Tech-invite  3GPPspace  IETFspace
21  22  23  24  25  26  27  28  29  31  32  33  34  35  36  37  38  4‑5x
Top   in Index   Prev   Next

TR 33.898
Study on Security and Privacy of
Artificial Intelligence/Machine Learning (AI/ML)-based services and applications in 5G

3GPP‑Page  
V18.0.1 (Wzip)  2023/06  17 p.
Rapporteur:
Mr. Wong, Marcus
OPPO

full Table of Contents for  TR 33.898  Word version:  18.0.1

Here   Top
1 Scope  2 References  3 Definitions of terms, symbols and abbreviations  4 Key issues  5 Solutions  6 Conclusions  A Classification and protection of AI/ML data transmitted between 5GC and AF  $ Change history 

 

1  Scope  p. 6

The present document will study, based on requirements as specified in clauses 6.40 and 7.10 of TS 22.261 and architecture and key issues captured in TR 23.700-80, 5GS assistance to support Artificial Intelligence (AI) / Machine Learning (ML) model distribution, transfer, training for various applications, e.g. video/speech recognition, robot control, automotive, etc.
The scope of this study is on how to provide security and privacy to the AI/ML-based service and applications in 5G based on the following objectives of identifying key issues, potential threats, requirements, and solutions to enable:
  1. 5G system assistance for the security management which requires data transmission support for application layer AI/ML operation over the 5G system.
  2. The authentication and authorization involving data collection and sharing among UE, AF and the network to take part in application layer AI/ML operation, i.e., UE and network privacy protections to support application AI/ML services over 5G system.
  3. UE and 5G system to secure AI/ML based services and operations.
  4. Secure provisioning of the external parameter required for AI/ML (e.g., expected UE activity behaviors, expected UE mobility, etc.)
Up

2  References  p. 6

3  Definitions of terms, symbols and abbreviations  p. 6

3.1  Terms  p. 6

3.2  Symbols  p. 6

3.3  Abbreviations  p. 7

4  Key issues  p. 7

4.1  KI #1: Privacy and authorization for 5GC assistance information exposure to AF  p. 7

4.1.1  Key issue details  p. 7

4.1.2  Security threats  p. 7

4.1.3  Potential security requirements  p. 7

5  Solutions  p. 7

5.1  Solution #1: Reusing existing mechanism for authorization of 5GC assistance information exposure to AF  p. 7

5.1.1  Introduction  p. 7

5.1.2  Solution details  p. 8

5.1.3  Evaluation  p. 8

5.2  Solution #2: UE profile based 5GC assistance information exposure authorization  p. 8

5.2.1  Introduction  p. 8

5.2.2  Solution details  p. 8

5.2.3  Evaluation  p. 9

5.3  Solution #3: Reusing existing authorization mechanism for internal or external AF  p. 9

5.3.1  Introduction  p. 9

5.3.2  Solution details  p. 9

5.3.3  Evaluation  p. 9

5.4  Solution #4: Authorization for 5GC assistance information exposure to external AF  p. 9

5.4.1  Introduction  p. 9

5.4.2  Solution details  p. 10

5.4.3  Evaluation  p. 10

5.5  Solution #5: Authorization for 5GC assistance information exposure to internal AF  p. 11

5.5.1  Introduction  p. 11

5.5.2  Solution details  p. 11

5.5.3  Evaluation  p. 12

5.6  Solution #6: New solution to privacy protection for 5GC assistance information exposure to AF  p. 12

5.6.1  Solution overview  p. 12

5.6.2  Solution details  p. 12

5.6.3  Evaluation  p. 13

6  Conclusions  p. 13

6.1  Key Issue #1: Privacy and authorization for 5GC assistance information exposure to AF  p. 13

A  Classification and protection of AI/ML data transmitted between 5GC and AF  p. 14

A.1  General  p. 14

$  Change history  p. 17

Up   Top