Research on Optimization of UAV Communication Network Security Protection Strategy Based on Advanced Encryption Technology

With the wide application of UAVs in various applications, the security, spectrum, and energy efficiency of their communication networks have become increasingly prominent. This paper proposes a joint optimization strategy based on deep reinforcement learning for drone swarm communication networks. First, a model is constructed that takes into account security threats, spectrum sharing, and energy consumption. Intelligent agents are then trained through deep reinforcement learning to dynamically select the best spectrum allocation and energy strategy to improve spectrum and energy efficiency while maintaining network security. Based on encryption technology, this paper studies resource optimization strategies for UAV security communication in different scenarios. Aiming at the incomplete certainty of multiple eavesdropping positions and the problem of the no-fly zone during UAV flight, a joint optimization algorithm is proposed to optimize UAV trajectory, interference power, and transmission power of ground base stations so as to maximize the minimum average security capacity of the system in the worst case. To solve the problem of the LoS link of UAV being quickly blocked in the city and secondary users easily causing excessive interference to primary users, intelligent reflectors are introduced to assist the UAV in secure communication. IRS can be used to reconfigure channel parameters to control the propagation direction of UAV communication links, enhance the channel quality of the primary link, and weaken the channel quality of eavesdropping links and interference links. Simulation results show that the optimization scheme improves the channel quality of UAV in crowded scenarios, inhibits the eavesdropping effect of eavesdroppers on secondary security users, and reduces the spectrum multiplexing interference of primary users, thus significantly enhancing the security capacity of the system. When the interference power of UAVs is increased, the value and growth rate of the security capacity of security users are significantly increased. The increase is 20%. Through a large number of simulation experiments, it has been proved that this method has excellent performance in improving communication security, spectrum utilization, and energy efficiency and has obvious advantages over the traditional baseline and average allocation DQN-wrap method.