伪线性卡尔曼滤波器广泛应用于纯方位目标跟踪中,其递归性非常适合于硬件实现。结合实际项目,在对数据流进行分析的基础上,使用xilinx公司virtex-4系列芯片提供的DSP48专用乘法单元,以及自行设计的高效除法单元,通过简化标量运算和充分利用并行流水机制,最终实现了伪线性卡尔曼滤波器。仿真验证了数据的精确性。综合报告显示此系统的最高频率为123.8MHz,满足实时性要求。
Pseudo-linear kalman filter is widely used in bearings-only target tracking and it's likely to be implemented by hardware because of it's recursion property. This paper presents an implementing of pseudo-linear kalman filter on FPGA by predigestion algorithm and parallel process. On analyzing dataflow speciality, DSP48 digital processing unit provided by virtex-4 series of xilinx and special designed very-high speed division are used in this paper. Simulation shows that the data in processing of algorithm based on FPGA is precise. Synthesis report shows that the highest frequency is 123.8 Mhz, so this system is real-time.