Due to the open nature of wireless communication, malicious electromagnetic jamming has long been a severe threat to the establishment and stability of communication links. To address this anti-jamming problem, a Markov decision process (MDP) with a two-dimensional action space consisting of transmit frequency and power is proposed in this paper, modeling the interaction between a normal communication link and the presence of malicious jammers in a frequency hopping (FH) communication system. Furthermore, we also prove the existence of the deterministic optimal policy of the proposed model theoretically. To obtain a policy for the communication link to avoid being jammed, the Dual Action Network-Based Deep Reinforcement Learning Algorithm, and Action Feedback Mechanism are proposed. The energy consumption and frequency switching overhead are considered and evaluated in both the proposed model and the algorithm. Finally, the proposed model and algorithm are verified not only in a virtual simulation environment but also in the field testing environment. The result suggests that the proposed algorithm is of great practical value for solving anti-jamming problems.

Increasing the hopping frequency speed and integrating artificial intelligence (AI) technologies are currently two of the most effective strategies for enhancing the anti-jamming performance of frequency hopping (FH) systems. However, due to the complexity of the decision-making process in intelligent agents, the system cannot complete decisions within the intervals between hops in fast frequency hopping (FFH) systems. As a result, there is no existing strategy for directly applying AI technologies to FFH systems. In this work, we introduce the concept of the available frequency set (AFS) and apply deep reinforcement learning (DRL) methods to FFH systems, enabling them to retain their inherent advantages while also gaining adaptability to dynamic environments. Building on this, we propose an improved multi-action deep recurrent Q-network (MA-DRQN) algorithm to determine the AFS for hopping sequence generation. Finally, the proposed method is shown to outperform both traditional FFH systems and advanced intelligent FH systems in handling blind and active jammers. Moreover, the hopping sequences generated based on AFS exhibit strong unpredictability.

Water‐Filling (WF) is widely applied in power allocation in multichannel wireless communications. By mathematically linearising the optimal WF expression, the authors observe an intrinsic parallel‐shift property of WF, based on which a fast and efficient WF algorithm is proposed. Compared with the conventional WF algorithms, it greatly simplifies WF execution by removing the Lagrange multiplier (or water‐level) searching process. The authors further apply the proposed parallel‐shift WF to solve the power allocation problem in orthogonal frequency division multiplexing (OFDM)‐based underlay cognitive radios (CRs), with the objective of maximising the secondary user's throughput over all OFDM sub‐channels under the transmit power and the interference constraints. To this end, the existing algorithm adopts an iterative binary searching process to find the solution, where the conventional WF algorithm with Lagrange multiplier searching is invoked in each iteration. In contrast, the authors propose a new power allocation algorithm to remove the iterative binary searching process. It runs the simplified parallel‐shift WF only once, and then directly calculates the final solution using a power adjustment process (with the parallel‐shift property as the underlying enabling mechanism). Numerical results show that both the proposed parallel‐shift WF and the OFDM‐based CR power allocation algorithms can run multiple times faster than the existing counterparts, and the gap on the running time increases with the total number of OFDM sub‐channels in the system.

Most channel estimation techniques for orthogonal frequency division multiplexing (OFDM) system are not directly applicable to filter bank multi-carrier with offset quadrature amplitude modulation (FBMC-OQAM) system because of the imaginary interference of FBMC-OQAM system. Thus, a novel support vector regression (SVR) algorithm for fading channel estimation in FBMC-OQAM system is proposed. We propose a nonlinear complex SVR algorithm (NCSVR) adapted to two auxiliary pilots (AP)-based FBMC-OQAM signal, namely as AP-NCSVR. The practicality of our method is proved by simulation results obtained for Jakes fading model and vehicular A model. Simulation results reveal the accuracy of the proposed AP-NCSVR algorithm to estimate channel frequency response coefficients and show that the bit error rate performance of the proposed AP-NCSVR algorithm is better than the conventional interpolation method for Jakes fading model and vehicular A model.