inter net things
TRANSCRIPT
![Page 1: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/1.jpg)
Fuzzy-based Sensor Search in the Web of Things
by
Cuong Truong
University of Lübeck, Germany
1
![Page 2: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/2.jpg)
The Vision of the Internet of Things
real world objects will be uniquely identifiable and connected to the Internet
2
![Page 3: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/3.jpg)
The Vision of the Web of Things
mashing up sensors and actuators with services and data available on the Web
3
![Page 4: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/4.jpg)
Sensor Search in WoT: Start-of-the-art
4
Internet
sensor capteur 傳感器
publish
a textual description
GSN
![Page 5: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/5.jpg)
5
Sensor Search in WoT: Start-of-the-art
complex for end user!
![Page 6: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/6.jpg)
• What to type in search engine?
• How to describe search criteria?
Not easy! Hmm!
Places that have similar climate and oceanic condition to Key West in the
last year?
Pick a climate sensor in Key West,
and search for similar sensors
Sensor Similarity Search: An Illustration
Key West Marathon
Fishery owner 6
![Page 7: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/7.jpg)
Sensor Similarity Search: Architecture
local database
Internet
crawls
crawls crawls
search for:
7
![Page 8: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/8.jpg)
Questions to be addressed
8
I. How to define and compute similarity between two sensors?
II. How to construct a fuzzy set from historical sensor readings?
III. How to minimize the cost of storing such fuzzy sets?
IV. How to efficiently compute a similarity score between a pair of sensors?
V. How to objectively evaluate the approach?
![Page 9: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/9.jpg)
I. Similarity Definition
9
10
20
11
21
12
125
similar
different
(2) similar reading ranges
(1) similar reading curves
what about me?
![Page 10: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/10.jpg)
Degree of membership of elements of fuzzy set
FK(38) = 0.9
FL(38) = 0.6
Key idea: Same value, different degree of memberships in different fuzzy sets
x
28
48 kitchen
time x
0
1 F K
28 48 35 44
F L 35 44
library
time
x 0.9
0.6
38
10
I. Similar Reading Curves: Captured by Fuzzy Set
![Page 11: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/11.jpg)
The reading 38 is likely read by sensor in kitchen:
FK(38) = 0.9 > 0.6 = FL (38)
Given a sensor S with set of readings X = {x}, S is likely located in kitchen if:
x
28
48 kitchen
time x
0
1 F K
28 48 35 44
F L 35 44
library
time
x 0.9
0.6
38
11
I. Similar Reading Curves: Captured by Fuzzy Set
![Page 12: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/12.jpg)
I. Similar Reading Ranges
12
captured by the reading range difference
![Page 13: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/13.jpg)
Given a sensor V, and a sensor S whose set of readings is X = {x}
Combining the two above mentioned similarity conditions:
Similar reading curves (defined by fuzzy set)
Similar reading ranges (defined by reading range difference)
13
I. Similarity Computation
![Page 14: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/14.jpg)
Questions to be addressed
14
I. How to define and compute similarity between two sensors?
II. How to construct a fuzzy set from historical sensor readings?
III. How to minimize the cost of storing such fuzzy sets?
IV. How to efficiently compute a similarity score between a pair of sensors?
V. How to objectively evaluate the approach?
![Page 15: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/15.jpg)
II. Fuzzy Set Construction
24
00:00 23:59
time
x
15
30 S
Temperature sensor S has been monitoring a room for 24 hours from 00:00 -> 23:59
FS(x)
x
15 30
1
0 24
15
+ + +
+ + +
![Page 16: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/16.jpg)
Questions to be addressed
16
I. How to define and compute similarity between two sensors?
II. How to construct a fuzzy set from historical sensor readings?
III. How to minimize the cost of storing such fuzzy sets?
IV. How to efficiently compute a similarity score between a pair of sensors?
V. How to objectively evaluate the approach?
![Page 17: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/17.jpg)
III. Efficient Fuzzy Set Storage: Approximation
Fuzzy set‘s storage overhead
Membership function is smooth
Approximation using set of line segments
17
+
+
![Page 18: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/18.jpg)
Questions to be addressed
18
I. How to define and compute similarity between two sensors?
II. How to construct a fuzzy set from historical sensor readings?
III. How to minimize the cost of storing such fuzzy sets?
IV. How to efficiently compute a similarity score between a pair of sensors?
V. How to objectively evaluate the approach?
![Page 19: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/19.jpg)
III. Efficient Similarity Score Computation
19
x
f(x)
![Page 20: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/20.jpg)
Questions to be addressed
20
I. How to define and compute similarity between two sensors?
II. How to construct a fuzzy set from historical sensor readings?
III. How to minimize the cost of storing such fuzzy sets?
IV. How to efficiently compute a similarity score between a pair of sensors?
V. How to objectively evaluate sensor similarity search?
![Page 21: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/21.jpg)
V. Evaluation: Approach
For a search, a list of sensors is returned
Ranked by decreasing similarity score
Similar sensors are ranked on top
Issue: „Similarity“ is highly subjective! no ground truth
Fact: Sensors close to each other have similar readings
Approach: Group sensors based on location and annotated group with its location
21
![Page 22: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/22.jpg)
kitchen
bedroom perform search
List ranked by similarity score
22
V. Evaluation: Approach
![Page 23: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/23.jpg)
V. Ranked List: Degree of Accuracy
23
![Page 24: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/24.jpg)
V. Evaluation: Multiple Real Data Sets
For each data set, group sensors based on location, and define a search trial as
Picking a sensor and perform search
Compute DOA value of the obtained ranked list
For each sensor
Last 24 hours of readings are used
Evaluation is done on a PC
Java VM
Intel Core i5 CPU at 2.4 Ghz clock rate
24
![Page 25: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/25.jpg)
IntelLab Data Set
http://db.csail.mit.edu/labd
ata/labdata.html
12 sensors in 3 groups
1500 data points/24 hours
Performance: 222 μs / pair
4505 sensors / second
(brute force)
25
![Page 26: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/26.jpg)
NOAA Data Set http://tidesandcurrents.no
aa.gov/gmap3
23 sensors in 5 groups
200 data points/24 hours
Performance: 28 μs / pair 35741 sensors / second (brute force)
26
![Page 27: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/27.jpg)
MavPad Data Set http://ailab.wsu.edu/m
avhome/index.html
8 sensors in 2 groups
500 data points / 24 hours
Performance: 70 μs / pair 14285 sensors / second (brute force)
27
![Page 28: inter net things](https://reader031.vdocuments.net/reader031/viewer/2022030206/58abec6c1a28ab504e8b58e1/html5/thumbnails/28.jpg)
Summary
Sensor similarity search and distributed architecture to realize it
Fuzzy-based approach to efficiently compute similarity score
Evaluation metric for ranked list
Accurate results of evaluation
Outlook: Scalability
Paralellize search
More efficient similarity computation
Index and lookup of fuzzy sets at server side
Incremental search accuracy 28