Neural Network Testing Framework for Microcontrollers

This paper analyzes and evaluates the performance of neural network testing using microcontrollers. The finite precision of microcontrollers registers and the computation time were considered for evaluation. The paper illustrates a general framework in which a neural network application can be easily integrated. This framework is based on Contiki IoT operating system that ensure support for wireless communication and efficient implementation of processes in a resource constrained system. As a practical example, a human activity recognition system, based on Long Short-Term Memory recurrent neural network, was implemented, and evaluated using a modified evaluation board from Analog Devices. The obtained system accuracy is very good, very close to the accuracy obtained in infinite precision simulation, the computation time is low enough and the system works in real time.