On accurate and efficient statistical counting in sensor-based surveillance systems

S. Guo, T. He, Mohamed Mokbel, J. A. Stankovic, T. F. Abdelzaher

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Sensor networks have been used in many surveillance systems, providing statistical information about monitored areas. Accurate counting information (e.g., the distribution of the total number of targets) is often important for decision making. As a complementary solution to double-counting in communication, this paper presents the first work that deals with double-counting in sensing for wireless sensor networks. The probability mass function (pmf) of target counts is derived first. This, however, is shown to be computationally prohibitive when a network becomes large. A partitioning algorithm is then designed to significantly reduce computation complexity with a certain loss in counting accuracy. Finally, two methods are proposed to compensate for the loss. To evaluate the design, we compare the derived probability mass function with ground truth obtained through exhaustive enumeration in small-scale networks. In large-scale networks, where pmf ground truth is not available, we compare the expected count with true target counts. We demonstrate that accurate counting within 1%-3% relative error can be achieved with orders of magnitude reduction in computation, compared with an exhaustive enumeration-based approach.

Original languageEnglish
Pages (from-to)74-92
Number of pages19
JournalPervasive and Mobile Computing
Volume6
Issue number1
DOIs
Publication statusPublished - 1 Feb 2010
Externally publishedYes

Fingerprint

Surveillance
Counting
Sensor
Sensors
Count
Enumeration
Complex networks
Target
Sensor networks
Wireless sensor networks
Decision making
Relative Error
Communication
Sensor Networks
Wireless Sensor Networks
Partitioning
Sensing
Decision Making
Evaluate
Demonstrate

Keywords

  • Duplicate counting
  • Graph partitioning
  • Target detection
  • Wireless sensor networks

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Applied Mathematics

Cite this

On accurate and efficient statistical counting in sensor-based surveillance systems. / Guo, S.; He, T.; Mokbel, Mohamed; Stankovic, J. A.; Abdelzaher, T. F.

In: Pervasive and Mobile Computing, Vol. 6, No. 1, 01.02.2010, p. 74-92.

Research output: Contribution to journalArticle

Guo, S. ; He, T. ; Mokbel, Mohamed ; Stankovic, J. A. ; Abdelzaher, T. F. / On accurate and efficient statistical counting in sensor-based surveillance systems. In: Pervasive and Mobile Computing. 2010 ; Vol. 6, No. 1. pp. 74-92.
@article{a9848f86b234489eba0b86bbe7fba3e4,
title = "On accurate and efficient statistical counting in sensor-based surveillance systems",
abstract = "Sensor networks have been used in many surveillance systems, providing statistical information about monitored areas. Accurate counting information (e.g., the distribution of the total number of targets) is often important for decision making. As a complementary solution to double-counting in communication, this paper presents the first work that deals with double-counting in sensing for wireless sensor networks. The probability mass function (pmf) of target counts is derived first. This, however, is shown to be computationally prohibitive when a network becomes large. A partitioning algorithm is then designed to significantly reduce computation complexity with a certain loss in counting accuracy. Finally, two methods are proposed to compensate for the loss. To evaluate the design, we compare the derived probability mass function with ground truth obtained through exhaustive enumeration in small-scale networks. In large-scale networks, where pmf ground truth is not available, we compare the expected count with true target counts. We demonstrate that accurate counting within 1{\%}-3{\%} relative error can be achieved with orders of magnitude reduction in computation, compared with an exhaustive enumeration-based approach.",
keywords = "Duplicate counting, Graph partitioning, Target detection, Wireless sensor networks",
author = "S. Guo and T. He and Mohamed Mokbel and Stankovic, {J. A.} and Abdelzaher, {T. F.}",
year = "2010",
month = "2",
day = "1",
doi = "10.1016/j.pmcj.2009.07.013",
language = "English",
volume = "6",
pages = "74--92",
journal = "Pervasive and Mobile Computing",
issn = "1574-1192",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - On accurate and efficient statistical counting in sensor-based surveillance systems

AU - Guo, S.

AU - He, T.

AU - Mokbel, Mohamed

AU - Stankovic, J. A.

AU - Abdelzaher, T. F.

PY - 2010/2/1

Y1 - 2010/2/1

N2 - Sensor networks have been used in many surveillance systems, providing statistical information about monitored areas. Accurate counting information (e.g., the distribution of the total number of targets) is often important for decision making. As a complementary solution to double-counting in communication, this paper presents the first work that deals with double-counting in sensing for wireless sensor networks. The probability mass function (pmf) of target counts is derived first. This, however, is shown to be computationally prohibitive when a network becomes large. A partitioning algorithm is then designed to significantly reduce computation complexity with a certain loss in counting accuracy. Finally, two methods are proposed to compensate for the loss. To evaluate the design, we compare the derived probability mass function with ground truth obtained through exhaustive enumeration in small-scale networks. In large-scale networks, where pmf ground truth is not available, we compare the expected count with true target counts. We demonstrate that accurate counting within 1%-3% relative error can be achieved with orders of magnitude reduction in computation, compared with an exhaustive enumeration-based approach.

AB - Sensor networks have been used in many surveillance systems, providing statistical information about monitored areas. Accurate counting information (e.g., the distribution of the total number of targets) is often important for decision making. As a complementary solution to double-counting in communication, this paper presents the first work that deals with double-counting in sensing for wireless sensor networks. The probability mass function (pmf) of target counts is derived first. This, however, is shown to be computationally prohibitive when a network becomes large. A partitioning algorithm is then designed to significantly reduce computation complexity with a certain loss in counting accuracy. Finally, two methods are proposed to compensate for the loss. To evaluate the design, we compare the derived probability mass function with ground truth obtained through exhaustive enumeration in small-scale networks. In large-scale networks, where pmf ground truth is not available, we compare the expected count with true target counts. We demonstrate that accurate counting within 1%-3% relative error can be achieved with orders of magnitude reduction in computation, compared with an exhaustive enumeration-based approach.

KW - Duplicate counting

KW - Graph partitioning

KW - Target detection

KW - Wireless sensor networks

UR - http://www.scopus.com/inward/record.url?scp=74849104581&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74849104581&partnerID=8YFLogxK

U2 - 10.1016/j.pmcj.2009.07.013

DO - 10.1016/j.pmcj.2009.07.013

M3 - Article

AN - SCOPUS:74849104581

VL - 6

SP - 74

EP - 92

JO - Pervasive and Mobile Computing

JF - Pervasive and Mobile Computing

SN - 1574-1192

IS - 1

ER -