Projects

GP-Regression-Extended-Target-Tracking

time: December, 2022

code implementation: https://github.com/zhiyuan-yang/GP-Regression-Extended-Target-Tracking

Based on paper [1], we use Gaussian process regression to recursively get the measurement function of extended target and then use Extended Kalman Filter(EKF) to track it. Future work can be done by subsituting the EKF with more advanced non-linear filter such as Rao-Blackwellised Particle filter, Variation Bayes Filter, etc.

[1] N. Wahlström and E. Özkan, ‘Extended Target Tracking Using Gaussian Processes’, IEEE Transactions on Signal Processing, vol. 63, no. 16, pp. 4165–4178, Aug. 2015, doi: 10.1109/TSP.2015.2424194.

Radar Emitter Identification based on CNN

time: May, 2022

code implementation: https://github.com/zhiyuan-yang/Radar-Emitter-Identification-based-on-CNN

In this project, we use Pseudo Wigner-Ville Distribution to extract the time-frequency feature of different modulated radar signals and then using convolution neural network(CNN) to classify the radar signals. The article has been received by the 14th Chinese Institute of Electroincs(CIE) DSP Conference.

[1]杨志远,张玉玺,孙进平等. 基于CNN的雷达辐射源信号开集识别方法[C]//中国电子学会数字信号处理专家委员会.第十四届全国DSP应用技术学术会议论文集.2022:5.DOI:10.26914/c.cnkihy.2022.066299.