A Study On The Selection Mechanism Of WSNs In A Favorable Location Priority

Wireless sensor networks consist of hundreds sensor nodes. Hierarchical clustering, since 21 century, plays a remarkable role in fields such as machine lifespan monitor, health care monitor, battlefield monitor, and wise irrigation, enabling the intellectualization and automation of our daily life.LEACH protocol serves as the benchmark of hierarchical clustering. These days, researchers from every walk improve it in various aspects, presenting SEP, DEEC, PSCND protocols as improved editions.Nevertheless, protocols above are born with respective drawbacks. LEACH ignores the rationality of cluster head distribution. SEP merely differentiates the cluster head probability into two categories, neglecting detailed concentration on residual energy and sensor location. DEEC overlooks node position information and terminates network operation prematurely. FLP solution squanders too much time on cluster head selection, while LEACH-C and PSCND are not competent for considering data-transmitting energy dissipation from cluster head to base station.This paper strives to hedge drawbacks of above protocols and combine their advantages, proposing a reasonable cluster head selection mechanism. Three major revised points have been listed as follows:1. Timely update the G value of cluster head candidates. This ensures that there are at least optimal number cluster head candidates in each round, eliminating the scenario that no cluster head emerges during some specific rounds.2. Offer a more accurate value in terms of average residual energy. Then, the value generates a more precise cluster head probability. By the way, this guarantees that WSNs could operate till the last node dies.3. Pay attention to distance among cluster heads as well as distance from the would-be cluster head to the base station when judging the reasonability of cluster head position. My approach, while punishing a node in disadvantage position, makes much easier a node in favorable location to serve as cluster head.This paper pursues a sub-optimal solution for cluster head location determination, featuring relatively low time cost. The experiment reveals that my protocol achieves an aggressive goal on extending network lifespan (notably stable period) as well as slashing weak sensing period.