基于软件无线电的机会信号实时盲分离系统

在复杂的电磁环境中,往往存在多种频率交叠的机会信号,为利用机会信号进行定位增加了难度。针对这个问题,提出了一种基于软件无线电的机会信号实时盲分离系统。系统的硬件部分是在FPGA中实现等变自适应盲分离算法的,实现了AD采集后的复数域混合信号的分离。分离后的信号通过USB3.0数据线上传到上位机中进行存储、识别和实时显示。算法的FPGA实现采用流水线架构和浮点运算设计来提高信号的实时处理能力和数据计算精度。实测结果表明,该系统能够实时分离多路复数域混合机会信号。

Real-time Blind Signal Separation System for Signals of Opportunity Based on Software Radio

In a complex electromagnetic environment, there are often multiple signals of opportunity (SOP) with overlapping frequencies, which makes it difficult to use SOP for location. To solve this problem, this paper proposes a software radio based real-time blind source separation (BSS) system of SOP. The hardware part is to implement the EASI (Equivariant Adaptive Separation via Independence) algorithm in FPGA to perform real-time BSS processing on the complex domain mixed signals collected by AD. And the separated signal is uploaded to the upper computer through the USB 3.0 data cable for storage, identification and real-time display. The FPGA implementation of the algorithm uses a pipeline architecture and floating-point arithmetic design to improve the real-time processing capabilities of the signal and the accuracy of data calculations. The experimental results show that the system can separate mixed SOP in complex domains in real time.