The accuracy of radar emitter signal sorting nowadays deteriorates due to the high flexibility and complexity of modern radar pulse streams and the density of crowded electromagnetic environment. In modern radar signal sorting based on pulse repetition interval, conventional methods usually fail to achieve acceptable accuracy and lack stable performance for two main reasons: (1) Conventional methods require a large number of pulses in the stream, which is not practical in many applications. (2) These methods are sensitive to pulse loss and random noise pulses. These two reasons are the main problem that is addressed in this paper. Our proposed model is a machine learning architecture called Unified Residual Recurrent Neural Network (URRNN). In this architecture, residual neural network and recurrent neural network are combined and modified to alleviate the forementioned shortcomings of traditional approaches and enhance the model performance in both classification and deinterleaving tasks. This aim is achieved due to the fact that URRNN extracts both spatial and temporal features, which means more information about processed stream that is exploited to enhance model performance. Three different architectural combinations of URRNN, which show high accuracy and reasonable processing time, are built and trained. The structural and functional description are provided for each architecture. Simulation demonstrates high accuracy and reliable performance of the proposed methods in different circumstances. The results are compared with the results obtained by other conventional machine learning techniques.
The analysis of intercepted multifunction radar (MFR) signals has attracted considerable attention in the field of cognitive electronic reconnaissance. The agility of pulse parameters makes it difficult to recognize their behavior states. Currently, most existing approaches for recognizing MFR behaviors heavily depend on prior information, which can hardly be obtained in a non cooperative way. This study develops a contrastive self supervised method to adaptively segment and cluster MFR pulse sequences into radar words with little prior information. First, a convolutional neural network (CNN) is trained in a self-supervised manner to differentiate between adjacent and nonadjacent pulse group embeddings to learn robust feature representations from sequences of pulse descriptor words (PDWs). Next, based on the learned embeddings, test statistics are calculated for each PDW sequence to detect change points, which are in turn used to segment the sequence into multiple subsegments, each containing only one radar word. Finally, feature vectors are produced for clustering the subsegments into groups of radar words. Simulation results show that without using any labeled data, the proposed method can effectively extract the radar words of a hypothetical MFR under corrupted and overlapped pulse parameters and performs only slightly worse than the state-of-the-art fully supervised method.
The conventional method of centralizing information fusion is commonly employed for sorting radar signals in reconnaissance receivers. However, challenges arise when the distance between reconnaissance receivers and the fusion center is distant, or when the fusion center is compromised by hostile forces. To address these issues, this paper proposes a novel distributed information fusion method. In this method, each reconnaissance receiver is restricted to accessing adjacent nodes within an undirected graph for information transmission and local computation. The distributed Dempster’s combination rule and the cautious conjunctive rule are implemented using weight functions and consensus algorithms. Furthermore, an innovative outlier detection algorithm is incorporated into the fusion process to enhance its robustness. Experimental results demonstrate that the proposed method effectively improves the accuracy of radar signal sorting. When the sorting accuracy of a single reconnaissance receiver is equal to or higher than 60%, both fusion rules achieve a sorting accuracy of 100%. Even when the sorting accuracy of a single reconnaissance receiver is as low as 50%, the fused result still maintains a sorting accuracy of over 97%.
22 April 2024 in IEEE Transactions on Aerospace and Electronic Systems
The analysis of intercepted multifunction radar (MFR) signals and the assessment of potential threats are critical in the field of radar countermeasures. The complexity of MFR control mechanisms poses serious challenges in modeling and recognizing their internal states. MFRs, being discrete event systems (DESs), can be well described by formal languages. This paper presents a generalized version of the previous syntactic modeling of MFRs, utilizing the proposed latent variable controlled stochastic context-free grammar (LVCSCFG). The enhanced model incorporates a mapping between the external environment and the production probabilities, offering a more effective representation of the resource management process. Additionally, we introduce a neural-based approach for maximum likelihood estimation of the MFR's internal states, which are represented by a grammar tree. LVCSCFG exhibits enhanced representation of MFR dynamic behaviors, and simulation results indicate that the neural network outperforms traditional statistical parsing algorithms in terms of speed and precision under various non-ideal channel conditions.