Geographic routing has been widely hailed as the most promising approach to generally scalable wireless routing. However, the correctness of all currently proposed geographic routing algorithms relies on idealized assumptions about radios and their resulting connectivity graphs. We use testbed measurements to show that these idealized assumptions are grossly violated by real radios, and that these violations cause persistent failures in geographic routing, even on static topologies. Having identified this problem, we then fix it by proposing the Cross-Link Detection Protocol (CLDP), which enables provably correct geographic routing on arbitrary connectivity graphs. We confirm in simulation and further testbed measurements that CLDP is not only correct but practical: it incurs low overhead, exhibits low path stretch, always succeeds in real, static wireless networks, and converges quickly after topology changes. 1

In Wireless Sensor Networks, many applications like structure monitoring require collecting all data without loss from motes. End-to-end retransmission which is used in Internet for reliable transport layer, does not work well in Wireless Sensor Networks, since wireless communication, and constrained resources give new challenges. We looked at factors affecting reliability, and searched possible options. Information redundancy like retransmission, erasure code, and thick path are candidates. However, if loss is not randomly distributed, those methods does not work well. For example, when link fails, but routing table is not updated, all packets through that path will be dropped. Route fix, which tries alternative next hop after some failure, reduces correlated consecutive drops, so that information redundancy can perform well. Experiment on real testbed in Soda Hall shows that route fix with erasure code provides good reliability. And encoding and decoding of erasure code is efficient. More investigation of data (overhead, delay) will give deeper insight in comparing options. Keywords wireless sensor networks, reliable transfer, retransmission, erasure code, route fix 1.

In today’s Internet core, routers store forwarding state proportional to the number of edge networks. As the Internet grows and core line rates increase, routers require memories that are increasingly fast and large—and are correspondingly increasingly expensive and difficult to engineer. In this paper, we present Reduced-State Routing (RSR), in which core routers require state only concerning the network topology within a two-hop radius, and thus of a size independent of the total number of Internet edge networks. RSR achieves this feat by routing geographically using two sets of node addresses: virtual coordinates, that are assigned to reflect the link costs within an autonomous system; and geographic coordinates, that correspond to nodes ’ physical locations. RSR routes greedily on virtual coordinates, and falls back to face routing on geographic coordinates when greedy progress is impossible on virtual coordinates. Unlike previous geographic routing schemes, RSR works on Internetlike graphs (rather than only on wireless-like graphs), and supports policy routing. By simulating RSR on real tier-1 ISP topologies, we demonstrate that RSR achieves low path stretch, comparable to that caused by policy routing in today’s Internet. 1.

Abstract—Multi-hop communication objectives and constraints impose a set of challenging requirements that create difficult conditions for simultaneous optimization of features such as scalability and performance. We have developed field division routing (FDR), a distributed and nonhierarchical routing protocol that aims to coordinated addressing of scalability, topology alternations, latency, throughput, energy efficiency, and local storage requirements. FDR is based upon two optimization mechanisms: a reactive and focused diffusion that collects only network topology information directly required for making localized routing decisions, and a protocol for sharing routing information among neighboring nodes. Routing table initialization and maintenance are scalable in terms of both storage and overhead traffic. FDR provides guaranteed connectivity while providing near-optimal all-node-pairs message delivery. The protocol is also power-efficient to a wide spectrum of topology changes that induce relatively few messages to update routing tables networkwide. We analyzed FDR both theoretically and using simulation. Index Terms—Ad-hoc networks, routing, multi-hop communication. I.

Copyright © 2010 Milenko Drinić et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Multihop communication objectives and constraints impose a set of challenging requirements that create difficult conditions for simultaneous optimization of features such as scalability and performance. Routing in wireless multihop networks represents a crucial component of the overall network efficacy because it is a lower layer that enables the actual functionality of networks. We have developed field division routing (FDR), a distributed and nonhierarchical routing protocol that aims to coordinated addressing of scalability, topology alternations, latency, throughput, energy efficiency, and local storage requirements. FDR is based upon two optimization mechanisms: a reactive and focused diffusion that collects only network topology information directly required for making localized routing decisions, and a protocol for sharing routing information among neighboring nodes. Routing table initialization and maintenance are scalable in terms of both storage and overhead traffic necessary for building routing tables. FDR provides guaranteed connectivity while providing near-optimal all-node-pairs message delivery. The protocol is also power-efficient to a wide spectrum of topology changes that induce relatively few messages to update routing tables networkwide. We analyzed the new routing protocol both theoretically and using simulation. 1.

Abstract-Multi-hop communication objectives and constraints impose a set of challenging requirements that create difficult conditions for simultaneous optimization of features such as scalability and performance. We have developed field division routing (FDR), a distributed and nonhierarchical routing protocol that aims to coordinated addressing of scalability, topology alternations, latency, throughput, energy efficiency, and local storage requirements. FDR is based upon two optimization mechanisms: a reactive and focused diffusion that collects only network topology information directly required for making localized routing decisions, and a protocol for sharing routing information among neighboring nodes. Routing table initialization and maintenance are scalable in terms of both storage and overhead traffic. FDR provides guaranteed connectivity while providing near-optimal all-node-pairs message delivery. The protocol is also power-efficient to a wide spectrum of topology changes that induce relatively few messages to update routing tables networkwide. We analyzed FDR both theoretically and using simulation. Index Terms-Ad-hoc networks, routing, multi-hop communication. I.