A data fusion anti-tampering method based on input data compression representation association analysis

Abstract

The invention provides a data fusion anti-tampering method based on input data compression representation association analysis, the method includes a network node data feature compression mechanism, which compresses the original input data input to represent data features, ensures data representation accuracy, and shields different data The difference of types; the method includes a sub-dataset collection method based on the Boost principle, which divides the data set reasonably with equal probability of replacement to form the input data vector for training sub-classifiers; the method includes a method based on Adaboost The sub-classifier and tamper detection model construction method, using the divided sub-data sets to train the sub-classifier, and finally combining the sub-classifiers to build an AdaBoost classifier to predict the type of data vectors of related network nodes, and apply the voting principle to make the final prediction result , and use this mechanism to quickly identify tampered data fusion results and prevent malicious data fusion operations.

Description

technical field [0001] The invention belongs to the technical field of Internet of Things data security, and in particular relates to a data fusion anti-tampering method based on input data compression representation association analysis. Background technique [0002] As an extension and expansion of the Internet, the Internet of Things is another major change in the field of information technology. The basic characteristics of the Internet of Things are comprehensive information perception, reliable transmission and intelligent processing. Physical information is obtained through sensing devices and transmitted to relevant service nodes through the network to realize the interconnection of people-things and things-things, and realize the intelligent identification, positioning, tracking, monitoring and management of entities in the physical world. [0003] The Internet of Things is a heterogeneous architecture composed of various supporting technologies; heterogeneous netw...

AI Technical Summary

Problems solved by technology

However, this approach has obvious disadvantages: in the process of data fusion, relevant nodes need to process text according to specific encryption and decryption algorithms, and the efficiency of data fusion and processing is low

Some security fusion mechanisms to resist some malicious attacks require a large amount of redundant data processing and transmission, which is inefficient

Images