Neural Network for Routing in a Directed and Weighted Graph

In this paper, we use a neural network based algorithm to find the best path in a directed and weighted graph. In this algorithm, we define a suitable energy function; the minimum of this function correspond to the best path. By using gradient descent method, the energy is minimized at the convergence of neural network. Simulation results show that this method finds the correct path between source and destination and because neurons act in parallel, the performance is comparable with other methods.In general, parameters of a learning algorithm are achieved by trial and error, but here we suggest some formulas to find the value of parameters. Upper trigger point for neurons being on is also calculated; designing neural network base on this point, gives the better functionality of the network. This algorithm can be implemented in hardware or software; the software implementation will be inspected.