面向DSP平台的CiSSA-CSP特征提取算法的移植与优化

为实现便携式信号二分类解析系统的在线实时处理，采用DSP平台完成CiSSA-DSP特征提取算法的嵌入式移植。CiSSA-CSP特征提取算法具有出色的时-频-空域特征提取性能，适合于提取实时二分类系统中非平稳信号的特征。相比于PC机，嵌入式系统具有小型化、便携性、低功耗和低延时的特点，而嵌入式平台处理器的计算资源和内存受到限制，必须优化移植特征提取算法，才能保证二分类解析系统的分类精度和低延时。通过优化CiSSA-CSP算法流程，使用编译器优化、关键字和库函数等手段提高编译效率，将CiSSA-CSP特征提取算法移植到TMS320C6678DSP嵌入式平台，并利用公共数据库数据验证了其用于实时分类系统的有效性。相比于PC机的Matlab实现，DSP平台实现的二分类系统分类准确度下降小于0.5%，且单次实验信号解析耗时少于0.15s。

Transplant and Optimization of CiSSA-CSP Feature Extraction Algorithm on DSP

In order to realize the on-line real-time processing of portable signal binary classification analysis system, the embedded transplantation of CiSSA-DSP feature extraction algorithm is implemented on DSP platform. The CiSSA-CSP feature extraction algorithm has excellent performance of time-frequency-spatial feature extraction, and is suitable for extracting non-stationary signals in real-time binary classification system. Compared with PC, embedded implementation has the characteristics of miniaturization, portability, low power consumption and low delay, while the computing resources and memory of embedded platform processor is limited, and the algorithm must be optimized to ensure the binary classification accuracy and low delay. By optimizing the flow of CiSSA-CSP algorithm, using compiler optimization, keywords and library functions to improve the compilation efficiency, CiSSA-CSP feature extraction algorithm is transplanted to the TMS320C6678 embedded DSP platform, and its effectiveness for real-time classification system is verified by public dataset. Compared with the Matlab on PC, the classification accuracy of the binary classification system implemented by DSP platform is reduced by less than 0.5% and the computing time is less than 0.15s.