Towards an Automated Approach for Testing IoT Devices

Towards an Automated Approach for Testing IoT Devices

연구 분야: Strategies

학회: SERP4IoT '24: Proceedings of the ACM/IEEE 6th International Workshop on Software Engineering Research & Practices for the Internet of Things

The Internet of Things (IoT) comprises a network of physical devices embedded with sensors and software to collect and exchange data with other devices and systems via the Internet. IoT devices vary from small devices to complex industrial appliances. Despite the increase in the number of IoT devices, there is a lack of proper testing for these devices, which can impact the functionality of IoT systems. This study focuses on an automated approach for testing IoT systems that use Android-based devices. We propose an approach to generate test cases (TCs) and execute them in physical devices as instrumented tests. Our methodology uses source code as input. We analyze the source code and create an Abstract Syntax Tree (AST). We navigate the AST to identify classes, methods, input parameters, and return types. We manually create a test case (TC) template and use a heuristic search algorithm to generate the test data for each unit test. We populate the TC templates with information extracted from the AST and data generated by a heuristic search algorithm to generate executable TCs. We assess the quality of generated TCs using the mutation analysis technique. The experiment demonstrates that the proposed approach effectively generates executable TCs for conducting functional tests for IoT devices. This study can be beneficial for practitioners, researchers, and device manufacturers towards improvement in the way IoT devices are tested.