Underlay频谱共享方式下信号参数估计和调制识别的方法研究

Signals parameter estimation and modulation recognition have the important application value in cognitive radio. Aiming at the problems that the signals parameters is difficult to estimate with the interference temperature measuring in the underlay spectrum sharing mode of cognitive radio system, the project will research on the parameter estimation and modulation recognition methods of time-frequency overlapped signals. The number of components for time-frequency overlapped signals is estimated based on the generalized cyclic cumulants and extended Shannon entropy, and carrier frequency for time-frequency overlapped signals is estimated by cyclic covariance spectrum and norm optimization. The signal to noise ratio estimation method based on generalized cyclic cumulants and modern spectral estimation for time-frequency overlapped signals is also studied and the project try to do the research on the performance bound of parameters estimation for time-frequency overlapped signals. Meanwhile, the modulation recognition method of time-frequency overlapped signals in underlay spectrum sharing mode is proposed by combining the generalized cyclic cumulants feature extraction theory and the least error square sum classifier theory. Based on the research, the project will also give a set of complete system scheme of parameter estimation and modulation recognition for time-frequency overlapped signals in Underlay spectrum sharing mode and evaluate the performances of the proposed methods in various application environments by computer simulation, which provide the basis of theory and application for signals parameter estimation and modulation recognition in cognitive radio system.

信号参数估计及调制识别技术在认知无线电中有重要的应用价值。针对目前认知无线电系统中以Underlay频谱共享方式接入时干扰温度测量所需参数难以估计的问题，本项目研究Underlay频谱共享方式下时频重叠信号参数估计和调制识别方法，包括探索基于广义循环累积量和扩展Shannon熵的时频重叠信号的分量个数估计方法、基于循环共变谱和范数优化的时频重叠信号的载波频率估计方法以及基于广义循环累积量与现代谱估计的时频重叠信号信噪比估计方法，并尝试研究时频重叠信号参数估计的性能界。此外，结合广义循环累积量特征提取理论和最小误差平方总和分类器理论，探索Underlay频谱共享方式下时频重叠信号的调制识别方法。在此基础上，给出完整的Underlay频谱共享方式下时频重叠信号参数估计及识别的系统方案，通过计算机仿真评估算法在各种应用环境中的性能，为信号参数估计和调制识别在认知无线电系统的应用提供理论基础。

信号参数估计及调制识别技术在认知无线电中有重要的应用价值。针对目前认知无线电系 统中以Underlay频谱共享方式接入时干扰温度测量所需参数难以估计的问题，本项目研究了Underlay频谱共享方式下时频重叠信号参数估计和调制识别方法，包括时频重叠信号的分量个数估计方法、时频重叠 信号的载波频率估计和码元速率估计方法以及时频重叠信号信噪比估计方法，并研究了时频重叠信号参数估计的性能界。此外，研究了Underlay频谱共享方式下时频重叠信号的调制识别方法。 在此基础上，给出了完整的Underlay频谱共享方式下时频重叠信号参数估计及识别的系统方案， 通过计算机仿真评估算法在各种应用环境中的性能，为信号参数估计和调制识别在认知无线电 系统的应用提供理论基础。本项目通过3年的研究，发表学术论文21篇，其中SCI检索论文7篇，EI检索论文12篇；申请国家发明专利30项，其中获得授权国家发明专利13项；培养博士生1名，硕士研究生5名。