Design And Implementation Of The Key Software Modules About IoT Multi-UE Tester

With the rapid development of Io T technology, its global commercial process also gradually began. According to Machina Research statistics, in 2011, the number of terminals worldwide equipped with Io T applications is 20 billion, while in 2020 this figure is expected to reach 12.2 billion. Due to the mobile communication network's sound infrastructure, wide range of network coverage, large users, Io T equipment as long as access to mobile communication network can realize each other's network connection.With the continuous development of the Internet of Things, demand for Io T base stations that based on a mobile communication network is growing, and demand for Io T Tester for testing those base stations is also increasing. Io T multi-UE(User Equipment) Tester as an indispensable tool to test Io T base stations, is widely used to test a variety of Io T Base stations.This paper is derived from a project that collaborates with the enterprise based on narrowband networking technology named Io T multi-UE Tester prototype R & D. The main work of this paper includes:1. Multi-UE tester real-time scheme design and and verification. As the multi-UE tester requests strict real-time performance, while the Linux system does not support hard real-time, we decide to use the open-source Linux RTAI patch to solve the real-time problem. We design a real-time scheme with the RTAI corresponding interfaces to satisfy the high real-time performance of the project. 2. Multi-UE tester software debugging technique design and implementation. Since the tester has too many frame formats and formats are too complex, Log debug mode is difficult to locate the problem. We develop Wireshark plug-in, by means of Wireshark platform functionality it can be easily to analyze the frame format. 3. Multi-UE tester channel soft decoding implementation and optimization. The tester allows users over 10,000 to communicate in real time, while Linux open source decoding scheme can not meet this requirement, so we need to implement a highly efficient Viterbi decoding algorithm independently, and use vector instructions and multithreading technology to optimize and accelerate it.