Vehicular ad-hoc networks (VANETs) have been gained importance for the inter-vehicle communication that supports local communication between vehicles without any expensive infrastructure and considerable configuration efforts. How to provide light-weight and scalable location management service which facilitates geographic routing in VANETs remains a fundamental issue. In this paper we will present a novel peer-to-peer location management protocol, called PLM, to provide location management service in VANETs. PLM makes use of high mobility in VANETs to disseminate vehicles’ historical location information over the network. A vehicle is able to predict current location of other vehicles with Kalman filtering technique. Our theoretical analysis shows that PLM is able to achieve high location information availability with a low protocol overhead and latency. The simulation results indicate that PLM can provide fairly accurate location information with quite low communication overhead in VANETs.. [Article copies are available for purchase from InfoSci-on-Demand.com]
TopIntroduction
Vehicular Ad-hoc Networks (VANETs) have received considerable attention in recent years due to its potential in supporting vast value-added and customized applications. By leveraging low cost and high bandwidth wireless interface installed on each vehicle, a VANET can be deployed quickly and economically. It provides high data rates, and is more cost-effective than systems that need pre-installed infrastructures (e.g. cellular networks). Hence a wide-range of applications such as increasing road safety, providing passengers with business information and entertainment, and improving traffic flow and efficiency on the roads, can be provided using VANETs as the communication infrastructure (Riva, O., Nadeem, T., Borcea, C., & Iftode, L., 2007; Morris, R., Jannotti, J., Kaashoek, F., Li, J., & Decouto, D., 2000; Dikaiakos, M., Florides, A., Nadeem, T., & Iftode, L., 2007; Festag, A., Fler, H., Hartenstein, H., Sarma, A., & Schmitz, R., 2004). Because of the short communication range of each network interface, multi-hop data transmission may be needed for end-to-end data delivery in VANETs. Recent study in multi-hop routing in VANETs (Zhao, J., & Cao, G., 2006; Mo, Z., Zhu, H., Makki, K., & Pissinou, N., 2006; Naumov, V., Baumann, R., & Gross, T., 2006; Sun, W., Yamaguchi, H., Yukimasa, K., & Kusumoto, S., 2000; Lochert, C., Hartenstein, H., Tian, J., Fler, H., Herrmann, D., & Mauve, M., 2003) has shown that, with the global position system (GPS) and digital map (e.g. Map Mechanics, 2005), geographic routing, in which data packets are forwarded from the source to the destination with the aid of nodes’ location information, has high end-to-end packet delivery ratio, low end-to-end delay and low control overhead. In Zhao et. Al studied the carry-and-forward scheme, named VADD, which delivers packets in sparse VANETs where disconnection happens frequently. In Mo, Z. et. Al (2006) a multi-hop routing protocol, called MURU, was proposed to set up robust end-to-end path for urban VANETs. An advanced greedy forwarding (AFG) (Naumov, V., et. Al, 2006) based on GPSR was proposed to increase packet delivery ratio in VANETs. All these protocols assume an efficient location management service is available to provide the source node with the destination’s location. Therefore, it is fundamentally important to design a good location management scheme in VANETs to support geographic routing and other location-based applications.
Up to date some location management schemes have been proposed in mobile wireless ad hoc networks (MANETs) (Li, J., Jannotti, J., Decouto, D., Karger, D., & Morris, R., 2000; Xue, Y., Li, B., & Nahrstedt, K., 2001; Kiess, W., Fler, H., Widmer, J., & Mauve, M., 2004; Basagni, S., Chalamtac, I., Syrotiuk, V. R., & Woodward, B. A., 1998; Sasson, Y., Cavin, D., & Schiper, A., 2005) and most of them are grid-based. That is: they divide the network area into ordered grids and select location servers based on a particular grid mapping algorithm. As all these schemes adopt the client-server architecture, they usually involve two phases: location update and location query. In the location update process, mobile nodes periodically send their up-to-date location information to one or more location servers to update their locations. In the location query process, a node queries target node’s location servers for target node’s location information.