| Popis: |
In the wireless sensor network (WSN), how to efficiently utilize the energy of each sensor node is the key to prolong the network lifetime. The balance of energy consumption among the nodes located at different areas, e.g. the central and edge areas, also plays another important role in the scenario of lifetime prolongation. Thus, the energy efficiency of the whole network can be severely reduced if the geography of network topology is ignored in the cluster head (CH) selection. To our best knowledge, the geography effect of Poison point process (PPP) has not yet been investigated in the existing cluster head (CH) selection algorithm. In this paper, considering the PPP geography effect, the weighted residual energy and distance (WRED) algorithm is proposed to favor the nodes positioned near the cluster center with more energy in the CH selection. On top of this algorithm, the edge sub-clustering (ESC) scheme is further developed to alleviate the effect of non- uniform node distribution on the edge. Additionally, without reforming the clusters and sub-clusters, frame-by-frame CH and sub-CH reselection is also designed to reinforce the effectiveness of the proposed WRED and ESC schemes. In this fashion, not only the energy conservation but also the balance of energy consumption can be significantly improved, which can contribute to longer network lifetime and lower outage probability. |
