The Reconfiguration Optimization Mechanism For Bio-inspired Self-repairing Hardware

The digital circuit of mechanical and electrical equipment faces complex and harsh working environment tests in the field of deep space and sea exploration and nuclear application.So it is prone to failure and is difficult or even impossible to maintain by manpower.How to improve its environmental adaptability is the key to improve the mission reliability of mechanical and electrical equipment.Bio-inspired self-repairing hardware based on the reconfigurable electronic cell array structure is an effective way to improve the reliability of the digital circuit.It can automatically and dynamically remove and replace the faulty cells to realize the self-repairing function.In view of the research,it is found that the existing reconfiguration mechanism for bio-inspired self-repairing hardware has some problems in engineering applications,such as low utilization of spare cells and the degradation of the sequential performance of the reconstructed circuit.Therefore,this paper is based on the application requirements,focusing on the performance optimization mechanism to study.The main research work and content are as follows:1.Based on the analysis of the bio-inspired self-repairing hardware structure,the reconfiguration principle is elaborated by the set,graph and mapping theory.And the reconfiguration optimization mechanism studied from placement and routing is prosed.2.Given the placement problem of bio-inspired self-repairing hardware,the off-line placement algorithm is used for reference,and a placement mechanism based on variable neighborhood search algorithm is proposed.The optimization objective and neighborhood structure are designed and the specific implementation steps of the algorithm are described.3.Given the routing problem of bio-inspired self-repairing hardware,a dynamic routing mechanism based on real-time flow is proposed by using the Pathfinder routing algorithm for reference.The mechanism is taking the maximum reuse of the original net wire as the main idea and is designed in detail from the two stages of path removal and reconstruction.4.The MCNC20 benchmark circuit file is parsed,and the simulation platform is built.The reconfiguration optimization mechanism proposed in this paper is compared with the row and column removal mechanism from two aspects of spare cell utilization and the total length of the net wire to verify its effectiveness and feasibility.