Cost-driven repair optimization of reconfigurable nanowire crossbar systems with clustered defects

With the recent development of nanoscale materials and assembly techniques, it is envisioned to build high-density reconfigurable systems which have never been achieved by the photolithography. Various reconfigurable architectures have been proposed based on nanowire crossbar structure as the primitive building block. Unfortunately, high-density systems consisting of nanometer-scale elements are likely to have many imperfections and variations; thus, defect tolerance is considered as one of the most exigent challenges. In this paper, we evaluate three different logic mapping algorithms with defect tolerance to circumvent clustered defective crosspoints in nanowire reconfigurable crossbar architectures. The effectiveness of inherited redundancy and configurability utilization is demonstrated through extensive parametric simulations. Then, costs associated with the repair process are analyzed and a method to find the most cost-effective repair solution is presented.