Research On Reliability Optimization Of All-Terminal Network Based On Improved Ant Colony Algorithm

For the effect of computer networks in the social development and the importance of people's lives, scholars from various countries have put a lot of energy research on and analyze the performance of computer network, and optimize its topology structure. Especially in recent years has been "network"everywhere in society, people rely on computer networks at the same time pay more attention to its reliability, computer network collapse will lead to government, business, transportation, medical institutions and other departments of different degree of paralysis. Therefore, the reliability of computer network reliability has become a hot research field.Considering that the computer network design is the core of the backbone network design, and the terminal is the key index of reliability performance, based on the above analysis, this paper aims to solve the problem of all-terminal networks reliability optimization. For the operator maximum benefit demand, put forward the all-terminal network cost optimization problem, and for customer safety highest demand, put forward all-terminal networks reliability optimization problem; at the same time in order to balance the interests of the two common demands, put forward multi-objective all terminal networks reliability optimization problems. In the optimization model, combined with the modeling idea of Thailand scholar Kanyapat Watcharasitthiwat in solutions of network optimization, the characteristic of the problem put forward specific model.Domestic and foreign scholars have been confirmed that all terminal networks reliability calculation problem is a NP-hard problem, due to its complexity, the traditional heuristic method and enumeration method is not applicable to large-scale network optimization design problem, but new intelligent algorithm to make up for this deficiency, and a large number of literature suggests, intelligent algorithm in solving combinatorial optimization problems has certain advantages. The ant colony algorithm in solving combinatorial optimization problems, showed strong searching ability and good performance, and in many areas has been widely used. But the traditional ant colony algorithm also has this some drawbacks, mainly embodied in the slow convergence speed; the computing time is long, easy to fall into local optimal problems. In view of the above problems, in this paper, the basic ant algorithm is improved and applied to the model of optimization; the results show that the improved ant colony algorithm in cruise speed and optimization effect is improved obviously.The simulation experiment results show that the improved ant colony algorithm to solve the optimization problem of all-terminal networks reliability is very effective. For telecommunications operators to design a reliable backbone network provides an executable program for reference.