A Vision of Fog Systems with Integrating FPGAs and BLE Mesh Network

摘要

This paper presents a novel integrated field-programmable gate arrays (FPGA) and Bluetooth Low Energy (BLE) mesh system in the field of fog computing. Combining the merits of 1) programmability of FPGAs, and 2) wide-area communication and low-energy consumption of BLE mesh technology, this platform will create a diverse range of applications such as low-latency computation at the network edge, as well as showing a demo for Internet-of-Things (IoT) connections to cover an entire building. By integrating FPGAs, more important, the fog node can offer a great potential for flexible acceleration of many workloads and improve power efficiency with more hardware parallelism than CPUs and GPUs. These advantages are going to move FPGAs into the mainstream of fog computing for the foreseeable future. The preliminary results of a BLE mesh network is demonstrated on http://sceweb.sce.uhcl.edu/xiaokun/#-> project. Generally we provide three test cases: to control the mesh network by local device, local server, and cloud server. It can be observed that the feedback control from the local device achieves the lowest latency, and the response delay from the cloud server is the highest. Therefore, it is necessary to pre-process more than 90% raw data from sensor network at the proximity of the network edge/fog, particularly with the number of IoT devices in tens of billions levels.