the channel variation by optimizing a criterion exploiting both the GI and NT redundancies and using a fast Generalized Eigen Value (GEV) tracking algorithm. Finally, the optimization of the previous criterion is achieved in such a way we control the Target Impulse Response (TIR) quality and we improve the overall system performance. Simulation results are provided to illustrate the performance of our method and assess our theoretical derivations.