基于自适应压缩感知的信道估计与窄带干扰检测算法研究

发布时间：2019-02-15 17:52

【摘要】：正交频分复用(OFDM)因为其具有较高的频带利用率、较强的抵抗频率选择性衰落等特点而被广泛应用于目前移动通信领域,成为了第四代通信系统的核心技术之一。然而目前OFDM技术仍旧面临许多技术问题。众所周知,依据奈奎斯特采样定律,对OFDM系统中数字信号问题的处理需要极高的采样率,然而目前的前端模数转换器(ADC)较难满足其要求。因此我们迫切需要寻求一种新的方法用于解决OFDM系统中数字化信号的处理问题。近几年来压缩感知理论为该问题提供了一种有效的解决方案,它可以同时实现信号的采样和压缩,然后通过特定的重构算法从少量的信息中准确地恢复出原始信号,可大大降低了通信系统系统的采样率和硬件成本。首先本文为了改善OFDM通信系统的性能,在深入分析了 OFDM通信系统的信道特征基础上,结合OFDM系统信道估计技术的特点,将经过加窗优化的自适应匹配追踪算法应用到通信系统的信道估计中,抑制了由于信道截断造成的冲击响应泄露现象,有效消除了冲击响应泄露所产生的噪声,保证了信道的稀疏性,并在此基础上进一步构造了基于窗函数的观测矩阵,然后将自适应匹配追踪算法用于信道的恢复中。本文算法可在未知信道稀疏度的条件下,通过自适应步长合理地调整候选集中原子的个数,进而较为精确的估计出信道的冲击响应。仿真结果表明,与目前信道估计算法相比,本文所提算法能够进一步改善信道估计的性能,有较好的实际应用以及推广价值。其次,本文针对目前OFDM系统中窄带干扰检测所存在的问题,在对压缩感知理论及其重构算法深入研究的基础上,利用基于自适应压缩感知理论的自适应匹配追踪算法解决通信系统的窄带干扰检测问题,该方法可在窄带干扰稀疏度未知的情况下,通过选择合适的补偿自动地调整候选集中原子的个数,在低于奈奎斯特采样率下,快速实现单个及多个窄带干扰信号检测。仿真结果表明,与现有的干扰检测算法相比,本文算法能够有效地实现窄带干扰检测且运行速度较快快,有效地改善OFDM通信系统的性能。
[Abstract]:Orthogonal Frequency Division Multiplexing (OFDM) has been widely used in the field of mobile communication due to its high bandwidth efficiency and strong resistance to frequency selective fading. It has become one of the core technologies of the fourth generation communication system. However, OFDM still faces many technical problems. It is well known that according to Nyquist's sampling law, the processing of digital signal in OFDM system requires a very high sampling rate. However, the current front-end analog-to-digital converter (ADC) is difficult to meet its requirements. Therefore, we urgently need to find a new method to solve the problem of digital signal processing in OFDM system. In recent years, the theory of compression perception provides an effective solution to this problem. It can simultaneously sample and compress signals, and then recover the original signal accurately from a small amount of information by a specific reconstruction algorithm. It can greatly reduce the sampling rate and hardware cost of the communication system. Firstly, in order to improve the performance of OFDM communication system, based on the in-depth analysis of the channel characteristics of OFDM communication system, combined with the characteristics of OFDM system channel estimation technology, A windowed adaptive matching tracking algorithm is applied to the channel estimation of communication systems. The leakage of impulse response caused by channel truncation is suppressed and the noise caused by the leakage of impulse response is effectively eliminated. The sparsity of the channel is ensured, and the observation matrix based on the window function is further constructed, and then the adaptive matching tracking algorithm is applied to the channel recovery. Under the condition of unknown channel sparsity, the proposed algorithm can reasonably adjust the number of atoms in the candidate set by adaptive step size, and then estimate the impulse response of the channel accurately. The simulation results show that the proposed algorithm can further improve the performance of channel estimation compared with the current channel estimation algorithm and has good practical application and popularization value. Secondly, aiming at the existing problems of narrowband interference detection in OFDM systems, this paper deeply studies the theory of compression sensing and its reconstruction algorithm. An adaptive matching tracking algorithm based on adaptive compression sensing theory is used to solve the problem of narrowband interference detection in communication systems. This method can be used to detect narrowband interference when the sparse degree of narrowband interference is unknown. By selecting the appropriate compensation to automatically adjust the number of atoms in the candidate set, the detection of single and multiple narrowband interference signals can be realized quickly under the Nyquist sampling rate. The simulation results show that compared with the existing interference detection algorithms, the proposed algorithm can effectively realize narrowband interference detection and run faster, and improve the performance of OFDM communication system effectively.
【学位授予单位】：哈尔滨工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.53

【参考文献】