In recent years, indoor localization has become a hot research topic with some sophisticated solutions reaching accuracy on the order of ten centimeters. While certain classes of applications can justify the corresponding costs that come with these solutions, a wealth of applications have requirements that can be met at much lower cost by accepting lower accuracy. This paper explores one specific application for monitoring patients in a nursing home, showing that sufficient accuracy can be achieved with a carefully designed deployment of low-cost wireless sensor network nodes in combination with a simple RSSI-based localization technique. Notably our solution uses a single radio sample per period, a number that is much lower than similar approaches. This greatly eases the power burden of the nodes, resulting in a significant lifetime increase. This paper evaluates a concrete deployment from summer 2012 composed of fixed anchor motes throughout one floor of a nursing home and mobile units carried by patients. We show how two localization algorithms perform and demonstrate a clear improvement by following a set of simple guidelines to tune the anchor node placement. We show both quantitatively and qualitatively that the results meet the functional and non-functional system requirements.

1. Embracing Localization Inaccuracy:
A Case Study
Usman Raza
Amy L. Murphy
Gian Pietro Picco
1

2. Motivation
 Localization in wireless sensor networks has been studied for
a decade now
 Increasing complex localization techniques to achieve tens of
cm accuracy
 Use costly specialized hardware (UWB, Antenna arrays)
 Experiences from the real environments are still limited!
2

3. Goals of This Work
 Not to propose yet another localization technique
 Evaluate localization techniques in a real-world nursing home
 Requirement are representative of diverse localization
systems
 Unveil the relationship between system level performance and
application level objectives
 What a WSN geek want? vs. What an end user want?
 To give guidelines to improve both the system level performance
and end user satisfaction
3

4. Higher quality of life for impaired and elderly in nursing homes
Monitoring to the medical support staff
A single floor of a nursing home in Trento, Italy
4 • 10 Public spaces, 20 Patients, 4 Nurses

5. Monitoring in Nursing Homes - Services
 Doctors
 Offline evaluation of patient movement
 To assess the general health of the patient
 To diagnose the progression of Alzheimer's disease
 Nurses
 Real time use of approximate patient location
 To find the patient
 To raise an alarm if patient leaves the facility
5

6. WSN Localization - Architecture
Mobile
Anchors(x,y)
Proximity Detection Localization Technique
6
Localization

7. WSN Localization - Architecture
Mobile
Anchors(x,y)
Proximity Detection Localization Technique
7
Localization

8. WSN Localization - Architecture
Mobile
Anchors(x,y)
(x,y) (x,y)
(x,y)
Proximity Detection Localization Technique
8
Localization

9. WSN Localization - Architecture
Mobile
Anchors(x,y)
(x,y) (x,y)
(x,y)
Proximity Detection Localization Technique Mobile(x,y)
9
Localization Visualization

10. WSN Localization - Architecture
Mobile
Anchors(x,y)
WSN geek – How far is the Nurse – Does the visualization allows
estimate to the actual position? me to find the patient?
(x,y) (x,y)
(x,y)
Proximity Detection Localization Technique Mobile(x,y)
10
Localization Visualization

11. Localization Localization Technique
Step 1– Proximity Detection
 System Design
 Objective  A custom proximity detection
 Identify the proximity of the mobile protocol
patient to an anchor
Anchor Sleep Box-MAC-LPL
 Requirement – Low maintenance
 Energy Efficiency
Mobile Sleep Sleep
Visualization Refresh Interval
Lifetime Anchors - 45 days (2 AA batteries)
11 Mobile - 5 days (1 coin battery)

12. Localization Proximity Detection
Step 2– Localization Techniques
 Max-RSSI Localization
 Localizes patient at the anchor
detecting proximity with
maximum signal strength
Patient
12

13. Localization Proximity Detection
Step 2– Localization Techniques
 Max-RSSI Localization
 Localizes patient at the anchor
detecting proximity with
maximum signal strength
 Relative Span Exponential
Weighted Localization (REWL)
 Localizes patient at weighted
centroid of all anchor coordinates Patient
13

14. Experimental Evaluation
 System level: WSN geek’s perspective
 Accuracy: Percentage of time the patient is correctly detected
in its current area/room
 Application level: Nurse's Perspective
 Satisfaction level : moderate, fair, excellent!
14

15. System Level Evaluation - Accuracy
Bedroom 94%
87% 77% 89% 92% 72%
Max-RSSI
89%
82% Corridor 85%
Bathroom
- Reasonable accuracy
Living Area Living Area - No clear winner
- REWL outperforms in
Bedroom 95%
key areas e.g. living and
exit areas
REWL
78% 72% 97% 91% 60%
91%
95% Corridor 88%
Bathroom
15
Living Area Living Area

16. Application Level Evaluation -
Methodology
 Operator was asked to evaluate multiple localization and
visualizations
 Find the patient
 Raise an alarm
Patient Icon
Patient Tracking GUI
Nurse
16

17. Application Level Evaluation - Results
Max RSSI - at anchor Max RSSI- at room center REWL - at x,y
Moderate Fair Excellent!
Quality of localization system depends heavily on the
17 visualization

18. Application Level Evaluation - Results
 Experience with our Initial deployment:
 Low accuracy
 Positive qualitative evaluation
Bedroom
65% 52% 84% 79% 57%
REWL 34%
72% Corridor 91%
Bathroom
Living Area Living Area
Low Accuracy ≠ Unacceptable solution to end user
18

19. Guidelines for Anchor Placement
 Principle: Minimize the likelihood of signal reception across the
monitored areas
Maximize the distance
between anchor nodes
deployed in the adjacent Use the radio shieling of
areas obstacle to your advantage!
Place the nodes near the
center of the monitored area
19

20. Conclusion
 Simple low cost localization system is enough
for many applications
 Low accuracy can still be acceptable to end
user
20

21. Thank You!
Usman Raza, Amy Murphy, Gian Pietro Picco, “Embracing Localization Inaccuracy – A Case
Study ”, April 2013, IEEE International Conference on Intelligent Sensors, Sensor Networks
and Information Processing (ISSNIP), Melbourne, Australia
Full Text (Pre-print) : http://disi.unitn.it/~raza/Papers/DISI_TR_12_038.pdf
21