A new scheme of target detection for pulse Doppler radar in interrupted sampling repeater jamming environment

Research on the anti‐jamming target detection for pulsed Doppler (PD) radar is intensively important to the normal work of radar in complicated and volatile electromagnetic environments. Generally, traditional anti‐jamming target detection methods involve several stages, each one of these stages is based on a model. Due to the issues of poor flexibility caused by multi‐stage processing and model mismatch caused by model‐based methods, the detection performance will be severely decreased. Therefore, a new target detection scheme for PD radar in an Interrupted‐Sampling Repeater Jamming (ISRJ) environment is proposed, which is a dual‐driven method of model and data. This innovative solution first achieves energy accumulation on the Doppler channel by a model‐driven method, then proposes a series of preprocessing methods to process radar signals to adapt to a data‐driven Convolutional Neural Network (CNN) detector. Finally, a CNN detector structure is proposed. The effectiveness of this novel scheme is demonstrated by a series of performance and visualisation experiments.