Data packet detecting method based on cloud system

Abstract

The invention provides a data packet detecting method based on a cloud system. The data packet detecting method comprises the steps as follows: step A, compressing each matching string in a matching string set, and constructing a DFA (Deterministic Finite Automata) for pattern matching by using the compressed matching string set; step B, compressing character strings in a data packet to generate new character strings, inputting the new character strings to the DFA for matching; and step C, classifying data which are not filtered, and then selecting a suitable matching string to accurately match the data according to result of classification. The data packet detecting method based on the cloud system disclosed by the invention has the beneficial effects that: during detection of the data packet, the data packet is subjected to a step of inaccurate matching firstly, in this way, partial non-attacking data are filtered out, and only the data which are not filtered out are accurately matched; by adopting such a way that attacking data are accurately found from partial non-attacking data and all attacking data, the time for matching the partial non-attacking data is shortened, and the efficiency is improved.

Description

technical field [0001] The invention relates to a data packet detection method, in particular to a cloud system-based data packet detection method. Background technique [0002] Deep packet inspection originated from the explicit string matching method, that is, matching the existing string sequence from the target data sequence, and as explicit strings are gradually replaced by regular expressions, target feature matching based on finite state machines becomes The main form of deep packet inspection. Existing research can roughly divide deep packet inspection technology into three categories: string matching algorithms based on heuristics; matching algorithms based on filters; matching algorithms based on state machines. [0003] 1. The heuristic-based string matching algorithm mainly uses BM algorithm, WM algorithm and their various improved algorithms. [0004] 2. Filter-based matching algorithms mainly use filters such as bloom filters for matching. [0005] 3. The st...

AI Technical Summary

Problems solved by technology

Such packet detection makes a large number of non-attack data packets also match each byte, which will waste a lot of time in the detection of non-attack data packets, which is not conducive to the improvement of time efficiency

Images