Wireless Networks and Mobile Communications for Internet of Things
View this Special IssueResearch Article  Open Access
Yang Yang, Yufei Wang, Dechang Pi, Ruchuan Wang, "Optimization of SelfDirected Target Coverage in Wireless Multimedia Sensor Network", The Scientific World Journal, vol. 2014, Article ID 416218, 9 pages, 2014. https://doi.org/10.1155/2014/416218
Optimization of SelfDirected Target Coverage in Wireless Multimedia Sensor Network
Abstract
Video and image sensors in wireless multimedia sensor networks (WMSNs) have directed view and limited sensing angle. So the methods to solve target coverage problem for traditional sensor networks, which use circle sensing model, are not suitable for WMSNs. Based on the FoV (field of view) sensing model and FoV disk model proposed, how expected multimedia sensor covers the target is defined by the deflection angle between target and the sensor’s current orientation and the distance between target and the sensor. Then target coverage optimization algorithms based on expected coverage value are presented for singlesensor singletarget, multisensor singletarget, and singlesensor multitargets problems distinguishingly. Selecting the orientation that sensor rotated to cover every target falling in the FoV disk of that sensor for candidate orientations and using genetic algorithm to multisensor multitargets problem, which has NPcomplete complexity, then result in the approximated minimum subset of sensors which covers all the targets in networks. Simulation results show the algorithm’s performance and the effect of number of targets on the resulting subset.
1. Introduction
Wireless multimedia sensor network (WMSN) [1], an advancing form of wireless sensor network (WSN) [2], is a multihop selforganizing distributed sensing network which is constituted by battery of multimedia sensor nodes with ability of calculation, storage, and wireless communication. Through multimedia sensor in nodes, this advanced network can find, collect, and handle many kinds of media information, such as audio, video, and image, in surrounding environment. Then the multimedia data will be sent to information gathering center by multihop trunking scheme. After data analysis in information gathering center, overall and effective environment monitoring will be accomplished [1, 3].
One of the most important issues in WSN is sensing coverage, which is also measurement for service quality in WSN [4]. In traditional sensor network, the sensor coverage is usually hypothesized as omnidirectional and predefined as a crude round [1]. But, in WMSN, the coverage method is very different with its traditional sensor network. The nodes in WMSN, such as video and image sensor nodes, catch a directed visual field, named as field of view (FoV) [5]. Usually the hypothesis is that direction of multimedia sensor node is adjustable [6]. When the node is randomly deployed into malicious environment, it can cover the target and the area selfdirectly through cooperation with neighbor nodes.
In this paper, the coverage expectation value from node to target is delineated through probabilistic models of FoV disk, which is the ground for selfdirection wireless multimedia sensor node and better coverage of target. The optimization algorithms for singlenode singletarget, multinode singletarget, and singlenode multitarget are given to solve the coverage problem. Also, the selfdirected direction of target covered by FoV disk for every node is set as candidate sensor direction. Genetic algorithm is used to discriminate the selfdirected coverage optimization in multinode multitarget circumstance and to discover the minimum set of nodes for coverage of all targets.
2. Relevant Work
Coverage issue of randomly deployed sensor node is the hotspot in this field. In traditional WSN, most researches hypothesise that the omnidirection sensor node which covers area is a circle with its center in node. But, in WMSN, the sensor cover area is usually hypothesized as sector, which is not applicable for traditional coverage algorithm [6, 7]. Coverage issue can be divided into two categories, the area coverage which ensures that the whole area is covered by sensor node and the target coverage which ensures that every interested target in district is covered by at least one sensor node.
New distributed algorithm has been raised after research in selfdirected WMSN [6]. This algorithm realizes node optimal sensing direction, minimal node coverage redundancy in sensing field, and maximal multimedia coverage. Directional sensor network deployment with connectivity and coverage guarantee has been considered in recent study [8]. In this study, minimal directional sensor network was deployed to form a connected network covering interested area. For the target coverage in directional sensor network, different priority was assigned to target [7]. Taking into account the distance between the target and node, a problem of target coverage based on priority has been suggested and a minimal node set was tried to be chosen for all targets monitoring.
In target coverage problem of directional sensor network, the direction and rotation of angle in node directional compensation were not considered [9]. Also, for resolving multinode multitarget coverage optimization, a battery of fixed sensing direction was set up for every sensor node, which was related to node angle of sensing [9]. But rotatable selfdirected sensor node has 360° sensing direction. In this research, the selfdirected direction and angle of rotation are inducted into selfdirection process of multimedia sensor network node, which could decrease the adjustment of sensor node to save energy while optimizing target coverage.
3. Network Hypothesis and Multimedia Coverage FoV Model
The network field in our research is twodimensional Euclidean field with randomly distributed multimedia sensor nodes and a certain number of interested targets, which means that the sensing direction and position of all nodes are random and independent. The same as the hypothesis in recent study [6], the image senor is equipped with multimedia sensor node to provide FoV with angle value , which could bilaterally rotate to redefine sensing direction in twodimensional Euclidean area. Simultaneously, it is hypothesized that monitoring target and multimedia sensor node can achieve their position information through lightweight positioning technology of WSN [10]. Also the nodes in network are hypothesized to have same sensing model, which means that they have identical sensing radius and angle. Since the WMSN is selforiented, the sensor nodes constituting network can regulate their sensing angle, which means that inexpensive multimedia node can rotate around its vertical axis [6].
In a WMSN, is the set of multimedia sensor nodes, is sensing radius of node, and interested target battery is . With regard to sensing model of multimedia sensor node, the definition is provided as follows.
Definition 1. 2D FoV: 2D FoV is a directional sensing area of multimedia sensor node, which is hypothesized as a proximate sector in twodimensional space (Figure 1). 2D FoV of node is defined as tetrad , with as sensor, as sensing radius, as vertex angle of sector, and as current sensing direction of node which divide sector into two halves.
Definition 2. 2D FoV disk: the 2D FoV disk of multimedia sensor node is defined as a set of all possible 2D FoV of node, which should be a round area with radius as (Figure 2).
Definition 3. Target coverage of multimedia sensor node : equation (1) shows the coverage of node to target : If the interested target was located in one FoV disk of some node and not covered by this node in present, there is the possibility that this target could be covered by this node when proceeding with its selfdirection. Taking into account decline of monitoring quality followed by increasing distance between interested target and node, along with the deflected angle between the interested target and current sensing direction of node ( as shown in Figure 2), the coverage expectation value of node to interested target is defined.
Definition 4. Expectation value of multimedia sensor node to target : equation (2) shows coverage expectation value of node to target : where means current sensing direction to interested target anticlockwise, with value between and parameter , , which defined the adjustable weight of diverted angle and distance between target and node in coverage expectation.
4. SelfDirected Target Coverage Optimization
4.1. SingleNode SingleTarget SelfDirection
At first, singlenode single interested target coverage, the simplest circumstance, is discussed. The hypothesis is that interested target was located in FoV disk of node but not covered by node, which needs selfdirected adjustment to be covered (Algorithm 1). As long as the target was located in FoV disk, it does not need to regard the distance between target and node in singlenode singletarget selfdirection model. The purpose is the minimal rotated angle of node to realize optimal coverage to interested target which is located in , sensing direction of node.

4.2. Multinode SingleTarget SelfDirected Coverage
The problem of multinode singletarget selfdirection is existence of multiple multimedia sensor nodes. The hypothesis is that interested target is not covered by any node while all nodes can cover the interested target through selfdirection. According to the coverage expectation value of multimedia sensor node to target, the node with maximal coverage expectation value is chosen to proceed with selfdirection and cover interested target, which is located in sensing direction of node, in optimal angle. The distance between interested target and node and the angel adjusted during selfdirection must be considered (see Algorithm 2).

4.3. SingleNode Multitarget SelfDirected Optimal Coverage
The problem of singlenode multitarget direction is how to cover maximal interested targets in node selfdirection in the circumstance of multiple interested targets randomly located in FoV disk of some node (see Algorithm 3). The central idea of the algorithm is how to select the interested target subset, which is the maximal subset of the difference between any two targets deflection angles less than or equal to FoV vertex angle, as the target of node selfdirected coverage optimization when measured by deflection angle of interested target to node current sensing direction.

4.4. Multinode Multitarget SelfDirected Coverage Optimization
The problem of multinode multitarget selfdirected coverage optimization in WMSN is how to find the minimal node set covering a definite interested target set after selfdirection. According to coverage corresponding relationship after optimization, the selfdirection is proceeded to cover the target node. For expediently showing the selfdirection (direction and angle of rotation) of every node, the variance is introduced.
Node rotates to direction in angle of , 1 as clockwise, 0 as anticlockwise, or 0
The multinode multitarget selfdirected coverage optimization can be described as follows: where is falling in , which means that interested target is covered after node set selfdirection. Objective function (3) is minimal subset of multimedia sensor node which can cover all targets after selfdirection. Objective function (4) is reciprocal value of coverage expectation for minimal node subset to all targets. Restrictive condition (5) promises that every interested target can be covered by node set after selfdirection. Restrictive condition (5) promises that every sensor node can only rotate to one direction at a given angle at any time.
To be described more vividly, a dilatation figure of multinode multitarget selfdirected coverage optimization based on simple example in Figure 3 is shown in Figure 4. In literate [11, 12], the potential sensing direction of every node is fixed. But, in this algorithm, every node can define its potential sensing direction according to node covered by its FoV disk, which is named as selfdirection. The top of figure includes multimedia sensor node, selfdirection action, and interested target. The left side column represents node set . The middle column is selfdirection action when every node is covering interested target in its own FoV disk. The right column is interested target set . Every multimedia sensor node is connected with its own selfdirection action. Every node can only choose one side because every multimedia sensor node can only do one selfdirection action at one time. Every interested target connects with corresponding selfdirection action, which means some node can cover this target after selfdirection.
Genetic algorithm is used. According to the methods used in [11, 13], the random algorithm in Algorithm 4 is used at first to generate initiate population. It means to generate all possible paths of node set covering all targets through selfdirection action based on dilatation figure.

Random algorithm showed in Algorithm 4 may generate empty set. But through executing this algorithm repeatedly, a bigger set of feasible paths in dilatation can be generated, which can be used as initiate group for genetic algorithm. Discriminate algorithm based on genetic algorithm is shown in Algorithm 5.

5. Simulation Analysis
Simulation analysis is performed to multinode multiinterested target selfdirection optimal coverage discriminate algorithm based on genetic algorithm [14]. The MATLAB 7.6.0 and genetic algorithm toolbox function are used. multimedia sensor nodes and interested targets are randomly sown into area. Sensing direction of node is randomly distributed in and every interested target should be covered by FoV disk of at least one node. While simulation, the number of sensor nodes is set as 100, sensing angle of sensing node is set as , and sensing radius is set as 100 . Ability of algorithm with 10, 20, and 30 interested targets is investigated individually. Algorithm runs 100 times under every circumstance. Figure 5 shows the stability of multitarget selfdirection coverage optimization by this algorithm under three circumstances. Floating error range of optimal results is less than 3 nodes. The floating range of algorithm running results increased slightly with augmented interested target.
Figure 6 shows optimal solution of target function and ability tracking of algorithm in different number of interested targets. It can be discerned that our algorithm can generate optimal solution with fewer iteration frequencies, which would increase with augmented interested target. Figure 7 shows the relationship between number of interested targets and optimal solution of target function. The size of minimal node set which can totally monitor interested targets increases linearly with increasing in number of interested targets. Every data in figure means average value of algorithm ran 100 times.
(a) 10 interested targets
(b) 20 interested targets
(c) 30 interested targets
6. Conclusion
In this paper, the factor of rotation angle, rotation direction, and distance between node and interested target when multimedia sensor node selfdirected cover interested target in selfdirected MSN is discussed. Also we investigate singlenode singletarget, multinode multitarget, and singlenode multitarget selfdirected coverage optimization, so node can selfdirect in maximal coverage expectation value, which means covering more closer targets with lesser rotation angle. For multinode multitarget selfdirected coverage optimization, the hypothesis that all nodes have fixed sensing direction in literature [11] is abandoned. The potential selfdirected sensing direction in node is defined by number of interested targets in its FoV disk, which would more fit the feature of randomly deployed WMSN. Then the problem is abstracted into optimization problem with restrictive conditions. Genetic algorithm is used to discriminate. Also simulation analysis is performed. Further study may focus on how to realize multinode multitarget selfdirected coverage optimization with induction of coverage expectation based on minimal node covering interested target.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
Foundation item: This paper is sponsored by Qing Lan Project and the National Natural Science Foundation of China (nos. 61170065, 61373017, 61171053, 61103195, and 61203217).
References
 I. F. Akyildiz, T. Melodia, and K. R. Chowdhury, “A survey on wireless multimedia sensor networks,” Computer Networks, vol. 51, no. 4, pp. 921–960, 2007. View at: Publisher Site  Google Scholar
 I. F. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, “Wireless sensor networks: a survey,” Computer Networks, vol. 38, no. 4, pp. 393–422, 2002. View at: Publisher Site  Google Scholar
 M. Huadong and T. Dan, “Multimedia sensor network and its research progresses,” Journal of Software, vol. 17, no. 9, pp. 2013–2028, 2006. View at: Google Scholar
 A. Ghosh and S. K. Das, “Coverage and connectivity issues in wireless sensor networks: a survey,” Pervasive and Mobile Computing, vol. 4, no. 3, pp. 303–334, 2008. View at: Publisher Site  Google Scholar
 U. M. Erdem and S. Sclaroff, “Optimal placement of cameras in floorplans to satisfy task requirements and cost constraints,” in Proceedings of the 5th Workshop on Omnidirectional Vision, Camera Networks and NonClassical Cameras (Omnivis '04), Prague, Czech Republic, May 2004. View at: Google Scholar
 N. Tezcan and W. Wang, “Selforienting wireless multimedia sensor networks for occlusionfree viewpoints,” Computer Networks, vol. 52, no. 13, pp. 2558–2567, 2008. View at: Publisher Site  Google Scholar
 J. Wang, C. Niu, and R. Shen, “Prioritybased target coverage in directional sensor networks using a genetic algorithm,” Computers and Mathematics with Applications, vol. 57, no. 1112, pp. 1915–1922, 2009. View at: Publisher Site  Google Scholar
 M. Cardei, D. MacCallum, X. Cheng et al., “Wireless sensor networks with energy efficient organization,” Journal of Interconnection Networks, vol. 3, no. 34, pp. 213–229, 2002. View at: Publisher Site  Google Scholar
 D. Tian and N. D. Georganas, “A coveragepreserving node scheduling scheme for large wireless sensor networks,” in Proceedings of the 1st ACM International Workshop on Wireless Sensor Networks and Applications (WSNA '02), pp. 32–41, Atlanta, Ga, USA, September 2002. View at: Google Scholar
 P. Wan and C. Yi, “Coverage by randomly deployed wireless sensor networks,” IEEE Transactions on Information Theory, vol. 52, no. 6, pp. 2658–2669, 2006. View at: Publisher Site  Google Scholar
 M. Cardei and J. Wu, “Energyefficient coverage problems in wireless adhoc sensor networks,” Computer Communications, vol. 29, no. 4, pp. 413–420, 2006. View at: Publisher Site  Google Scholar
 P. K. Sahoo, J. P. Sheu, and W. S. Lin, “Dynamic coverage and connectivity maintenance algorithms for wireless sensor networks,” in Proceedings of the 2nd International Conference on Communication System Software and Middleware and Workshops (COMSWARE '07), pp. 1–9, Bangalore, India, January 2007. View at: Publisher Site  Google Scholar
 X. F. Han, X. Cao, E. L. Lloyd, and C. C. Shen, “Deploying directional sensor networks with guaranteed connectivity and coverage,” in Proceedings of the 5th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON '08), pp. 153–160, San Francisco, Calif, USA, June 2008. View at: Publisher Site  Google Scholar
 T. He, C. Huang, B. M. Blum, J. A. Stankovic, and T. Abdelzaher, “Rangefree localization schemes for large scale sensor networks,” in Proceedings of the 9th Annual International Conference on Mobile Computing and Networking (MobiCom '03), pp. 81–95, San Diego, Calif, USA, September 2003. View at: Publisher Site  Google Scholar
Copyright
Copyright © 2014 Yang Yang 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.