Two FCA-Based Methods for Reducing Energy Consumption of Sensor Nodes in Wireless Sensor Networks

Lei, Yuxia; Qu, Meiyan; Lei, Chen; Kong, Zhiqiang; Tian, Jingying; Wang, Shi

doi:https://doi.org/10.1155/2022/8520447

Scientific Programming

On this page

Abstract Introduction Related Work Analysis Conclusion Data Availability Conflicts of Interest Acknowledgments References Copyright Related Articles

Special Issue

Intelligent Mobile Edge Computing for Smart Internet of Things: Architecture, Algorithm, and Application

View this Special Issue

Research Article | Open Access

Volume 2022 | Article ID 8520447 | https://doi.org/10.1155/2022/8520447

Two FCA-Based Methods for Reducing Energy Consumption of Sensor Nodes in Wireless Sensor Networks

Yuxia Lei,¹Meiyan Qu,¹Chen Lei,¹Zhiqiang Kong,¹Jingying Tian,²and Shi Wang³

Academic Editor: Ying Chen

Received08 Apr 2022

Revised04 Jun 2022

Accepted13 Jun 2022

Published05 Jul 2022

Abstract

In the design of wireless sensor networks (WSNs), it is important to reduce energy consumption and extend the service life of the sensors. Selecting one of the minimum sensor combinations (MSCs) that can monitor all areas, while the other MSCs are asleep, can effectively extend the lifetime of WSNs. This paper proposes two algorithms based on Formal Concept Analysis (FCA) for extracting some MSCs, to minimize the energy consumption and meet the coverage requirement. These two methods firstly extract the concept lattice from a monitor-areas context, and then it is simple to extract sensor nodes monitoring the overlapping area from the concept lattice. The algorithms consist of three steps as follows: the first step is to transform sensors and monitoring areas into a context, the second is to extract the concept lattice and implications among areas, and the third is to extract some different MSCs that can monitor all areas. Thus, some strategies are designed to awaken different MSCs to achieve the purpose of reducing energy consumption. Experimental results show that these methods have played a positive effect on extracting different MSCs and extending the lifetime of sensor networks.

1. Introduction

With the popularity and rapid development of the fields of Internet of Things, there are many meaningful research directions such as point of interest recommendation [1], reducing the energy consumption of WSNs [2, 3], dynamic task offloading [4], and admission control for edge computing [5]. In recent years, WSNs have been more and more widely used in economic life, which are usually made up of a large number of sensor nodes, and are scalable [6, 7]. But because of their limited power and computing power, they are often placed in areas that are hard for people to reach [2]. Energy is a valuable resource in WSNs, and the energy consumption of sensor nodes is the main energy consumption. Therefore, sensor power supply capability is the main obstacle that limits the application of sensor network technology [8]. However, the power of sensor nodes is limited and difficult to replace [9–11], how to reduce the energy consumption of sensor nodes becomes the core problem of WSNs [12–14].

FCA is proposed by Wille R. in 1982, which focuses on the lattice structure induced by a binary relation between a pair of sets (respectively called objects and attributes), known as the Galois lattice or the concept lattice of the relation [15–18]. A node of concept lattice is a pair of objects and attributes, called a formal concept. It consists of two parts: the extent and the intent. The extent is the set of objects which have all attributes in the intent. Similarly, the intent is the set of attributes common to the objects in the extent. A process of generating the concept lattice from a binary relation is a process of clustering, and the line diagram corresponding to a concept lattice vividly shows the generalization and specialization relationship among formal concepts. FCA can be combined with techniques in many fields like big data [19], Internet of Things [20–22], data mining [23], and social computing. If it is combined with deep learning methods to apply to the field of big data and edge computing [24–26], the accuracy of data mining and resource allocation will be optimized. If it is applied to social networks [27, 28], the relationship between nodes can be strengthened to a certain extent and the structure of social networks can be improved. To reduce the energy consumption of WSNs and achieve optimal deployment of nodes [29], this paper uses FCA to extract the minimum sensor combinations to monitor the entire area.

According to the research of existing methods, this paper proposes two algorithms based on FCA for extracting some MSCs. One is called implications-based algorithm for extracting MSCs. This method can find the minimum sensor combination, but the number of rules affects the efficiency of this algorithm. To reduce the rule set, we propose the second method for extracting the stem base of region implications from the concept lattice. The implementation process of these two algorithms can be roughly divided into three parts. Firstly, transform sensors and monitoring areas into a context. Secondly, extract the concept lattice from the context and implications among areas. Lastly, extract sensor nodes that monitor the overlapping area from the concept lattice. Thus, some strategies are designed to awaken different MSCs to reduce energy consumption and extend the lifetime of the network. Experimental results show that different MSCs can be extracted. These two methods have achieved good results in practical applications.

At present, the mechanisms for reducing energy consumption are broadly classified into the following three categories:(1)Sleeping mechanism: This mechanism mainly solves the problem of energy waste caused by idle monitoring [30]. When there is no event of interest occurring around the node, the computing and communication module will be in the idle module, turning them off or turning them to a lower power, that is, the dormant state [31]. This mechanism is important for extending the lifetime of sensor nodes. For example, dynamic voltage scaling (DVS) and dynamic power management (DPM) manage system power in a power-aware manner to reduce energy consumption and extend the lifetime of the nodes.(2)Data compression and data fusion mechanism: In WSNs with high coverage, there is redundancy in the information passed by neighbor nodes. And each node consumes too much energy to transmit data alone, shortening the lifetime of the network accordingly [32]. Data compression and fusion can effectively reduce the amount of raw data and the number of transmissions of sensor nodes. The data compression method reduces the energy consumption of nodes by compressing the transmitted data in advance [33]. Data fusion technology can merge the data information collected by multiple sensors and eliminate redundant and useless information [34, 35].(3)Self-contained energy replenishment device: The sensor node is equipped with the energy harvesting and conversion device to collect energy like solar energy [36–38]. However, renewable energy fluctuates with the environment and the energy density is low. Therefore, this energy supplementation method has great limitations in practical applications. To solve these problems, the relevant research proposes to add the rechargeable sensor node of the wireless charging device to form a wireless rechargeable sensor network [39, 40]. And wireless power transmission (WPT) is used to replenish energy for rechargeable sensor nodes [41, 42].

However, there are still many disadvantages in the existing mechanisms. The sleeping mechanism requires an effective scheduling algorithm, and unreasonable arrangement of the dormant and working state of nodes will lead to idle waste. Besides, the transition between the dormant and working state of nodes also consumes some energy. Thus, frequent state transitions can also lead to excessive energy consumption. The data compression process requires more powerful processing power to handle compression algorithms [43]. Although data fusion can substantially reduce energy consumption, it usually needs specific nodes to collect data before proceeding to the next stage of transmission, which increases the network delay. Moreover, the process of data compression and fusion may cause the loss of detailed information and reduce the quality of data transmission. For wireless rechargeable networks, the current wireless charging technologies are mostly used in a short range. The farther the transmission distance, the lower the energy transmission efficiency. And the charging speed and efficiency of wireless charging still need to be further improved.

The main contributions of our work in this paper are summarized as follows: we propose two algorithms based on FCA for extracting some MSCs and waking up a group of sensor nodes by periodic loop mechanism and prediction mechanism to minimize the energy consumption on condition that the sensors monitor all areas. The specific idea is to extract MSCs by proposed algorithms and sort the combinations, and then put a group of sensor nodes in quasi-wake state by using the periodic loop mechanism. Finally, wake up the sensor nodes that are predicted to have a task schedule. The motivation diagram of this study is shown in Figure 1.

The remainder of this paper is organized as follows. Section 2 presents the related work, followed by some basic notations in formal concept analysis in Section 3. Section 4 provides two FCA-based algorithms called implications-based algorithm and the stem base-based algorithm for extracting different MSCs. Section 5 provides some experimental results to verify the validity and feasibility of the algorithms. Finally, we conclude and present the future work in Section 6.

In recent years, with the rapid transformation and development of wireless communications in various fields [44, 45], a large number of researchers have been committed to studying various technologies that reduce the energy consumption of WSNs to extend the service life. In [46], the authors studied a data compression algorithm that determines the compression level of each sensor node to reduce total energy consumption based on the average energy level of neighboring sensor nodes. This technique reduces the energy consumption to transmit and receive packets and ultimately extends the entire network life. The literature [47] proposed a mean filtering algorithm based on node data images, which divides nodes into active and inactive nodes. This method clusters nodes and analyzes data obtained from sensor networks to eliminate data redundancy of sleep nodes. In [3], the authors introduced an area segmentation model based on optimal energy-saving constraints. According to the energy attribute of the sensor nodes, it optimally segments the coverage area of the wireless sensor network to improve its coverage, reduce the energy cost of a single node and realize the optimal networking of WSNs. To fully balance the energy, the literature [48] proposed to use power supply lines to connect nodes. Based on energy balance, a data transmission method with optimal hop count is proposed, which fully reduces the power consumption of data transmission. To determine the optimal monitoring sensor nodes and the information flow paths to the destination and receivers to reduce the energy consumption between the node and the receiver, Bat algorithm (BA) is proposed in literature [49]. In [2], the authors developed two different system models that use optimal node placement strategies compared with traditional equidistant placement strategies to minimize energy consumption in linear wireless sensor networks (LWSNs). Based on improved sparrow search algorithm (ISSA) optimized self-organizing maps (SOM), a cluster head selection strategy used in heterogeneous wireless sensor network (HWSN) is proposed in the literature [7]. This strategy comprehensively considers the residual energy, distance, and the times the node becomes a cluster head. It uses the adaptive learning of improved competitive neural networks to optimize the cluster head node and extend the life cycle of the network. The literature [50] proposed a power management method to reduce energy consumption in an idle state. Moreover, they studied a fine-grained power mode (FGPM) with five states. This mode can adjust the power consumption according to the communication state of the sensor nodes, reducing the power consumption of each node.

The strategies to wake up nodes will also affect the energy consumption of the sensor nodes, and an effective wake-up strategy can reasonably switch the sleep and working state of the sensors and reduce the energy loss of each sensor. In [51], a low duty cycle energy-efficient MAC protocol is proposed, which can adaptively update based on the prediction wake-up time of nodes. In this method, each node has a neighbor node information table including the target node address. By this table, the sending node will predict the wake-up time of receiving node. In [52], the authors proposed a self-adaptive sleep/wake-up scheduling approach based on the reinforcement learning technique. Each node can decide its working state independently in each time slot. The literature [53] found a multitarget sensing wake-up control method based on swarm intelligence, which studies the optimization of the wake-up strategy of dynamic targets to reduce the number of wake-up nodes when multiple targets are crossing. This method avoids waking up excessive nodes and extends the service life of WSNs.

3. Basic Notations in Formal Concept Analysis

In FCA [15], a (formal) context is defined as a triple , where and are sets and is a binary relation. The elements of and are respectively called objects and attributes. For any and , implies that the object r possesses the attribute . For a set of objects, is the set of attributes common to the objects in . It is defined as fd1.

Similarly, for a set , is the set of objects having all attributes in . It is defined as fd2.

Given a context , for any and Y , the pair is called a (formal) concept if and , where and are respectively called the extent and the intent of the concept. There are two kinds of special concepts: object concept and property concept. Given an object , the pair is a concept, called the object concept of . The object concept is the smallest concept having in its extent. Correspondingly, given an attribute , the pair is also a concept, called the attribute concept of , which is the greatest concept having in its intent.

The set of all concepts in the context is represented as . If , are concepts, then is called a subconcept of , provided that (which is equivalent to ). The comparison expression is represented as . In this case, is a superconcept of . Therefore, we can get fd3.

The relation ‘’ is an order on , which is called the hierarchical order of the concepts. The hierarchical order produces a lattice structure in called the concept lattice of the context , which is also represented as . is also a complete lattice in which infimum and supremum are given by: if is an index set and for every , is a concept, then the infimum of the set is defined as , which represents the largest common subconcept of the concept . Accordingly, the supremum of the set represents the smallest common subconcept.

The labelling can be simplified considerably by putting down each object and attribute only once, namely at the circle for the respective object or attribute concept. Thus, the concept of lattices can be described by the line diagrams with reduced labelling. In a line diagram, the name of an object is always attached to the circle that represents the smallest concept with in its extent. And the name of an attribute is always attached to the circle that represents the largest concept with in its intent. This allows us to read the map from the diagram: an object has an attribute if and only if there is an ascending path from the circle labeled by to the circle labeled by . The extent of a concept consists of all tuples whose labels are below in the diagram. The intent consists of all properties attached to the concepts above in the hierarchy. For example, in Figure 2 in section 5, the intent of the concept labeled by the attribute is MSC, and its extent is . Similarly, the extent of the concept labeled by the is , and its intent is .

(a)

(b)

(c)

(d)

4. FCA-Based Algorithms for Extracting Different MSCs

In this section, we propose two FCA-based algorithms for extracting MSCs. Moreover, we will explain the specific algorithm flow of these two algorithms and strategies to reduce the energy consumption of sensors in detail. The algorithms consist of three steps as follows: the first step is to transform sensors and monitoring areas into a context, the second is to extract the concept lattice and implications among areas, and the third is to extract the minimum sensor combinations that can monitor all areas. The core idea of these two algorithms is that when the sensor set we have found covers all areas, end loop and output the minimum sensor set.

4.1. Transforming Sensors and Their Coverage Areas into a Context

In a wireless sensor network with many nodes, each sensor node can be considered as an object, and each area monitored by sensors can be considered as an attribute of the corresponding object. Assuming that some sensors are scattered in one area, we divide the area into some smaller areas. According to the positioning and operating radius of these sensors, the areas that can be monitored by each sensor can be obtained. Thus, we can transform sensors and monitoring areas into a table. The row represents the sensor number, the column represents the divided monitoring area, and the symbol “+” at the intersection indicates that the sensors can monitor the corresponding areas. For example, there are eight sensors, and the area is divided into eight smaller areas. Each sensor and corresponding monitoring area are as shown in Table 1, which is a context.

By constructing a concept lattice from the context, it is simple to see which sensor nodes are monitoring the overlapping area. Each sensor can perceive and record the surrounding sensors. Communication between sensors is achieved by finding all the sensor nodes that can be reached through the path on the concept lattice. Then we can make some strategies to awaken the corresponding sensors to achieve the purpose of reducing energy consumption and extending network life.

4.2. Implications-Based Algorithm for Extracting MSCs

The basic idea of this algorithm is to extract the region implications among regions from the concept lattice, and then extract different minimum sensor sets according to the rules. To describe this algorithm, we provide several definitions such as implication [15–18].

Definition 1. Given a context , for any , if every object with attributes in also has the attributes in , then is called an implication of . and are respectively called antecedent and consequence of the implication.

Definition 2. Given a concept , if , it matches an implication .

Definition 3. Given a concept , if is both an attribute concept and an object concept, then it is called .

Definition 4. Given a concept , if is only an attribute concept, then it is called .
Then we describe the algorithm flow of implications-based Algorithm 1 for extracting MSCs as below.

	Input: a context , where and is a set of sensors and domains respectively.
	Output: the minimum sensor combinations that monitor all areas in .
(1)	Generate the concept lattice with reduced labelling of and implications in
(2)	Let be the set of of , be the set of implications in , ,
(3)	If
(4)	Then For the first concept in
(5)	If it matches the implication
(6)	then
(7)
(8)
(9)	If is a
(10)	then // is a sensor in the reduced extent of
(11)	else for a sensor in
(12)	if
(13)	then
(14)
(15)	Return 3
(16)	Output MSC

4.3. The Stem Base-Based Algorithm for Extracting MSCs

The basic idea of this algorithm is to extract the stem base of region implications from concept lattice, and then extract different minimum sensor sets according to the rules. To describe this algorithm, we provide several definitions such as the stem base of the attribute implications [15–18].

Definition 5. If each subset of respects also respects , then the implication follows from a set of implications between attributes.

Definition 6. If every implication follows from , then a set of implications of a context is called complete.

Definition 7. If none of the implications follows from the others, then a set of implications of a context is called nonredundant.

Definition 8. is called the pseudointent of if and only if and holds for every pseudointent .

Theorem 1. The set of implications is nonredundant and complete, which is called the stem base of the attribute implications.

Then we define the stem base-based Algorithm 2 as follows:

	Input: a context , where and is a set of sensors and domains, respectively.
	Output: the minimum sensor combinations that monitor all areas in .
(1)	Generate the concept lattice with reduced labelling of and the stem base of implications
(2)	Let be the set of MScs. Standalone , be the stem base in , ,
(3)	If
(4)	Then if there is an implication satisfying
(5)	then // is a sensor in the support set of
(6)
(7)	else for the first concept in
(8)	if it matches the implication
(9)	then
(10)
(11)
(12)	If is a 1
(13)	then // is a sensor in reduced extent of
(14)	else for a sensor in
(15)	if
(16)	then
(17)
(18)	Return 3
(19)	Output MSC

4.4. Strategies for Reducing Sensor Energy Consumption

There are several ways to reduce the energy consumption of sensor nodes [54, 55], such as periodic awakening. One way is to make one node run out of its energy and then make the other overlapping nodes wake up. Another way is to calculate the size of the overlapping area of the nodes and decide whether they’re awake or not.

In wireless sensor networks, there are the following ways to wake up the nodes:(1)Full wake-up mode: In this mode, all nodes in the wireless sensor network wake up at the same time to detect and track targets in the network. Although this mode can obtain high tracking accuracy, it is at the cost of huge network energy consumption.(2)Random wake-up mode: In this mode, the nodes in the wireless sensor network are randomly awakened by a given wake-up probability .(3)Select the wake-up mode by the prediction mechanism: In this mode, the nodes that have a greater return on tracking accuracy can be selectively awakened according to the needs of tracking tasks. Then predict the state of the target at the next moment through the information of this beat and wake up the nodes.(4)Task cycle wake-up mode: In this mode, the nodes in the wireless sensor network are in the awake state periodically. Nodes in this mode can coexist with nodes in other working modes and assist these nodes to work.

In this study, we choose the period loop mechanism and prediction mechanism to wake up nodes. After the processing of the proposed algorithms, we can obtain different MSCs. Then sort the sensor combinations and use the period loop mechanism to put each group of nodes into quasi-wake state in order. The quasi-wake state is a state between wake-up and sleep, which means the node is ready to be awakened. The principle of the period loop mechanism is that only one group of sensor nodes is put into the quasi-wake state in each cycle until each group of nodes has been processed once, and then start again with the first set of nodes and repeat this process. Lastly, wake up the sensor nodes that may have a task arrangement by using a prediction mechanism. In this mechanism, the sending node calculates and predicts the wake-up time of receiving node by retaining the address of the target node in the neighbor node information table [51]. This step ensures the receiving nodes wake up timely and avoid the problems like collision caused by waking up all nodes simultaneously.

5. Simulation Results and Analysis

5.1. Extraction of Implications Rules from Table 1

In this section, we conduct some experiments with specific example to verify the effectiveness of the proposed algorithms. By using lattice-miner or ConExp, we can obtain the following concept lattices in Table 1, as shown in Figure 2. The concept lattice tools can be downloaded from the websites https://sourceforge.net/projects/lattice-miner/ and https://sourceforge.net/projects/conexp/.

From Figure 2(a), we can obtain the intents and extents of concepts in the concept lattice constructed according to the example. Furthermore, we conduct several experiments to simplify the labelling. Therefore, the concept of lattices can be described by the line diagrams with reduced labelling. By reducing objects labelling, the corresponding line diagram is shown as Figure 2(b). Likewise, the line diagram in Figure 2(c) is built by reducing attributes labelling. And the line diagram in Figure 2(d) is built by reducing labelling.

According to the concept lattices in Figure 2, we can obtain the concepts including the intents and extents of nodes in Table 2.

5.2. Extracting MSCs from Table 1 by Implications-Based Algorithm

We find out the implications and do some experiments to verify its effectiveness. The results including support and confidence are shown in Table 3.

From Figure 2(d), we can know that (sensor-4, h) and (sensor-7, d) are 1/2 concepts, and (a′, a″), (b′, b″), (c′, c″), (e′, e″), (f′, f″), (, ) are 2 concepts. The specific steps of the implications-based algorithm are as follows:

Step 1. Initialization.
Let the concept set C = {(sensor-4, h), (sensor-7, d), (a′, a″), (b′, b″), (c′, c″), (e′, e″), (f′, f″), (, )}, W = {rule set}, MWZ = empty.

Step 2. Cycle.(1)First cycle: Select the first concept (sensor-4, h) from . If h matches the 12th rule in W, then let C = C/{sensor-4, h}, M = M – {b, f, , h} = {a, c, d, e}, MWZ = {sensor-4};(2)Second cycle: Select the first concept (sensor-7, d) from C. If d matches the 14th rule in W, then let C = C/{sensor-7, d}, M = M – {b, c, d} = {a, e}, MWZ = {sensor-4, sensor-7};(3)Third cycle: Select the first concept (e′, e″). If e matches the 6th rule in W, then let C = C/{(e′, e″)}, M = M – {e, c} = {a}. It is necessary to select one sensor from sensor-5, sensor-6 and sensor-8 covering e and put it into MWZ;(4)Fourth cycle: Select the first concept (a′, a″). If there is no matching rule, then let C = C/{(a′, a″)}, M = M – {a} = empty set. We just need to select one sensor from sensor-1, sensor-2, sensor-3, sensor-5 and sensor-6 covering a and put it into MWZ. If it has covered the area covered by the sensors in MWZ, then MWZ remains unchanged, otherwise the sensor will be added to MWZ;

Step 3. Let M = empty set. End loop and output MWZ. According to the analysis of the example, MWZ can be expressed by the following combination of sensor nodes: MWZ = {sensor-4, sensor-7, sensor-5} or {sensor-4, sensor-7, sensor-6} or {sensor-4, sensor-7, sensor-8, sensor-1} or {sensor-4, sensor-7, sensor-8, sensor-2} or {sensor-4, sensor-7, sensor-8, sensor-3}.The implications-based algorithm can find the minimum-maximum sensor combination, but a constant comparison is made between the selection and the rules. The number of rules affects the efficiency of the algorithm. Therefore, we will further reduce the number of rules in the rule set.

5.3. Extracting MSCs from Table 1 by Stem Base-Based Algorithm

In the same way, we conduct some experiments on the second algorithm. The implications and experimental results including support and confidence are shown in Table 4.

The specific steps of the stem base-based algorithm are as follows:

Step 4. Let the concept KC = {(sensor-4, h), (sensor-7, d), (a′, a″), (b′, b″), (c′, c″), (e′, e″), (f′, f″), (, )}, KI = {The rules set}, MWS = {}, M = {a, b, c, d, e, f, , h}.

Step 5. Cycle.(1)The first cycle: For any implication in Table 4, there is not M ⊆ A∪B. Select the first concept ((sensor-4, h), h) from KC and h matches the fifth rule in W; Then let KC = KC\ {(sensor-4, h), (b′, b″), (f′, f″), (, )} = {(sensor-7, d), (a′, a″), (c′, c″), (e′, e″)}, M = M – {b, f, , h} = {a, c, d, e}, KI = {1, 2, 3, 4, 6}, MWZ = {sensor-4};(2)The second cycle: For any implication in KI, there is not M ⊆ A∪B. Select the first concept ((sensor-7, d), d) from KC, and d matches the sixth rule in Table 4; Then let KC = KC\ {(sensor-7, d), (b′, b″), (c′, c″)} = {(a′, a″), (e′, e″)}, M = M – {b, c, d} = {a, e}, KI = {1, 2, 3, 4}, MWZ = {sensor-4, sensor-7};(3)The third cycle: There is an implication (a, c)⟶{e}, which satisfies {a, e}⊆{a, c, e}. MWZ = {sensor-4, sensor-7, sensor-5} or {sensor-4, sensor-7, sensor-6}.

6. Conclusion

In wireless sensor networks, the energy storage and power supply capabilities of sensors are the main problems that limit the application of sensor network technology. Therefore, how to extend the service life of sensor networks and reduce the energy consumption of sensor nodes is a key issue in the research of wireless sensor networks. To solve this problem, this paper proposes two FCA-based methods to reduce the energy consumption of nodes, which are different from other commonly used methods. One is an implications-based algorithm for extracting MSCs. This method can find the minimum-maximum sensor combination, but it has to constantly compare the selection and the rules. The number of rules affects the efficiency of this algorithm. Therefore, we need to reduce the rule set. The second method is stem base-based algorithm, which can find the base of rules. These two methods are based on FCA, and both help solve the following problems. The first is how to choose a sensor. The second is how to optimize the sensor layout. And the third is how to wake up the corresponding sensors. According to experimental results, these two methods have achieved good results in practical applications.

The methods proposed in this paper are aimed at the research and discussion of the energy consumption of WSNs. The concept lattice of nodes in the wireless sensor network can express the relationship between nodes more effectively, and it is more helpful to choose nodes. With the deepening of research, we will propose more algorithms with concept lattice methods for other fields like radar sensor networks [56]. We believe there will be better effective solutions to such problems.

Data Availability

The data used to support this study are included within the article.

Conflicts of Interest

The authors declare no conflicts of interest regarding the publication of this paper.

Acknowledgments

This study was funded by Shandong Province Teaching Reform Project (project no. S2018Z022).

References

J. Huang, Z. Tong, and Z. Feng, “Geographical POI recommendation for Internet of Things: a federated learning approach using matrix factorization,” International Journal of Communication Systems, 2022.
View at: Publisher Site | Google Scholar
A. Hussein, A. Elnakib, and S. Kishk, “Linear wireless sensor networks energy minimization using optimal placement strategies of nodes,” Wireless Personal Communications, vol. 114, no. 4, pp. 2841–2854, 2020.
View at: Publisher Site | Google Scholar
X. Chen and T. Wu, “Region segmentation model for wireless sensor networks considering optimal energy conservation constraints,” Cluster Computing, vol. 22, no. S3, pp. 7507–7514, 2019.
View at: Publisher Site | Google Scholar
Y. Chen, F. Zhao, Y. Lu, and X. Chen, “Dynamic task offloading for mobile edge computing with hybrid energy supply,” Tsinghua Science and Technology, vol. 10, 2021.
View at: Google Scholar
J. Huang, B. Lv, Y. Wu, Y. Chen, and X. Shen, “Dynamic admission control and resource allocation for mobile edge computing enabled small cell network,” IEEE Transactions on Vehicular Technology, vol. 71, no. 2, pp. 1964–1973, 2022.
View at: Publisher Site | Google Scholar
H. Yetgin, K. T. K. Cheung, M. El-Hajjar, and L. Hanzo, “A survey of network lifetime maximization techniques in wireless sensor networks,” IEEE Communications Surveys & Tutorials, vol. 19, no. 2, pp. 828–854, 2017.
View at: Publisher Site | Google Scholar
L. Cao, Y. Yue, and Y. Zhang, “A data collection strategy for heterogeneous wireless sensor networks based on energy efficiency and collaborative optimization,” Computational Intelligence and Neuroscience, vol. 2021, Article ID 9808449, pp. 1–13, 2021.
View at: Publisher Site | Google Scholar
Z. Zhang, J. Willson, Z. Lu, W. Wu, X. Zhu, and D.-Z. Du, “Approximating maximum lifetime $k$ -coverage through minimizing weighted $k$ -cover in homogeneous wireless sensor networks,” IEEE/ACM Transactions on Networking, vol. 24, no. 6, pp. 3620–3633, 2016.
View at: Publisher Site | Google Scholar
C. Catlett, P. Beckman, N. Ferrier et al., “Measuring cities with software-defined sensors,” Journal of Social Computing, vol. 1, no. 1, pp. 14–27, 2020.
View at: Publisher Site | Google Scholar
J. Man, H. Dong, L. Jia, and Y. Qin, “GGC: gray-granger causality method for sensor correlation network structure mining on high-speed train,” Tsinghua Science and Technology, vol. 27, no. 1, pp. 207–222, 2022.
View at: Publisher Site | Google Scholar
S. Yan, Z. Zhou, Y. Yang, Q. Leng, and W. Zhao, “Developments and applications of tunneling magnetoresistance sensors,” Tsinghua Science and Technology, vol. 27, no. 3, pp. 443–454, 2022.
View at: Publisher Site | Google Scholar
X. Wan, J. Wu, and X. Shen, “Maximal lifetime scheduling for roadside sensor networks with survivability $k$,” IEEE Transactions on Vehicular Technology, vol. 64, no. 11, pp. 5300–5313, 2015.
View at: Publisher Site | Google Scholar
M. Kariman-Khorasani, M. A. Pourmina, and A. Salahi, “Energy balance based lifetime maximization in wireless Sensor networks employing joint routing and asynchronous duty cycle scheduling techniques,” Wireless Personal Communications, vol. 83, no. 2, pp. 1057–1083, 2015.
View at: Publisher Site | Google Scholar
M. Yang, D. Kim, D. Li, W. Chen, and A. O. Tokuta, “Maximum lifetime suspect monitoring on the street with battery-powered camera sensors,” Wireless Networks, vol. 21, no. 4, pp. 1093–1107, 2015.
View at: Publisher Site | Google Scholar
B. Ganter and R. Wille, Formal Concept Analysis: Mathematical Foundations, Springer Science & Business Media, Germany, 2012.
J. Ma, W. Zhang, and Y. Qian, “Dependence space models to construct concept lattices,” International Journal of Approximate Reasoning, vol. 123, pp. 1–16, 2020.
View at: Publisher Site | Google Scholar
T.-M. Liaw and S. C. Lin, “A general theory of concept lattice with tractable implication exploration,” Theoretical Computer Science, vol. 837, pp. 84–114, 2020.
View at: Publisher Site | Google Scholar
W. Conradie, S. Frittella, K. Manoorkar et al., “Rough concepts,” Information Sciences, vol. 561, pp. 371–413, 2021.
View at: Publisher Site | Google Scholar
A. K. Sandhu, “Big data with cloud computing: discussions and challenges,” Big Data Mining and Analytics, vol. 5, no. 1, pp. 32–40, 2022.
View at: Publisher Site | Google Scholar
C. Chi, Y. Wang, X. Tong, M. Siddula, and Z. Cai, “Game Theory in Internet of Things: A Survey,” IEEE Internet of Things Journal, 2021.
View at: Publisher Site | Google Scholar
M. Azrour, J. Mabrouki, A. Guezzaz, and Y. Farhaoui, “New enhanced authentication protocol for internet of things,” Big Data Mining and Analytics, vol. 4, no. 1, pp. 1–9, 2021.
View at: Publisher Site | Google Scholar
Y. Chen, W. Gu, and K. Li, “Dynamic task offloading for Internet of Things in mobile edge computing via deep reinforcement learning,” International Journal of Communication Systems, Article ID e5154, 2022.
View at: Publisher Site | Google Scholar
E. Shemis and A. Mohammed, “A comprehensive review on updating concept lattices and its application in updating association rules,” WIREs Data Mining and Knowledge Discovery, vol. 11, no. 2, Article ID e1401, 2021.
View at: Publisher Site | Google Scholar
Y. N. Malek, M. Najib, M. Bakhouya, and M. Essaaidi, “Multivariate deep learning approach for electric vehicle speed forecasting,” Big Data Mining and Analytics, vol. 4, no. 1, pp. 56–64, 2021.
View at: Publisher Site | Google Scholar
Y. Chen, Z. Liu, Y. Zhang, Y. Wu, X. Chen, and L. Zhao, “Deep reinforcement learning-based dynamic resource management for mobile edge computing in industrial internet of things,” IEEE Transactions on Industrial Informatics, vol. 17, no. 7, pp. 4925–4934, 2021.
View at: Publisher Site | Google Scholar
F. Li, X. Yu, R. Ge, Y. Wang, Y. Cui, and H. Zhou, “BCSE: blockchain-based trusted service evaluation model over big data,” Big Data Mining and Analytics, vol. 5, no. 1, pp. 1–14, 2022.
View at: Publisher Site | Google Scholar
C. Jiang, A. D’Arienzo, W. Li, S. Wu, and Q. Bai, “An operator-based approach for modeling influence diffusion in complex social networks,” Journal of Social Computing, vol. 2, no. 2, pp. 166–182, 2021.
View at: Publisher Site | Google Scholar
P. D. Waggoner, R. Y. Shapiro, S. Frederick, and M. Gong, “Uncovering the online social structure surrounding COVID-19,” Journal of Social Computing, vol. 2, no. 2, pp. 157–165, 2021.
View at: Publisher Site | Google Scholar
H. Niu, Z. Chu, Z. Zhu, and F. Zhou, “Aerial intelligent reflecting surface for secure wireless networks: secrecy capacity and optimal trajectory strategy,” Intelligent and Converged Networks, vol. 3, no. 1, pp. 119–133, 2022.
View at: Publisher Site | Google Scholar
L. Rajesh and C. B. Reddy, “Efficient wireless sensor network using nodes sleep/active strategy,” International Conference on Inventive Computation Technologies (ICICT), vol. 2, pp. 1–4, 2016.
View at: Publisher Site | Google Scholar
R. Wan, N. Xiong, and N. T. Loc, “An energy-efficient sleep scheduling mechanism with similarity measure for wireless sensor networks,” Human-centric Computing and Information Sciences, vol. 8, no. 1, p. 18, 2018.
View at: Publisher Site | Google Scholar
P. Zhang, L. Li, K. Niu, Y. Li, G. Lu, and Z. Wang, “An intelligent wireless transmission toward 6G,” Intelligent and Converged Networks, vol. 2, no. 3, pp. 244–257, 2021.
View at: Publisher Site | Google Scholar
C. J. Deepu, C.-H. Heng, and Y. Lian, “A hybrid data compression scheme for power reduction in wireless sensors for IoT,” IEEE transactions on biomedical circuits and systems, vol. 11, no. 2, pp. 245–254, 2017.
View at: Publisher Site | Google Scholar
D. Izadi, J. Abawajy, S. Ghanavati, and T. Herawan, “A data fusion method in wireless sensor networks,” Sensors, vol. 15, no. 2, pp. 2964–2979, 2015.
View at: Publisher Site | Google Scholar
L. Qi, C. Hu, X. Zhang et al., “Privacy-aware data fusion and prediction with spatial-temporal context for smart city industrial environment,” IEEE Transactions on Industrial Informatics, vol. 17, no. 6, pp. 4159–4167, 2021.
View at: Publisher Site | Google Scholar
M. Getahun, M. Azath, D. P. Sharma, A. Tuni, and A. Adane, “Efficient energy utilization algorithm through energy harvesting for heterogeneous clustered wireless sensor network,” Wireless Communications and Mobile Computing, vol. 2022, pp. 1–17, 2022.
View at: Publisher Site | Google Scholar
T. V. Tran and W.-Y. Chung, “High-efficient energy harvester with flexible solar panel for a wearable sensor device,” IEEE Sensors Journal, vol. 16, no. 24, pp. 9021–9028, 2016.
View at: Publisher Site | Google Scholar
C. Wang, J. Li, Y. Yang, and F. Ye, “Combining solar energy harvesting with wireless charging for hybrid wireless sensor networks,” IEEE Transactions on Mobile Computing, vol. 17, no. 3, pp. 560–576, 2018.
View at: Publisher Site | Google Scholar
K. Li, W. Ni, L. Duan, M. Abolhasan, and J. Niu, “Wireless power transfer and data collection in wireless sensor networks,” IEEE Transactions on Vehicular Technology, vol. 67, no. 3, pp. 2686–2697, 2018.
View at: Publisher Site | Google Scholar
M. Zhang and W. Cai, “Data collecting and energy charging oriented mobile path design for rechargeable wireless sensor networks,” Journal of Sensors, vol. 2022, pp. 1–14, 2022.
View at: Publisher Site | Google Scholar
S. He, J. Chen, F. Jiang, D. K. Y. Yau, G. Xing, and Y. Sun, “Energy provisioning in wireless rechargeable sensor networks,” IEEE Transactions on Mobile Computing, vol. 12, no. 10, pp. 1931–1942, 2013.
View at: Publisher Site | Google Scholar
Z. Zhang, H. Pang, A. Georgiadis, and C. Cecati, “Wireless power transfer-an overview,” IEEE Transactions on Industrial Electronics, vol. 66, no. 2, pp. 1044–1058, 2019.
View at: Publisher Site | Google Scholar
T. A. Kohler, D. Bird, and D. H. Wolpert, “Social scale and collective computation: does information processing limit rate of growth in scale?” Journal of Social Computing, vol. 3, no. 1, pp. 1–17, 2022.
View at: Publisher Site | Google Scholar
M. Z. Siddiqi and T. Mir, “Reconfigurable intelligent surface-aided wireless communications: an overview,” Intelligent and Converged Networks, vol. 3, no. 1, pp. 33–63, 2022.
View at: Publisher Site | Google Scholar
M. Jian, G. C. Alexandropoulos, E. Basar et al., “Reconfigurable intelligent surfaces for wireless communications: overview of hardware designs, channel models, and estimation techniques,” Intelligent and Converged Networks, vol. 3, no. 1, pp. 1–32, 2022.
View at: Publisher Site | Google Scholar
S. Kim, C. Cho, K.-J. Park, and H. Lim, “Increasing network lifetime using data compression in wireless sensor networks with energy harvesting,” International Journal of Distributed Sensor Networks, vol. 13, no. 1, Article ID 155014771668968, 2017.
View at: Publisher Site | Google Scholar
X. Fan, W. Wei, M. Wozniak, and Y. Li, “Low energy consumption and data redundancy approach of wireless sensor networks with bigdata,” Information Technology and Control, vol. 47, no. 3, pp. 406–418, 2018.
View at: Publisher Site | Google Scholar
X. Liu and J. Wu, “A method for energy balance and data transmission optimal routing in wireless sensor networks,” Sensors, vol. 19, no. 13, p. 3017, 2019.
View at: Publisher Site | Google Scholar
A. K. Sangaiah, M. Sadeghilalimi, A. A. R. Hosseinabadi, and W. Zhang, “Energy consumption in point-coverage wireless sensor networks via Bat algorithm,” IEEE Access, vol. 7, pp. 180258–180269, 2019.
View at: Publisher Site | Google Scholar
S. You, J. K. Eshraghian, H. C. Iu, and K. Cho, “Low-power wireless sensor network using fine-grain control of sensor module power mode,” Sensors, vol. 21, no. 9, p. 3198, 2021.
View at: Publisher Site | Google Scholar
D.-g. Zhang, S. Zhou, and Y.-m. Tang, “A low duty cycle efficient MAC protocol based on self-adaption and predictive strategy,” Mobile Networks and Applications, vol. 23, no. 4, pp. 828–839, 2018.
View at: Publisher Site | Google Scholar
D. Ye and M. Zhang, “A self-adaptive sleep/wake-up scheduling approach for wireless sensor networks,” IEEE Transactions on Cybernetics, vol. 48, no. 3, pp. 979–992, 2018.
View at: Publisher Site | Google Scholar
J. Qi, L. Pan, S. Ren, F. Chang, and R. Wang, “SMTS: a swarm intelligence-inspired sensor wake-up control method for multi-target sensing in wireless sensor networks,” Wireless Networks, vol. 26, no. 5, pp. 3847–3859, 2020.
View at: Publisher Site | Google Scholar
J. Huang, C. Zhang, and J. Zhang, “A multi‐queue approach of energy efficient task scheduling for sensor hubs,” Chinese Journal of Electronics, vol. 29, no. 2, pp. 242–247, 2020.
View at: Publisher Site | Google Scholar
A. Varmaghani, A. Matin Nazar, M. Ahmadi, A. Sharifi, S. Jafarzadeh Ghoushchi, and Y. Pourasad, “DMTC: optimize energy consumption in dynamic wireless sensor network based on fog computing and fuzzy multiple attribute decision-making,” Wireless Communications and Mobile Computing, vol. 2021, Article ID 9953416, pp. 1–14, 2021.
View at: Publisher Site | Google Scholar
M. R. Khosravi and S. Samadi, “Mobile multimedia computing in cyber-physical surveillance services through UAV-borne video-SAR: a taxonomy of intelligent data processing for IoMT-enabled radar sensor networks,” Tsinghua Science and Technology, vol. 27, no. 2, pp. 288–302, 2022.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2022 Yuxia Lei et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

277

Downloads

408

Citations

Scientific Programming

Intelligent Mobile Edge Computing for Smart Internet of Things: Architecture, Algorithm, and Application

Two FCA-Based Methods for Reducing Energy Consumption of Sensor Nodes in Wireless Sensor Networks

Abstract

1. Introduction

2. Related Work

3. Basic Notations in Formal Concept Analysis

4. FCA-Based Algorithms for Extracting Different MSCs

4.1. Transforming Sensors and Their Coverage Areas into a Context

4.2. Implications-Based Algorithm for Extracting MSCs

4.3. The Stem Base-Based Algorithm for Extracting MSCs

4.4. Strategies for Reducing Sensor Energy Consumption

5. Simulation Results and Analysis

5.1. Extraction of Implications Rules from Table 1

5.2. Extracting MSCs from Table 1 by Implications-Based Algorithm

5.3. Extracting MSCs from Table 1 by Stem Base-Based Algorithm

6. Conclusion

Data Availability

Conflicts of Interest

Acknowledgments

References

Copyright