低轨卫星微弱信号检测技术与方法OA

With the rapid development of integrated space-air-ground networks and 6G communications,Low Earth Orbit(LEO)satellites,leveraging their advantages of global coverage and low latency,have become a core support for ubiquitous communications.However,satellite-ground links often face challenges such as ultra-low Signal to Noise Ratio(SNR),strong Doppler frequency shift,and limited terminal resources,making weak signal detection a key bottleneck.The development of weak signal detection technologies is systematically reviewed,and a comparative analysis of the principles and applicability of traditional signal processing methods and Artificial Intelligence(AI)-driven technologies is conducted.Traditional methods offer high computational efficiency and ease of deployment,but lack the adaptability to high-dynamic and non-stationary noise.AI-driven technologies achieve feature self-learning,enhancing the detection accuracy in complex scenarios,while with higher requirements for annotated datasets and computing power.Key technologies such as signal enhancement,dataset construction and feature extraction,lightweight models,real-time optimization algorithm design,and hardware-software collaboration are discussed,core challenges are summarized,and the future detection paradigm integrating intelligent sensing and edge computing is prospected,providing references for algorithm optimization and engineering deployment of LEO satellite communication systems.