What is Data Compression?

 What is Data Compression?

Data Compression
Data Compression

In WSN most of the components consume maximum energy. Hence, it is essential to use data compression methods that can increase the computational energy thereby reducing the number of packet transmissions.

Following are the WSN features that support effective data compression protocols:

1. WSNs use logical topology that is similar to trees. This enables correlation between sensor nodes is easy towards the sink node.

2. The data gathered in the neighboring nodes are correlated, mostly if the sensor nodes are deployed in a dense network.

3. The information about the occurrence of an event can be helpful to obtain the data regarding the event.

4. The application semantic can enable data fusion or data aggregation.

5. If there are data errors in applications then sensor nodes can reduce the sampling frequency. 

The compression methods comprise the following:

Data aggregation-based compression methods like tiny aggregation services for Adhoc sensor networks. TAG uses semantic-based aggregations like MIN, MAX, SUM. The location of the aggregation point is the main drawback of this method. For applications that do not support semantic-based aggregations, this method is not useful.

Information theoretic-based methods like DISCUS (Distributed Source Coding Using Syndromes). WSNs comprise sensor nodes that are arranged in a tree topology. The sink node is the root. The data at each node is encoded or compressed along with its correlation with the parent node. It employs Slepian Wolf coding as the sensor network is dense. The sink or source sensor nodes then decompress the encoded or compressed data. 

A sampling of the random process: If a WSN application tolerates error up to some level, the sensor nodes may decrease the sampling frequency.

Post a Comment

Previous Post Next Post