Abstract
Abstract In this paper we present a neural-based algorithm for topological via-minimization (TVM) problem in two-layer channels. TVM problem requires not only assigning wires or nets between terminals without an intersection to one of the two layers, but also a minimization of the number of vias, which are the single contacts of nets between two layers. The proposed algorithm which is designed to embed the maximum numbers of nets without an intersection, uses gradient ascent learning of the coefficients to help the Hopfield network escape from local minima and find a global minimum. The proposed algorithm is applied to the split rectangular TVM (RTVM) problem and simulations are performed. The experimental results show that the proposed algorithm generates much better solutions than other existing algorithms for this problem.
