认知无线电系统中稀疏信道估计的最优导频序列研究

发布时间：2018-08-25 14:04

【摘要】：正交频分复用是一种无线环境下的多载波调制技术,它具有抗多径衰落、抗干扰能力强和频谱利用率高等特点,是新一代移动通信系统的核心技术之一。认知无线电技术可以有效的解决频谱资源闲置和浪费的问题,OFDM自身的优点使其成为认知无线电技术的最佳调制方式,不连续的正交频分复用技术NC-OFDM应运而生,随之产生的诸多关键问题成为研究热点。其中信道估计直接影响系统的通信品质。基于导频的信道估计由于其估计误差小、收敛速度快、高传输速率等优点成为主流的信道估计技术,基于导频的信道估计受到两个方面的制约,一是信道估计器的设计；二是导频序列的设计,因此NC-OFDM系统导频序列的设计研究具有十分重要的意义。本文详细介绍了OFDM系统基本原理以及无线信道的传输特性,系统分析了最优导频序列设计的常用设计准则和算法,针对稀疏信道环境,提出了NC-OFDM系统导频序列位置和功率设计方法。针对传统NC-OFDM系统信道估计导频序列设计方法计算复杂度高、效率低下、数据利用率低的问题,提出了一种NC-OFDM稀疏信道环境下导频序列位置设计方法,该方法以最小测量矩阵互相关为目标函数,通过求解一个带约束的最优化问题并简化搜索方法得到最优导频位置分布,该方法不仅在信道估计MSE上优于传统等间隔以及随机搜索的导频,在系统BER性能以及计算复杂度方面均有明显改善。传统OFDM系统的导频功率分配方法通常考虑等功率限制或总发射功率限制因素,由于NC-OFDM的特殊性,该方法不能直接用于NC-OFDM系统导频功率分配,为此,提出了一种NC-OFDM系统导频子载波功率分配算法,该方法利用最小测量矩阵互相关准则,并满足以下几种约束条件：干扰温度约束、总功率约束、最大功率约束、子信道功率约束,通过求解多约束条件最优化问题得到最优的导频子载波功率分配。保证了信道的估计精度,提高了数据传输利用率,加强了系统的传输有效性和可靠性。与其他导频序列设计算法相比,能够有效降低系统误码率(BER, Bit Error Rate)和信道估计(MSE, Mean Squared Error),从而在整体上提高了系统性能。
[Abstract]:Orthogonal Frequency Division Multiplexing (OFDM) is a kind of multi-carrier modulation technology in wireless environment. It has the characteristics of anti-multipath fading, strong anti-interference ability and high spectrum efficiency. It is one of the core technologies of the new generation mobile communication system. Cognitive radio technology can effectively solve the problem of idle and wasteful spectrum resources. The advantages of OFDM make it the best modulation mode of cognitive radio technology. The discontinuous orthogonal frequency division multiplexing (NC-OFDM) technology emerges as the times require. Many key problems have become the focus of research. The channel estimation directly affects the communication quality of the system. Channel estimation based on pilot has become the mainstream channel estimation technology because of its small error, fast convergence and high transmission rate. The channel estimation based on pilot is restricted by two aspects, one is the design of channel estimator; The second is the design of pilot sequence, so the design of pilot sequence in NC-OFDM system is of great significance. In this paper, the basic principle of OFDM system and the transmission characteristics of wireless channel are introduced in detail. The common design criteria and algorithms of optimal pilot sequence design are analyzed systematically. The position and power design method of pilot sequence in NC-OFDM system is presented. Aiming at the problems of high computational complexity, low efficiency and low data utilization rate in traditional channel estimation pilot sequence design method for NC-OFDM systems, a pilot sequence position design method in NC-OFDM sparse channel environment is proposed. In this method, the minimum measurement matrix cross-correlation is taken as the objective function, and the optimal pilot position distribution is obtained by solving an optimization problem with constraints and simplifying the search method. This method not only outperforms the traditional equal-interval and random search pilots in channel estimation MSE, but also improves the BER performance and computational complexity of the system. Traditional pilot power allocation methods in OFDM systems usually consider equal power limit or total transmit power limiting factors. Because of the particularity of NC-OFDM, this method can not be directly applied to pilot power allocation in NC-OFDM systems. In this paper, a pilot subcarrier power allocation algorithm for NC-OFDM systems is proposed. The method uses the minimum measurement matrix cross-correlation criterion and satisfies the following constraints: interference temperature constraint, total power constraint, maximum power constraint, subchannel power constraint. The optimal pilot subcarrier power allocation is obtained by solving the multi-constrained optimization problem. The channel estimation accuracy is guaranteed, the data transmission efficiency is improved, and the transmission efficiency and reliability of the system are enhanced. Compared with other pilot sequence design algorithms, BER (BER, Bit Error Rate) and channel estimation (MSE, Mean Squared Error), can be effectively reduced, thus improving the overall performance of the system.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN925

【相似文献】