Dynamic Sleeping In WSN

Ashwath Narayan
Entry date: 05-Oct-2013
Final Submission: 01-Jul-2014
A wireless sensor network (WSN) consists of spatially distributed autonomous sensors to monitor physical or environmental conditions. and to cooperatively pass their data through the network to a main location.
One of the problems in WSN is the longevity of the sensor as it is placed in a remote location and runs on a ba

Project details


Wireless Sensor Networks(WSN) are used to gather information of the surroundings, these sensors are deployed in the environment and they transmit the data back to the base station. They're used by the military to monitor sensitive regions for movement of troops and also used in buildings like nuclear powerplants and hospitals to monitor the surroundings. Over the past decade, the use of WSN's has increased dramatically because of the advances in technology, but the main drawback of WSN's still persists, and that is the longevity of the sensors. They usually operate on batteries and are in remote loccations and hence cannot be recharged. This is one of the major limitations of WSN's. 

 Interdependent Node Structure for WSNA sensor network with base station


To help improve the longevity of the sensors, we can use a method of dynamically putting the sensors to sleep. This is a already known technique, but i would like to propose a more efficient and reliable way to introduce this concept. In this concept, each of the interconnected nodes usecd to survey the environment can be put to sleep based on the time of the day, the activity in the region around the node and also the deliberately by the user. 

The node(sensor) can be put to sleep by using a specific signal and be woken up using a specific signal, the other criteria required to decide whether a node can be put to sleep or not depends on the prediction of how long will the node be inactive. If the sleep time of the node is below a threshold value, then the node need not be put to sleep, because putting the node to sleep and making it active again costs some resources too. 

The activity of the node can be monitored over a period of time and detect the activity pattern, this can be used to allocate sleep times for the node. This can be achieved by performing a probablity distribution of the activity of the node. The node can be put to sleep by the base station and can be woken up by any of the nodes, this feature helps the node become active incase of an emergency or a sudden change in the environment. This can be achieved only if the nodes are interconnected and communicating with each other. 

The Dynamic sleep can be a simple software that can be installed in the WSN's to help improve their longevity.

The best way to implement this system would be by designing an algorithm based on probability theory and finding out the range in which the sensor is active. The standard deviation of the above result can be used to generate a pattern to put a sensor to sleep. 

In the initial stages, a base node is used to set up the instruction for the nodes. THe base node collects the information from the nodes to get patterns of each nodes activity. Once this is done, the base node gives the instructions to the other nodes and then becomes part of the mesh. The Flow Chart for the base node is described below

Flow Chart For Base Node


Applications:Mesh Network(independent WSN )

-This concept can be best implemented if the number of nodes in the surrounding are high.

- Can be used on any WSN  

Comments Hide

Ashwath Narayan

Received the kit... will start working on it as soon as exams are up.