Placement with self-organising neural networks

This paper introduces a neural network approach to node placement in an arbitrarily shaped rectilinear boundary based on self-organising principle. An abstract specification of the design is converted to a set of appropriate input vectors fed to the network at random. At the end of the process, the map shows the rectilinear shape 2-dimensional plane of the design in which the modules with higher connectivity to each other and also to some external ports are placed close to each other, hence minimising total connection length in the design