Fuzzy Logic-based Battle Field Mobility Prediction Method in War-game Simulation

War-game simulation models the situation of a battle field and has been used for analyzing the operation or the occupation of a troop. It is necessary to develop a fast and accurate mobility prediction method of a weapon or equipment with mobility in a battle field. However, communication resources are limited and links are frequently disconnected. A mobile node moving in war-game simulation is greatly influenced by various environmental factors. This paper proposes a fuzzy logic– based battle field mobility prediction method in war-game simulation in order to predict the mobility of a mobile node and provide accurate location information with mobile nodes according to change of location. A base station communicates with mobile nodes and detects their locations. In this study, we use fuzzy logic control and perform a velocity inference on the basis of a number of rules by assigning an intelligent agent to each base station. The location error associated with FLMP is lesser than that with existing models such as the mobile distance-based mobility prediction model and the mobile velocity-based mobility prediction model. Experimental results indicate that the proposed method improves the efficiency of mobility prediction in hybrid wireless network. 1. INTRODUCTION Defense modeling and simulation has been used for evaluating fighting power or analyzing the occupation of a troop in the field of war-games. Especially it is very important to change an operation or communicate information through a weapon or equipment with mobility such as a tank and IP radio. In a battle field, the communication between a base station and weapons is done through a network environment such as a hybrid wireless network. There has been recent research on hybrid wireless networks [1] that combine a cellular network [2] with a mobile ad hoc network [3] through a gateway. In a hybrid wireless network, mobile nodes can communicate with each other, and communication between a base station and a central control station can yield a considerable amount of information about the network. Therefore, the network