Design And Implementation Of Quantum Classification Circuit And Its Application In Network

With the continuous development of big data business and the extensive application of artificial intelligence and 5G technology,the scale of data has grown explosively,how to efficiently handle data has become an urgent problem to be solved.Quantum computing takes advantage of quantum superposition and entanglement to handle computing tasks,which has obvious advantages in speeding up solving specific problems.In recent years,it has become a research trend to apply quantum computing to classical machine learning algorithms in order to seek the quantum acceleration effect,so as to handle data more efficiently.However,most of the current researches on quantum machine learning algorithms are theoretical derivation,and rarely involve the research on the design methods of quantum circuits used to implement algorithms.Therefore,three improved quantum classification algorithms and corresponding quantum classification circuits are proposed in this thesis,and applied in the task of network flow classification.The work of this thesis can be summarized as the following three points:First,by analyzing the shortcomings of the classical k-nearest neighbor algorithm,the classical k-means algorithm and the classical perceptron algorithm in big data classification problems,we quantize the process of the above algorithms,and design the corresponding quantum classification circuits based on the modified quantum classification algorithms in this thesis.Through the analysis of the design method and performance of the quantum classification circuits,it is proved that the quantum classification circuits designed in this thesis have significant advantages in reducing storage requirements and algorithm complexity compared with the classical classification algorithms.Second,in order to alleviate the pressure of efficient classification of network flows caused by the massive data traffic faced by cloud data center networks,we conduct the simulation experiments of network flow classification on the simulation platform of IBM quantum cloud platform based on the three modified quantum classification algorithms.The experimental results show that the three modified quantum classification algorithms can achieve a high classification accuracy of network flows.At the same time,the experiments also verify the feasibility and practicability of the quantum classification circuits designed in this thesis.Third,by using the real quantum computing equipment provided by the IBM quantum cloud platform,three modified quantum classification algorithms and corresponding quantum classification circuits are verified experimentally in this thesis.Although the experimental results are affected by noise,they are in line with expectations,which proves the effectiveness of the quantum circuits designed in this thesis in real environment.