基于FPGA的二维谱峰搜索算法硬件架构设计

Hardware Architecture Design of Two-dimensional Spectrum Peak Search Algorithm Based on FPGA

当前基本采用现场可编程门阵列(field programmable gate array,FPGA)来实现一维多重信号分类(multiple signal classification,MUSIC)算法,为了使MUSIC算法更适合应用于实际环境中入射信号的波达方向(direction of arrival,DOA)估计,采用FPGA来实现二维MUSIC算法的谱峰搜索。合理设计各个步骤的并行结构以提高实时性,使用分步搜索法减少计算量和计算复杂度,通过模块复用降低资源消耗,采用流水线设计提高计算效率,从而在满足估计精度的同时使得谱峰搜索的实现时间更少。仿真试验结果表明,该硬件实现方案在100 MHz工作频率的FPGA芯片上实现二维谱峰搜索时,估计精度为1°,耗时为3.77 μs。该二维谱峰搜索实现结构的主要优点在于速度快,实时性高,具有较高的工程应用价值。

At present, Field Programmable Gate Array (FPGA) is mostly used to implement one-dimensional multiple signal classification (MUSIC) algorithm. In order to make MUSIC algorithm more suitable for direction of arrival (DOA) estimation of incident signal in real environment, this paper uses FPGA to implement the spectrum peak search of two-dimensional MUSIC algorithm. The experiment is arranged as follows. First, reasonably design the parallel structure of each step to improve real-time performance. Second, use a two-step search to reduce computational amount and complexity. What's more, reduce resource consumption through module reuse and use pipeline design to improve computational efficiency. So, the estimation accuracy is met and the spectrum peak search time becomes less. The simulation results show that the hardware implementation scheme takes 3.77 μs to realize two-dimensional spectrum peak search with the estimation accuracy of 1 degree on FPGA chip at 100 MHz. The main advantages of the two-dimensional spectrum peak search architecture include fast speed, high real-time performance and high engineering application value.