基于 GPU 的外辐射源雷达信号处理实时实现方法

摘要

随着外辐射源雷达技术的发展，实时处理能力成为制约外辐射源雷达系统性能的重要因素。现有的处理手段仅适用于较小采样率信号以及低速目标的情况，对于数字电视信号等大带宽高采样率的外辐射源雷达信号以及高速小目标的情况很难满足实时性的要求。提出了一种基于图形处理器的外辐射源雷达信号处理实时实现方法，能够提高分块最小均方误差算法和 Keystone 算法的并行处理能力。通过真实数据验证表明基于所提结构的实现方法能够满足外辐射源雷达信号处理实时性的需求。%With the development of passive radars,real-time processing ability has become an important factor of the system performance.The exiting processing means are only suitable for low sample rate signals or low-speed targets.When it comes to high sample rate signals (such as digital TV signals)or small high-speed targets real-time processing is difficult to achieve.A real-time implementation of signal processing for passive radars based on the graphic processing unit (GPU)is presented,which can enhance the parallel processing capa-bility of the block least mean square (BLMS)algorithm and the Keystone algorithm.Furthermore,the per-formance of the proposed implementation is verified by the processing of real data.