| A typical Wireless Sensor Network(WSN) consists of large numbers of tiny, cheap and low-power sensor nodes which can self-organize into an ad hoc network. WSNs are usually deployed in specific areas for the purpose of information sensing and gathering about the diverse monitoring objects. As the information ligament conneting human society and physical world, they send information to end users by adopting a multi-hop communication method. Compared with traditional wireless networks, sensor nodes in WSNs have restricted resources, such as bandwidth and storage capacity. Especially, the limited energy resource directly affects network lifetime and information quality, and it is usually inconvenient to replace the battery for most applications. Under this case, how to improve energy efficiency and enhance network lifetime is one of the researching hotspots currently. Since data transmission contributes the majority of energy dissipation for a sensor node, adopting energy-efficient routing protocols helps to reduce energy consumption and prolong network lifetime. Through systematically studying and summarizing current routing protocols, it is found that the cluster-based routing protocol outperforms other ones in terms of network scalability and energy efficiency. Then this thesis mainly focuses on how to construct cluster-based routing protocols for different types of WSNs. The main contents and innovative contributions include:(1)How to construct single-hop clustering protocols for traditional homogenous WSNs is considered in this thesis, and a Hybrid, Game Theory based and Distributed clustering(HGTD) protocol is presented. In traditional single-hop clustering protocols for WSNs, each cluster head(CH) consumes energy quickly as it transmits data to base station(BS) directly for a long distance by the single-hop communication method. How to select the optimal CHs so that the energy among sensor nodes can be well balanced and the network lifetime is enhanced as well is a research emphasis. Game theory is introduced in this thesis to solve the problem of CH selection in single-hop clustering, and each sensor node is modeled as a player which can get an equilibrium probability to be CH by playing a clustering game with the sensor nodes within its close neighbor. In the clustering game, the payoffs of each node when choosing different strategies are specifically defined, where both node degree and the distance to BS are considered so that each node can achieve a good trade-off between minimizing energy consumption and providing required network services. In addition, an iterative algorithm is also designed to select final CHs from the potential CHs, which can achieve a good balance of energy consumption among sensor nodes and avoid the case that two neighbor nodes become the CH simultaneously. Simulation results show that HGTD can improve network lifetime obviously.(2) As the interference of some external factors, sensor nodes are usually heterogeneous about their energy after being deployed into the sensor field for a time. Then how to construct the multi-hop clutering protocols for heterogeneous WSNs is considered, and a protocol named Unequal Cluster-based Routing scheme for WSNs with multi-level energy Heterogeneity(UCR-H) is proposed. In multi-hop clustering protocols for WSNs, all CHs in the network cooperatively transmit data to BS by the multi-hop communication method. Thus, the CHs closer to BS have heavier inter-cluster relay traffic that results in the “energy hole” problem. To avoid this problem when adopting multi-hop clustering in energy heterogeneous networks, the sensor field is divided into a number of equal-size rectangular units. Firstly, the number of CHs in each unit is calculated by balancing energy consumption among CHs in different units. Secondly, the optimal number of units is achieved by minimizing the total energy consumption of inter-cluster data forwarding. At last, the cluster size for a node when serving as CH in each unit is designed based on node's energy level and the number of clusters in this unit. In addition, a round threshold is also deduced to avoid excessive punishment to the nodes with higher energy level. Simulation results show that UCR-H can effectively mitigate the “energy hole” problem and enhance the network lifetime obviously.(3)Recently, sensor nodes are equipped with energy harvesting devices in some WSN applications, but the ability to harvest energy from the environment is very limited. In this thesis, how to construct multi-hop clustering protocols for energy harvesting WSNs is studied, and a protocol called Multi-hop Energy Neutral Clustering(MENC) is presented. The main objective of our protocol is to achieve perpetual network operation by effectively managing the energy harvested from environment. Through combining the energy consumption and energy status of sensor nodes, an energy neutral constraint for each node is concluded. Under this constraint, each node can work in an energy neutral state, which in turn provides perpetual network operation and avoids network failure. In addition, on the premise of guaranteeing that each node can work in an energy neutral state, the minimum network data transmission cycle is mathematically derived using convex optimization techniques while the amount of data packet that successfully received by BS is maximal. Simulation results show that MENC can achieve infinite network lifetime that ensures continuous data collection and transmission, and also improve the network throughput vastly. |
PDF Full Text Request