Relay node placement for building wireless sensor networks with reconfigurability provision☆
Introduction
In pace with the developments of wireless communications and microelectronics, Wireless Sensor Networks (WSNs) have gained tremendous advances, which can collect various information from the physical world. WSNs have been adopted in a wide range of applications, e.g., environment monitoring, battlefield surveillance, healthcare, smart grid and industrial automation [1], [2], [3].
Sensor Nodes (SNs) in WSNs are spatially distributed in target regions to gather certain information of interest, and the locations of SNs are generally predetermined and fixed [4], [5]. However, the Base Stations (BSs) are commonly installed near the boundaries of external networks with stable power supply [6]. Due to the limited communication range and power supply of SNs, Relay Nodes (RNs) are placed to build the connected network topology under various constraints (e.g., connectivity [7], [8], [9], [10], [11], [12], [13], [14], energy efficiency [15], [16], [17], [18], [19], [20], fault tolerance [21], [22], [23], deployment locations [24], [25] and delay [26], [27], [28], [29]). As network topology significantly impacts the performance of WSNs, the Relay Node Placement (RNP) problem, which is NP-hard [7], has aroused great attention [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29].
Due to the merits of low-cost, handy-deployment and easy-maintenance, WSNs have been progressively used in factory automation [30], [31]. To build network topologies and transmit data reliably and in time, the Delay Constrained Relay Node Placement (DCRNP) problem has been studied in depth recently. Furthermore, the conception of intelligent manufacturing is proposed and has gained tremendous attention in these years. The implementation of this conception needs the manufacturing systems to provide flexibility and reconfigurability so as to respond to dynamic market demands promptly [32]. This requires the underlying information systems (e.g., industrial WSNs) to work correctly across different configurations and constraints [33], [34], [35].
Existing works regarding the RNP problem assume that the network configurations (e.g., transmit power, QoS requirements, etc.) are fixed and cannot be reconfigured during the entire network lifetime. As a result, these algorithms fail to support reconfigurability. To this end, this paper aims to build a network topology that is stably connected with respect to constraints across different network configurations through relay node placement, which is termed as the RNP problem with Reconfigurability provision (RRNP). In addition, the two-tiered architecture, in which SNs only send packets to their 1-hop neighbor RNs and the network connectivity relies on the mesh network composed of only RNs, brings many advantages such as energy efficiency and scalability. Therefore, this paper solves the RRNP problem in the two-tiered architecture [11].
We first propose a framework for the RRNP problem. This framework decomposes the RRNP problem into some subproblems that are instances of the DCRNP problems, and the RNs in each subproblem are placed for one network configuration. With the repetition of this framework, an adaptive WSN under different network configurations will be deployed. We also prove that the proposed framework promises an explicit approximation ratio and a tight time complexity through rigorous mathematical analysis.
Moreover, the objective of existing algorithms regarding the DCRNP problem is only to minimize the number of deployed RNs, providing no improvement on the network performances of end-to-end delay and reliability. Thus, a set of Differential Evolution (DE) based algorithms are presented to optimize the overall network performance (including end-to-end delay, reliability and the number of deployed RNs). Our main contributions are listed as follows:
- •
An RNP framework with supporting network reconfiguration is proposed. In addition, we prove that this framework ensures an explicit approximation ratio and a tight time complexity through rigorous mathematical analysis.
- •
To provide an efficient solution to each subproblem during the execution of the framework, the DE-based algorithms are presented. Different from existing algorithms, the DE-based algorithms not only try to save the deployment cost but also try to reduce the end-to-end delay and improve the Packet Reception Rate (PRR).
- •
We adopt a more realistic wireless channel model to estimate the wireless link quality during the node deployment instead of the ideal 0–1 disk channel model.
- •
Extensive simulations are conducted to evaluate the performances of the proposed algorithms. Simulation results verify the effectiveness of these algorithms.
The remaining parts of this paper are organized as follows: Section 2 reviews the related work regarding the RNP problem. Section 3 describes the system model and formulates the RRNP problem. Section 4 describes and analyzes the proposed framework for the RRNP problem. Section 5 presents the differential-evolution-based algorithm. Section 6 shows the simulation results. Finally, Section 7 concludes the paper.
Section snippets
Related work
The basic RNP problem, which only considers connectivity, is formulated as the Steiner minimum tree with the minimum number of Steiner points and bounded edge length problem (SMT-MSP) [7]. Most researchers resort to approximation algorithms to solve this problem since the scale of WSNs is typically large. Lin and Xue [7] prove that SMT-MSP is NP-complete and present an approximation algorithm with a 5-approximation ratio. Chen et al. [8] prove that the algorithm in [7] actually has a
Communication model
This paper estimates the wireless link quality according to the method in [38], [39] instead of the widely used ideal geometric disk model in existing works. The link quality is measured in terms of Packet Reception Rate (PRR). In this paper, we assume that wireless nodes employ the physical and MAC layer protocols defined by IEEE 802.15.4 standard.
Let u and v denote two wireless nodes, and d(u, v) denote the Euclidean distance between them. When u sends messages to v, the average received
Algorithm description
Given S, C, and a set of network configurations, this framework solves the RRNP problem gradually under each network configuration. To be specific, at the ith step, this framework first builds a HCG under a network configuration (), where R is the set of RNs placed previously. Then, this framework starts a procedure that places a set of RNs in two phases: (1) the covering phase, which places RNs to cover SNs; (2) the connecting phase, which places RNs to
Differential-evolution-based algorithm for DCRNP
From the descriptions in Section 4 we know that the performance of the proposed frame is greatly impacted by the algorithm for solving the DCRNP problem. However, the objective of existing algorithms for the DCRNP problem is to save deployment cost (i.e., deployed RNs) with respect to hop constraints, providing no methods to optimize end-to-end delay and reliability. To address this limitation, we design a Differential-Evolution-based delay-constrained Relay node placement (DER) algorithm. The
Simulations
In this section, extensive simulations are conducted to evaluate the proposed algorithms. The algorithms are implemented by MATLAB on a computer equipped with a 2.9 GHz Intel Core i7-3520M CPU and an 8 GB RAM. In this paper, we use the method provided in [38], [39] to simulate the wireless channel due to its well estimation accuracy, and we develop the simulator according to this method. In these simulations, packet length, total packets to be sent by each SN, path loss exponent, data rate,
Conclusion
In this paper, we have studied the RRNP problem to build WSNs with provision of network reconfigurability through placing additional RNs. To address this problem, we have proposed a framework with two phases, i.e., the covering phase and the connecting phase. Through rigorous mathematical analysis, we have proved that the proposed framework could guarantee an explicitly approximation ratio. Furthermore, we have designed the DE-based algorithms in the connecting phase. Finally, we have conducted
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Chaofan Ma received the B.S degree in Electrical Engineering and Automation from Henan Polytechnic University in 2008 and the M.S. degree in Circuits and Systems from Dalian University of Technology in 2013. He received the Ph. D. degree in Detection Technology and Automatic Equipment at Shenyang Institute of Automation, Chinese Academy of Sciences, China, in 2017. He is currently serving as a lecturer of Zhongyuan University of Technology. His research interests are in wireless sensor
References (40)
- et al.
Deploying sensor networks with guaranteed fault tolerance
IEEE/ACM Trans. Netw.
(Feb. 2010) - et al.
Steiner tree problem with minimum number of Steiner points and bounded edge-length
Informat. Proc. Lett.
(1999) - et al.
Approximations for Steiner trees with minimum number of Steiner points
Optimization
(2000) - et al.
Relay node placement in wireless sensor networks
IEEE Trans. Comput.
(Jan. 2007) - et al.
One-step approach for two-tiered constrained relay node placement in wireless sensor networks
IEEE Wirel. Commun. Lett.
(Aug. 2016) - et al.
On energy provisioning and relay node placement for wireless sensor networks
IEEE Trans. Wirel. Commun.
(2005) - et al.
Fault-tolerant relay node placement in wireless sensor networks: problems and algorithms
Proc. IEEE INFOCOM’07
(2007) - et al.
Optimal relay node placement in delay constrained wireless sensor network design
Eur. Res.
(2014) - et al.
Delay constrained relay node placement in two-tiered wireless sensor networks: a set-covering-based algorithm
Comput. Appl.
(Sep. 2017) - et al.
Relay node placement in wireless sensor networks with respect to delay and reliability requirements
IEEE Syst. J.
(Sep. 2019)
Reconfigurable wireless networks
Proc. IEEE
A wireless sensor network border monitoring system: deployment issues and routing protocols
IEEE Sens. J.
High-performance wireless networks for industrial control applications: new targets and feasibility
Proc. IEEE
A survey on reliability protocols in wireless sensor networks
ACM Comput. Surv.
Corona based deployment strategies in wireless sensor network: a survey
Comput. Appl.
Relay node position optimization in complex environment
Proc. IEEE WCNC’17,
Relay sensor placement in wireless sensor networks
Wirel. Netw.
Relay node placement in large scale wireless sensor networks
Comput. Commun.
Novel relay node placement algorithms for establishing
Connect. Topol.
Optimal placement of relay nodes in wireless sensor network using artificial bee colony algorithm
Comput. Appl.
Cited by (5)
The optimal 5G base station location of the wireless sensor network considering timely reliability
2023, Reliability Engineering and System SafetyRelay node placement in wireless sensor network for manufacturing industry
2023, Bulletin of Electrical Engineering and InformaticsTowards Enabling Fault Tolerance and Reliable Green Communications in Next-Generation Wireless Systems
2022, Applied Sciences (Switzerland)Modeling and solution of the romote support network design problem with relays in an environment with multiple uncertainty demands
2021, Xi Tong Gong Cheng Yu Dian Zi Ji Shu/Systems Engineering and Electronics
Chaofan Ma received the B.S degree in Electrical Engineering and Automation from Henan Polytechnic University in 2008 and the M.S. degree in Circuits and Systems from Dalian University of Technology in 2013. He received the Ph. D. degree in Detection Technology and Automatic Equipment at Shenyang Institute of Automation, Chinese Academy of Sciences, China, in 2017. He is currently serving as a lecturer of Zhongyuan University of Technology. His research interests are in wireless sensor networking, and embedded systems.
Yan Li was born in Taian, Shandong, China in 1983. He received the B.E. degree in communication engineering from the Shandong University of Technology, Zibo, in 2006 and the Ph.D. degree in electronic and electrical engineering from Pusan National University, Busan, Korea, in 2013. From 2013 to 2015, he was an Assistant Professor with Shenyang Institute of Automation, Chinese Academy of Sciences. Since 2016, he has been an Associate Professor with Shenyang Institute of Automation, Chinese Academy of Sciences. He is the author of more than 40 articles. His research interests include robot vision technology, image processing, intelligent underwater robotic systems, neural signal processing, and so on.
Bo Yang received the BS degree in computer science from China Agricultural University and the MS degree in computer science from Shenyang Aerospace University, in 2010 and 2013, respectively. He received the PhD degree in control science and engineering from the University of Chinese Academy of Sciences in 2017. He is currently a research associate at the Institute of Computing Technology, Chinese Academy of Sciences. His research interest span wireless networks, algorithmic number theory and artificial intelligence.
Yuying Zhang received the B.S degree in Computer Science from zhengzhou University in 2008 and the M.S. degree in Computer Science from Otto Von Guericke University Magdeburg, Germany in 2012. She is currently serving as a lecturer of Zhongyuan University of Technology. Her research interests are in Data Base.
Furan Guo received the B.S degree in Electrical Engineering from Zhengzhou University in 2008 and the M.S. degree in Electrical Engineering from Otto Von Guericke University Magdeburg, Germany in 2011. He is currently serving as a Engineer of Research Institute of Economics and Technology in State grid of Henan. His research interests are in Electrical Engineering.