As electromagnetic environments are increasingly complex, there are more kinds of radar jamming signals. Active jamming recognition has problems of the low recognition accuracy and the high computational complexity, especially under a low jamming‐to‐noise ratio (JNR). Herein, a deep learning network based on bilinear EfficientNet and attention mechanism is proposed to recognise and classify eight kinds of jamming signals automatically. Firstly, the one‐dimensional interference signal is transformed into a two‐dimensional time–frequency image by the time–frequency analysis. Based on the transfer learning of EfficientNet‐B3, the effective features of the time–frequency image are automatically extracted by the two‐way network with attention mechanism, and the active interference classification is realised. The experimental results show that the method's overall recognition rate for eight kinds of interference signals is more than 97.5% when the JNR is −8 dB and is close to 100% at −2 dB.

Due to the increase in the number of antennas in massive Multiple-Input Multiple-Output (MIMO) systems, traditional MIMO detection algorithms need to be improved. In this letter, we introduce trainable variables in Broyden Quasi-Newton method to obtain Broyden-Net that avoids the high-dimensional matrix inversion of the linear Minimum Mean Square Error (MMSE) detector and also breaks through the performance limitations of the linear detector. To further simplify the complexity of the algorithm, we use the search direction consistency of the special form Broyden-Fletcher-Goldfarb-Shanno (BFGS) in Broyden to further combine the BFGS Quasi-Newton method with Deep Learning (DL), to propose BFGS-Net that is more adapted to massive MIMO detection. Numerical results show that Broyden-Net and BFGS-Net effectively reduce the computational complexity of the MMSE detector and can achieve good detection performance for massive MIMO systems in spatially correlated scenarios.

This paper presents the joint estimation of the direction of arrival (DOA) and the time difference of arrival (TDOA) for multipath signals, which is the central issue for mobile communications. The estimator decouples the multi-dimensional problem for realizing the high resolution of DOA and TDOA based on the matrix pencil method without the computation of the correlation matrix. First, the data are transformed into the frequency domain to deconvolution by the known channel model, which maps delays into phase shifts in the frequency domain. Then, a two-dimensional matrix pencil algorithm yields the joint DOA and TDOA estimation based on the double Vandermonde structure data obtained. A closed-form solution without searching is given with the DOA and TDOA of incidence that can be paired automatically without an additional matching algorithm. The simulation results show that the proposed algorithm is not only suitable for the case that the elements' number is less than the source number but also for the case that multipath rays have equal DOA or TDOA.

With the development of the radar technology, it is required that a radar has wider bandwidth and can transmit multiple signals simultaneously. The polyphase digital channelised transmitter structure can transmit multiple signals simultaneously and increase the bandwidth of the output signal. However, in order to reduce the frequency aliasing and improve the spectrum utilisation, the prototype filter needs to have sharp transition bandwidth. The design of the prototype filter directly affects the performance of the radar transmitter. Using the complex‐exponential modulation frequency‐response masking (CEM‐FRM) approach to design the prototype filter can obtain both sharp transition bandwidth and low complexity. The application of a polyphase structure can further reduce the complexity of the filter bank and make the sampling rate of the output signal several times the field programmable gate array operating frequency. The proposed CEM‐FRM‐based low complexity polyphase channelised radar transmitter is verified by MATLAB simulation and the theoretical complexity of the proposed CEM‐FRM‐based channelised radar transmitter structure is compared with other methods.

Modern radar systems often face various interference signals in complex and rapidly changing electronic environments. The task of suppressing this interference in the radar echo signal to extract vital information is challenging. A radar interference suppression method is proposed based on a generative adversarial network (GAN). This method effectively recovers the target signal from the echo signal, which contains interference and noise, by leveraging the powerful fitting ability of GAN. Specifically, this method was tested using coherent suppression interference, smart noise interference, and noise frequency modulation suppression interference. We compared the proposed GAN method with recurrent neural network, short‐time Fourier transform time‐varying filtering, short‐time fractional Fourier transform time‐varying filtering algorithms and RNN approach. The results show that the interference suppression algorithm based on GAN is superior to the other three algorithms.