Research On Key Technologies Of Iterative Receiver At Low Signal-to-noise Ratios

Recently, high data rate transmission technology in DRSS becomes an active research field in wireless communication, and is one of the key problems of our country’s spatial information systems. With the development of high rate data transmission technology, many encoding and decoding technologies have been used, which ofthen work at low signal-to-noise ratios, making a great challenge for traditional synchronization algorithms. Fore main subjects are discussed in this thesis: code-aided symbol timing, carrier synchronization and frame synchronization thchnique for LDPC-coded system at low signal-to-noise ratios(SNR), together with the FPGA implementation of traditional synchronization algorithms for high data rate transmission systems.In the first section, the thesis investigates the carrier recovering techniques for broad-band quadrature modulation signals at low SNR.Firstly the impact of the timing offset on system performance, the performance of the traditional symbol timing algorithms at low SNR, and the necessity of the code-aided timing algorithms are anylized. Secondly, since the Newton algorithm to solve the maximum value of the EM algorithm is of high complexity, a new approach based on the steepest descent method is proposed in this thesis. The simulation results show that the algorithm complexity is reduced, but the performance is still able to meet the requirements. Finally, to solve the problem of biased estimation suffered by both the traditional NDA symbol timing methods and the code-aided symbol timing algorithms, we propose a method which combines the M-value searcher and the code-aided fine estimator. The simulation results show that the algorithm performance is excellent with moderate complexity.In the second section, the thesis investigates the carrier recovering techniques for LDPC coded systems at low SNR.Firstly the impact of the carrier frequency offset and phase offset on system performance are analyzed, providing a designing criterion for carrier synchronization of broadband quadrature modulation signals. Secondly, we summarize the traditional feedforward carrier synchronization algorithms, and propose a univesal form for an code-aided frequency offset eatimator based on autocorrelation function of the samples for the tradeoff between complexity and performance. Finally, since the estimation range of the code-aided frequency offset estimation algorithms based on the EM framework is very small, we propose a method which utilizes a coarse synchronization based on the M-value algorithm, making the residual frequency offset reduced to the estimating range of the EM algorithm. Morever, we also use a simple interpolator and threshold to furtherly reduce the searching times of the M-value method and greatly reduces the complexity of the algorithm.In the third section, we analyze the drawbacks of a traditional frame synchronizer and existent code-aided frame synchronization algorithms. A novel code-aided frame synchronizer is proposed, which can utilize both the code information and the frame synchronization marker. Simulation results show that this improved methd can approach ideal synchronization with low complexity and high spectral efficiency.In the last section, we introduce how to implement the traditional synchronization algorithms on FPGA for high data rate transmission systems. A joint code-aid structure including symbol timing, carrier synchronization and frame synchronization is finally presented.