surfi: detection surveillance camera looping attacks with
TRANSCRIPT
![Page 1: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/1.jpg)
SurFi: Detecting Surveillance Camera Looping Attacks with
Wi-Fi Channel State Information
Nitya Lakshmanan*, Inkyu Bang+, Min Suk Kang*, Jun Han*, Jong Taek Lee#
*National University of Singapore, +Agency for Defense Development, #Electronics and Telecommunications Research Institute
+ research done while working in NUS
![Page 2: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/2.jpg)
2
Surveillance cameras are now everywhere
![Page 3: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/3.jpg)
Surveillance camera looping attack
3
security guard
Surveillance systemVideo shows
a normal activity!
Place of interest
valuable
security guard
Surveillance systemVideo shows
a normal activity!
video feed
![Page 4: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/4.jpg)
security guard
Surveillance systemVideo shows
a normal activity!
Place of interest
valuable
Surveillance camera looping attack
4
security guard
Surveillance systemVideo shows
a normal activity!
video feed
No activity
![Page 5: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/5.jpg)
security guard
Surveillance systemVideo shows
a normal activity!
Place of interest
valuable
Surveillance camera looping attack
5
video feed
security guard
Surveillance systemVideo shows
a normal activity!
looped!
No activityLooped
![Page 6: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/6.jpg)
security guard
Surveillance systemVideo shows
a normal activity!
Place of interest
valuable
Surveillance camera looping attack
6
security guard
Surveillance systemVideo shows
a normal activity!
looped!
video feed
No activityLooped
Robbery
Reality Seen by the guard
![Page 7: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/7.jpg)
Exploiting Surveillance Cameras Like a Hollywood Hacker
BlackHat 2013
Looping Surveillance Cameraslike in the movies
DefCon 2015
7
Surveillance camera looping is a reality now
Live video Replayed image
Modified timestamp
![Page 8: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/8.jpg)
Incur prohibitive cost
Live (3 pm) This morning (10 am)
Surveillance camera with integrity protection
Video frame comparison
8
Mitigation of camera looping attack is hard
Not robust against an adversary who can manipulate the video
Can we mitigate the camera looping attack effectively at no extra hardware cost?
![Page 9: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/9.jpg)
security guard
Place of interest
valuable
Surveillance system
Video shows a normal activity!
9
video feed
looped!
SurFi
Compare
channel state information (CSI)
Low false alarms
SurFi (Surveillance with Wi-Fi) detects camera looping attack
Video and CSI shows a
normal activity!
Wi-Fi receiver
No extra hardware cost
SurFi achieves attack detection accuracy of 98.8% and false positive rate of 0.1%
![Page 10: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/10.jpg)
System model: indoor space under video surveillance
Field-of-view✓ Place of interest such as bank or
jewelry store
Place of interest
✓ Field-of-view of the camera
10
✓ CSI measurement cannot be compromised
![Page 11: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/11.jpg)
Field-of-view
Place of interest
Threat model: adversary can loop surveillance video feed
11
✓ Manipulate video feed
✓ Evade detection of his unauthorized activities
![Page 12: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/12.jpg)
✓ Displacement of body keypoints (e.g., wrist, elbow)
✓ Amplitude of subcarriers
Video
Challenge: video and CSI signals are different
CSI
How to find common attributes for reliable comparison of two different sensing modalities?
![Page 13: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/13.jpg)
13
Start time End time
Prominent frequency
Main intuition: Both signals capture the similar timing and frequency components
• Timing components: Start and end time of the activity
• Frequency component: Prominent frequency
Reliable detection observed consistently across different activities, people, and times
![Page 14: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/14.jpg)
OpenPose
Denoise
Data Pre-processing module
Live video feed
Wi-Fi CSI signal
14
System design of SurFi
CSI event detector module
New Event (i) detected
Video attributes
CSI attributes
Attribute extractionmodule
looped or
not?
Comparison module
Computesimilarity
score (S(i))
…
Event(N), S(N)
Event(i), S(i)
Event(1), S(1)
…
Decision module
![Page 15: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/15.jpg)
15
1) Data preprocessing module: Preprocesses the raw video and CSI signals
Video OpenPose
![Page 16: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/16.jpg)
16
Video
CSI
1) Data preprocessing module: Preprocesses the raw video and CSI signals
Raw video signal
OpenPose
Processed video signal
✓ Filter high frequency noises
Processed CSI signal
Raw CSI signal
Denoise
![Page 17: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/17.jpg)
17
2) CSI event detector module: Uses the motion energy to detect the start of a new event
1) Data pre-processing module
2) CSI event detector module
Start
![Page 18: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/18.jpg)
18
3) Attribute extraction module:Extracts common attributes
Video
CSI
Time
Time
Start time End time
Frequency
Prominent frequency
Frequency
![Page 19: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/19.jpg)
19
4) Comparison module:Computes the per-event similarity score of a single event
Video
CSI
Time
Time
Start time End time
Frequency
Prominent frequency
Frequency
∑Per-event similarity score S(i) [0, 3]
Compare
0
1
0
1
0
1
Per-attribute threshold
![Page 20: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/20.jpg)
Event 1, S(1)
S(i) are averaged
20
Decision threshold
Looped
Not loopedCompare
5) Decision module:Outputs looped or not after observing multiple events
The more the events seen, the higher the confidence for the final decision
Event 2, S(2)
Event 3, S(3)
Event 4, S(4)
Event 5, S(5)
![Page 21: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/21.jpg)
• Redmi Note 4 phone camera (13-Megapixel)
• Wi-Fi transmitter receiver pair set up on Thinkpad laptops running Linux 802.11n CSI tools
participant
Place of interest
receiver
Wi-Fi
transmitter
Wi-Fi
21
4.9-meter
2.6-meter
Experiment setup
![Page 22: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/22.jpg)
22
(E1)stand/arm waving
Three events
(E2)sit/fist thumping
(E3)sit/clapping
![Page 23: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/23.jpg)
Legit: High similarity score
Attack: Low similarity score
23
Clear difference in the per-event similarity
![Page 24: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/24.jpg)
24
Multiple events are observed for a duration of time
Example:
10 sec 15 sec 13 sec 25 sec
Time
25 sec 15 sec 30 sec 23 sec 10 sec
![Page 25: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/25.jpg)
25
1 event36%
5 events98.8%
Attack detection accuracy increases with more events
![Page 26: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/26.jpg)
26
Future improvements
• Stronger adversary• Performs criminal activities while replicating start +
end times, prominent frequency of legitimate events
• Future work: Investigate more attributes
• Multiple events in sequence• Future work: Activity recognition techniques
![Page 27: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/27.jpg)
Wi-FiWi-Fi
behind-the-wall activities
Deployment consideration
• Threshold calibration• Adjust to the new environment
• Placement of the receiver• Strategically placing the receiver
way from the wall
Not looped
0
1
0
1
0
1
Looped
27
![Page 28: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/28.jpg)
• First practical system to detect surveillance camera looping attack in real-time
•Defense technique requiring no additional hardware deployment
• Attack detection accuracy of 98.8% with false positive rate of 0.1%
• Future work: more diverse events, sophisticated adversary model
28
Conclusion
![Page 29: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/29.jpg)
29
Questions?
security guard
Place of interest
valuable
Surveillance system
Video shows a normal activity!
Wi-Fi receiver
video feed
looped!
SurFi
Compare
channel state information
(CSI)
![Page 30: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/30.jpg)
Extra Slides
30
![Page 31: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/31.jpg)
Activities behind-the-wall may degrade the performance of SurFi
transmitter
Wi-Fi
receiver
Wi-Fi
Near
receiver
Wi-Fi
Middle
receiver
Wi-Fi
Far
behind-the-wall
Conduct experiments to test behind-the-wall activities.
31
![Page 32: SurFi: Detection Surveillance Camera Looping Attacks with](https://reader031.vdocuments.us/reader031/viewer/2022012515/618f73184c5ca755e61282c6/html5/thumbnails/32.jpg)
Strategically placing the receiver at a certain distance from the wall will minimize false alarms
✓ Activities are not detected since the corresponding motion energy is close to zero.
✓ Varying motion energy may lead to false detection of an activities.
32