Cognitive radar (CR) systems have garnered significant attention for their ability to adapt and optimize radar performance in dynamic and uncertain environments. Game theory (GT) provides a powerful framework for modeling and analyzing the interactions between CR systems and their environment. Reinforcement learning (RL) can serve as an effective decisionlearning algorithm for dynamic interference suppression in CR systems interacting with their environment. This paper reviews the application of these two solid foundational theories, GT and RL, in various directions of game modeling and algorithm research for CR systems. Through extensive research and development by scholars and engineers, both theories have gained recognition in academia and industry. Furthermore, utilizing GT as the framework for CR interference suppression game modeling, and introducing RL as the decision-making mechanism for interactive learning between CR systems and their environment, aligns well with the operational mechanisms of CR. Therefore, by integrating the core content of GT, establishing RL algorithms based on game equilibrium for application in CR systems, theories and algorithms can be developed to address real-world cognitive electronic warfare scenarios. The review aims to provide insights into the current state of research in this field and to inspire further advancements in CR technology through the integration of GT principles and RL algorithms.

Radar and jammer can be considered as two players in the countermeasure game of electronic warfare (EW). A game framework for joint beamforming and power allocation is investigated for multistatic radars and multiple jammers in this paper. Under a certain target detection criterion, the major aim of every radar is to reduce its power consumption and suppress various jammings. Conversely, the multi‐agent jammers determine their transmitting power strategies to interfere with the multistatic radar network, which is based on the transmitting power of the radars. According to the countermeasure game, both existence and uniqueness are strictly confirmed about the Nash Equilibrium (NE) of the proposed game. Besides, the receive beamformer weight vector can be obtained by minimum variance distortionless response (MVDR). And the corresponding transmit beamformer weight vector is also obtained. An algorithm for joint beamforming and power allocation is proposed, which is iteratively convergent. Finally, the convergence and validity for the proposed algorithm and the interference rejection capability of the multistatic radars are supported by the numerical results.

To deal with the threat of the new generation of electronic warfare, we establish a non-cooperative countermeasure game model to analyze power allocation a

The normal operations of radar systems and communication systems under the condition of spectrum coexistence are facing a huge challenge. This paper uses game theory to study power allocation problems between multistatic multiple-input multiple-output (MIMO) radars and downlink communication. In the case of spectrum coexistence, radars, base station (BS) and multi-user (MU) have the working state of receiving and transmitting signals, which can cause unnecessary interferences to different systems. Therefore, when they work together, they should try to suppress mutual interferences. Firstly, the signal from BS is considered as interference when radar detects and tracks targets. A supermodular power allocation game (PAG) model is established and the existence and uniqueness of the Nash equilibrium (NE) in this game are proved. In addition, the power allocation problem from BS to MU is also analyzed, and two Stackelberg PAG models are constructed. It is proved that the NE of each game exists and is unique. Simultaneously, two Stackelberg power allocation iterative algorithms converge to the NEs. Finally, numerical results verify the convergence of the proposed PAG algorithms.

This paper develops a Bayesian game-theoretic framework for joint power allocation and beamforming in a multistatic radar system operating in point clutter with multiple adversarial targets. To address the inherent incomplete information in radar–target interactions, a Bayesian game model is formulated, which enables joint optimization of transmit power and beamforming under clutter interference. Under the proposed framework, each radar directs its mainlobe toward the target and suppresses clutter through sidelobe nulling, thereby ensuring reliable detection. Due to variations in the angle belief probability, the power iteration algorithm exhibits stable oscillatory convergence. Power trajectories fluctuate within a bounded range around the Nash equilibrium without divergence, confirming numerical stability. Increasing the number of antenna elements leads to a linear rise in required transmit power. The minimum signal-to-clutter-plus-noise ratio (SCNR) constraint is positively correlated with transmit power; however, beyond a certain threshold, the rate of power increase diminishes. Simulation results validate the effectiveness and robustness of the proposed algorithms in complex multi-target, multi-clutter environments, providing a novel theoretical foundation for intelligent radar detection under incomplete information.