Two novel approaches for determining the minimum number of transmissions in multicast packet radio networks

This paper describes two practical techniques which can be implemented to compute optimum or near optimum paths from a single source to multiple destinations in a packet radio network (PRN) environment. This problem is usually solved by making copies of the packet, then addressing and sending them along paths which have been computed independently without concern for a global optimization. This form of solution requires a greater communications channel bandwidth yet is frequently tolerated because determination of an optimal solution yielding a minimal number of transmissions is NP-complete. This paper proposes two alternative solutions to the problem, one of which is based on neural networks and the other employing heuristic algorithms. Extensive simulation results show that both methods compare very favorably with the exact solution method (exhaustive search).