Joint Optimization Strategy of Unmanned Aerial Vehicle Group Trajectory and Symbol-Level Precoding in an Adversarial Communication Environment

Unmanned aerial vehicle （UAV） base stations have the characteristics of low cost， fast deployment， wide coverage， high mobility， and high variability. Moreover， they play increasingly important roles in wireless communication systems. When facing a complex electromagnetic interference environment， the interference avoidance trajectory optimization technique based on the concept of UAV obstacle avoidance suffers from serious intra-user interference and high optimization overhead. To solve these problems， this paper proposes a joint optimization scheme of UAV swarm trajectory and symbol-level precoding. The optimization goals were to minimize the average flight distance of multiple UAVs and maximize the minimum signal-to-interference-plus-noise ratio received by ground terminals. Thus， a time slot-by-time slot optimization sampling synchronous iteration algorithm was designed， and fast exploration optimization sampling random trees were applied between different time slots to effectively plan the trajectories of multiple UAVs. In addition， to realize the joint utilization of the geographical and electromagnetic domains of UAVs， a fitness function was constructed using the minimum received signal-to-interference-plus-noise ratio of ground terminals and the average distance of multiple UAVs from the end point to ensure anti-interference communication. In the process of multi-UAV trajectory optimization， the collision states between multiple UAVs were judged by the velocity obstacle method， thereby generating a dynamic collision avoidance velocity between multiple UAVs to ensure safe flight between multiple UAVs. Simulation results showed that compared with traditional zero forcing （ZF） precoding， singular value decomposition （SVD） precoding， minimum mean square error precoding （MMSE）， and Tomlinson-Harashima Precoding （THP）， the proposed symbol-level precoding and trajectory joint optimization strategy significantly reduced the average bit error rate and outage probability. Among them， the outage probability was reduced by up to 59.46%， and the effective spectral efficiency was increased by up to 1.4 bps/Hz. In conclusion， the proposed method can perform collision-free trajectory interference avoidance planning for multiple UAVs and therefore has strong theoretical and engineering application value.