1. Probabilistic Coverage in Wireless Sensor Networks Authors : Nadeem Ahmed, Salil S. Kanhere, Sanjay Jha Presenter : Hyeon, Seung-Il

2. Contents Introduction Related Work Technical Preliminaries Probabilistic Coverage Algorithm Simulation Setup Conclusion and Future Work 2008 Advanced Ubiquitous Computing 2

3. Introduction Wireless sensor networks differ from ad-hoc networks in several ways Introduction of the sensing component  Performing two demanding tasks simultaneously, sensing and communicating with each other The sensing coverage is usually assumed uniform in all direction The binary detection model Unrealistic assumption The sensing capabilities are affected by environmental factors It is imperative to have practical considerations at the design 2008 Advanced Ubiquitous Computing 3

4. Introduction This paper presented, Explore the problem of determining the coverage Capture the real world sensing characteristics of sensor nodes using the probabilistic model  Based on the path loss log normal shadowing model Propose the Probabilistic Coverage Algorithm (PCA)  Extension of the perimeter coverage algorithm Simulation results  Coverage calculated using probabilistic coverage algorithm is more accurate than the idealistic binary detection model 2008 Advanced Ubiquitous Computing 4

5. Related Work Most of the coverage related protocols assume uniform sensing ranges Different from above research in several ways Computational geometry based approach Coverage is calculated at perimeter of each node sensing circles This approach is truly distributed  Added advantage of being scalable and robust to failures Differ from the perimeter coverage algorithm Sensing capabilities in all directions is always probabilistic The detection probability depends on the relative position of the event/target from the sensor 2008 Advanced Ubiquitous Computing 5

6. Technical Preliminaries Using the log-normal shadowing model The path loss PL(in dB) at a distance d is n and X σ can be measured experimentally Also, PL(d0) can be measured experimentally 2008 Advanced Ubiquitous Computing 6

7. Technical Preliminaries Each sensor has a receive threshold value γ Describes the minimum signal strength For a given transmit power and receive threshold value, we can calculate the probability of receiving a signal at a given distance ( =d ) using Equations (2) and (4) 2008 Advanced Ubiquitous Computing 7

8. Technical Preliminaries The decrease in detection probability for a sensor based on shadowing model for parameters shown in Table 1 2008 Advanced Ubiquitous Computing 8

9. Technical Preliminaries A point in the target region can be covered by more than a single sensor Find the product of the individual detection probabilities of all sensors receiving the event occurring at that point 2008 Advanced Ubiquitous Computing 9

10. Probabilistic Coverage Algorithm The coverage depends on these: The sensing capabilities of the sensor The event characteristics The transmit power and the receive threshold of sensors are known value PT (Table 2) can be precomputed using Equation 1-4 2008 Advanced Ubiquitous Computing 10

11. Probabilistic Coverage Algorithm Definition 1 Effective coverage range, R effec, of a sensor S i is defined as distance of the target from the sensor beyond which the detection probability is negligible R effec is taken as the distance at which the probability of detection falls below 0.1 2008 Advanced Ubiquitous Computing 11

12. Probabilistic Coverage Algorithm The cumulative detection probability for two neighbors in a region, for parameters listed in Table 1 The cumulative detection probability is higher if neighbor sensors are located near each other 2008 Advanced Ubiquitous Computing 12

13. Probabilistic Coverage Algorithm Definition 2 A location in region A is said to be sufficiently covered if its cumulative detection probability, due to sensors located within the effective coverage range R effec of this location, is equal to or greater than DDP, the detection probability desired by the application Objective is to check whether all locations in the given region are sufficiently covered or not 2008 Advanced Ubiquitous Computing 13

14. Probabilistic Coverage Algorithm Make the following assumptions for this work Sensors are randomly deployed in the field Location information is available to each sensor node Communication range of sensors is at least twice the effective coverage range, R effec Sensors can detect boundary of the region if the boundary is within a sensor’s R effec Transmit power of target Pt and receive threshold γ for a sensor are known and γ is the same for all the sensors Mean values of path loss component n and shadowing deviation σ are assumed for all the sensors 2008 Advanced Ubiquitous Computing 14

15. Probabilistic Coverage Algorithm PCA A node S i receives location information from all of its one hop communication neighbors  S i has two sensing circles with radius d reqd and d eval Node S i first detects whether it is within vicinity of the region boundary  The segments on perimeter that lie outside the region boundary are assigned detection probability of 1  Sensor do not need to calculate coverage for this part Next, neighbor contribution towards detection probability is calculated  Paper only consider neighbor contribution from nodes within a distance of 2 * d eval 2008 Advanced Ubiquitous Computing 15

16. The value of distance increment being a tradeoff between the computational time and detection granularity 2008 Advanced Ubiquitous Computing 16

17. Probabilistic Coverage Algorithm 2008 Advanced Ubiquitous Computing 17

18. Probabilistic Coverage Algorithm 2008 Advanced Ubiquitous Computing 18

19. Probabilistic Coverage Algorithm Line 10-14 explained by Figure 5 Definition 3 If the perimeter of a sensor S i circle with radius d eval is covered by cumulative detection probability ρ reqd, the region inside the circle is sufficiently covered with detection probability at least ρ reqd 2008 Advanced Ubiquitous Computing 19

20. Probabilistic Coverage Algorithm Theorem 1 The whole region A is sufficiently covered by ρ reqd if all sensors in the region has perimeter, of circle with radius d eval (> d reqd ), sufficiently covered with detection probability ρ ≥ DDP Following Theorem 1, If all the sensors report sufficiently covered perimeters at C i (d eval ), the whole region is sufficiently covered The information from all sensors describe the current state of area coverage supported by the sensor network  It can be utilized to deploy more sensors in the topology Coverage hole detection  or to guide mobility capable redundant nodes to specific locations to satisfy the detection probability constraint 2008 Advanced Ubiquitous Computing 20

21. Simulation Setup Implemented in NS2 simulator d reqd is 6m for ρ reqd 0.9, d eval is 9m for ρ eval 0.655 The PCA provides a more granular and accureate estimate of the coverage and detection probability 2008 Advanced Ubiquitous Computing 21

22. Conclusion and Future Work Proposed a probabilistic coverage algorithm To estimate area coverage in a randomly deployed WSN The algorithm Adopts a probabilistic approach Simulation shows effectiveness of the algorithm Relax some of the assumptions Mean values of path loss component, n The shadowing deviation, σ n and σ varies spatially as well as temporally due to changing environments Multiple coverage constraint 2008 Advanced Ubiquitous Computing 22