欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对时变幅度LFM干扰下的单通道通信信号与干扰盲分离问题，本文提出了一种基于遗传粒子滤波的单通道扰信盲分离新算法。该算法首先建立了受扰信号的状态空间模型，并利用遗传重采样粒子滤波得到通信码元和未知参数的最大后验估计。针对标准粒子滤波中存在的粒子退化现象，本文引入了遗传进化操作来迭代估计优质粒子，在减少了所需粒子数量的同时，加快了算法的收敛速度，使新算法在时变幅度LFM干扰的影响下具有较好的分离效果。非扩频通信信号仿真实验表明，新算法在干信比小于15dB，信噪比大于16dB的条件下，可以有效地从单路接收的受扰信号中分离出通信信号与LFM干扰。
英文摘要：
A new approach is proposed for single-channel blind separation of communication signal and co-channel time- varying amplitude LFM interference. The proposed algorithm aims to obtain the maximum a posterior （MAP） estimates of communication symbols and the unknown parameters using genetic evolution resampling particle filtering by establishing the state space model for the interfered signal. Specially, in order to overcome the sample impoverishment problem, genetic op- erations are introduced into the re-sampling process of particle filtering and estimate iteratively high quality particles that have more important weigh. In such a way, the number of needed particle set is reduced and the convergence rate of itera- tive estimation is speeded up during the sequential importance sampling process. Simulation results of nonspread spectrum communication signal show that new algorithm is effective to separate communication signal and LFM interference form sin- gle-channel received signal when the Interference-Signal-Ratio （ISR） is less than 15dB and SNR is more than 16dB.
同期刊论文项目
同项目期刊论文
Generalized generating function with tucker decomposition and alternating least squares for underdet
期刊信息
位置：