Localization

1
Localization
2
Learning Objectives

• Understand why WSNs need localization protocols
• Understand localization protocols in WSNs
• Understand secure localization protocols

3
Prerequisites

• Basic mathematics knowledge
• Basic concepts in network protocols

4
The Problem

• The determination of the geographical locations
  of sensor nodes
• Why do we need Localization?
• Manual configurations of locations is not
  feasible for large-scale WSNs
• Location information is necessary for some
  applications and services, e.g. geographical
  routing
• Providing each sensor with localization hardware
  (e.g., GPS) is expensive in terms of cost and
  energy consumption

5
Localization

• In some applications, it is essential for each
  node to know its location
• Global Positioning System (GPS) is not always
  possible
• GPS cannot work indoors
• GPS power consumption is very high

6
Solutions

• Range-based
• Use exact measurements (point-to-point distance
  estimate (range) or angle estimates)
• More expensive
• Ranging the process of estimating the distance
  between the pair of nodes
• Range-free
• Only need the existences of beacon signals
• Cost-effective alternative to range-based
  solutions

7
Localization Algorithms in WSNs

• Beacon Nodes know their locations
• Range-based Algorithms
• Sensor nodes need to measure physical
  distance-related properties
• How to measure distance
• RSSI (Received Signal Strength Indication)
• ToA (Time of Arrival)
• TDOA (Time Difference of Arrival)
• How to estimate location
• MMSE (Minimum Mean Square Estimation)
• Range Free Algorithms
• Do Not involve distance estimation

8
Localization Algorithms in WSNs
9
Range-based Solutions - MMSE

• MMSE
• Minimum Mean Square Estimation

10
Range-based Solutions - MMSE

• Ideally, ei should be 0

11
Range-based Solutions - MMSE

• Rearrange the previous equations, we have

• We have N equations

12
Range-based Solutions - MMSE

• Eliminate , we get the following
  N-1 equations

• Hx z

13
Range-based Solutions - MMSE

• H

14
Range-based Solutions - MMSE

• z

15
Range-based Solutions - MMSE

• x

• Solution

16
Range-free Approach - Centroid

• RefLoc_1, Section 2.1

17
Security Concerns in WSNs

• Secure Localization Problem

• Secure Localization Solutions

18
Secure Localization

• Attack-resistant Minimum Mean Square Estimation

• RefLoc_2

19
Attack-resistant Minimum Mean Square Estimation
20
Minimum Mean Square Estimation

• The more inconsistent a set of location
  references is, the greater the corresponding mean
  square error should be

• RefLoc_2, Section 2

21
Impact of Malicious Beacons
22
Impact of Malicious Beacons
23
Minimum Mean Square Estimation

• t is important Depend on many factors

24
How to Decide the set of Consistent Location
References?

• Given a set L of n location references and a
  threshold t
• Optimal solution
• Greedy solution

25
How to decide t?

• Measurement error model
• How to obtain?
• Study the distribution of the mean square error
  when there are no malicious attacks

26
Voting-based Location Estimation Basic Ideas
27
Iterative Refinement

• The larger the number of cells
• More state variables need to be kept
• The smaller each cell will be precision
• Iterative Refinement
• Initially, the number of cells is chosen based on
  memory constraints
• After the first round, the node may perform the
  voting process on the smallest rectangle that
  contains all the cells having the largest vote