Optimized Handover Algor ithm Based on Stackelberg Games in CBTC Systems for Urban Rail Transit

Abstract

Communications Based Train Control (CBTC) systems have become indispensable for train operation to improve the safety and efficiency of train running to the utmost extent at present. In this paper, the problem of train handover in wireless LAN of CBTC systems for urban rail transit was taken into consideration, which was formulated as a market competition with Stackelberg games and cooperative diversity to obtain the reasonable resource allocation for trains. Under the environment of continuous train-to-wayside communications with high reliability, the successful train handover probability was analyzed in detail. Firstly, the relay triggering during handover was introduced, and the price strategies were determined in terms of the bandwidth requested by trains. Then, the utility function of train, based on price and revenue, was designed, and the existence of Nash equilibrium was proved. To achieve the optimal bandwidth strategies of train and network revenue, a distributed iterative learning scheme was proposed to figure out Nash equilibrium. Finally, the train handover process was further analyzed. The results of our study showed that the proposed strategy can motivate relay nodes to participate in cooperation for reasonable resource allocation of networks, the train handover time was less than 50ms and the successful train handover probability was remarkably higher than that of the traditional handover of RSS (Received Signal Strength).