Base Station-AI Server IPSec Tunnel for Secure RAN Data Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems lack secure methods for transmitting user-plane and control-plane data between base stations and AI servers, which are crucial for training AI models and network automation.
Innovation Solution
Establishing a secure data transmission protocol using Internet Protocol Security (IPSec) and Internet Key Exchange (IKE) to create a data transport tunnel between base stations and AI servers, encrypting user-plane data, and exchanging cryptographic keys for secure control-plane data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data transmission between base station and AI server is performed without encryption, then transmission speed and system complexity are improved, but security and data protection are worsened
Solution Approach 1:
The patent introduces IPSec as an intermediary security protocol that operates at the network layer, mediating between the existing UP/CP signaling protocols and the underlying network infrastructure. This intermediary layer provides encryption and security functions without requiring modifications to the existing signaling protocols, thus improving security while maintaining protocol simplicity.
Solution Approach 2:
The patent segments the data transmission into two distinct channels: control-plane signaling for configuration commands and user-plane signaling for measurement and training data. Each plane can be secured independently using IPSec, allowing selective application of security measures where needed without complicating the entire system.
2Reliability
If IPSec protocol is implemented for secure transmission, then data protection is improved, but processing overhead and energy consumption are worsened
Solution Approach 1:
The patent applies IPSec encryption selectively only to the control-plane signaling channel, which carries critical configuration commands, while user-plane signaling for measurement data can use lighter security mechanisms or none at all. This partial application of strong encryption minimizes processing overhead and energy consumption while still protecting the most sensitive data.
3Reliability
If cryptographic key exchange is performed for each connection, then security is improved, but connection establishment time is worsened
Solution Approach 1:
The patent performs cryptographic key exchange and IPSec security association establishment as preliminary actions during the initial connection setup phase. Once established, these security parameters remain valid for the duration of the connection, eliminating the need for repeated key exchanges during data transmission and thus reducing overall time loss.
4Reliability
If network slicing isolation is implemented, then security between different networks is improved, but device complexity and configuration complexity are worsened
Solution Approach 1:
The patent implements a universal IPSec configuration framework that can be applied across multiple network slices and base stations using standardized parameters and procedures. This universal approach allows the same security mechanism to serve multiple isolation requirements, reducing overall system complexity while maintaining strong network separation security.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods and systems for securely sending user plane data from a base station to an Artificial Intelligence (AI) server via a mobile telecommunication network are disclosed herein. In one embodiment, a method performed by the base station located in a Radio Access Network (RAN) includes: sending an interface setup request to the (AI) server, receiving an interface setup response from the AI server, establishing, by exchanging cryptographic keys using an Internet Key Exchange (IKE) protocol, a data transport tunnel between the base station and the AI server based on an Internet Protocol Security (IPSec) network protocol, and sending the user plane data from the base station to the AI server for training artificial intelligence based models.