水声跳频通信的信道估计与CPM解调算法研究

发布时间：2018-04-03 15:50

本文选题：水声通信　切入点：自适应跳频　出处：《南京理工大学》2017年硕士论文

【摘要】：随着通信领域的迅猛发展,以常规的跳频技术为基本方法的自适应跳频技术渐渐成为通信研究的热点之一。在水声自适应跳频通信系统中,干扰与噪声的影响以及通信频率的随机跳变,使得信道质量估计变得尤为重要。另外,对于通信系统,调制解调技术是关键部分,影响着通信系统的整体性能。因此,合适的调制解调技术与信道质量估计算法是实现跳频通信的关键所在。本文在总结国内外学者的研究基础上,研究基于水声跳频通信的信道估计算法与CPM信号解调译码算法,以算法的准确度与复杂度为衡量标准,对经典算法进行了相应的优化,并且提出了一些新的改进算法。首先在深海信道中,对经典最大似然估计算法进行推导,并且在不同的数据长度与调制方式下进行理论仿真实验分析;根据CPM信号频谱特性提出一种新的信道估计算法,利用功率谱中信号与噪声的相对关系进行信噪比估计,相比于最大似然估计算法,新算法在中高信噪比的均方根误差低0.1dB左右,且算法复杂度大大减少;对经典卡尔曼滤波估计信噪比算法进行改进,使用支持向量回归模型对卡尔曼滤波结构进行修正,相比于原经典算法,改进算法解决了高信噪比时性能变差的问题,并且估计的均方根误差至少降低了 0.08dB。然后在浅海信道中,对CPM频谱信噪比估计算法优化推导,扩大算法的应用范围,理论仿真实验分析得到该算法在浅海信道下的均方根误差只比深海信道差了0.08dB,达到了 0.18dB左右,并且复杂度很低,适合实时估计;对经典卡尔曼滤波估计信噪比算法进行改进,改进后的算法适用于浅海信道,并且均方根误差在0.15dB左右,但是在高信噪比时性能变差。最后进行基于信道估计的CPM信号低复杂度解调译码算法的改进。在加性高斯白噪声信道下,详细介绍了差分减状态序列检测算法,并且在不同的条件下进行理论仿真实验;提出一种改进的CPM解调译码算法,仿真证明改进算法的误码率明显低于差分减状态算法。
[Abstract]:With the rapid development of communication field, adaptive frequency-hopping technology, which is based on conventional frequency-hopping technology, has gradually become one of the hotspots of communication research.In underwater acoustic adaptive frequency-hopping communication system, the influence of interference and noise and the random jump of communication frequency make the channel quality estimation become more and more important.In addition, the modulation and demodulation technology is a key part of the communication system, which affects the overall performance of the communication system.Therefore, appropriate modulation and demodulation technology and channel quality estimation algorithm are the key to achieve FH communication.Based on the research of domestic and foreign scholars, the channel estimation algorithm based on underwater acoustic frequency hopping communication and the demodulation decoding algorithm of CPM signal are studied in this paper. The classical algorithm is optimized according to the accuracy and complexity of the algorithm.Some new improved algorithms are also proposed.At first, the classical maximum likelihood estimation algorithm is deduced in deep-sea channel, and the theoretical simulation experiment is carried out under different data length and modulation mode. According to the spectrum characteristics of CPM signal, a new channel estimation algorithm is proposed.Compared with the maximum likelihood estimation algorithm, the new algorithm has a lower root mean square error (0.1dB) than the maximum likelihood estimation algorithm, and the complexity of the algorithm is greatly reduced.The classical Kalman filter estimation SNR algorithm is improved and the support vector regression model is used to modify the Kalman filter structure. Compared with the original classical algorithm, the improved algorithm solves the problem of poor performance when the SNR is high.The estimated root mean square error is reduced by 0.08dB at least.And the complexity is very low, suitable for real-time estimation; the classical Kalman filter estimation SNR algorithm is improved, the improved algorithm is suitable for shallow water channel, and the root mean square error is about 0.15dB, but the performance becomes worse when the SNR is high.Finally, the low complexity demodulation and decoding algorithm of CPM signal based on channel estimation is improved.In the additive Gao Si white noise channel, the differential subtraction state sequence detection algorithm is introduced in detail, and the theoretical simulation experiment is carried out under different conditions, and an improved CPM demodulation decoding algorithm is proposed.The simulation results show that the BER of the improved algorithm is obviously lower than that of the differential subtraction algorithm.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.3

【参考文献】