04-05-2011, 04:38 PM
Abstract
Using location information to help routing is often proposed as a means to achieve scalability in large mobile ad hocnetworks. However, location-based routing is difficult when there are holes in the network topology and nodes are mobile or frequentlydisconnected to save battery. Terminode routing, presented here, addresses these issues. It uses a combination of location-basedrouting (Terminode Remote Routing, TRR), used when the destination is far, and page link state routing (Terminode Local Routing, TLR),used when the destination is close. TRR uses anchored paths, a list of geographic points (not nodes) used as loose source routinginformation. Anchored paths are discovered and managed by sources, using one of two low overhead protocols: Friend Assisted PathDiscovery and Geographical Map-based Path Discovery. Our simulation results show that terminode routing performs well in networksof various sizes. In smaller networks, the performance is comparable to MANET routing protocols. In larger networks that are notuniformly populated with nodes, terminode routing outperforms existing location-based or MANET routing protocols.Index Terms—Restricted random waypoint, mobility model, ad hoc network, scalable routing, location-based routing method,robustness to location inaccuracy.
1 INTRODUCTION
Location-Based Routing for Scalability. Many existingrouting protocols (DSDV [30], WRP [27], OLSR [21], FSR[20], LANDMAR [12], DSR [8], AODV [29], TORA [28],CBRP [22]), proposed within the MANET [26] workinggroup of IETF, are designed to scale in networks of a fewhundred nodes. They rely on state concerning all links inthe network or links on a route between a source and adestination. This may result in poor scaling properties inlarger mobile ad hoc networks or when nodes frequentlydisconnect to save battery. More recently, there has been agrowing focus on a class of routing algorithms that relylargely, or completely, on location (and possibly mobility)information. These algorithms improve network scalabilityby reducing the total routing overhead. The idea is to uselocation information in order to reduce propagation ofcontrol messages (LAR [37]), to control packet flooding(DREAM [2]), to reduce intermediate system functions or tomake simplified packet forwarding decisions (GPSR [23],GFG [7], and GRA [33]).Issues with Existing Location Based Routing. LAR is anon-demand routing protocol where location information isused to reduce the search space for a desired route. Thesource uses the last known destination location in order toestimate the zone in which the destination is expected to befound. This is used to determine a request zone, as a set of nodes that should forward route requests. DREAM proactivelymaintains location information at each node inrouting tables and data packets are partially flooded tonodes in the direction of the destination. GPSR [23], GFG[7], and GRA [33] use only neighbor location informationfor forwarding data packets. Routing is done in a greedyway by forwarding the packet to a neighbor closer to thephysical location of the destination. This local optimalchoice repeats at each intermediate node until the destinationis reached. When the greedy process fails, GPSR andGFG route the packet around the problem region usingperimeter mode packet forwarding. Perimeter mode forwardsthe packet using a planar graph traversal. Theknowledge of locations of its one-hop neighbors is sufficientfor a node to determine its local view of the planar graph.An issue with perimeter mode is that it may give a very badpath in large networks when the source and destination arenot well connected along a straight line.With GRA, when the greedy method fails, a distributedbreadth-first or depth-first route discovery method isinvoked to find an acyclic path to the destination. Theproblem with this method is that the discovery andmaintenance of such paths can result in large overheadfor large mobile ad hoc networks.Further, in location-based routing protocols, sourcesshould know destination locations accurately enough forpackets to reach, or come close to their destination.However, it is very difficult for the location managementservice to maintain accurate location information at alltimes. This is especially true if nodes are close and theirrelative locations change frequently. Existing location-basedrouting protocols do not address how to cope with locationmanagement inaccuracies.
Download full report
http://googleurl?sa=t&source=web&cd=1&ve....1.113.705%26rep%3Drep1%26type%3Dpdf&ei=OjPBTb-ROMnhrAftrYz0Aw&usg=AFQjCNERo_66Rck0HtFoX3J22Dk7J9K0Zg&sig2=y9XY97JEwtf_21y2cEx7Rw