In radar systems, target tracking errors are mainly from motion models and nonlinear measurements. When we evaluate a tracking algorithm, its tracking accuracy is the main criterion. To improve the tracking accuracy, in this paper we formulate the tracking problem into a regression model from measurements to target states. A tracking algorithm based on a modified deep feedforward neural network (MDFNN) is then proposed. In MDFNN, a filter layer is introduced to describe the temporal sequence relationship of the input measurement sequence, and the optimal measurement sequence size is analyzed. Simulations and field experimental data of the passive radar show that the accuracy of the proposed algorithm is better than those of extended Kalman filter (EKF), unscented Kalman filter (UKF), and recurrent neural network (RNN) based tracking methods under the considered scenarios.

Among all types of unwanted signals in high-frequency (HF) surface wave (HFSW) over-the-horizon (OTH) radars, radio-frequency interference (RFI) is dominant since HF band is shared by many radio services. In observation data, there are two types of common RFI. The most common one is the conventional RFI which presents vertical stripe paralleling to range axis in range-Doppler spectrum (RDS) and has been exhaustively reported by previous papers. Meanwhile, a new type of RFI characterized by sloping stripes (called RFISS) in RDS is also frequently observed. This work concentrates on the new RFISSand establishes a unified model for the above two types of RFI. Based on this generalized model, a time-domain RFI suppression algorithm is proposed here. Benefiting from a closed-form approximate maximum likelihood estimator, the proposed algorithm exhibits excellent performance and is computationally efficient. Its operational performance is evaluated using the field data recorded by experimental HFSW OTH radar of Wuhan University.