Abstract:
Wireless sensor networks have emerged as a propitious solution to monitor human activity, such as pedestrian and vehicular movement, as well as detection of unauthorized ...View moreMetadata
Abstract:
Wireless sensor networks have emerged as a propitious solution to monitor human activity, such as pedestrian and vehicular movement, as well as detection of unauthorized persons. This can be especially advantageous in military applications such as tracking, surveillance, and security. For such systems, it is essential to implement efficient algorithms for onboard computation that save time and memory. In view of this, the work presented in this paper aims to implement classification and tracking of targets with reduced computation time and lower dimension feature vectors using Symbolic Dynamic Filtering - a wavelet based process. The extracted features follow particular trends that were distinct enough to be classified with the help of simple algorithms. This eliminates the need of an external classifier which saves quite a lot of computational time. Targets, here, vehicles were classified into two classes, namely tracked and wheeled using their seismic signatures. The results of tracking can also be used to monitor the heath of the sensors.
Published in: 2018 9th International Conference on Computing, Communication and Networking Technologies (ICCCNT)
Date of Conference: 10-12 July 2018
Date Added to IEEE Xplore: 18 October 2018
ISBN Information: