Các thuật toán xấp xỉ tìm đường xâm nhập có khả năng bị phát hiện nhỏ nhất trong mạng cảm biến dây (approximate algorithms for solving the minimal exposure path problems in wireless sensor networks) tt tiếng anh

MINISTRY OF EDUCATION AND TRAINING HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGY NGUYEN THI MY BINH APPROXIMATE ALGORITHMS FOR SOLVING THE MINIMAL EXPOSURE PATH PROBLEMS IN WIRELESS SENSOR NETWORKS Major: Computer Science Code: 9480101 SUMMERY OF COMPUTER SCIENCE DISSERTATION Hanoi − 2020 This dissertation was finalized at Hanoi University of Science and Technology Supervisors: Assoc Prof Dr Huynh Thi Thanh Binh Assoc Prof Dr Nguyen Duc Nghia Reviewer 1: Reviewer 2: Reviewer 3: This dissertation was accepted to be defensed by the University Doctoral defense committee at Hanoi University of Science and Technology Date and time: This dissertation could be found at: Ta Quang Buu Library- Hanoi University of Science and Technology National Library INTRODUCTION WSNs have great potential for various applications in many scenarios from military target tracking and surveillance, natural disaster relief, biomedical health monitoring, hazardous environment exploration, seismic sensing, industrial diagnosis, and so on In military target tracking and surveillance, a WSN can assist in intrusion detection and identification Specific examples include spatially-correlated and coordinated troop and tank movements With natural disasters, sensor nodes can sense and detect the environment to forecast disasters before they occur In biomedical applications, surgical implants of sensors can help monitor a patient’s health For seismic sensing, ad hoc deployment of sensors along the volcanic area can detect the development of earthquakes and eruptions To evaluate the efficiency of a WSN, coverage measurement is considered as one of the fundamental problems By this means, the coverage measurement is called coverage problems which can be classified into three categories based on the coverage type: point-coverage, area coverage and barrier coverage In point coverage, the sensor nodes are deployed to cover all given specified points, whereas the goal of area coverage is that the whole region of interest is covered by the WSN Unlike these coverage types, in barrier coverage, the subjects to be covered are not fixed or known before node deployment The main focus of barrier coverage is on dynamic targets tracking and surveillance Due to the mobility of targets, barrier coverage is considered to be complicated as compared to the static ones such as area coverage and point coverage Moreover, to detect intruders penetrating the ROI, it is not necessary to guarantee that every point in the ROI is covered by one or multiple sensor nodes Therefore, full area coverage model is not suitable for intruder detection anymore In contrast, barrier coverage was proposed specifically for intruder detection in WSNs where sensing regions of sensor nodes form one or multiple barriers so that every intruder crossing the ROI will be detected Compared to full area coverage, barrier coverage can efficiently detect intruders with much fewer sensor nodes Figure Examples of target coverage (a), area coverage (b), and barrier coverage (c) Motivation This dissertation investigates the minimal exposure path (MEP) problem in WSNs which is the fundamental barrier coverage problem The MEP is a good performance metric, which can be used to measure the quality of surveillance systems or coverage quality of the sensor networks The MEP problem aims at finding a path on which an intruder can penetrate through the sensing field with the lowest probability of being detected Through the found MEP, the defenders or network designers can estimate the worst-case coverage provided by a sensor network, because objects moving through a sensor field along this path is the most difficult to be detected Therefore, the knowledge of MEP can be used in managing, optimizing and maintaining WSNs Furthermore, the applications of MEP are not only restricted to WSNs, but also in many other fields Figure Illustration of a general barrier coverage problem The MEP problem in WSN strongly depends on various factors such as type of sensors, sensing coverage model, deployment strategies, deployment environments, approach methods solving the problem, etc It is certain that almost the related works did not regard to several aspects as follows: ❼ Moving ability of sensor nodes or mobile wireless sensor networks ❼ Realistic sensing coverage model ❼ Heterogeneous wireless networks ❼ Deployment environment with obstacles These problems have become a motivation for this dissertation to further research and come up with a more efficient solutions for the MEP problem Methodology The methodology of this dissertation are as follows: ❼ Theoretical study of barrier coverage problems as optimal problems is done ❼ Analyses the related works to the barrier coverage problems Especially, the minimal exposure path problem is comprehended and considered ❼ Propose the practical and useful models of the minimal exposure path problem in wireless sensor networks ❼ Proffer efficient algorithms to solve the proposed minimal exposure path problems Scope of research The scope of dissertation is to solve the MEP problem which is a typical type of the barrier coverage problems in WSNs The MEP problem is a practical optimization problem with high dimension, non-differentiation and non-linearity To efficient cope with these characteristics, the dissertation mainly investigates the strength of metaheuristic algorithms Besides, the MEP problem becomes more useful when it is considered under some challenging scenarios as follows: ❼ Mobile wireless sensor networks ❼ Practical sensing coverage model ❼ Heterogeneous wireless sensor networks ❼ Deployment environments with obstacles Contributions This dissertation explores the barrier coverage problems in WSNs, especially, the minimal exposure path problem which is a well-known method for evaluating the coverage quality of WSNs, and is ultimately useful for sensor network designers Our main contributions in this dissertation are as follows: ❼ Presenting a rigorous theoretical analysis, and devise the formulation of the MEP problem in mobile wireless sensor networks, called MMEP Based on the characteristics of the MMEP problem, two efficient meteheuristic algorithms (GAMEP, GPSO-MMEP) are proposed to solve it ❼ Formulating a minimal exposure path problem under practical sensing coverage model, i.e the probabilistic coverage model with noise in a WSN, called PM-based-MEP A new definition of exposure measure for this model is also introduced We then propose the GB-MEP algorithm to obtain the solution based on the traditional grid-based method incorporated with several improvements To enhance the search space and more efficiently solve the problem, we design a new individual representation, an efficient crossover and a suitable mutation operator to form a genetic algorithm Conduct experiments in various scenarios to examine the proposed algorithms Quality of solutions and computation time are compared with existing methods and analyzed to give insights into the use of each algorithm in the PM-based-MEP ❼ Establishing mathematical models to represent the MEP problem in heterogeneous wireless multimedia sensor networks, called HM-MEP We also propose two efficient meta-heuristic algorithms: HEA - a hybrid evolutionary algorithm in combination with local search and GPSO - a novel particle swarm optimization based on the gravity force theory Analysis, evaluate and compare the experimental results and show that our proposed algorithms outperform the previous methods for most cases regarding quality solution and computation time ❼ The dissertation investigates a systematic and generic MEP problem under real-world deployment environment networks with presenting obstacles called OE-MEP Based upon its characteristics, we devise an elite algorithm namely FEA for solving the OE-MEP An extension to a custom-made simulation environment to integrate a variety of network topologies as well as obstacles are created Experimental results on numerous instances indicate that the proposed algorithm is suitable for the converted OE-MEP problem and performs better in both solution accuracy and computation time than existing approaches Dissertation organization The dissertation is organized as follows: Chapter presents background knowledge about the barrier coverage problem such as wireless sensor networks, optimal problems, approximate algorithms especially heuristics and metaheuristics Chapter focuses on the MEP problems in omni-directional sensor networks Chapter highlights the MEP problem in heterogeneous directional sensor networks Chapter investigates the MEP problem under real-world deployment environment sensor networks with presenting obstacles Finally, the summary and evaluation of the achieved results of the dissertation are included Additionally, the future work is also described briefly CHAPTER BACKGROUND 1.1 Wireless Sensor Networks Wireless Sensor Networks have revolutionized the IoTs industry by building a reliable and efficient communication system With the rapidly growing technology of sensors, WSNs play an important role in implementing IoTs Recently, wireless communications, computing and sensor technology have enabled the rapid development of low-cost, small-size sensor nodes that integrate sensing, data processing and wireless communication Although sensor nodes are usually resource limited, such as limited battery, memory and computation capacities, they can collaborate with each other to accomplish big tasks efficiently A typical WSN consists of thousands of sensor nodes deployed in the region of interest, which can be used to monitor physical phenomena of the ROI 1.1.1 Sensors A sensor is a device, module, machine, or subsystem whose purpose is to detect events or changes in its environment (such as heat, light, sound, pressure, magnetism, etc.) and send the information to other electronics, frequently a computer processor 1.1.2 Sensor nodes A sensor node is a node in a sensor network that is capable of performing some processing, gathering sensory information and communicating with other connected nodes in the network A typical of a sensor node consists of sensor unit, communication unit, micro-controller unit, and memory and power unit 1.1.3 Sensor coverage model Sensor coverage models are used to reflect the sensing capability and quality of sensors Commonly, most sensing functions share two aspects in common ❼ Sensing ability decreases as distance increases ❼ Due to diminishing effects of noise bursts in measurements, sensing ability can improve as the allotted sensing time (exposure) increases Assume sensor si is deployed at point (xi , yi ) For any target point at location l(x, y), the Euclidean distance between si and the target point as: (xi − x)2 + (yi − y)2 d(si , l) = (1.1) The Boolean disk coverage model f (d(si , l)) = if d(si , l) ≤ r otherwise (1.2) where d(si , l) is the Euclidean distance between a sensor si and a space target point l, and the constant r > is called sensing range.Figure 1.1(a) demonstrates a Boolean disk coverage model The truncated attenuated coverage model The coverage measure becomes very small when the distance between a space point and a sensor becomes very large In such cases, the coverage measure might be ignored, and some approximations can be made by truncating the coverage measure for larger values of distance in Equation 1.3 Ce−αd(s,z) if d(si , l) ≤ r otherwise f (d(si , l)) = (1.3) where α is a parameter representing the physical characteristics of the sensor, and r the sensing range Figure 1.1(b) demonstrates the truncated attenuated coverage model r r (a) (b) Figure 1.1 Illustration of (a) the Boolean disk coverage model in which the red stars are the target points respectively belonging inner and outer the green sensing area of a sensor, (b) the truncated attenuated coverage model 1.1.4 Sensing field intensity models The sensing field intensity models specify the collaboration of sensors in the sensing field There are two types of sensor field intensity are usually applied: The all sensing field intensity N I(l) = f (d(si , l)) (1.4) i=1 where N is the number of sensor nodes in the sensor network, i.e the sensor network consists of N sensors {s1 , s2 , , sN }, The closest sensing field intensity I(l) = f (d(s∗i , l)) |d(s∗i , l) ≤ d(sk , l) ∀k = 1, 2, , N (1.5) where s∗i is the sensor that has the smallest Euclidean distance to the target point at location l 1.1.5 Terminologies Region of interest In barrier coverage, the ROI is often a long belt domain, which is the boundary of the monitored area Sensors are deployed in the ROI to detect intruders that attempt across the belt region into the monitored area The ROI is usually assumed to be a two-dimensional belt region that is bounded by two parallel lines The ROI can be a closed belt or an open belt region Penetration path A penetration path, or a crossing path is a path that connects two the opposite sides in the ROI, where the entry point and the exit point reside on two opposite sides of the region For a two dimensional belt, orthogonal crossing paths are straight lines, whose length is equal to the width of the belt Static sensor node A static sensor node has the ability to collect sensed data, send or receive messages, process data and messages, and other types of computation in static WSNs Typically, these sensor nodes not move once they are deployed Mobile sensor node A mobile node not only has all the characteristics of the static sensor nodes, but also has some mobility A mobile sensor node can move act as a router when it is in a low or even no coverage area, and accomplish the recovery task 1.1.6 Wireless sensor network scenarios Sensor nodes are deployed in a ROI to monitor some physical phenomena Such a field is called a sensor field, and the sensor nodes form a sensor network CHAPTER MINIMAL EXPOSURE PATH PROBLEMS IN OMNI-DIRECTIONAL SENSOR NETWORKS 2.1 2.1.1 Minimal exposure path problem in mobile wireless sensor networks Problem formulation The MMEP problem can be briefly stated as follow: given N mobile sensors S = {s1 , s2 , , sN } are randomly deployed in the ROI The sensor si has the initial location si0 and the location at time t is denoted as sit Each sensor moves in a preset trajectory at constant speed Any object penetrating the monitoring region always starts at the source point and targets to reach the destination point Time interval for which the object stays inside the sensing field is [0, T ] Speed of the object is limited at a maximum value and we assume that the object always moves at its max speed since exposure will get higher if the object stays longer in the sensing field The goal of the MMEP problem is to find out a penetration path ℘ from the beginning point B to the ending point E such that the exposure value of the path ℘ is minimum More precisely, the MMEP problem is formulated as follows: Input ❼ W , L: the width and the length of the ROI ❼ N : the number of mobile sensors ❼ {s1 , s2 , , sN }: set of mobile sensor nodes deployed in ❼ si0 : the initial location of sensor ❼ Ri : trajectory of sensor si ❼ vs : the speed of sensors, all sensor nodes have the same speed ❼ vI : the max speed of the intruder ❼ (0, yB ): the coordinates of the source point B ❼ (L, yE ) : the coordinates of the destination point E Output: A path ℘ in connects from the source point B to the destination point E 11 Objective: The exposure of path ℘ is minimized, that means: T/ ∆t N k=1 i=1 E(℘, T ) ≈ f (d(si (∆t), p(k.∆t)))∆t → (2.1) Constrains: ❼ The intruder always moves towards the right side of intruder to the right border of , i.e distance from get smaller over time ❼ The intruder always moves at its max speed vI 2.1.2 2.1.2.1 Proposed algorithms The GAMEP for solving the MMEP problem Individual representation Initialization Crossover operator Mutation operator Selection operator 2.1.2.2 The HPSO-MMEP algorithm for solving the MMEP problem Individual representation Control-point initialization Integration of evolutionary operators into PSO to form HPSO-MMEP Crossover operator Mutation operator 2.1.3 Experimental results 2.1.3.1 Experimental settings 2.1.3.2 Computation results Performance of HPSO-MMEP in the presence of evolutionary operators Performance of HPSO-MMEP using control-point initialization method Comparison between HPSO-MMEP and GAMEP 12 Comparison between HPSO-MMEP and HPSO Effects of several factors of the MMEP problem on the solution 2.2 2.2.1 Minimal exposure path problem in probabilistic coverage model Problem formulation The PM-based-MEP can be briefly state as follows: given a total number N of sensor nodes to be deployed randomly in the ROI , each sensor is affected by noise environment ηi , and there are two arbitrary points that lie on opposite sides of respectively, called the source point and the destination point The goal is to find out a penetration path ℘ from the source point to the destination point such that any object moving through along path ℘ has the least detection probability Then, based on the above definition of exposure under the probabilistic model with noise, the PM-based-MEP is formulated more precisely as below Input ❼ W , L: the width and the length of the sensor field ❼ N : the number of sensors ❼ (0, yB ): coordinates of the source point B of the object ❼ (W, yE ): coordinates of the destination point E of the object ❼ σi : standard deviation of noise of sensor si (i = 0, N ) ❼ A: the detection threshold ❼ C: constant ❼ λ: attenuate exponent Output: A set L℘ of ordered points in the ROI forming a path ℘ that connects B and E Objective: The exposure of path ℘ is the smallest, which means: − ln P l ds → E (℘) = (2.2) l∈℘ Constraint: ❼ The object always moves within from B to E with a constant speed (i) ❼ The intruders are eager to cross the sensing field as quickly as possible and they never go backwards (ii) 13 2.2.2 Proposed algorithms 2.2.2.1 Grid-based algorithm for solving the MEP problem As described, the main idea of this approach is to partition the region of interest into a uniform grid Then, paths connecting the source and destination point are formed through grid points Among such paths, which has the best objective value is the solution for the problem 2.2.2.2 Genetic algorithm for solving the MEP problem Individual representation Individual initialization Crossover operators Mutation operator Selection operator 2.2.3 Experimental results 2.2.3.1 Experimental settings 2.2.3.2 Computation results The saw-tooth degree Performance of GB-MEP and GA-MEP by using different subintervals Exposure value when using different values of threshold A Computation time of GB-MEP in comparison to original method Performance of GA-MEP using different ratio of crossover operators Comparison between GB-MEP and GA-MEP Comparison between GA-MEP and HGA-NFE 2.3 Conclusion This Chapter focuses on the MEP problem in omni-directional sensor networks, and have achieved some signification results as follows: For the MEP problem in MWSNs, which is the subject of investigation, has significant meaning for identifying the weaknesses regarding coverage of WSNs The MMEP problem is converted into a numerical functional extreme problem The efficient two algorithms are proposed for solving the problem The first one is GAMEP which bases on GA To improve the quality of solution, the second one is devised, HPSO-MMEP algorithm is combined the strength of GA and PSO Different experimental instances are generated with the variation of the sensor of number, sensor 14 distribution and sensor trajectories to examine algorithm Effects of different factors on the performance of proposed algorithms as well as the network quality are thoroughly analyzed and explained after conducting numerous experiments The results show that our proposed algorithms address successfully and is more efficient than previous algorithm Furthermore, the two proposed algorithms are not restricted in the scope of this work as it is also applicable to a variety of MEP models or even more general optimization problems For the MEP problem under probabilistic coverage model, we study takes into account environmental factors such as noise, temperature, humidity and vibration that affect the sensing ability of WSNs We apply the probabilistic coverage model to the MEP problem (PM-based-MEP), and convert PM-based-MEP into a numerical function extreme problem A Grid-based method (GB-MEP) and a new genetic algorithm (GA-MEP) are also proffered and implemented to solve the PM-based-MEP problem Various experimental scenarios with different number of sensors as well as their deployment methods are designated for the simulations of our algorithms The results show that developed algorithms are highly suitable to the model and more effective than previous approaches The experimental results also prove that the solutions by GA-MEP are overall better than the solutions by GB-MEP but with a higher cost of the computation time with the same value of ∆s Depending on demanding for the solution (better quality or less computation time), one can decide which algorithm to use The PM-based-MEP with environment factor as noises have solved successfully Several essential parameters of the algorithms along with the coverage model such as subinterval ∆s and threshold A are also experimented In addition, the proposed algorithms can not only improve the results and computation time but also be applied to the extreme case with large-scale WSNs Research results in this Chapter have been published at [1, 2, 5] in Publication list 15 CHAPTER MINIMAL EXPOSURE PATH PROBLEM IN WIRELESS MULTIMEDIA SENSOR NETWORKS 3.1 Motivations Heterogeneous wireless multimedia sensor networks (HeWMSN) in IoTs has been receiving extensive attention of the research community in recent years due to its advantages for security applications One of typical wireless multimedia sensor networks is wireless directional sensor network (WDSN), in which each sensor is usually the directional sensing model Each type of sensors in the WDSNs including camera sensor, radio sensor, ultrasound sensor, or radar sensor, has its own distinctive characteristics The WDSNs can retrieve higher levels of sensor information or richer information form such as audio, image and video, thus providing more detailed information about the environment However, the WDSNs have some unique characteristics, such as limited sensing angle, directional sensing, communicating range, and line of sight These features cause the majority of existing coverage control theories and methods of traditional omni-directional WSNs can not be directly applied to WDSNs Therefore, this Chapter focuses on the MEP problem in HeWMSN 3.2 Problem formulation The MEP problem under the assumption of a directional sensing coverage model, HM-MEP problem, can be briefly described as follows: Given a set of heterogeneous sensors S with T different types, randomly deployed in the region of interest and two arbitrary points on opposite sides of the ROI , respectively the source point and the destination point The goal is to find out a penetration path from the source point B on one side to the destination point E on opposite side of sensor field such that an object moving through along the path has minimal exposure value More precisely, the HM-MEP problem is formulated as follows Input ❼ W , L: the width and the length of sensor field ❼ N : the number of sensors ❼ T : the number of sensor types 16 ❼ ti : the number of sensors that belong to type i (i = 1, 2, , T ), such that: T ti = N i ❼ (xj , y j ): the position of sensor sj − − → ❼ W d: the working direction ❼ ri : the sensing radius of type i (i = 1, 2, , T ) ❼ αi : sensing angle of the sensor type i (i = 1, 2, , T ) ❼ (0, yB ): coordinates of the source point B ❼ (L, yE ): coordinates of the destination point E Output: ❼ A set L℘ of ordered points in forming a path that connects B and E Objective: The exposure of path ℘ is the smallest, i.e E(I, ℘) = I(P )dl → ℘ (3.1) N f (d(si , O))dl → E(I, ℘) = ℘ i=1 Constraint: ❼ The object always moves within the sensor field from B to E with a constant speed (*) ❼ The object cannot go backwards (**) 3.3 Proposed algorithms The two proposed algorithms share individual representation and the fitness function in common Regarding evolutionary operators, HEA is based on evolution as a result of crossover and mutation in a genetic process, while GPSO is an improved version of standard PSO which is combined optimization method based on the social behaviors of birds flocking or fish schooling and gravitational forces 17 3.3.1 Individual representation 3.3.2 Individual initialization 3.3.2.1 HEA individual initialization 3.3.2.2 GPSO individual initialization 3.3.3 Evolutionary operators 3.3.3.1 Evolutionary operators of HEA 3.3.3.2 Evolutionary operators of GPSO 3.4 Experimental results 3.4.1 Experimental settings 3.4.2 Computation results 3.4.2.1 Comparison between HEA and GPSO 3.4.2.2 Comparison between GPSO and the previous HPSO 3.5 Conclusion This Chapter focuses on the MEP problem for HWMSNs in IoTs called HMMEP problem, which has significant meaning in identifying the vulnerability regarding the worst-case coverage path for HWMSNs The problem is very meaningful for designing and deploying HWMSN in large-scale However, the HM-MEP problem is inherently hard to solve in large-scale HWMSN, so it is converted into a numerical function extreme problem Two efficient meta-heuristic algorithms are also proffered and implemented to solve this problem Various experimental scenarios are designated for the simulations of our algorithms The essential parameters of the algorithms such as ∆s, c1 , c2 and c3 are also experimented The results extrapolate that our developed algorithms are highly suitable to the model and more effective than previous approaches The experimental results also evidence that the solutions by HEA are better than the solutions by GPSO but with a higher cost of computation time with the same value of ∆s Depending on our requirement for the solution (better quality or less computation time), we can choose the suitable algorithm Furthermore, the proposed algorithms not only can improve the results as well as computation time but also can be applied to the extreme case with large-scale HWMSNs Therefore, the proposed algorithms can produce high quality results in an efficient way and can be used as the worst-case path coverage analysis tool in HWMSNs The HM-MEP problem has been solved successfully by meta-heuristic algorithms Research results in this Chapter have been published at [4] in PUBLICATION list 18 CHAPTER OBSTACLES-EVASION MINIMAL EXPOSURE PATH PROBLEM IN WIRELESS SENSOR NETWORKS 4.1 Motivations Wireless sensor networks are expected to be deployed in environments with certain terrain complexity or even hostile areas wherein some parts can be void of sensors Thus, modeling WSN with the occurrence of obstacles certainly has a great impact on the MEP problem formulation as well as the design, simulation, and performance evaluation of the algorithms in solving the MEP problem Modeling networks with obstacles certainly helps to improve the reliability and credibility of the analytical results (using network models and or simulations) which are vital for network designers to apply obtained findings in predicting the detection performance of the sensor networks (in barrier coverage scenarios) without the expense of realworld deployment and test Thus, this Chapter investigates the obstacles-evasion MEP problem in WSNs 4.2 Problem formulation The OE-MEP problem can be briefly described as follows: given a set of truncated directional sensors S = {s1 , s2 , , sN } randomly deployed in the sensor field, a set of obstacles Z = {ZO1 , ZO2 , , ZOH } and two arbitrary points on opposite sides of the field, respectively the source point and the destination point The goal is to find out a penetration path from the beginning point B to the ending point E such that an object moves through along path without crossing any obstacles has minimal exposure value More precisely, the OE-MEP problem is formulated as follows Input ❼ W , L: the width and the length of sensor field ❼ N : number of sensors ❼ The sensor si (i = 1, 2, , N ): – (xi , yi ): position of sensor si − − → – W di : working direction of sensor si 19 – ri : sensing radius of sensor si – αi : sensing angle of sensor type si ❼ H: number of obstacles ❼ The obstacle ZOl (l = 1, 2, , H): – A list of vertex points GOl of obstacle ZOl – νl : the absorption coefficient of the obstacle ZOl ❼ (0, yB ): coordinates of the beginning point B of the object ❼ (L, yE ): coordinates of the ending point E of the object Output: ❼ A set L℘ of ordered points in forming a path that connects B and E Objective: The exposure of path ℘ is the smallest, i.e K N E(I, ℘) ≈ f (si , Tj )∆s → M in (4.1) j=0 i=1 where K is the number points included in L℘ Constraints: ❼ The object always moves within the sensor field from B to E with a upper- bounded speed (i) ❼ The path can not cross any area of obstacles (ii) 4.3 Proposed algorithm The proposed FEA is proposed to try the best to reduce the possibility of getting trapped in the local optima without increasing the computation time, by adding the Family System into the population 20 4.3.1 Individual representation 4.3.2 Individual initialization 4.3.3 Family pairing 4.3.4 Crossover operator 4.3.5 Mutation operator 4.3.6 Update operator 4.3.7 Selection 4.4 Experimental results 4.4.1 Experimental settings 4.4.2 Computational results 4.4.2.1 The performance of FEA when using different A and D values 4.4.2.2 The performance of FEA when using different pmin and pmax values 4.4.2.3 Comparison between FEA and previous algorithm 4.5 Conclusion This Chapter proposes to investigate the OE-MEP problem in WSN where obstacles are presented in their arbitrary shapes, and can be used as a tool to determine the weaknesses in coverage level of a given sensor network The goal of the OE-MEP problem is to find out a path that penetrates through the field with the minimal exposure value and does not cross any obstacles.This problem is substantially beneficial for network designers who can apply established formulas to evaluate the quality of coverage of the provided WSN without costly deployment and test The OE-MEP is formulated and presented as a generic mathematical model which then is converted into an optimization problem with constraints We create a family system based evolutionary algorithm (FEA) in an attempt to solve this OE-MEP problem efficiently We consider obstacles with arbitrary shapes (to match realistic scenarios) and we model these obstacles as convex polygons and create random data sets to effectively measure the performance of the OE-MEP approach We then conduct numerous systematic simulations to test the performance of our proposed FEA algorithm with a variety of network scenarios and obstacles The results show evidence that FEA is strongly suitable for solving the OE-MEP problem and more efficient than prior approaches regarding solution quality and computational time 21 CONCLUSIONS AND FUTURE WORKS Contributions This dissertation surveys the approximation algorithms for dealing with the barrier coverage problem in wireless sensor networks Especially, it has studied and proposed some new models of the MEP problem which is roof of barrier coverage problem, devised the efficient metaheuristic algorithms to solve the proposed MEP problems The dissertation has been achieved the highlight research results as follows: The main results of the dissertation include: For omni-directional mobile wireless sensor networks, the MEP problem in mobile wireless sensor networks with several different sensor coverage models (MMEP problem) is studied and has been obtained some emphasized results ❼ Formulate the MMEP problem in MWSNs with several different sensing coverage models ❼ Devise two efficient metaheuristic algorithm: GAMEP, HPSO-MMEP in which GAMEP is a genetic algorithm and HPSO-MMEP is integrated strength of genetic operators into PSO algorithm to significantly improve the performance of original PSO and GAMEP ❼ Design a new individual representation and propose a strategy for generating individuals called controlled-point (CP) initialization which ensures the diversity of the generated population ❼ Conduct intensive numerous experiments in various scenarios to evaluate the performance of GAMEP and HPSO-MMEP The experimental results are thoroughly analyzed to give insights into the affects of different factors Regarding omni-directional static wireless sensor networks, we have investigated the MEP problem based on probabilistic coverage model with noise (PM-based-MEP), and accomplished some results as follows: ❼ Formulating a minimal exposure path problem under the probabilistic coverage model with noise in a WSN, called PM-based-MEP A new definition of exposure measure for this model is also introduced ❼ Converting the PM-based-MEP into an optimization problem with a objective function and constraints which permit the use of mathematical optimization methods to solve 22 ❼ Proposing the GB-MEP algorithm to obtain the solution based on the traditional grid-based method incorporated with several improvements To enhance the search space and more efficiently solve the problem, we design a new individual representation, an efficient crossover and a suitable mutation operator to form a genetic algorithm called GA-MEP ❼ Conducting experiments in various scenarios to examine the proposed algorithms Quality of solutions and computation time are compared with existing methods and analyzed to give insights into the use of each algorithm in the PM-based-MEP In term of heterogeneous directional wireless sensor networks, the MEP problem in HWDSNs (HM-MEP) is focused, and has been obtained the main results: ❼ Establish mathematical models to represent the HM-MEP problem ❼ Propose two efficient meta-heuristic algorithms: HEA - a hybrid evolutionary algorithm in combination with local search and GPSO - a novel particle swarm optimization based on the gravity force theory ❼ Analysis, evaluate and compare the experimental results and show that our proposed algorithms outperform the previous methods for most cases regarding quality solution and computation time With respect to the MEP problem in real-world WSN scenarios, the MEP problem in WSNs with arbitrary-shape obstacles is addressed successfully, and has been achieved the results as follows: ❼ Formulate a generic mathematical model to represent the arbitrary shape obstacles-evade MEP problem in WSNs, called OE-MEP and convert OEMEP into an optimization problem ❼ Model the obstacles as convex polygons to match realistic scenarios and devise a method to randomly generate obstacles in different forms The newly created data set can serve as an effective measurement for the performance of OEMEP approaches ❼ Propose a new algorithm called Family System based Evolutionary Algorithm (FEA) for solving the OE-MEP problem efficiently ❼ Conduct a number of systematic simulations to study the performance of FEA under a variety of network scenarios as well as obstacles Limitations lthough the MEP problems in WSNs have been studied under several sensing coverage models efficiently, the dissertation has still some limitations as follows: 23 ❼ The MEP problem has not addressed in WSNs under modern sensing coverage models such as full-view sensing coverage, k-angle sensing coverage yet ❼ The MEP problem has not solved in 3-Dimensions WSNs yet ❼ The MEP problem has not taken into account rotating capability of sensor nodes Future works Although barrier coverage has been actively studied by academic community, there is still a long way to go before it can be applied into large-scale practical systems In our future work, we would like to focus on applying barrier coverage into real systems Three-dimension barrier coverage Most prior researches studied on the barrier coverage problem in two-dimensional spaces The gap between theory and practice will cause that laboratory researches cannot be applied in large-scale practical systems However, in many practical scenarios, sensors deployed in the atmosphere, in underwater or in the sea, and the outer space may need to guarantee the barrier covered in three-dimensional space, or referred to shell coverage It is not straightforward to extend the approaches proposed for two-dimensional barrier coverage to adapt to three-dimensional barrier coverage, and new algorithms need to be designed Practical sensing coverage model Most of existing barrier coverage solutions, the directional sensing model models are simple and ideal such as binary sensing model, attenuated sensing model, etc Recently, practical sensing coverage model, full-view coverage model, and some application-oriented barrier coverage, bring new reflection and enlightenment to the design of barrier coverage solutions Sensor with rotating capabilities In directional sensing coverage models, a sensor can only sense in the direction of its orientation A rotating directional sensor can change the orientation of its sensor at a certain rotational speed to provide good coverage When these sensors are randomly deployed in the ROI, one interesting problem is how to select appropriate rotating directional sensors and their working orientations to guarantee barrier coverage 24 PUBLICATIONS [1] Binh, N.T.M., Thang, C.M., Nghia, N.D and Binh, H.T.T., 2017, Genetic algorithm for solving minimal exposure path in mobile sensor networks In 2017 IEEE Symposium Series on Computational Intelligence (SSCI) (pp.1-8) [2] Binh, H.T.T., Binh, N.T.M., Ngoc, N.H., Ly, D.T.H and Nghia, N.D., 2019 Efficient approximation approaches to minimal exposure path problem in probabilistic coverage model for wireless sensor networks Applied Soft Computing, 76, pp.726-743 (SCIE, Q1, IF 4.873) [3] Binh, N.T.M., Binh, H.T.T., Le Loi, V., Nghia, V.T., San, D.L and Thang, C.M., 2019, An efficient approximate algorithm for achieving (k − ω) barrier coverage in camera wireless sensor networks In Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications (Vol 11006, p 1100613) International Society for Optics and Photonics [4] Binh, N.T.M., Binh, H.T.T., Van Linh, N and Yu, S., 2020, Efficient meta-heuristic approaches in solving minimal exposure path problem for heterogeneous wireless multimedia sensor networks in Internet of Things Applied Intelligence, 50, pp 1889–1907 (SCIE, Q2, IF 3.325) [5] Binh, N.T.M., Abdelhamid Mellouk, Binh, H.T.T., Loi, L.V, San, D.L, Anh, T.H, 2020, An elite hybrid particle swarm optimization for solving minimal exposure path problem in mobile wireless sensor networks Sensors, 20(9), p.2586-2611 (SCIE, Q1, IF 3.275) ... arbitrary points on opposite sides of the field, respectively the source point and the destination point The goal is to find out a penetration path from the beginning point B to the ending point E... dissertation 10 CHAPTER MINIMAL EXPOSURE PATH PROBLEMS IN OMNI-DIRECTIONAL SENSOR NETWORKS 2.1 2.1.1 Minimal exposure path problem in mobile wireless sensor networks Problem formulation The MMEP problem... the beginning point B to the ending point E such that the exposure value of the path ℘ is minimum More precisely, the MMEP problem is formulated as follows: Input ❼ W , L: the width and the length