| Channel estimation and carrier frequency offset (CFO) estimation are two critical issues for Orthogonal Frequency Division Multiplexing (OFDM) systems to accommodate high rate wireless communications. Conventional OFDM systems are based on sufficient cyclic prefix (CP) to facilitate simple receiver implementation, however, at the cost of significant channel utilization loss that may be the over-riding constraint for future high speed services. Thus, there exists increasing interest in OFDM systems with less CP or no CP, for which previous algorithms assuming sufficient CP may not be appropriate. This dissertation is focused on blind (i.e. non-data-aided) parameter estimation for such systems. The redundancy introduced by receive diversity or virtual carriers (VCs) are exploited to fulfill new algorithms that give the OFDM systems the potential to achieve higher channel utilization.; First, a subspace based blind channel estimator is developed for OFDM systems without CP, motivated by the resemblance of the multichannel signal model resulting from over-sampling or use of multiple receive sensors of the received OFDM signal to that in conventional single carrier system. The proposed algorithm distinguishes itself from many previously reported methods by the elimination of the CP.; Next, another source of redundancy—VC, is explored to accomplish a new blind channel estimator. The method is attractive since it is applicable to conventional OFDM systems with CP and those without CP. For the former, the exploitation of VC brings additional performance gain to the previously reported blind estimators. Further, the proposed method can be employed for non-CP systems, thereby potentially achieving higher channel utilization.; The last part of the dissertation addresses CFO estimation for OFDM systems over ISI channels. In this scenario, previous CFO estimators suffer from an irreducible error floor due to model mismatch. We introduce a modified signal model for this case. Benefited from the special structure in the OFDM signals due to the presence of CP and VCs, a novel subspace CFO estimator is proposed. The method is appealing for its low complexity by avoiding the expensive subspace decomposition computation and its higher channel utilization for the OFDM systems when keeping/decreasing the length of CP even in ISI channels. |
