Multi-hop Routing in Wireless Sensor Networks by Shalli Rani & Syed Hassan Ahmed

Author:Shalli Rani & Syed Hassan Ahmed
Language: eng
Format: epub
Publisher: Springer Singapore, Singapore

3.2.2 Query Based Protocols

In some applications the nodes respond only when data request is send to them. For example if node A1 needs some data d, it will forward its request (query) to the neighbor node A2 which will be entertained by that node in case availability of the same data (response) otherwise it can send report to the node A1 with negative message. It is also possible that query is broadcasted to all the nodes and node having the data of that query replies back. Many protocols fall under this category (Fig. 3.3), some of them are Directed Diffusion [14], TEEN [15], Enhanced APTEEN [16] etc. A message interest is a query , which describes the need of the user. Physical phenomenon is portrayed by the data and sensed phenomena by event and gradient describes the direction state where each node receives interest. Gradients are attributed by data rate, interval and expiration time. Tasks are names designed with the attribute and value pairs. It proposed the reactive routing techniques. Routes are established on demand only. Diffusion present in this protocol is different from the traditional network as it is centric and data communication is neighbor to neighbor. They do not require global identifiers. It is shown that it is different from ad-hoc reactive routing scheme as it uses directional flooding and redundant messages are sent on the multiple paths. After that this multiplicity is reduced based on the route performance. The naming technique of this protocol is used to save the energy by avoiding unnecessary operations of routing layer. When a route between node and the sink fails, then new path is identified by the reinforcement method that is done by searching the path between the nodes operating at the lower speeds. But this protocol is application specific as it works for the on demand query and is not suitable for the applications with the requirement of continuous transmission of data. Naming design is also application dependent, which should be known in advance. To match the query with data also requires extra control overheads and needs more processing. In some query based routing schemes, query is broadcasted to all the nodes and it works like flooding. But when data required by nodes is at small scale then flooding is not the good approach. So events, which are less in number, can be flooded rather than query that are large in number [17]. In this case query will be forwarded to the nodes, which have sensed some events. When some node detects some event then it is stored in the event table and an agent is generated which propagates information about local events to distant nodes. If the node has information of route of event then query will be transmitted on that path otherwise it is forwarded in the random direction until it reaches the target location. It proved better in the cases where geographical routing cannot be established. It shows its validation over event flooding. But overhead of generating of the agent during an event require more processing at the node’s end.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.