Candidate selection algorithms in opportunistic routing based on distance progress

Abstract

Abstract-Opportunistic Routing (OR) is a new class of routing protocols that selects the next-hop forwarder on-the-fly. In contrast to traditionally routing, OR does not select a single node as the next-hop forwarder, but a set of forwarder candidates. When a packet is transmitted, the candidates coordinate such that the best one receiving the packet will forward it, while the others will discard the packet. The selection and prioritization of candidates, referred to as candidate selection algorithm, has a great impact on OR performance. In this paper we propose and study two new candidate selection algorithms based on the geographic position of nodes. This information is used by the candidate selection algorithms in order to maximize the distance progress towards the destination. We compare our proposals with other well-known candidate selection algorithms proposed in the literature through mathematical analysis and simulation. We show that candidate selection algorithms based on distance progress achieve almost the same performance as the optimum algorithms proposed in the literature, while the computational cost is dramatically reduced.

Some Efficient Algorithms to Deal With Redundancy Allocation Problems

“…To solve our problem using the VCSP model and the UGF introduced above and compare it with the previous approaches, we adapted and extended the forward checking algorithm [19], which consists of constructing a solution, by considering assignments to variables in a particular order; an order where the constraints are satisfied. The vector in our case is a set of assignments of devices to subsystems.…”