MIMO Radar Waveform Design via Deep Learning

Radar systems are a topic of great interest, especially due to their extensive range of applications and ability to operate in all weather conditions. Modern radars have high requirements such as its resolution, accuracy and robustness, depending on the application. Noise Radar Technology (NRT) has the upper hand when compared to conventional radar technology in several characteristics. Its robustness to jamming, low Mutual Interference and low probability of intercept are good examples of these advantages. However, its signal processing is more complex than that associated to a conventional radar. Artificial Intelligence (AI)‐based signal processing is getting increasing attention from the research community. However, there is yet not much research on these methods for noise radar signal processing. The aim of the authors is to provide general information regarding the research performed on radar systems using AI and draw conclusions about the future of AI in noise radar. The authors introduce the use of AI‐based algorithms for NRT and provide results for its use.

“…For MIMO waveform design in ref. [104], a feedforward NN was proposed. GAs have also been used for waveform design since the beginning of the century [59,105,106] until recently in ref.…”

Section: Waveform Designmentioning

confidence: 99%

Artificial Intelligence applications in Noise Radar Technology

Sénica,

Marques,

Figueiredo

2024

“…Ref. [34] decoupled the non-convex optimisation problem into two sub-optimisations, which were solved separately by artificial neural networks and relaxed semi-positive definite programming methods before feature fusion in order to derive the optimal radar waveform. Ref.…”

Section: Related Workmentioning

confidence: 99%

Online unsupervised generative learning framework based radar jamming waveform design

Sun

Gong

Mao³

et al. 2023

The jamming effect on radar is dominated by the design of the jamming waveform directly. Traditional jamming waveforms are generated using the template‐based method and are not environmentally resilient. Without the precise radar centre frequency support, using the direct guidance of the radar signal power spectrum density (PSD), a novel jamming waveform design method with an online unsupervised generative learning framework is proposed. The proposed framework consists of two modules: the waveform online generation module and the waveform unsupervised optimisation module. For the waveform online generation module, the well‐designed neural network (NN) is used to generate the jamming waveform. While for the waveform unsupervised iterative optimisation module, two loss functions are developed as two learning tasks to guide back‐propagation online. Given transmit power, the jamming PSD is designed adaptively to focus on the radar band and fit the radar PSD well, which improves the receiver in‐band jamming‐to‐signal ratio (JSR). It is worth noting that NN is end‐to‐end trained online without extensive prior training data. The framework can generate jamming waveforms within reasonable physical parameter ranges, and numerical experiments have shown that the designed jamming waveform performed better jamming effects on radar target detection in comparison with the traditional one.

“…[17], the mean-square-error (MSE) is adopted as the MIMO radar waveform design metric to enhance the target estimation performance. Recently, with the rapid development of artificial intelligence [18], deep learning methods have been proposed for MIMO radar waveform design as well [19][20][21]. Moreover, there are also many attempts in designing waveforms based on information theoretic approaches.…”

Section: Introductionmentioning

confidence: 99%

Mutual information‐based constant‐modulus waveform design for multiple‐input‐multiple‐output radar in spectrally crowded environment

Wang

Tang

et al. 2023

This study considers designing constant-modulus waveforms for multiple-input-multipleoutput radar in spectrally crowded environment. The mutual information is used as the waveform design metric. A constant-modulus constraint is imposed on the waveforms to improve the hardware efficiency. To enhance the spectral compatibility of the transmit waveforms, a precise control on the interference energy produced on each reserved bandwidth is considered. To tackle the multi-spectrally constrained waveform design problem, a minorisation-maximisation based method is developed, and a quadratic function to minorise the objective function is derived. Then an alternating direction method of multipliers is used to tackle the quadratically constrained quadratic programming problem at each iteration. The effectiveness of the presented algorithm is demonstrated via simulation results. K E Y W O R D Salternating direction method of multipliers (ADMM), constant-modulus, MIMO radar, minorisationmaximisation (MM), mutual information, spectrally crowded environment, waveform designThis is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.