There has been a growing interest in developing data-driven, and particular deep neural network (DNN) based methods for modern communication tasks. For few popular tasks such as power control, beamforming, MIMO detection, these achieve state-of-the-art performance while requiring less computational efforts, resources acquiring channel state information (CSI), etc. However, it is often challenging approaches to learn dynamic environment. This work develops new approach that enables...

One emerging issue in modern electronic warfare is the competition between radar and jammer, which principle can be viewed as a noncooperative game with two players. In practice, interaction jammer involves multiple rounds well partial observation. This makes become dynamic imperfect information. Antijamming strategy design for such kind of still unclear. this work, frequency agile transmit/receive time-sharing considered. We utilize an extensive-form (EFG) information to model jammer. For...

This article considers the 3-D collaborative trajectory optimization (CTO) of multiple unmanned aerial vehicles to improve multitarget tracking performance with an asynchronous angle arrival measurements. The predicted conditional Cramér–Rao lower bound is adopted as a measure predict and subsequently control error online. Then, CTO problem cast time-varying nonconvex subjected constraints arising from dynamic security (height, collision, obstacle/target/threat avoidance). Finally,...

In this paper, a performance driven maneuvering resource allocation (MRA) scheme is developed for target tracking in airborne radar network (ARN). To exploit the degree of freedom ARN mobility tracking, we formulate MRA maximizing performance, under practical constraints on radar's speed and attitude variation rate, as well threat zone collision avoidance. We adopt Bayesian Cramér-Rao lower bound metric function to gauge build non-convex optimization problem. Instead using heuristic based...

Automated optimization modeling (AOM) has evoked considerable interest with the rapid evolution of large language models (LLMs). Existing approaches predominantly rely on prompt engineering, utilizing meticulously designed expert response chains or structured guidance. However, prompt-based techniques have failed to perform well in sensor array signal processing (SASP) area due lack specific domain knowledge. To address this issue, we propose an automated approach based retrieval-augmented...

An integrated trajectory planning and resource scheduling (ITPRS) scheme for multiple target tracking (MTT) in an airborne radar network is proposed this paper. The ITPRS seeks to improve overall MTT performance by exploring degrees of freedom power, beam, waypoints several radars. To quantify, forecast, control accuracy, the predicted conditional Cramer-Rao lower bound used as a measure. subsequently formulated nonconvex mixed integer nonlinear programming problem while satisfying total...

There has been a growing interest in developing data-driven and particular deep neural network (DNN) based methods for modern communication tasks. For few popular tasks such as power control, beamforming, MIMO detection, these achieve state-of-the-art performance while requiring less computational efforts, channel state information (CSI), etc. However, it is often challenging approaches to learn dynamic environment where parameters CSIs keep changing. This work develops methodology that...

To combat main lobe jamming, preventive measures can be applied to radar in advance based on the concept of active antagonism, and efficient antijamming strategies designed through reinforcement learning. However, uncertainties jammer, which will result a mismatch between test training environments, are not considered. Therefore, robust strategy design method is proposed this paper, frequency-agile jammer This problem first formulated under framework Wasserstein Then, imitation...

The competition between radar and jammer is one important issue in modern electronic warfare, which principle can be viewed as a non-cooperative game with two players. In this work, the frequency agile (FA) noise-modulated considered. As FA adopts coherent processing several pulses, hence multiple-round way where each pulse modeled round interaction jammer. To capture such property well imperfect information inside game, i.e., are unable to know upcoming signal, we propose an extensive-form...

With the recent development in radar technology, a multiple system (MRS) has become an attractive platform for target tracking. Technically speaking, data fusion among radars can definitely enhance tracking performance. However, enhancement may not always be significant, as improvement depends on several factors, such signal-to-noise ratio, deployment, and resolution of each radar. In this paper, benefit analysis MRS is developed. particular, whether, given environment, between two worthy to...

This work considers low-rank canonical polyadic decomposition (CPD) under a class of non-Euclidean loss functions that frequently arise in statistical machine learning and signal processing. These are often used for certain types tensor data, e.g., count binary tensors, where the least squares is considered unnatural. Compared to loss, losses generally more challenging handle. Non-Euclidean CPD has attracted considerable interests number prior works exist. However, pressing computational...

The problem of 3-D target tracking in asynchronous 2-D radar network is considered. To deal with this problem, paper presents a decentralized track-to-track fusion (DAT2TF) algorithm. DAT2TF algorithm implemented by reconstructing the optimal centralized result local estimates and their error covariance matrices. derivations show that actually operates sense, but not due to two approximations about motion model polar Cartesian measurement conversion procedure. evaluate estimation performance...

With the development of electronic warfare, jammer is becoming much smarter than before and its capability learning adapting to radar's transmission strategy poses a great challenge radar. The learning-based methods have been recently proposed improve anti-jamming performance ground-based radar when dealing with smart mainlobe self-protection jammers. Model-free reinforcement (MFRL) widely used method that easy implement can achieve good performance. However, MFRL low sample efficiency...

This paper proposes a multi-dimensional resource management (MRM) scheme for multifunctional radar systems that aims to improve multiple target tracking accuracy under dynamic jamming environments. The proposed constructs an optimization model coordinate the radar's power and frequency resources. To combat jammers, we present Reinforcement Learning based active anti-jamming Markov Decision Process model. resulting signal interference plus noise ratio Bayesian Cramer-Rao lower bound...

Machine learning has become successful in solving wireless interference management problems. Different kinds of deep neural networks (DNNs) have been trained to accomplish key tasks such as power control, beamforming and admission control. There are two state-of-the-art approaches train DNNs based models: supervised (i.e., fits labels generated by an optimization algorithm) unsupervised directly optimizes some system performance measure). However, it is no means clear which approach more...

A scheme of joint power and bandwidth allocation (JPBA) for centralized target tracking in multiple radar system is proposed this paper. The Bayesian Cramér-Rao lower bound gives a measure the best achievable performance tracking. Hence, it derived utilized as an optimization criterion JPBA strategy. Simulation results show that algorithm significantly outperforms equal allocation, terms root mean square error.

An important preliminary procedure in multi-sensor data fusion is <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">sensor registration</i> , and the key step this to estimate sensor biases from their noisy measurements. There are generally two difficulties bias estimation problem: one unknown target states which serve as nuisance variables problem, other highly nonlinear coordinate transformation between local global systems of sensors. In...

In this paper, we consider the problem of cooperative sensing via a system multi-unmanned aerial vehicles (UAVs), where each UAV is equipped with directional antenna to cooperatively perform detection tasks for several targets interest. To measure perception ability system, choose probability as metric, aiming maximize sum by jointly optimizing UAVs' deployment and orientations. tackle inherent nonconvexity formulated problem, first decompose it into two sub-problems, i.e., slave...