OFDM系统中基于压缩感知的双选择性信道估计方法研究

发布时间：2019-01-05 21:28

【摘要】：无线通信系统的性能在很大程度上受到无线信道的约束。无线信号传播会遭受各种复杂地物影响，接收端信号在幅度、相位和频率方面会发生不同程度的失真。正交频分复用(Orthogonal Frequency Division Multiplex, OFDM)技术虽能有效克服频率选择性衰落，但是对频率偏移非常敏感，因此，信道估计显得尤为重要。传统关于OFDM系统的信道估计方法通常是假设信道具有丰富多径，从而利用大量导频信号获取准确的信道状态信息，这极大地降低了系统资源利用率。为解决这一问题，本文基于压缩感知理论，对OFDM系统信道估计方法展开研究。依据压缩感知理论的三大关键技术，本文从信道系数的稀疏表示、导频序列设计和重构算法选取三个方面展开研究，针对现有稀疏信道估计算法存在的一些问题，给出相应解决方法。针对现有稀疏信道估计算法只考虑信道的频率选择性衰落，本文同时考虑信道的时间选择性衰落。在实际系统中，信道时延和多普勒频移通常不能被整数倍采样，由此造成的能量泄漏问题将极大减少等效信道稀疏性，本文针对这一问题通过提高信道离散精度，用过完备字典代替现有的傅里叶正交基提高等效信道系数在字典域的稀疏性，从而减少重构算法所需的观测值，即导频数量。仿真结果表明，无论是单天线还是多天线信道，过完备字典方法虽然增加一定的计算复杂度，但有效提高了信道估计精度，同时减少了对导频数量的需求。针对现有多天线稀疏信道估计算法只是简单将多天线信道划分为多个单对单信道的处理方式，本文分析多天线信道的联合稀疏特性，利用分布式压缩感知理论方法进行联合信道估计。仿真结果表明，联合稀疏信道估计方法利用信道间的互相关和自相关性，使得对联合稀疏支撑集估计更准确，，其性能要优于基于单对单稀疏的信道估计方法。
[Abstract]:The performance of wireless communication systems is largely constrained by wireless channels. Wireless signal propagation will be affected by various complex ground objects, and the amplitude, phase and frequency of the signal at the receiving end will be distorted to varying degrees. Orthogonal Frequency Division Multiplexing (Orthogonal Frequency Division Multiplex, OFDM) can overcome frequency selective fading effectively, but it is very sensitive to frequency offset, so channel estimation is very important. The traditional channel estimation methods for OFDM systems usually assume that the channel has abundant multipath, so a large number of pilot signals are used to obtain accurate channel state information, which greatly reduces the system resource utilization. In order to solve this problem, the channel estimation method of OFDM system is studied based on compressed sensing theory. According to the three key technologies of compressed sensing theory, this paper studies the sparse representation of channel coefficients, pilot sequence design and reconstruction algorithm selection, aiming at some problems of existing sparse channel estimation algorithms. The corresponding solutions are given. Since the existing sparse channel estimation algorithms only consider the frequency selective fading of the channel, the time selective fading of the channel is also considered in this paper. In practical systems, the channel delay and Doppler frequency shift can not be sampled by integer multiple sampling, and the energy leakage problem will greatly reduce the equivalent channel sparsity. In this paper, we improve the channel dispersion accuracy by improving the channel dispersion accuracy. The overcomplete dictionary is used to replace the existing Fourier orthogonal basis to improve the sparsity of the equivalent channel coefficients in the dictionary domain, thus reducing the observed values required for the reconstruction algorithm, namely, the number of pilots. Simulation results show that the over-complete dictionary method increases the computational complexity of both single antenna and multi-antenna channels, but effectively improves the channel estimation accuracy and reduces the need for the number of pilots. In this paper, the joint sparse characteristic of multi-antenna channel is analyzed in view of the fact that the existing multi-antenna sparse channel estimation algorithm is only a simple way to divide the multi-antenna channel into several single-pair single-channel. Joint channel estimation is based on distributed compressed sensing theory. Simulation results show that the joint sparse channel estimation method makes use of the cross-correlation and self-correlation between the channels to estimate the joint sparse support set more accurately and the performance of the joint sparse channel estimation method is better than that based on the single-pair sparse channel estimation method.
【学位授予单位】：重庆邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.53

【参考文献】