AHP based relay selection strategy for energy harvesting wireless sensor networks

Highlights

• In Energy Harvest Wireless Sensor Network (EHWSN), dynamic energy flow features of the sensor node is modeled the G/G/1/K model, and energy dynamic characteristics, such as energy state, energy outage probability, are analyzed mathematically.

• The cooperation probability of relay nodes is defined. Through the probability that a node can activate itself to participate in the cooperation, different transitions of node states can be realized. To save energy cost of cooperative transmission for the network system, the threshold of outage probability is used as the constraint to optimize the cooperative probability and control the relay nodes participating in cooperative transmission.

• The AHP based relay selection strategy was proposed, which selects the optimal nodes for cooperative transmission through a multi criteria relay selection strategy based on the solar energy state, energy balance of the network, signal to noise ratio and energy supply outage probability.

Abstract

In order to save energy consumption, improve network performance and prolong the life cycle of communication network, a reasonable relay selection algorithm for energy collection wireless sensor networks is proposed. Firstly, taking solar wireless sensor network as an example, the energy consumption and solar energy state of sensor nodes are mathematically analyzed. Then, the cooperation probability is optimized with the interruption probability threshold as the constraint, and the relay nodes participating in the cooperative transmission are controlled to save the energy of cooperative transmission in the system. According to the solar energy status, network energy balance, signal-to-noise ratio and outage probability of the relay node, a multi criteria relay selection strategy is adopted to select the optimal node. Compared with the comparison algorithm, this algorithm can save at least 47.1% power consumption and prolong the network lifetime by 25.9%.

Introduction

With the continuous development of wireless communication technology and the high popularity of wireless terminal equipment, the rapid development of mobile communication industry has stepped into the 5G era. However, there are some factors hindering the development of wireless communication network. In the harsh external environment and the communication environment with special mission requirements, it is difficult to arrange the special communication infrastructure on the network site. This requires a strong dynamic adaptability and survivability of the network to provide strong support for information communication. For example, smart city, smart medical, smart transportation and other applications are supported by wireless network.

Wireless Sensor Network (WSN) is a kind of wireless network formed by self-organization of sensor nodes. Sensor nodes can obtain part of the information parameters in the external environment, and then transmit the collected information data after simple operation. Because of its three advantages of self-organization, distributed and flexible, it has attracted great attention of researchers and played an irreplaceable key role in wireless communication. At present, the application of WSN has extended to many fields. In military applications, by randomly arranging a large number of sensors in the theater, the sensor nodes form a self-organizing network, and then complete the task of monitoring the target, timely help the army to obtain the battlefield situation and accurately locate the enemy target. In agriculture, the WSN can realize the real-time monitoring of crop growth environment. Specific monitoring targets can include air humidity, environmental temperature, light intensity, microbial content in soil, etc., which is convenient for growers to control the growth state of crops in time, so as to enhance the quality of agricultural products and promote the increasing of agricultural product yield [1]. In the natural disaster early warning, the WSN can real-time monitor water, fire, debris flow, landslides, volcanic eruptions and other hazards, so as to timely warn and evacuate the masses [2]. In terms of transportation, a real-time and high-efficiency traffic management system is built through the WSN to improve the efficiency of transportation, alleviate traffic congestion and reduce the occurrence of traffic accidents [3].

In WSN, the hardware structure of the sensor node contains high-sensitivity detection elements used to sense the external environment state (such as temperature, humidity, sound and other information), and sensor nodes have simple computing power [4]. In the vast majority of WSN, the storage power of the energy supply module in sensor nodes is very limited. In WSN, when the power of the node is exhausted, it will be unable to work, or even the network will be paralyzed. Therefore, how to reduce energy consumption as much as possible and extend the whole network life cycle is a challenging topic in the research of the WSN.

Energy harvesting technology is a new resource-saving technology, which is applied to the WSN. The energy harvested from the external environment is used to supply the nodes in the network, so as to prolong the service life time of the WSN. Renewable energy is introduced into the WSN to form Energy-Harvesting Wireless Sensor Network (EHWSN). In the network, nodes collect energy from the surrounding environment through energy acquisition module, i.e., solar, wind and heat, for the normal work of nodes. However, in EHWSN, limited by the size of the sensor node battery capacity, the energy from the external environment is limited. Therefore, how to use the harvested energy efficiently has become a key factor affecting the network performance, and even greatly affects the large-scale application and development of sensor networks.

Recently, it is a hot issue to enhance energy efficiency and prolong the life cycle of the network. At present, scholars focus on reducing the energy consumption of the WSN, such as relay selection algorithm, power allocation optimization, duty cycle technology, cross layer optimization algorithm, etc. These technologies will save energy and prolong the network life cycle for the network.

Bletsas et al. proposed an opportunistic relay selection algorithm. In the proposed strategy, the relay node with the best instantaneous channel state to transmit information cooperatively is selected. The algorithm is proved to be simple, efficient and easy to implement [5]. Shukla et al. constructed a hierarchical cluster framework, and proposed an effective relay selection algorithm considering the node density in the cluster, the relay node communication range, and shortest distance [6]. Zhang et al. constructed an adaptive relay selection algorithm considering the chance. To avoid the fast energy consumption of nodes, delay algorithm was introduced to balance the energy consumption of the node and improve the network performance [7]. Sidrah et al. proposed a relay selection scheme to enhance the energy efficiency by optimizing the energy efficiency of the system, but they ignores the impact of transmission rate on the system [8]. The minimum power routing cooperation algorithm selects the transmission path with the minimum total transmission power based on decoding and forwarding cooperation while ensuring the transmission rate [9]. Wang et al. described the problem of fair and effective resource sharing among cooperative users as a game theory. By introducing an auxiliary function to obtain the result of proportional fair resource allocation, the relay selection can be carried out, and the advantages of this method in efficiency and fairness are verified [10]. Atapattu et al. proposed two relay selection schemes, optimal relay selection and suboptimal relay selection for two-hop networks based on global channel state information and only source–destination pairs channel state information for multi-hop full-duplex relay networks [11]. Wang et al. proposes a GPSR-based routing protocol which incorporates the metric expected transmission delay for data dissemination with heterogeneous communication range. [12]. Rathore et al. proposed an innovative cluster head selection algorithm to minimize energy utilization and to enhance the lifetime of the networking by introducing the shortest path relay node concept [13]. To improve the spectrum efficiency of the considered system, Zhang et al. design a novel two-way communication protocol for simultaneous wireless information and power transfer enabled cognitive radio networks [14]. Guleria et al. proposed an enhanced energy proficient clustering to reduce the energy consumption of the entire sensor nodes in the field of tracking events [15]. Zhao et al. proposed a relay selection algorithm with average power allocation, which effectively avoids the overuse of nodes with high quality of channel and increases the probability of the node selection with general channel quality [16]. Du et al. present a time switching-based network coding relaying for a two-way relay system, which is emerged energy harvesting and network coding technologies [17].

In EHWSN, the relay selection of sensor network nodes not only considers the network performance such as interrupt probability, error rate, delay, information transmission rate, but also the energy performance of nodes such as network energy consumption equilibrium and network lifetime. At present, many scholars focus on the relay selection algorithm for the EHWSN. Gu et al. proposed the energy threshold based multi-relay selection [18]. In order to maximize the first-hop signal to noise ratios (SNRs), second-hop SNRs, and end-to-end SNRs, Son et al. designed the relay select criteria for a cooperative communication network with multi energy harvesting relays [19]. Liu et al. proposed a relay selection scheme. In this scheme, time switched relay protocol is considered to reduce the risk of wrong relay selection and avoid the mismatch between source relay and relay target channel conditions [20]. Zhang et al. Proposed a distributed relay selection and power control scheme. In this scheme, the sensors around the source node and sink node are used as relays to balance the residual energy of sensors in EHWSN [21]. Baidas et al. analyzed power allocation and relay selection for the networks without and with energy cooperation [22]. Lin et al. proposed a relay selection scheme considering both data buffer and energy buffer status [23]. The relay node selection method proposed by Sui et al. is to select the relay receiving information with the largest energy, and select the relay sending information, so as to reduce the risk of improper relay selection. [24]. Yuan et al. investigated the cooperative cognitive radio system based on the energy collection technology. [25]. Aiming at the bidirectional relay protocol based on energy harvesting in wireless ad hoc networks, two relay selection methods are proposed to improve the system outage performance [26].

In most existing relay selection algorithms, the tradeoff between energy saving and outage probability, bit error rate, system delay and information transmission rate is considered. Then, in these, the high frequency use of relay nodes with fine channel conditions will accelerate the death of nodes, while the simple use of energy rich nodes may reduce the reliability of information transmission, thus increasing the number of information retransmission to a certain extent. To save energy consumption and prolong the life cycle of communication network, we proposed relay selection strategy was proposed. In particular, the cardinal contributions are as follows:

(1) Taking the solar powered wireless sensor network as an example, the energy consumption and solar state of the sensor nodes are mathematically analyzed.

(2) The outage probability threshold is used as a constraint to optimize the cooperation probability, and the relay nodes participating in the cooperative transmission are controlled to save the energy of the cooperative transmission in the system.

(3) According to the relay node’s solar energy status, network energy balance, signal-to-noise ratio and outage probability, a multi-criteria relay selection strategy is proposed.

The remainder of this paper is organized as follows. In Section 2, the system model will be described in detail, and energy dynamic state is analyzed mathematically. In Section 3, the AHP based relay selection strategy was proposed. In Section 4, the simulation and experiment are carried out. Finally, some conclusions are drawn in Section 5.