SC-FDE系统中的均衡技术研究

发布时间：2018-05-24 16:23

本文选题：单载波频域均衡 + 码间干扰　；参考：《重庆大学》2014年硕士论文

【摘要】：单载波频域均衡（SC-FDE）技术创造性地将单载波技术和频域均衡技术结合起来，不仅可以有效地克服由多径效应带来的码间干扰（ISI），实现可靠的高速率通信，而且能够显著减小单载波时域均衡（SC-TDE）技术庞大的运算量和正交频分复用技术（OFDM）过高的峰均比。单载波频域均衡以其技术上的优势，逐渐成为近年来的研究热点，，同时也被纳入到IEEE802.16（WiMAX）标准中作为下一代宽带无线通信系统物理层的重要解决方案。论文重点研究SC-FDE系统中的频域均衡技术，并围绕现有均衡算法中存在问题展开，探讨如何在信道估计精度与计算复杂度两个方面取得最佳平衡，从而提高SC-FDE系统性能。论文的主要工作有： ①论文从介绍SC-FDE基本原理入手，归纳现有频域均衡的关键技术，包括均衡算法，信道估计算法和插值算法，并对其比较分析，总结各种算法在估计精度和运算复杂度上的优缺点，为论文后续的研究工作提供切入点； ②针对SC-FDE系统传统功率恒定的帧结构进行功率调整，提出在时域对帧结构中UW块和FFT块功率进行再分配的概念，通过对符号能量分配因子的仿真设计，解决功率的最优化分配问题，仿真结果表明，在数据帧总能量不变的情况下，功率优化后系统与优化前相比提高了误码性能； ③针对传统LS算法信道估计精度不高和IFFT/FFT滤波算法实现难度太大的问题，论文根据后者消除估计信道中时域噪声的思想，提出一种改进的自适应滤波LS算法。该方法在传统LS算法变换域插值的时域增加自适应滤波器，并根据信道能量和噪声方差设计出能够跟踪信道变化的滤波器系数，从而避免虚检和漏检现象，通过在多径信道中仿真结果的分析，自适应滤波LS算法在信噪比适中的区间能够得到比上述两种算法更高的信道估计精度； ④为了解决MMSE均衡中对信道噪声方差的需要，论文结合SC-FDE系统帧结构的特点，提出一种基于后向反馈结构的信噪比估计MMSE均衡算法，该方法在经过了ZF均衡的FFT块中，采用ML算法对其冗余的UW序列进行噪声方差的估计，并将估计值作为参数对FFT块再次进行MMSE均衡。仿真结果表明，该算法在不增加额外开销的情况下，能够达到近似传统MMSE算法的性能。
[Abstract]:Single carrier frequency domain equalization (SC-FDE) technology combines single-carrier technology with frequency-domain equalization technology creatively. It can not only effectively overcome the inter-symbol interference (ISI) caused by multipath effect, but also realize reliable high-speed communication. Moreover, it can significantly reduce the huge computational complexity of single carrier time domain equalization (SC-TDE) technology and the high peak to average ratio (PAPR) of orthogonal frequency division multiplexing (OFDM) technology. Single carrier frequency domain equalization has become a research hotspot in recent years due to its technical advantages, and has also been incorporated into the IEEE 802.16 WiMAX standard as an important solution to the physical layer of the next generation broadband wireless communication system. This paper focuses on the frequency-domain equalization technology in SC-FDE systems and discusses how to achieve the optimal balance between channel estimation accuracy and computational complexity around the existing equalization algorithms so as to improve the performance of SC-FDE systems. The main work of the thesis is as follows: 1. Starting with the introduction of the basic principle of SC-FDE, this paper summarizes the key technologies of frequency domain equalization, including equalization algorithm, channel estimation algorithm and interpolation algorithm, and compares and analyzes them. The advantages and disadvantages of various algorithms in estimation accuracy and computational complexity are summarized to provide a starting point for further research work in this paper. (2) aiming at the traditional frame structure of SC-FDE system with constant power, the concept of redistributing UW block and FFT block power in time domain is proposed. The simulation design of symbol energy allocation factor is carried out. In order to solve the problem of optimal power allocation, the simulation results show that the error performance of the optimized system is improved compared with that before the optimization when the total energy of the data frame is invariant. 3 in view of the problem that the channel estimation accuracy of the traditional LS algorithm is not high and the IFFT/FFT filtering algorithm is too difficult to realize, an improved adaptive filtering LS algorithm is proposed according to the idea of eliminating the time-domain noise in the channel estimation by the latter. This method adds adaptive filter in time domain interpolation of traditional LS algorithm transform domain, and designs filter coefficients which can track channel change according to channel energy and noise variance, so as to avoid false detection and miss detection. Through the analysis of the simulation results in multipath channel, the adaptive filtering LS algorithm can obtain higher channel estimation accuracy than the above two algorithms in the range of moderate SNR. 4 in order to solve the need of channel noise variance in MMSE equalization, this paper presents a MMSE equalization algorithm based on backward feedback structure, which is based on the characteristics of frame structure of SC-FDE system. This method is used in the FFT block of ZF equalization. ML algorithm is used to estimate the noise variance of its redundant UW sequence, and the estimated value is used as a parameter to MMSE equalize the FFT block again. Simulation results show that the performance of the proposed algorithm is similar to that of the traditional MMSE algorithm without additional overhead.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN92

【参考文献】