Discovering Application Level Semantics for Data Compression using HCT
Deepa. R1, K. John Peter2

1Deepa.R, Computer science and engineering, Vins Christian college of engineering, Anna University, Nagercoil, India.
2K. John peter, Computer science and engineering, Vins Christian college of engineering, Anna University, Nagercoil, India.

Manuscript received on July 01, 2012. | Revised Manuscript received on July 05, 2012. | Manuscript published on July 10, 2012. | PP: 24-29 | Volume-1 Issue-2, July 2012. | Retrieval Number: B0140061212 /2012©BEIESP
Open Access | Ethics and  Policies | Cite 
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Natural phenomena show that many creatures form large social groups and move in regular patterns. However, previous works focus on finding the movement patterns of each single object or all objects. I propose an efficient distributed mining algorithm to jointly identify a group of moving objects and discover their movement patterns in wireless sensor networks. This algorithm consists of the local mining phase and the cluster ensembling phase. The local mining phase adopts the VMM model together with Probabilistic Suffix Tree to find the moving patterns, as well as Highly Connected Component to partition the moving objects. The cluster ensembling phase utilizes Jaccard Similarity Coefficient and Normalized Mutual Information to combine and improve the local grouping results. The distributed mining algorithm achieves good grouping quality and robustness. In this paper, I extend it further, and propose a technique called hybrid compression technique based on the location information of nodes in the sensor network. A hybrid compression technique problem is formulated to reduce the amount of energy consumption and increases the lifetime of network. The experimental result shows that the technique have good ability of approximation to manage the sensor network and have high data compression efficiency and leverages the group movement patterns to reduce the amount of delivered data effectively and efficiently
Keywords: Clustering, hybrid, patterns, similarity