1 / 56

Distributed Deployment Schemes for Mobile Wireless Sensor Networks to Ensure Multi-level Coverage

Distributed Deployment Schemes for Mobile Wireless Sensor Networks to Ensure Multi-level Coverage. You-Chiun Wang( 王友群 ) and Yu-Chee Tseng( 曾煜棋 ) Department of Computer Science National Chiao-Tung University Present by C.T. Lee 2007 / 6 / 4. Outline. Introduction Problem Statement

gauri
Download Presentation

Distributed Deployment Schemes for Mobile Wireless Sensor Networks to Ensure Multi-level Coverage

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Distributed Deployment Schemes for Mobile Wireless Sensor Networks to Ensure Multi-level Coverage You-Chiun Wang(王友群) and Yu-Chee Tseng(曾煜棋) Department of Computer Science National Chiao-Tung University Present by C.T. Lee 2007 / 6 / 4

  2. Outline • Introduction • Problem Statement • K-coverage Sensor Placement Schemes • Distributed Sensor Dispatch Schemes • Experimental Results • Conclusions • Comments

  3. I. Introduction (1/9) • In this paper, authors solve the k-coverage sensordeployment problemto achieve multi-level (k) coverage of the area of interest I • k-coverage placement problem • The placement problem asks how to determine the minimum number of sensors required and their locations in I to guarantee that I is k-covered and the network is connected • distributed dispatch problem • The dispatch problem asks how to schedule mobile sensors to move to the designated locations according to the result computed by the placement strategy

  4. Introduction (2/9) • Their solutions to the placement problem consider both the binary and probabilistic sensing models, and allow an arbitrary relationship between the communication distance and sensing distance of sensors • For the dispatch problem, authors propose a competition-based and a pattern based schemes • Their proposed schemes are efficient in terms of the number of sensors required and are distributed in nature. Simulation results are presented to verify their effectiveness

  5. Introduction (3/9) • The region is k-covered if every point in that region can be monitored by at least k sensors, where k is a given parameter. • A large amount of applications may impose the requirement of k > 1. • k ≥ 2 to avoid leaving uncovered holes when some sensors are broken. • Positioning protocols using triangulation require at least three sensors (i.e., k ≥ 3) to detect each location where an object may appear.

  6. Introduction (4/9) • In this paper, authors address the sensor deployment problem with the following requirements: • Multiple-level coverage of the area of interest is required • Connectivity between sensors (in terms of their communications) should be maintained • The area of interest may change over time • Sensor nodes are autonomous and mobile and thus can bedispatched to desired locations • The region of interest can be “self-deployed” by these sensors

  7. Introduction (5/9) • In Sun et al.[9], a hexagon-like placement is proposed to guarantee the sensing field to be k-covered, under the assumption that the communication distance of sensors rc is no smaller than twice of their sensing distance rs. • The work in Chakrabarty et al.[18] models the sensing field by grids and considers two kinds of sensors with different costs and sensing capabilities to be deployed in the sensing field. The objective is to make every grid point k-covered and the total cost is minimum. • Both [9] and [18] do not address the relationship between rc and rs. • In Wang, Xu et al.[25-26], the sensing field is partitioned into grids, and sensors are moved from high-density grids to low density ones to achieve more uniform coverage. • These studies do motivate them to investigate the sensor dispatch problem.

  8. Introduction (6/9) • Authors consider two types of sensing models: • Binary sensing model • a location can be either monitored or not monitored by a sensor, depending on whether the location is within the sensor’s rsrange. • Probabilistic sensing model • a location will be monitored by a sensor according to some probability function

  9. Introduction (7/9) • Authors first consider the binary sensing model of sensors and propose two solutions to the placement problem • The first one is based on an intuitive duplication idea • The second one is based on a more complicated interpolating scheme and thus can save more sensors • Then, authors adapt these solutions to the probabilistic sensing model by properly adjusting sensing distances of sensors

  10. Introduction (8/9) • For the sensor dispatch problem, authors propose two distributed schemes to let sensors move to the designated locations (computed by the placement result) on their own.

  11. Introduction (9/9) • Major contributions • First, their schemes allow changeof the monitoring region and coverage level of the WSN in an autonomous and distributed manner. • Second, their deployment solutionsare helpful in conditions where the precise initial deployment (e.g., by humans) is almost impossible because the region of interest is very dangerous or even inaccessible to people.

  12. Outline • Introduction • Problem Statement • K-coverage Sensor Placement Schemes • Distributed Sensor Dispatch Schemes • Experimental Results • Conclusions • Comments

  13. II. Problem Statement (1/4) • Authors are given a sensing field A, an area of interest Iinside A, and a set of mobile sensors Sresident in A • Each sensor has a communication distance rcand a sensing distance rs. Sensors are homogenous, but the relationship of rc and rs can be arbitrary. • For connectivity, authors assume that two sensors can communicate with each other if their distance is no larger than rc.

  14. Problem Statement (2/4) • Authors consider both the binary and probabilistic sensing models of sensors. • binary sensing model • a point can be monitored by a sensor if it is within the sensor’s sensing region. In this way, a point in A is defined as k-covered if it is within k sensors’ sensing regions, where k is a given parameter.

  15. Problem Statement (3/4) • probabilistic sensing model • the detection probability of a sensor will decay with the distance from the sensor to the monitored object. In particular, the detection probability of a point u by a sensor sican be evaluated as [14], [31]: • where ε is a parameter indicating the physical characteristics of the sensor and d(u, si) is the distance between u and si. In this way, a point in A is considered as k-covered if the probability that there are at least k sensors which can detect this point is no smaller than a predefined threshold pth, where 0 < pth< 1. • The model conveys the intuition that the closer a location is to the node, a higher signal-to-noise ratio is expected, resulting in a higher confidence level of that location being detected.

  16. Problem Statement (4/4) • Given an integer k, the k-coverage sensordeployment problemcan be divided into two subproblems: • K-coverage sensor placement problem • minimum number of sensors • Distributed sensor dispatch problem • the total energy consumption of sensors due to movement is minimized, i.e., • where emove is the energy cost for a sensor to move in one unit-distance and di is the total distance that sensor i has moved.

  17. Outline • Introduction • Problem Statement • K-coverage Sensor Placement Schemes • Distributed Sensor Dispatch Schemes • Experimental Results • Conclusions • Comments

  18. III. k-Coverage Sensor Placement Schemes (1/11) A. The Simple Duplicate Placement Scheme • For the 1-coverage placement, authors adopt the method proposed in Wang et al.[17], which has been proved to be able to use the minimum number of sensors to achieve 1-coverage and connectivity (Bai et al.[32]).

  19. k-Coverage Sensor Placement Schemes (2/11) • According to the relationship of rcand rs, authors separate the discussion into two cases, as shown in Fig. 1. (Bai et al.[32])

  20. k-Coverage Sensor Placement Schemes (3/11) • After determining the 1-coverage placement, authors can duplicate k sensors on each location to ensure k-coverage of the whole area. • Note that in the case of rc < √3rs, since the distance between sensors on adjacent rows is larger than rc, it is necessary to add some extra columns of sensors, where sensors on each column are separated by a distance no larger than rc, to connect adjacent rows.

  21. k-Coverage Sensor Placement Schemes (4/11) B. The Interpolating Placement Scheme • The following interpolating placement scheme will try to balance the coverage levels of sub-regions. • According to the relationship of rcand rs, authors separate the discussion into three cases.

  22. k-Coverage Sensor Placement Schemes (5/11) • Case of rc ≤√3rs/2 • Fig. 2 shows the coverage level of the sensing field will directly become three. • To summarize, the previous duplicate scheme uses 3x rows of sensors to ensure 3-coverage of a belt-like area of width (x−1)rs+(x+1)·√(r2 s − r2 c /4) , while this interpolating scheme uses only 2x+ 1 rows of sensors to ensure 3-coverage of the same region. • In general, fork >(>=)3, this interpolating placement requires only (∟k/3」 (2x + 1)+ (k mod 3) · x) rows of sensors. + +

  23. k-Coverage Sensor Placement Schemes (6/11) • Case of √3rs/2 < rc ≤ (2+√3rs)/3 • In general, fork > 3, this interpolating placement requires only (∟k/3」 (2.5x + 1)+ (k mod 3) · x) rows of sensors.

  24. k-Coverage Sensor Placement Schemes (7/11) • Case of rc > (2+√3rs)/3 • In this case, this interpolating placement will not save sensors compared with the duplicate placement, so authors adopt the duplicate scheme in this case.

  25. Adapting to the Probabilistic Sensing Model Adaptation of the Duplicate Scheme: With the aforementioned argument, if authors replace rs by rpswhen executing the duplicate placement scheme, authors can make sure that the duplicate scheme can still guarantee that I is k-covered under the probabilistic sensing model. k-Coverage Sensor Placement Schemes (8/11) pmin: minimum k-covered probability pth: this point is no smaller than a predefined threshold

  26. k-Coverage Sensor Placement Schemes (9/11) • Adaptation of the Interpolating Scheme: Case of < k ≥ 3 is a multiple of three k is not a multiple of three, we will add extra (k mod 3) sensors on each location in old rows in Fig. 2.

  27. k-Coverage Sensor Placement Schemes (10/11) • Adaptation of the Interpolating Scheme: Case of

  28. k-Coverage Sensor Placement Schemes (11/11) • Adaptation of the Interpolating Scheme: Case of

  29. Outline • Introduction • Problem Statement • K-coverage Sensor Placement Schemes • Distributed Sensor Dispatch Schemes • Experimental Results • Conclusions • Comments

  30. IV. Distributed Sensor Dispatch Schemes • After determining the locations to be placed with sensors, the next issue is how to move existing sensors in the sensing field A to the designated locations in Isuch that the energyconsumption of sensors due to movement is minimized.

  31. The Competition-based Dispatch Scheme (1/9) • In this scheme, authors assume that there is a sink that can calculate all locations in I to be placed with sensors and then broadcast these designated locations to all sensors.

  32. The Competition-based Dispatch Scheme (2/9) • Once receiving the notification from the sink, sensors will compete with each other to move toward these locations. This competition-based scheme involves the following rules. • Rule 1: The sink first calculates a set of locations L = {(x1, y1, n1), (x2, y2, n2), · · · , (xm, ym, nm)}to be placed with sensors, according to their placement schemes in Section III. The sink then broadcasts L to all sensors.

  33. The Competition-based Dispatch Scheme (3/9) • Rule 2: On receiving L for the first time, each sensor siconstructs a table OCC[1..m] such that every entry OCC[j] = {(sj1, dj1), (sj2, dj2), . . . , (sjα, djα)}, α ≤ nj.

  34. The Competition-based Dispatch Scheme (4/9) • Rule 3: When the state of a sensor si is free, it will check its OCC[1..m] table to select a location in L as its destination. The selection is as follows: • The first priority is to consider uncovered locations. • If all locations in L are already covered (i.e., |OCC[j]| = nj, ∀j = 1..m), then siwill select a location (xj, yj) such that there exists a record (sk, dk) ∈ OCC[j] and d(si, (xj, yj)) < dk. • When sensor sibecomes engaged, it will begin moving toward its destination. Otherwise, si will enter the terminated state because it does not need to cover any location.

  35. The Competition-based Dispatch Scheme (5/9) • Rule 4: For the maintenance purpose, each sensor siwill periodically perform the following two actions: • Update the contents of its OCC[1..m] table. Specifically, for each (sjβ, djβ) ∈ OCC[j], j = 1..m, decrease djβby the expected moving distance of sjβduring the past period of time, unless djβ= 0. • Broadcast si’s current status to its one-hop neighbors, including its (1)ID, its (2)OCC[1..m] table, and (3)its current position and state. • The aforementioned actions can be controlled by setting two timers Tupdate_OCCand Tbroadcast.

  36. The Competition-based Dispatch Scheme (6/9) • Rule 5: When a sensor si receives an update message from another sensor sk, two actions will be taken: • First, si has to update its OCC[1..m] table as follows. Let us denote by OCCi[1..m] and OCCk[1..m] the tables of siand sk, respectively. For each j = 1..m, authors calculate the union: • Uj = OCCi[j] ∪ OCCk[j]. • If |Uj| ≤ nj , we will replace OCCi[j] by Uj. Otherwise, it means that there are too many sensors scheduled to cover (xj, yj), in which case we will truncate those records in Uj that have longer moving distances, until the size |Uj | = nj.(sensors’ state from engaged to free and then execute the rule 3) • After the above merge, if siremains engaged, say, with (xj, yj) as its destination, then authors will do the following optimization. authors will check if • d(si, (xl, yl)) + d(sk, (xj, yj)) < d(si, (xj, yj)) + d(sk, (xl, yl)), • where (xl, yl) is the destination of sk. If so, it means that the total moving distance of si and sk can be reduced if we exchange their destinations. • In the above steps, if any entry in OCCi[1..m] table has been changed, si will broadcast the modified content to its direct neighbors.

  37. The Competition-based Dispatch Scheme (7/9) • Rule 6: When a sensor si is in the engaged state, it will keep moving toward its destination (xj, yj). However, in case that the record (si, d(si, (xj, yj))) is removed from its current OCCi[j] entry as specified in the rule 5, it will stop moving and change its state to free (so that the rule 3 will be executed again).

  38. The Competition-based Dispatch Scheme (8/9) • Rule 7: When an engaged sensor si arrives at its destination (xj, yj), it will change its state to terminated and begin its monitoring job at the designated location. Meanwhile, it will still execute the maintenance actions in the rule 4, until the sink commands it to stop. Since the sink will eventually observe that I is k-covered (by receiving the sensing reports from sensors), it can notify all sensors to exit from this dispatch algorithm.

  39. The Competition-based Dispatch Scheme (9/9) • Theorem 1: Given an area I ⊆ A, the competition-based dispatch scheme guarantees that I will be eventually k-covered if there are sufficient mobile sensors inside A.

  40. The Pattern-based Dispatch Scheme (1/6) • The aforementioned competition-based scheme assumes that every sensor has the full knowledge of all target locations inside I. • Authors propose a pattern-based dispatch scheme which allows sensors to derive the target locations on their own, thus relaxing the above limitation.

  41. The Pattern-based Dispatch Scheme (2/6) • In the duplicate placement scheme, sensors will be placed in a hexagon-like fashion. Thus, each sensor at the location (x, y) can derive its potential six neighbors’ positions according to Table II.

  42. The Pattern-based Dispatch Scheme (3/6) • When the interpolating placement scheme is adopted, the pattern will be changed according to the relationship of rcand rs. In particular, there are three cases to be discussed:

  43. The Pattern-based Dispatch Scheme (4/6)

  44. The Pattern-based Dispatch Scheme (5/6) • Each sensor initially maintains a set of seed locations L’ = {(x1, y1, n1, δ1), (x2, y2, n2, δ2), · · · , (xα, yα, nα, δα)}, which is a partial list of locations to be placed with sensors in I, where δj is the pattern used by the sensor at location (xj, yj). Clearly, L’ can be considered as a subset of L. • Each sensor then executes the competition-based scheme to contend for their closest locations in L’. However, the original rules 3 and 7 in the competition-based scheme should be modified as follows: • New rule 3: A free sensor si will try to select a location in L’ as its destination among these newly-derived locations. if si cannot calculate any new location from its current L’ (which means that L’ = L), then si will enter the terminated state because it does not need to cover any location. • New rule 7: When an engaged sensor si arrives at its destination, it will derive some new locations from its current L’and add the corresponding new entries in its OCC[·] table.

  45. The Pattern-based Dispatch Scheme (6/6) • Corollary 1: Given an area I ⊆A, the pattern-based dispatch scheme guarantees that I can be k-covered if there are sufficient mobile sensors inside A.

  46. Outline • Introduction • Problem Statement • K-coverage Sensor Placement Schemes • Distributed Sensor Dispatch Schemes • Experimental Results • Conclusions • Comments

  47. V. EXPERIMENTAL RESULTS • The evaluation includes three parts. • First, authors measure the numbers of sensors required by different placement schemes discussed in Section III. • Second, authors verify the effectiveness of their sensor dispatch schemes proposed in Section IV. • Finally, authors study the effect of seed locations on the pattern-based dispatch scheme.

  48. A. Evaluations of the Sensor Placement Schemes Authors set the sensing distance rsto 15 m, 11.55 m, 10 m, and 8.04 m, respectively The interpolating scheme can save approximately 19.5% ∼ 32.8% and 9.6% ∼ 16.1% sensors

  49. B. Performances of the Sensor Dispatch Schemes (1/2)

  50. Performances of the Sensor Dispatch Schemes (2/2)

More Related