论文部分内容阅读
机载雷达通常要求具备气象探测、防撞、火控甚至目标搜索、跟踪和识别功能,而这些功能的实现依赖于雷达数据处理的实时性。在现有的各种处理器架构中,多核DSP在数据处理方面具有比较优秀的表现。本文以工程化设计为出发点,在现有的基于DSP的数据处理硬件平台上进行多核数据处理流程整体框架的设计,对平台的实际数据处理能力进行量化评估,与单核流水线式数据处理方式相比,大大缩短了数据处理时间,提高了雷达数据处理的实时性。
Airborne radars usually require weather detection, collision avoidance, fire control and even target search, tracking and identification functions, and the realization of these functions depends on the real-time performance of radar data processing. In the existing variety of processor architecture, multi-core DSP in data processing has a relatively good performance. This paper takes engineering design as the starting point, designs the overall framework of multi-core data processing flow on the existing DSP-based data processing hardware platform, quantitatively evaluates the actual data processing capability of the platform, compares it with the single-core pipeline data processing method Than, greatly reducing the data processing time and improve the real-time radar data processing.