非协作FSK和CPM信号解调方法和实现

发布时间：2018-01-17 18:04

  本文关键词：非协作FSK和CPM信号解调方法和实现　出处：《电子科技大学》2014年硕士论文　论文类型：学位论文

  更多相关文章： 连续相位调制 参数盲估计 复合网格解调

【摘要】：在调制相关信息未知的情况下,接收机对接收信号进行解调,称之为非协作信号盲解调。关于非协作FSK信号的盲解调问题已有大量的研究,调制参数估计和解调算法都很成熟。为了避免了FSK振荡器突发式切换产生较大的频谱旁瓣,由FSK演变出连续相位键控(CPFSK)。它是后续CPM信号的先驱。连续相位调制(CPM)显著特点是包络恒定,同时相位是连续变化的,为它带来了高的频谱利用率。CPM信号调制方式是非线性的。同时CPM信号随着设置不同的调制参数,例如调制指数、相关长度,各类CPM信号表现出不同的性质。所以针对这些特点研究相关的解调技术,实现非协作CPM信号盲解调很有意义。本论文将CPM信号盲解调方法分为调制参数估计算法和解调算法,对它们都进行了深入的分析和研究。主要工作内容有：1、根据循环平稳特性,分析了针对载频／调制指数／符号周期估计算法的代价函数,研究了针对CPM信号相位展开的方法,为接下来调制指数调整奠定了基础,仿真验证了该算法的性能,表明该算法估计性能稳定。2、分析推导了多进制CPM信号的二进制表达形式和其与二进制CPM信号的频域关系。研究了基于信号的循环谱,根据它的离散谱线结构进行调制阶数估计。仿真验证该算法不受调制指数和频率成形脉冲限制,具有较高的正确识别率。3、利用CPM信号符号之间有相关性,从而根据信号自相关函数实现相关长度估计。利用先前估计出的调制阶数,仿真验证了适用于任意调制指数的相关长度估计算法性能,表明该算法实现简单,估计精度高。4、在相关解调算法的基础上,重点研究了利用复合网格的思想,进行非相下解调的算法。该算法引入旋转网格图和状态耦合转移,实现了解调中对多普勒频移的有效跟踪。理论分析和仿真表明,该算法可以得到优于差分的非相干解调性能,可以很好的接近最佳解调性能,同时它具有抗初相和频偏跟踪的优点。5、在完成调制参数估计和解调算法分析之后,设计了非协作CPM信号盲解调器结构,对盲解调器进行模块划分,针对DSP实现给出了程序流程和算法代码优化方法。同时分析了复合网格解调算法中关于Viterbi算法和网格之间耦合的算法复杂性,提出了基于多核DSP的复合网格解调方案,利用多个处理器核的并行运算,实现大量数据的并行处理。
[Abstract]:In the case of unknown modulation related information, the receiver demodulates the received signal, which is called non-cooperative signal blind demodulation. There has been a lot of research on the blind demodulation of non-cooperative FSK signal. Modulation parameter estimation and demodulation algorithms are very mature in order to avoid the FSK oscillator burst switching to generate large spectrum sidelobe. The continuous phase keying (FSK) is the precursor of the subsequent CPM signal. The continuous phase modulation (CPM) is characterized by constant envelope and continuous phase variation. It brings high spectral efficiency. CPM signal modulation mode is nonlinear, and CPM signal with different modulation parameters, such as modulation index, correlation length. All kinds of CPM signals show different properties. Therefore, the demodulation technology is studied according to these characteristics. It is very meaningful to realize blind demodulation of non-cooperative CPM signal. In this paper, the blind demodulation method of CPM signal is divided into modulation parameter estimation algorithm and demodulation algorithm. The main work is: 1. According to the characteristics of cyclic stationary, the cost function of carrier frequency / modulation index / symbol period estimation algorithm is analyzed. The method of phase unwrapping for CPM signal is studied, which lays a foundation for the modulation exponent adjustment. The performance of the algorithm is verified by simulation, which shows that the estimation performance of the algorithm is stable. 2. The binary expression of multiary CPM signal and its frequency domain relationship with binary CPM signal are analyzed and deduced. The cyclic spectrum based on the signal is studied. The modulation order is estimated according to its discrete spectral line structure. The simulation results show that the algorithm is not limited by modulation exponent and frequency shaping pulse and has a high recognition rate of 3. 3. The correlation length is estimated according to the autocorrelation function of the signal by using the correlation between the symbols of the CPM signal, and the modulation order is estimated by the previous estimation. The performance of the correlation length estimation algorithm suitable for arbitrary modulation exponents is verified by simulation. It shows that the algorithm is simple in implementation and has high estimation accuracy. 4. The algorithm is based on the correlation demodulation algorithm. The algorithm of non-phase demodulation based on the idea of composite mesh is studied, which introduces the rotating grid graph and the state coupling transfer. The theoretical analysis and simulation show that the proposed algorithm can obtain better performance than the differential non-coherent demodulation and approach the best demodulation performance. At the same time, it has the advantages of anti-initial phase and frequency offset tracking. After completing the modulation parameter estimation and demodulation algorithm analysis, a non-cooperative CPM signal blind demodulator structure is designed, and the blind demodulator is divided into modules. The optimization method of program flow and algorithm code for DSP implementation is given, and the complexity of Viterbi algorithm and grid coupling algorithm in complex mesh demodulation algorithm is analyzed. A multi-core DSP based complex mesh demodulation scheme is proposed. The parallel processing of a large number of data is realized by using the parallel operation of multiple processor cores.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.3
，

本文编号：1437339