Efficient pilot-aided frequency synchronization and channel estimation in wireless OFDM transmission

Abstract

A critical issue in wireless orthogonal frequency division multiplexing (OFDM) transmission systems is their ability to synchronize, and mitigate the fading propagation channel effects. Since pragmatic wireless links entail channel-induced interference, as well as frequency offsets, it is necessary to account for these effects when designing wireless communication systems. To remove these effects, pilot-aided scheme makes sense since it simplifies the challenging task of receiver design for unknown channels. Therefore, we design pilot symbol assisted modulation for frequency synchronization and channel estimation. Because we separate carrier frequency offset (CFO) and channel estimation from symbol detection, the novel pilot patterns, especially targeting multi-input multi-output OFDM (MIMO-OFDM) system, lead to low-complexity CFO and channel estimators. In addition, we develop the pilot pattern, which is not only suitable for single- and multi-carrier systems, but also for synchronization and channel estimation in wireless sensor networks. Finally, assuming the perfect synchronization, we introduce a new iterative channel estimator combined with low-density parity-check (LDPC) decoding for OFDM systems, where pilot symbols are encoded by an LDPC systematic encoder. Since exploiting the channel coding scheme can result in significantly enhanced channel estimates, and can lower the required amount of pilots for a given performance level, it is well motivated to perform iterative channel estimation and decoding. In this dissertation, we will show that pilot symbols inserted along with information data improve the performance of an LDPC decoder in frequency selective channels. The resulting iterative receiver also can give a good channel estimate with the small number of pilots. The performance of our algorithms is investigated analytically, and then compared with an existing approach by simulations.