Information Security Risk Assessment Method Study Based On QPSO Wavelet Neural Network

The face of today's growing security needs of information system security, has not simply on technology tofundamentally solve the problem of information systems security, but should be from the perspective of systems engineering of information system security issues. Systems engineering in the information system security risk assessment is the basis and premise, it plays a very important place position. Through the information security risk assessment, we can understand the current and future information systems risk assessment of these risks may be caused by security threats and the impact for the establishment of information systems to determine the security policy and ensure the safe operation of the system to provide the fundamental basis. Therefore, today's information security risk assessment has become an increasingly pressing issue, developed countries attracted great attention, they believe: the lack of effective information security risk assessment will result in demand for information security and information security solutions, the serious out of touch, risk assessment must be institutionalized.This paper introduces the domestic and international information security risk assessment and concept development status, and then analyzes the relationships between the elements of risk assessment and evaluation process, introduces several of the most representatives of the risk assessment methodology, summarizes the advantages of these assessment methods shortcomings. Particle swarm optimization is proposed wavelet neural network risk assessment methods, mainly based on information system security risk assessment study to fuzzy mathematics, neural networks, wavelet analysis, such as particle swarm theory as a tool for systematic study of traditional wavelet-based neural network assessment methods and the use of quantum particle swarm algorithm to train fuzzy neural network, and its key technologies in-depth research in this rich and developed based on wavelet neural network, and explore to improve the information system security risk assessment results new ways, the proposed risk assessment methodology for information system security risk assessment in full swing with a very important significance.In this paper, research in the following three aspects: (1) Based on Quantum Particle Swarm Optimization Fuzzy Wavelet Neural Network Neural network can be viewed as a link artificial neural network, based on the wavelet function as the base function. The traditional neural network training algorithm deficiencies, but also on improved particle swarm optimization algorithm, first proposed fuzzy quantum particle swarm optimization algorithm, and used for training neural networks, propose a new neural network model. Propose a quantum particle swarm optimization-based neural network training algorithm, the parameters of wavelet neural network to form a multi-dimensional vector, as the algorithm in the evolution of particles, which in the feasible solution space search for optimal solutions within.(2) And other risk assessment methods are comparedIn this paper, Matlab7.1 environment on the quantum particle swarm optimization method for the assessment of neural network to do the simulation experiments, and were from the convergence of the method, the training accuracy and prediction accuracy of the three aspects of the traditional algorithm based on BP neural network method are compared. Simulation results show that the quantum particle swarm optimization of wavelet neural network convergence speed and training accuracy, prediction effect, has great advantages(3) A quantum particle swarm optimization based on fuzzy wavelet neural network algorithm for information systems security risk assessment modelInformation system security risk assessment based on the actual situation, proposed a quantum particle swarm fuzzy wavelet neural network evaluation model. An information system and risk assessment, evaluation results and the system compares the actual security situation, with good consistency. The results showed that: this approach is feasible; the evaluation method is scientific, objective, and reasonable.