A Light-Weight Moving Preferences BasedDynamic Location Management Scheme Using Road Map and GPS

• Main Authors: Hsiang-Tsai Chung, 鍾享材
• Other Authors: Eric Hsiaokuang Wu
• Format: Others
• Language: en_US
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/91647906156400977613

碩士 === 國立中央大學 === 資訊工程研究所 === 92 === ABSTRACT Future microcellular Personal Communication Service (PCS) networks are expected to be high user density and high mobility. Location managements in the PCS networks will incur a large amount of signaling traffic. Several dynamic location management schemes based on different strategies are proposed to reduce the signal costs both of location update (LU) and paging (PG). The velocity-based scheme, an extension of distance-based, is known as the best of conventional dynamic location update schemes. However, due to lack of effective mobility prediction capabilities, these schemes which consider only the optimal size of the LA regardless the optimal shape of the LA will perform inefficiently when mobile terminals move in a higher speed. To satisfy user requirements of high mobility mentioned above, we proposed a light-weight dynamic location management scheme called “Moving Preference” for mobility prediction in this paper. On the other hand, since MTs carried in vehicle are the ones demonstrate high mobility, an idea of incorporation of location-based information (i.e. road map & GPS) is introduced in the proposed scheme. The “Moving Preference” which is a mobility prediction technique derived from mobile terminal’s long-term moving history. In our basic idea, we are more interested in the destinations (moving preferences) a mobile terminal is likely to leave or stay rather than the sequence of cells or road segments a mobile terminal is likely to pass in transit. Our scheme features light-weight which does not need to keep track all possible nodes and edges in the moving history of a pecified MT. Thus, provides better flexibility and scalability. Moreover, the mobility prediction of the proposed scheme is effective not only within a short distance, but also a longer one. Based on these features, we introduced a new cost model, both the size and shape of a LA are taken into considerations, to find out an optimal location update threshold for signal costs minimizing. The simulation results indicate that our moving preferences based (MPB) scheme do have a great improvement over conventional ones.