Foundation Model Based Native AI Framework in 6G with Cloud-Edge-End Collaboration

Xiang Chen,Zhiheng Guo,Xijun Wang,Howard H. Yang,Chenyuan Feng,Junshen Su,Sihui Zheng,Tony Q. S. Quek

2023-10-26

Abstract:Future wireless communication networks are in a position to move beyond data-centric, device-oriented connectivity and offer intelligent, immersive experiences based on task-oriented connections, especially in the context of the thriving development of pre-trained foundation models (PFM) and the evolving vision of 6G native artificial intelligence (AI). Therefore, redefining modes of collaboration between devices and servers and constructing native intelligence libraries become critically important in 6G. In this paper, we analyze the challenges of achieving 6G native AI from the perspectives of data, intelligence, and networks. Then, we propose a 6G native AI framework based on foundation models, provide a customization approach for intent-aware PFM, present a construction of a task-oriented AI toolkit, and outline a novel cloud-edge-end collaboration paradigm. As a practical use case, we apply this framework for orchestration, achieving the maximum sum rate within a wireless communication system, and presenting preliminary evaluation results. Finally, we outline research directions for achieving native AI in 6G.

Information Theory,Distributed, Parallel, and Cluster Computing,Machine Learning,Networking and Internet Architecture,Signal Processing

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve

This paper aims to address a series of challenges in implementing Native Artificial Intelligence (Native AI) in the sixth-generation mobile communication (6G) networks. Specifically, the paper focuses on the following aspects:

1. **Data Perspective**:
- **Establishment of Expert Knowledge Base**: Due to the lack of widely available public datasets in the wireless AI field, the paper emphasizes the importance of establishing an expert knowledge base specifically for 6G networks. This knowledge base will include real-time data, synthetic data, and structured data to improve data efficiency and the accuracy of model training.

2. **Intelligence Perspective**:
- **Intent-aware Pre-trained Foundation Model (PFM)**: Although existing pre-trained foundation models (PFM) have achieved significant success in fields such as computer vision (CV) and natural language processing (NLP), their application in actual communication networks still requires customized design. The paper proposes an intent-aware PFM for efficiently identifying user intentions and fine-grained resource allocation.
- **Task-oriented AI Toolkit**: Since resource-constrained devices cannot deploy the entire PFM, the paper suggests developing lightweight task-oriented AI models and establishing a standardized task-oriented AI toolkit to achieve ubiquitous intelligence in 6G networks.

3. **Network Perspective**:
- **Collaborative Paradigm**: Current edge or terminal devices lack the capability to autonomously establish expert knowledge bases and task-oriented AI toolkits, and existing communication networks struggle to support efficient data and resource sharing among intelligent nodes. The paper proposes a new paradigm based on cloud-edge-terminal collaboration, combining task-oriented connections and device-oriented connections to achieve distributed resource sharing and collective intelligence of intelligent nodes.

### Summary

By analyzing the challenges from the perspectives of data, intelligence, and network, the paper proposes a 6G Native AI framework based on pre-trained foundation models. This framework leverages cloud-edge-terminal collaboration to achieve ubiquitous intelligence in 6G networks, improve Quality of Service (QoS) and Quality of Experience (QoE), and demonstrates its potential contributions through specific orchestration use cases.