Research on the Behavior Prediction Method of the Friendly Aircraft in the Environment

With the growing complexity of modern air combat and the increasing frequency of adversarial interference, the rejection environment has become a critical challenge for real-time situational awareness and tactical decision-making. In such environments, the degradation of radar, communication, and navigation systems severely affects the reliability of aircraft behavior prediction. This paper presents a multi-source information fusion-based behavior prediction framework, designed to enhance the robustness and accuracy of predicting friendly aircraft behavior under varying interference conditions. By integrating radar tracking data, communication signal analysis, and onboard sensor telemetry, our framework leverages Hidden Markov Models (HMM) and Long Short-Term Memory (LSTM) networks to model temporal dynamics and improve prediction stability. A decision-level fusion strategy is introduced to combine outputs from individual models, dynamically weighted by their confidence scores. Experimental results show that the proposed method achieves significantly higher accuracy compared to single-source models, especially under medium and high interference levels. This research contributes a systematic approach to aircraft behavior modeling in contested airspace and supports more effective tactical planning and mission execution.