Tasarsız mobil ağlar için SPGR temelli enerji verimli yeni bir yönlendirme tekniği

Wireless Sensor Networks is a group of a spatially dispersed and densely deployed sensors. When the sensor nodes of WSN become mobile and the network does not have the stable infrastructure, it becomes an ad hoc network. Ad hoc networks have become more popular nowadays, especially for people who always use laptops and other portable devices. Wireless networks enable us to have access to the internet regardless of where we are. With the development of mobile equipment, such as mobile phones and laptops, we want to have an uninterrupted service while using wireless networks, even when we are moving from one area to another. It is more convenient compared to the wired Ethernet or other network access. In this respect battery lifetime of these type equipments and energy-efficiency of ad hoc networks have been a crucial issue in recent years. In his work we have presented an energy-efficient geographic routing technique based on the shortest path based routing method. Firstly, geographic forwarding was studied in terms of energy-efficiency. The distinctive feature of the geographic forwarding is the ability to be aware of the geographic location of nodes. Using location information, the nodes can develop more reliable packet forwarding function. Then, a new energy-efficient geographic routing technique is introduced. According to the proposed technique the source and the destination are linked with a virtual Euclidean line. During the selection process of the next forwarder node among the neighbors, the node tries to choose the one which is nearest to the Euclidean line and also to the destination node. Selection method is the function of the distance between the neighbor and the sink; and the distance between the neighbor and the virtual Euclidean line. In other words, the node selects the neighbor, which the minimal distance to the sink and Euclidean line, this approach guarantees the delivering of the packets along the shortest path. Additionally, proposed technique takes into consideration battery lifetime of the network node. Selection function includes the energy level parameter of the neighbor nodes: the forwarder node sends the packet to the neighbor, which is not only in the shortest path to the sink, but also has a maximum energy level. Therefore, the closest neighbor selection function introduced combines the two parameters: distance and energy level. The method can be rearranged according using the weighted value according to the demands of the network. The offered technique also uses the backtracking method. If the current forwarder node cannot find the next one among its neighbors, which is in shortest path in a comparison with itself, it re-forwards the packet to the previous node. The technique was simulated on OPNET tool. We have used three scenarios which differs in mobility type. In the first scenario we have tested the technique employing absolutely static network. The second scenario is partially mobile: half of the network nodes are mobile, the rest is static. The third scenario is absolutely dynamic scenario, where all the nodes of the network are mobile. All scenarios were tested under the exponential and the Pareto traffic with the different parameters. We have compared the proposed technique with Geographic Routing Protocol (GRP) and Shortest Path Based Geographic Routing (SPGR), on which our method is based. In a development of the these techniques energy parameter of the network nodes was not taken into the consideration. All the techniques were compared in terms of the delay, average number of the received data packets. In order to analyze the energy consuming parameter of the methods mentioned above, we have integrated energy parameter to the GRP and SPGR. It is shown that the technique introduced has low delay and a large number of received packets compared to the GRP technique and SPGR technique. Also Average Number of Received Data packets are also considerable higher in our technique. To sum up, we have studied the geographic routing based techniques for ad hoc networks. Using the main idea of SPGR forwarding, we have integrated the energy-efficiency parameter and developed the neighbor selection function. The node chooses the neighbor with the maximum energy level and with the shortest distance to the sink considering the virtual line between the source and the sink. In the future work, we are planning to improve the performance of the Angular Routing Protocol. The integration of energy parameter will be studied extensively. Furthermore, the possible modification of the protocol to improve its performance results will also be considered. Multi-source transmissions will be considered as a future work.