Filtering Method of Stateful Network Protocol Fuzz Test Cases Based on qrnn

Abstract

The present invention provides a method for filtering stateful network protocol fuzzy test cases based on QRNN. The method includes: forming a byte stream vector X; throttle vector X ’ ; Convert the byte stream vector X, X ’ The byte stream vector with a small medium dimension is zero-filled at the end, and the two byte stream vectors with the same length are XORed. The byte stream vector after the XOR operation is recorded as P, and the byte stream vector P is input into the training A good stateful network protocol fuzzy test case filtering model; the trained stateful network protocol fuzzy test case filtering model outputs validity of test cases. According to the method of the invention, the matching degree between the test case and the state of the protocol entity is improved, the invalid execution time of the protocol entity is reduced, and the efficiency of stateful network protocol fuzzy testing is improved.

Description

technical field [0001] The invention relates to the field of network security, in particular to a QRNN-based stateful network protocol fuzzy test case filtering method. Background technique [0002] Network protocol vulnerability mining is an important means to ensure the security of network communication. Fuzz testing is a commonly used network protocol vulnerability mining method at present. Test cases are constructed by means of generation or mutation, which are used as the input of protocol entities, and protocol entities are monitored to find out Security flaws in network protocols. For a stateful network protocol, the packets received by the protocol entity are related, and the state of the protocol entity may change after processing the packets. Therefore, unlike stateless network protocol fuzzing, when the test case does not match the state of the protocol entity, the test case may be directly discarded by the protocol entity, causing the protocol entity to execute ...

AI Technical Summary

Problems solved by technology

However, the above studies are all aimed at ordinary binary programs, and the research on stateful network protocol fuzzy test case filtering has not yet formed an effective solution.

Images