A novel satellite target recognition method based on radar data partition and deep learning techniques is proposed in this paper. For the radar satellite recognition task, orbital altitude is introduced as a distinct and accessible feature to divide radar data. On this basis, we design a new distance metric for HRRPs called normalized angular distance divided by correlation coefficient (NADDCC), and a hierarchical clustering method based on this distance metric is applied to segment the radar observation angular domain. Using the above technology, the radar data partition is completed and multiple HRRP data clusters are obtained. To further mine the essential features in HRRPs, a GRU-SVM model is designed and firstly applied for radar HRRP target recognition. It consists of a multi-layer GRU neural network as a deep feature extractor and linear SVM as a classifier. By training, GRU neural network successfully extracts effective and highly distinguishable features of HRRPs, and feature visualization technology shows its advantages. Furthermore, the performance testing and comparison experiments also demonstrate that GRU neural network possesses better comprehensive performance for HRRP target recognition than LSTM neural network and conventional RNN, and the recognition performance of our method is almost better than that of other several common feature extraction methods or no data partition.

Frequency agile radar (FAR) has been widely used due to its good ability in antijamming and low probability of interception, which will benefit passive bistatic radar (PBR) using FAR as the illuminator. However, the traditional moving target detection (MTD) method cannot be utilized by FAR due to the random phase fluctuation of its echo signals. To solve this problem, we propose a modified MTD method of PBR for target detection using FAR as the illuminator. This method first removes the initial carrier frequency of echo signals. Then a bank of matched filters are constructed according to the echo signals with the initial carrier frequency removed. After being matched filtered, the phase fluctuation caused by frequency agility is eliminated. Consequently, coherent integration can be achieved by MTD processing under certain criteria. Furthermore, the factors that affect the performance of the proposed method are analyzed. Finally, numerical simulations are performed to verify the effectiveness and validity of the proposed method in target detection.