Contextual Factors in Mobile Security and Privacy Policy Enforcement - - PowerPoint PPT Presentation

▶
contextual factors in mobile security and privacy policy
SMART_READER_LITE
LIVE PREVIEW

Contextual Factors in Mobile Security and Privacy Policy Enforcement - - PowerPoint PPT Presentation

Aug 17, 2022 224 likes •547 views

Contextual Factors in Mobile Security and Privacy Policy Enforcement Mobile Services and Edge Computing Workshop, Helsinki, 28.7.2016 Markus Miettinen Technische Universitt Darmstadt 28.07.2016 | Fachbereich Informatik | Lehrstuhl

slide-1
SLIDE 1

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 1

Mobile Services and Edge Computing Workshop, Helsinki, 28.7.2016 Markus Miettinen Technische Universität Darmstadt

Contextual Factors in Mobile Security and Privacy Policy Enforcement

slide-2
SLIDE 2

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 2

About the Speaker

Alumnus of the University of Helsinki 13 years experience in industrial R&D at NOKIA Research Center Helsinki, Finland and Lausanne, Switzerland Researcher at Fraunhofer Institute for Secure Information Technology, Darmstadt Since 2013 Researcher at Technische Universität Darmstadt Areas of interest include Mobile Security, Context- Awareness, Data analysis for security applications and IoT Security

slide-3
SLIDE 3

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 3

Outline

Context-aware policy adaptation

  • Utilizing profiled information about the context to make

access control decisions Context-based Proofs-of-Presence (PoP)

  • Using context measurements to verify co-presence of two

devices

slide-4
SLIDE 4

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 4

What is Context?

In this presentation: Any properties of the physical ambient environment that mobile devices can sense with their on-board sensors.

slide-5
SLIDE 5

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 5

Context-Aware Policy Adaptation

Markus Miettinen, Stephan Heuser, Wiebke Kronz, N. Asokan and Ahmad-Reza Sadeghi ” Proceedings of the 9th ACM Symposium on Information, Computer and Communications Security (ASIACCS 2014) , June 2014.

slide-6
SLIDE 6

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 6

Security and Context

Rich sensing capabilities New context-aware apps and services All of these features need to be managed! Challenge: How to make security & privacy policy management

  • User-friendly
  • Personalized
  • Context-aware

6

slide-7
SLIDE 7

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 7

Challenge: Inflexible device lock

Many people feel device locks to be too difficult to use, leaving their device unprotected

 need for a better device locking mechanism

Goal: context-sensitive device locking:

  • Quick locking in high-risk contexts
  • Fewer passcode requests in low-risk contexts

7 Markus Miettinen

slide-8
SLIDE 8

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 8

Challenge: Sensory Malware

Mobile apps tend to ask for excessive permissions

  • Users often grant permissions automatically

Adversary: Sensory Malware

  • malicious software can use sensors to collect potentially

sensitive information from user’s context

  • e.g., audio, video, accelerometer, etc.

 Need for more fine-grained, context-sensitive permission management Goal: restrict apps’ access to sensors in sensitive contexts

slide-9
SLIDE 9

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 9

Legacy solution: user-specified, pre-defined policies

This has some Drawbacks:

  • Difficult to understand
  • Time-consuming
  • Likelihood of erroneous policies is

high A quick remedy: One preconfigured policy

  • Inflexible
  • Not personalized
  • May surprise users

9

  • M. Conti, V. Nguyen, and B. Crispo. CRePE: Context-Related Policy Enforcement for Android. In ISC 2011, volume 6531 of LNCS, pages 331-345.

Springer, 2011.

  • M. Covington, P. Fogla, Z. Zhan, and M. Ahamad. A context-aware security architecture for emerging applications. In Computer Security

Applications Conference, 2002. Proceedings. 18th Annual, pages 249-258, 2002.

  • M. L. Damiani, E. Bertino, B. Catania, and P. Perlasca. GEO-RBAC: A spatially aware RBAC. ACM Trans. Inf. Syst. Secur., 10(1), Feb. 2007.
slide-10
SLIDE 10

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 10

What security concerns do users have with regard to their smartphone?

User Perceptions

Questionnaires and on-line survey More than 150 participants

slide-11
SLIDE 11

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 11

User Perceptions

Two main user concerns: Concerns related to privacy exposure

  • Intrusive apps exfiltrating sensitive user information to

unauthorised parties Risk of device misuse

  • Someone stealing the user‘s device or using it without the

user‘s permission

slide-12
SLIDE 12

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 12

Main findings from the Survey

Perception of risk of device misuse depends on people present and their familiarity, not so much on the place  Estimate familiarity of people Perception of privacy exposure depends on the place itself, not so much on the people present  Estimate familiarity of places

slide-13
SLIDE 13

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 15

Our approach

Profile user‘s relevant places (= "contexts") Profile frequent social contacts (= devices) Create prediction model for access control based on profiles and sensed data

Relevant places Relations to other users Safe? Unsafe? Sensitive? Public?

slide-14
SLIDE 14

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 23

Context Features

Familiarity of Context (identified through GPS and WiFi)

  • Number of visits
  • Time spent in context

Familiarity of devices in vicinity (identified thorough Bluetooth)

  • Number of visible devices
  • Number of visible familiar devices
  • Average # of past encounters for familiar devices
  • Average time spent with familiar devices
slide-15
SLIDE 15

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 26

Results

Adaptive device lock: 70% TP rate at relatively moderate FP rate of 10% Number of passcode queries reduced by 70%!

Sensory malware protection: Random Forest and k-NN achieve 70% TP rate at very low FP rate of 2-3.5%

slide-16
SLIDE 16

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 33

Context-based Proofs-of-Presence

Markus Miettinen, N. Asokan, Farinaz Koushanfar, Thien Duc Nguyen, Jon Rios, Ahmad-Reza Sadeghi, Majid Sobhani, Sudha Yellapantula, „I know where you are: Proofs of Presence resilient to malicious provers” 10th ACM Symposium on Information, Computer and Communications Security (ASIACCS 2015), April 2015.

slide-17
SLIDE 17

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 34

Venue check-ins in OSN:s

“check-in” Location claim Incentives for location cheating

slide-18
SLIDE 18

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 36

Context-based Proofs-of-Presence

Context measurements 𝐷𝑊 = 𝑛1, 𝑛2, … Audio Luminosity

WiFi Bluetooth

𝐷𝑄 = 𝑛1′, 𝑛2′, …

PoP request, 𝐷𝑄 If 𝑒𝑗𝑡𝑢(𝐷𝑊, 𝐷𝑄) < Δ𝑢ℎ𝑠, accept PoP PoP accept Verifier 𝑊 Prover 𝑄

slide-19
SLIDE 19

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 37

Location Claim Verification

Machine learning-based classification model Trained with a set of annotated pairs of co-located and non- co-located measurements Classifier used to determine whether two measurements

  • riginate from co-located devices or not
slide-20
SLIDE 20

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 38

Context Guessing

Context 𝐷 Malicious prover A Verifier 𝑊

𝐷A(𝑢 − 𝑜)

𝐷𝑊(𝑢) Context replay: 𝐷A(𝑢 − 𝑜)

slide-21
SLIDE 21

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 41

Hardening of PoPs

Surprisal filtering

  • Reject easy-to-guess PoPs

Longitudinal ambient context modalities

  • Increase the inherent entropy of PoPs

41

slide-22
SLIDE 22

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 42

Surprisal of Context Measurements We use surprisal to measure how easy it would be for a malicious prover to guess a valid context

  • bservation in a context. The higher the surprisal

is, the more difficult it would be for the attacker to correctly guess such observations. The surprisal of a context measurement 𝐷 is defined as the self-information that measurement

𝐽 𝑃𝑌 = 𝐷 = log2( 1 𝑄 𝑃𝑌 = 𝐷 ) = − log2(𝑄(𝑃𝑌 = 𝐷))

42

slide-23
SLIDE 23

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 43

Types of Context Information

Static Dynamic WiFi, Bluetooth 𝑒1 𝑒2 𝑒3 𝑒4 𝑒6 𝑒7 𝑒5

slide-24
SLIDE 24

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 44

Surprisal Filtering

1. Profile the occurrence frequency of contextual elements (e.g. WiFi and BT devices) in the context 2. When receiving a PoP, evaluate the surprisal associated with the elements of the verifier’s context measurement. 3. If surprisal is too below surprisal threshold 𝐽𝑢ℎ𝑠, reject PoP.

44

slide-25
SLIDE 25

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 47

Effectiveness of Surprisal Filtering

Surprisal filtering significantly reduces False Positive rate of PoPs

𝑱𝒖𝒊𝒔 = 𝟓 𝒄𝒋𝒖𝒕 Unfiltered Bluetooth WiFi Average 27.7 %

  • 16.7 %
  • 5.5 %
  • Rel. change
  • 60.4 %
  • 20.0 %
slide-26
SLIDE 26

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 48

Longitudinal Ambient Context Modalities (Luminosity & Audio)

Goal: Increase entropy of PoP Approach:

  • 1. Measure ambient context modalities level and record

snapshots

𝑁 = {𝑛1, 𝑛2, … , 𝑛𝑜} Each measurement 𝑛𝑗 has length 𝑥 = 1 sec and 𝑜 = 60

Trade-off

  • Longer snapshot provides more entropy
  • Shorter snapshot provides better usability
  • Short measurements require accurate time

synchronisation

48

slide-27
SLIDE 27

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 49

Evaluation: Longitudinal Modalities

Attack Dataset False Positive Rate Luminosity 1.1 % Audio 0.4 % Luminosity + Audio 0.4 % Bluetooth 21.9 % WiFi 26.0 % Bluetooth + WiFi 23.5 % Luminosity + Audio + BT + WiFi 3.6 %

slide-28
SLIDE 28

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 50

Conclusion

Use of context enables many novel applications and services but poses also challenges w.r.t. privacy and manageability Utilizing context-profiling can help in tackling some of the manageability-related issues Context fingerprinting-based approaches enable new possibilities for utilizing context to construct entirely new security functionalities like proofs-of-presence

slide-29
SLIDE 29

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 53

Ongoing work

Utilizing deeper context-awareness to encounter sophisticated threats like relay attacks and context- manipulating adversaries Extending the use of context into IoT domain through, e.g., context-based pairing

slide-30
SLIDE 30

28.07.2016 | Fachbereich Informatik | Lehrstuhl Systemsicherheit | Prof. Ahmad-Reza Sadeghi | 54

Thank You!

markus.miettinen@trust.tu-darmstadt.de @mmietti

www.trust.informatik.tu-darmstadt.de/people/markus-miettinen

54