基于LLM引导与FSM动态推断的物联网模糊测试方法
By combining large language models with finite state machines, new states of IoT protocols are dynamically identified, solving the problems of high cost of manual modeling and limited exploration of deep paths in traditional fuzz testing, and achieving high-precision protocol vulnerability detection and efficient vulnerability mining.
Patent Information
- Application Number
- CN202610171299.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-06
- Publication Date
- 2026-07-17
- Estimated Expiration
- 2046-02-06
AI Technical Summary
Traditional fuzz testing methods for IoT protocols suffer from high costs of manual modeling, implementation disconnect from standards, and limited exploration of deep paths when dealing with complex protocols, resulting in low coverage and inaccurate state machine inference.
By combining Large Language Model (LLM) and Finite State Machine (FSM), an initial FSM model is constructed and the semantic reasoning capability of LLM is used to dynamically identify new states, thereby expanding the FSM structure and achieving effective coverage of deep states of the protocol.
It achieves high-precision automated construction of IoT protocol state machines, breaks through the bottleneck of deep protocol logic coverage, significantly improves the accuracy and efficiency of vulnerability discovery, and reduces the false positive rate.
Smart Images
Figure CN122093109B_ABST
Abstract
Citation Information
Patent Citations
Stateful protocol fuzzy test method and system based on large model construction state machine
CN121356909A
Adaptive management system for IoT networks utilizing dynamic fuzzy logic framework
US20250300904A1