The radio network is a large-scale communication network in which a large number of stations are distributed over a wide area. Each station can communicate directly by radio packets only with neighborhood stations. Packet communication between stations which cannot communicate directly is realized by passing packets through multiple stations between them. For this purpose, when a packet communication request arises, the route for the packet and the transfer timing must be chosen so that the total transfer time is minimized and the efficiency of the radio network is maximized. It should be noted that the reliability, representing the probability that the required performance is achieved, fluctuates among the stations in the network due to the differences in the time of installation, the designed performance, and communication congestion. Consequently, this paper proposes a two-stage algorithm in which the route is assigned so that its reliability is larger than the required value. The proposed algorithm is composed of the iterative reliability satisfaction extraction procedure, in which a set of path candidates satisfying the reliability condition for the given transfer request, and the greedy neural network procedure, in which the route minimizing the cost is selected from the set of path candidates. The effectiveness of the proposed algorithm is demonstrated by simulation for networks with up to 500 stations. © 2002 Scripta Technica, Electron Comm Jpn Pt 3, 85(6): 63–73, 2002

A multicast packet switching system can replicate a packet in the window of each input port to send out the copies from different output ports simultaneously. In order to maximize the throughput, a combinatorial optimization problem must be solved in real time of finding a switching configuration which does not only satisfy the constraints on the system, but also maximizes the number of copies under transmission demands. In this paper, we focus on the one-shot scheduling problem where all the copies of selected packets must be sent out simultaneously. We propose the neural network composed of W/spl times/N binary neurons for the problem in the W-window-N-input-port system. The motion equation is newly defined with three heuristic methods. We verify the performance through simulations in up to 3-window-1000-input-port systems, where our binary neural network provides the better performance than the existing methods so as to reduce the delay time under practical situations.