A novel approach to maximize network life time by reducing power consumption level using CGNT model

Abstract

A system having an embedded processor which can operate at minimum power with a small memory, where sensor having low data rate, battery, a GPS system and transceiver that operates at low data rate, collectively constitute a wireless device system. The power source in a wireless device system has a fixed value electro motive force (EMF) which is nothing but a battery. The EMF placed to any wireless sensor network is fixed and has deployed at the time of installation. After deployment it is almost difficult to manage and readjust the power source of each component individually in a network. Therefore it is our prime concernto optimize the energy consumption in each node level to have Maximum Network Life Time(MNLT). We have proposed a model name as Complementary Grouping of Network Traffic(CGNT) which gives a better solution for a load balance power aware routing by dissolving theprime uniformly distributed with coalition routing network into two standard group configurations without coalition routing network. The packets are forwarded from source to destination using shortest distance among nodes taking different path for each configuration to balance the power load and to avoid dead node. CGNT model approach consists of two complementary heterogeneous network groups, one which holds the corner nodes named as Corner Complementary Grouping of Network Traffic (CCGNT) and the other which holds the mid nodes named as Mid Complementary Grouping of Network Traffic (MCGNT). Our approach is to follow the motive of power aware routing along with the load balance use of power where individual source to destination node path of total network traffic are carried out differently in two groups, as a result of which the total variance of power level of all nodes in our proposed model CGNT comes up with a satisfactory minimal value, hence that certainly ensures a network's maximized life time as compared to all other existing models.