基于FPGA的传感数据加速处理

摘要

This paper presented a field-programmable gate arrays(FPGA)-based sensor data processing method to accelerate median operators, which included the calculation model, implementation process to fit the design constraints on the FPGA,and the appropriate integration strategy to connect the FPGA to the rest of the system. Experiments show that the proposed data processing method based on FPGAs has obvious advantages compared with the general-purpose CPUs in terms of parallel data stream evaluation and the power consumption strategy of sensor. It can be widely used in design and implementation for lightweight data pre-processing front-end node in the internet of things.%为解决物联网应用的实时响应问题,提出一种基于FPGA加速处理传感数据中位数计算的方法,包括计算模型、符合FPGA设计限制的实现过程,把FPGA连接到系统其他部分的融合策略等.实验表明,与通用CPU相比,提出的基于FPGAs的加速数据处理策略在功耗和并行传感数据流评价方面具有明显的优势,可以广泛地应用于物联网中前端轻量级数据预处理节点的设计与实现.