The main features of new ad hoc networks include large number of unattended nodes with varying capabilities, lack or impracticality of deploying supporting infrastructure, and high cost of human supervised maintenance. What is necessary for these types of networks is a class of algorithms which are scalable, tunable, distributed, easy to deploy, and most importantly easy to maintain. These large networks of low power nodes face a number of challenges: cost of deployment, capability and complexity of nodes, routing without the use of large conventional routing tables, adaptability in front of intermittent functioning regime, network partitioning and survivability. In all these networks, both basic network operations (routing, forwarding), and higher level functions (multicast, resource discovery) impose a tradeoff between communication overhead and infrastructure support. For example, a network may have high powered basestations to hold large routing tables, or it has to use flooding to discover routes on demand; it may have a GPS receiver in each node, or it has to spend some energy running a positioning algorithm. Another tradeoff is encountered in route management––either proactively maintain routes to all possible destinations, or reactively discover them when needed. Both approaches prove better than the other one under different mobility and communication conditions.
The scalability problem of very large ad hoc networks of small nodes, such as sensor networks, is that traditional routing schemes are either table based, or source based, both of which incur high costs on the network. A table based routing scheme can require a high amount of memory at each node, depending on the number of possible destinations. It can require a large amount of signaling to maintain the table in case of mobility. A source based, on demand scheme shifts the scalability problem to the discovery process, based on flooding, and to the packet overhead, sensitive to path length. A class of routing schemes that addresses these problems is using the Euclidean space as a complementary namespace for routing. These are position centric approaches, because the position of the destination functions as a delivery address, thus eliminating both the need of routing tables at forwarding nodes, and the need for complete path indication in each packet.
Position centric approaches normally assume that positions of nodes are known in the network, as node positions are used for both naming and forwarding decisions. In this paper we argue that Cartesian routing, trajectory based forwarding (TBF), and the applications they enable would work even in a network in which positions are not globally known, provided that some localized sensing of neighboring nodes is available. The central idea of localized positioning is that only nodes maintaining active communications need to be positioned with respect to each other. Positioning can be relative, that is, each group of communicating nodes maintains a different coordinate system for the purpose of that communication. A group usually means two end points and the intermediate nodes in the case of routing, but may also comprise of nodes involved in discovery, or flooding. A point to multipoint communication group for example, can use a coordinate system belonging to the source, provided that the intermediate nodes and the receivers are all registered to that coordinate system.
The necessary node functionality to support localized positioning is also required by most positioning algorithms, and includes the ability to measure ranges to neighbors, angle of arrival (AoA), or orientation of the node with respect to the North. Ranging is usually provided by estimating the distance to a neighbor by measuring the strength of the radio signal from that neighbor, or by time difference of arrival (TDoA), a method that sometimes employs radio and ultrasound signals. AoA is a method to estimate the direction from which a neighbor is sending data. It can be implemented either using an antenna array, or a combination of radio and ultrasound receivers. Digital compasses may be used in combination with AoA measurement to provide more information about a node’s orientation. All these methods have been previously used to provide or extend positioning capabilities in ad hoc networks.
If positions are not available to all nodes in a consistent fashion, due to lack or obstruction of positioning infrastructure, such as GPS satellites, the only option left is to run a distributed positioning algorithm. These algorithms however, come with their own disadvantages––they may either require separate infrastructure, like in the case of Cricket, requiring ultrasound beacons, or may require a preprocessing phase in which large portions of the network collaborates, thus imposing a high communication cost on all nodes, every time the network changes topology, like in the case of APS or AhLOS.
Instead of positioning all the nodes in the network, which may be either globally inconsistent due to mobility, or expensive due to high signaling, we propose a method to locally position only the nodes involved in point to point communication, in a coordinate system relative to the originator of the packet. This enables some position centric applications––flooding, discovery, or source–sink communication in networks in which global positioning is not available, or necessary.
The main sections of the paper are organized as follows: the next section reviews related work, Section 3 is a short review of trajectory based forwarding, Section 4 presents local positioning system (LPS), our method to position only the nodes involved in communication (forwarding the trajectory). Section 5 discusses simulation results, and we summarize with some concluding remarks in Section 6.
http://watchharrypotter7.jimdo.com/
http://watchharrypotteronlinefree.terapad.com
http://www.blurty.com/users/harrypotter7
http://www.thoughts.com/harrypotter7
http://harrypotter7full.sosblog.com/
http://www.blogtext.org/watchharrypotterandthedeathlyhal/
http://harrypotter7ful.insanejournal.com/
http://harrypottermovie2010.zoomshare.com/
http://watchharrypottermovie.wiki.zoho.com/Watch-Harry-Potter-and-the-Deathly-Hallows-Online-Free.html
http://watchharrypotteronlinefree.blog.friendster.com/
http://harrypottermovie7.xanga.com
http://harrypotter72.livejournal.com
http://harrypotterandthedeathlyhallowsonlinemovie.webs.com
http://harrypotter7fullonline.tripod.com/
http://watchharrypotterandthedeathlyhallowsonlinefree.wordpress.com
http://quizilla.teennick.com/stories/17634402/watch-harry-potter-and-the-deathly-hallows-online-free
