Peer-to-peer-like telematics service discovery using the distributed-region-server-based approach

Abstract

In this study, a Telematics service discovery mechanism called Distributed telematics peer-to-peer for mobile users based on the peer-to-peer-like paradigm is proposed. When mobile users travel on the road, they may request for some services from service providers to meet their demands. Mobile users in vehicles are assumed to go through a lot of regions, in which each region is associated with a region server. Related information of all service providers in a region is stored in its region server, and each region server is in charge of service discovery requests sent from all vehicles that are currently in its region. Since mobile users in vehicles are moving and thus will pass by many regions on their ways to their destinations, a region server decision scheme should be devised in order to maintain correct connectivity between mobile users and their associated region servers. Then, a telematics service discovery mechanism can have awareness of its region server and correctly send requests to its associated region server. An application layer handoff mechanism is proposed to handoff data between region servers when the mobile user moves from one region to another region. Performance analyses of the proposed schemes are also depicted in the paper.

On optimally reducing search costs for placing replicas in Symphony P2P networks

“…Particularly, Algorithm 1 maintains two two-dimensional (r +1)×d i tables, Q i (w, ℓ)a n dR i (w, ℓ), at each node n i except the root. In addition, this algorithm computes the number of queries contributed by a leaf node as well as the optimal replica placement in lines 2-10 and then does that for a non-leaf and non-root node in lines [11][12][13][14][15][16][17][18][19]. After all the tables are constructed following the postorder of the lookup tree, the minimum number of requests sent or forwarded by all nodes in the tree for searching a copy of the object can be obtained by calculating the value of the entry Q root (r,0) (line 20).…”

“…Chord [8], Tapestry [9], Pastry [10], CAN [11] and OpenDHT [12]) or randomised (e.g. Symphony [3]), are widely adopted for content‐sharing [13, 14], telematics [15] and data streaming [16, 17]. More importantly, such network systems can operate across different organisations [18].…”