A Study On Coverage Optimization Methods In Wireless Sensor Networks

Wireless sensor network(WSN) is involved with microprocessors, wireless communication, and perceptive technology, embedded technology, distributed information processing technology, and so on. In order to accommodate the development of the intelligent internet of things(IOT) system, WSN has integrated self-control ability and autonomous decision-making ability through multi-hops or cooperative communication ways. It is one of the key technologies in constructing intelligent cyber systems(ICS) and cognitive environment. The WSN technology has been applied in many fields, including military battlefield, intelligent transportation, environmental monitoring, bio-medical field, industrial control and so on. How to design the corresponding deployment strategy and covering algorithm according to the specific of application environment is always the key issue in WSN coverage research, which is directly impact the service quality of perception, monitoring, and communication, throughout the network life cycle.Researches on the problems of increasing the coverage quality take place in two aspects in this paper. On one hand, the isosceles trapezoid 2-coverage pattern(Iso-Tra C2) is used to avoid coverage holes and enhance networks’ coverage during networks deployment phase. On the other hand, efficient triangular coverage holes detection algorithm is proposed to maintain the static overlay network quality. And the distributed coverage optimization algorithms are designed to improve the quality of mobile networks coverage dynamincally during the networks usage phase.When wireless sensor network is used in a roadway, or a highway, even an underground mine, which are similar to be long strip areas, complete multi-coverage is needed for cooperative sensing application. Considering both coverage quality and deployment costs, we take 2-coverage into mainly concern here. An Iso-Tra C2 deployment pattern is put forward to make sure that any point within the long belt area should be covered by at least 2 nodes. Then the formulas of performance evaluation index for each deployment patterns have been derived. And the node density has been proved to be the quadratic function of the ratio between node radius r and belt width w. When r > w, the critical value of r/w ratio by each deployment pattern is worked out to achieve the smallest node density. In addition, when r/w=2/√3, the special case of isosceles trapezoid deployment can construct the honeycomb grid with complete 2-coverage. The analysis results show that isosceles trapezoid 2-coverage deployment pattern has advantages than the isosceles triangle deployment and the Equalization deployment, in terms of higher coverage efficiency and lower node density. The proposed deployment pattern guarantees that each point in the region is covered by double-redundant coverage with the lowest cost.In order to detect and repair coverage holes in static networks casued by random initial deployment or disasters, a triangle holes detection algorithm based on the Sum of a Polygonal Inner-angles(SPI) theorem has been proposed. Compared with traditional ones, our holes detection algorithm can overcome the dependence on the number of neighbor nodes. After detecting coverage holes, the holes repair strategies for different types of holes are designed accordingly. Also the positions calculation methods of new nodes which will be added into networks are figured out. Experimental and analysis results show that compared with the traditional adjacent nodes’ intersection-based hole detection algorithm and coverage arc-based hole detection algorithm, the SPI algorithm proposed in this paper has faster convergence speed and lower power consumption level while guaranteeing the same accuracy of detection. Our algorithm is not constrained by nodes’ radius or global information. Also its execution time hardly changes with the number of nodes and the size of monitoring area.To optimize the coverage of heterogeneous mobile network consist of nodes with different radius by stochastic initial deployment,an improved virtual force-based strategy is put forward. Combining the theory of computational geometry and graph theory, the conception of Voronoi centroid attractive virtual force is proposed to constrain nodes’ diffusion behavior. The proposed Voronoi centroid-based virtual force algorithm(VCVFA) overcomes the dependence on node density determined by network initialization in traditional algorithm, and optimizes the uniformity of coverage. The experimental simulation results show that compared with other heterogeneous network coverage optimization algorithms, VCVFA strategy has obvious advantage in promoting network coverage ratio. When the two algorithms take place under the same number of nodes, the total distance moved by the proposed algorithm is about 50% less than the distance moved by the VFA algorithm. This data illustrates that VCVFA algorithm has better energy efficiency than the traditional one. The coverage ratio by our algorithm is 10% larger than by the COSH optimization algorithm.In heterogeneous wireless sensor networks which consist of fixed sink nodes and mobile sensor nodes, the coverage of the whole Io A can’t be optimized globally by traditional virtual force algorithm. Because those mobile sensor nodes are limited by the virtual force of fixed sink nodes. To break through this constraint, the improved centroid-based attractive virtual force algorithm(CAVFA) has been proposed. Simulation experimental results show that after five step length, CAVFA algorithm performs better than the other two in improving coverage quality. The maximum coverage ratios by CAVFA, CBA, and VFA algorithm are all stabilized. But coverage ratio by VFA appears a decling trend with step length increases. Thus algorithm CAVFA integrates the advantages of the VFA and CBA algorithm, which takes advantage of the global search features by virtual force algorithm and the local unit optimization method. In this way, the proposed CAVFA algorithm overcomes the focusing phenomenon of local nodes, and supplies a more uniform coverage status and a higher coverage ratio than others.