Modeling and Analysis of Synchronization Schemes for the TDMA Based Satellite Communication System.

摘要

As a multiple access technique, time division multiple access (TDMA) is becoming more widespread in satellite communications. Although single-carrier (SC) modulation techniques has been mostly employed in the TDMA satellite communication system, multi-carrier modulation techniques, like orthogonal frequency division multiplexing (OFDM), are also promising candidates for the future high rate satellite communication due to its high bandwidth efficiency and robustness to fading. As is known, accurate synchronization is a key requirement for efficient data transmission in TDMA systems. Because of the channel impairment or propagation delay, the receiver may experience the carrier frequency offset (CFO) or symbol timing error (STE) that may result in difficulty of signal detection and estimation. Comparing with SC systems, OFDM systems are more sensitive to synchronization errors, because these errors will lead to inter-symbol interference (ISI) and inter-carrier interference (ICI). Although many synchronization algorithms have been disclosed in literature as the effective recovery methods to reduce detection error in the presence of CFO as well as STE, tradeoffs still have to be made between the estimation accuracy and estimation range (or implementation complexity). This motivates further research to improve the performances of synchronization methods for the TDMA based satellite communication system.;The main objectives of this research are modeling and mathematical analyses of the synchronization methods under various channel impairment scenarios towards the practical applications in TDMA satellite communication systems. To achieve these objectives, this research firstly investigates the performance degradation due to CFO for both SC and OFDM systems. The performance of SC system impaired by CFO is mathematically analyzed and computer simulation is performed under additive white Gaussian noise (AWGN). For OFDM system in the presence of CFO, the performance of binary phase shift keying/quadrature phase shift keying (BPSK/QPSK)-OFDM impaired by CFO is evaluated and simulated under AWGN, Rayleigh flat fading, and frequency selective fading. Because of the lack of analyses in higher order modulation or modulation with non-rectangular decision region, this research derives the closed form of symbol error probability expressions for 8-ary phase shift keying (8PSK)-OFDM and 16-ary quadrature amplitude modulation (16-QAM)-OFDM impaired by CFO under AWGN. All of the corresponding simulation results validate the correctness of the theoretical analyses and symbol error probability derivations.;In order to compensate the effect of CFO to SC systems, numerous methods have been proposed as the effective architecture for CFR. Thus, comparison between various existing synchronization algorithms under certain condition has been performed based on modeling and mathematical analysis. By the results, the major impact factors of different CFR algorithms have been identified. Then, a novel CFR algorithm, which is derived based on least square estimation method, is proposed and evaluated under AWGN. Finally, the proposed algorithm is simulated and compared with other exiting algorithms with respect to estimation range, estimation accuracy and computational complexity. The simulation results show that the proposed algorithm has the benefit of reduced computational complexity and high estimation accuracy.;In the case of OFDM systems, two novel data-aided (DA) carrier frequency and symbol timing recovery algorithms are proposed. These maximum likelihood (ML) based algorithms jointly estimate the CFO and STE for the OFDM systems. The performance of the proposed algorithms is evaluated and compared with existing estimation algorithms in terms of estimation range and accuracy under AWGN, Rayleigh flat fading and frequency-selective multipath fading, respectively. Simulation results confirm the effectiveness of the proposed algorithms under various levels of signal to noise ratios (SNRs) and the robustness of different lengths of cyclic prefix (CP) of the preamble.