Study Of Adaptive Reconfiguration Model For Enhancing Survivability Of System

The traffic system is heterogeneous, distributed, dynamic and it has massive information, which may lead to multiform traffic emergencies. But the current technology can't make the traffic service system work in the face of all kinds of accidents, failures and even man-made damages. This implies that, it can't meet survivable need of traffic service system. Therefore how to improve the survivable capability of traffic service system has become an important research topic in the area of intelligent transportation.Biological immune system as a complex, highly parallel, distributed and adaptive information-processing system can identify and eliminate intruding antigenic materials, while possessing the ability of learning, memory and adaptive adjustment. It can maintain inner stability of bodies. Based on immunology principle, this paper studies survivability of traffic coordinative service system, aim at simulating immune mechanism and architecture of biological immune system, using immune-related principles and algorithm, solving the survivability problems that can't be easily solved by traditional ways, so that the constructed traffic service system has the desired survivability features.This paper first analyzes the related research on survivability, forms systematism theoretical framework about survivability research, and summarizes key technologies in the design of the survivability, which provide the theoretical evidence for its related research. Then, biological immune system, bionic immune mechanism and artificial immune system are summarized. Combined with the characteristic of traffic coordinative service system, the paper proposes an adaptive reconfiguration model for enhanced survivability of system, and applies the model to traffic coordination service system. The model can detect the menaces that the system suffered in real time, including the external attack and internal failure. It can give an appropriate strategy to deal with the menaces, combined with the effect of disposal, learn and memorize this process, and reconfigurate system. When the same or similar menace arises, the system can respond more quickly. The whole process does not need people to participate, with autonomous and adaptive capacity. At last, each unit function realization and artificial immune algorithm which is suitable for traffic coordinative service system are given in this paper, the algorithm is applied in multi-objective optimization for service reconfiguration, the enhanced survivability technology in traffic service system is elementarily realized. The experimental results show that, when the model deals with the same type of attack, with the increasing of processing, reaction time has been greatly improved.