| Summary: | 碩士 === 國立成功大學 === 資訊工程學系碩博士班 === 94 === In the field of mobile ad-hoc networks, energy-awareness schemes are more complex and important. How to save the energy of devices is an important topic that is worth probing into. In relevant research, for instance, the way of GAF divides the simulation area in a lot of square grids. There are so many nodes but only a coordinator needs to wake up in the same grid. Coordinator helps other nodes transfer or handle the management packets so that other nodes can sleep in the same grid. Such this method saves a lot of energy.
GAF has really saved a large amount of energy, but there is still something we can improve. First is when the all nodes are inside the same grid, we hope that the nodes which have more residual energy can reduce its sleep time. Therefore they have a bigger chance to compete and be a coordinator. Similarly, we hope that the nodes which have less residual energy nodes can increase its sleep time. Therefore they can save energy because they avoid wasting energy on competition and keeping nodes hold on idle state. According this way, network lifetime is extended.
The second is we want to balance the average residual energy between grid and grid. If there is an average residual energy of all neighbor grids including myself grid and is greater than the average residual energy of all each neighbor grids. Coordinator assigns a node which just wakes up to the smallest average residual energy of a neighbor grid. Therefore the node can help the neighbor grid transfer data packets and handle management packets. If GAF can follow this way, network lifetime is extended in most cases.
The third is we hope when AODV sets up a routing path in mobile ad hoc network with GAF, it can also choose the coordinator by the ratio between timer interval and residual energy of all coordinators to set up a reverse path.
We have studied in EPR and find out that EPR can extend network lifetime efficiently according to balance the forwarding data between coordinators. Therefore, we adopt the concept of EPR and implement it into our algorithm. The modified method of the path routing we have talked above, it is still a lot of work to do about the issue of extending the network lifetime.
The goals we briefly described above are that the nodes which have more residual energy have more chances to be a coordinator. The nodes which have less residual energy sleep more time and have less chance to be a coordinator as well. Such methods balance the average residual energy inside grid and among grids, and eventually the lifetime of whole mobile ad-hoc network is extended.
|
|---|
