RFSM-Fuzzing a Smart Fuzzing Algorithm Based on Regression FSM

Wireless communication protocol is the core of wireless sensor networks. The wireless sensor networks transfer data wirelessly and are usually deployed in wild, moreover, the energy, computing power and storage capacity of its nodes are all limited, such characteristics make wireless communication protocols more vulnerable to be attacked. In order to ensure the security of the protocol, it´s necessary to use effective testing techniques for defect detection. As the test subject here, ZigBee protocol is widely used in low-rate wireless sensor networks. In this paper, a test algorithm called RFSM-Fuzzing (Regression Finite State Machine Fuzzing) is proposed which can automate test the robustness of wireless communication protocols and mine the potential vulnerabilities. The algorithm combines the advantages of conformance testing and intelligent Fuzz testing, so that it has high coverage and effectiveness. In this paper, Z-Stack Fuzzer is designed and through the application of RFSM-Fuzzing algorithm, testing of some test cases of MAC-layer of the ZigBee Protocol is realized. At last, the test results are obtained and the effectiveness of the algorithm is proved.