Cognitive electronic warfare is gaining increasing attention and will play an important role in warfare. Cognitive jamming decision-making, as a key technology, significantly influences battlefield dynamics. Modern radar can work in multiple directions at the same time, and most of the existing research is on jamming style decision-making of a single jammer, which is difficult to adapt to complex battlefield situations. Multi-agent coordination can effectively deal with multi-functional radar, so this paper investigates cooperative jamming decision-making algorithms. First, this paper establishes a cooperative jamming decision-making model, in which three jammers are assumed to jam the radar in three directions and are based on a centralized learning distributed execution architecture. Then, this paper extends the Advantage Actor-Critic(A2C) algorithm by proposing the Multi-Agent A2C(MA-A2C) algorithm and applying it to the model. In order to effectively solve the problems of high variance and environmental sensitivity in the MA-A2C algorithm, we propose two improvement methods and combine them to propose the Multi-Agent Heuristic Dueling A2C(MA-HDA2C) algorithm. We conducted simulation experiments on MA-A2C, two improved methods and MA-HDA2C algorithms under the same modeling conditions. The results show that the MA-A2C algorithm can effectively carry out cooperative jamming decision-making, but it has problems with low efficiency and poor stability. The MA-HDA2C algorithm has the advantages of fast convergence speed and stable convergence results. Compared with the MA-A2C algorithm, the MA-HDA2C algorithm has more significant superiority, which proves the effectiveness of the improvement method proposed in this paper.

With the advancement of multi-function radar (MFR) technology, miniaturized radar systems (MRSs) inevitably operate in complex electromagnetic environments (CEEs) dominated by MFRs as single-function radars are gradually being replaced by MFRs. MFRs can not only flexibly switch working states and generate diverse radar signal characteristics, but they can also acquire the MRSs’ position information, which has a significant impact on the execution of the MRSs’ close-range remote sensing missions. For resource-constrained MRS, selecting the optimal electromagnetic working mode in such environments becomes a critical challenge. This paper addresses the adaptive electromagnetic working mode decision-making (EWMDM) problem for MRS in CEE by establishing an EWMDM model and proposing a reinforcement learning (RL) method based on an improved soft actor–critic algorithm with prioritized experience replay (SAC-PER). First, we simulate the process of MRS receiving pulse description words (PDWs) from MFR waveforms and introduce noise into the PDWs to emulate real electromagnetic environments. Then we use a threshold to filter out uncertain recognition results to reduce the impact of noise on the MFR’s working state recognition. Subsequently, we analyze the limitations of the SAC-PER algorithm in noisy environments and propose an improved algorithm—SAC with alpha decay prioritized experience replay (SAC-ADPER)—to address the influence of environmental noise and stochasticity. Experimental results show that SAC-ADPER significantly accelerates the convergence speed of EWMDM in noisy environments and validate the effectiveness of the proposed method.