The first 6G Intelligent Wireless Communication System Competition

（Date 2022.12.26-2023.3.10）

Sponsor：IMT-2030(6G) Promotion Group

Organizer：China Academy of Information and Communications Technology (CAICT)

HUAWEI TECHNOLOGIES CO., LTD.

Guangdong OPPO Mobile Communications Corp., Ltd

Zhejiang University

1、Competition Background

The deep integration of communication and AI technologies has become one of the most important directions in the development of wireless communication systems. For 6G, the breadth and depth of communication and AI integration will be further expanded to meet "unlimited" possibilities. This competition is proposed at the key development stage of 6G research, aiming at promoting the visions, advanced technologies and concepts of 6G to all walks of life, and attracting talents around the world to study the impact of AI on future wireless communication systems, to find solutions to key problems in a systematic and multi-faceted manner. Hopefully, through this competition, we can facilitate technological breakthroughs in intelligent 6G and build industrial first-mover advantages.

With upholding public welfare, adhering to the principles of fairness, justice and openness, the main objective of the competition is to attract the global wireless communication technology research enthusiasts, researchers from enterprises and institutions, students and teachers from universities and research institutions to participate in the competition, and to find outstanding talents to help 6G research.

2、Competition Tasks

（Please refer to the official websites for the contents, rules, schedules, registration and other detailed event information.）

Competition Task 1：Intelligent Beam Prediction and Model Transfer

The wireless Massive-MIMO technology significantly increases the capacity of communication systems. In high carrier frequency scenarios, the intelligent beamforming with higher accuracy and lower overhead is the key to achieve even higher capacity increasing in future 6G communication systems. Proper MIMO beam selection is based on accurate beam measurement. However, as the number of beams increases, the overhead of scanning and measuring all beams is huge and unacceptable in real systems. A more practical solution would be to first perform sparse beam scanning measurements, then predict the remaining beams based on the measured results, and finally select the best beam. How to achieve beam prediction with a higher accuracy based on a limited measurement overhead by AI technology is a very important research topic in future 6G communications.

This task is related to the intelligent beam prediction, focusing on the key performance indicators such as prediction accuracy and generalization ability, and guiding the research direction of key technologies for 6G intelligent air interface.

Competition Task 2：Intelligent Distributed Radio Map Construction

Radio map is a key tool for wireless communication systems to satisfy requirements of potential 6G scenarios such as smart cities and smart factories. Radio maps have a wide scope of applications in base station siting, user scheduling, transmission policy optimization, etc. However, constructing precise radio maps often requires processing huge amount of data, so AI-based methods are more suitable than traditional methods.

This task expects to find the intrinsic connection between user location and radio signal strength with the learning ability of neural network, then complete the construction of radio maps. Furthermore, a distributed learning architecture will be considered to utilize the data from user devices distributed in different geographical locations, jointly completing the map construction. In such scenario, how to improve the training efficiency and model accuracy of the distributed learning system while at the cost of a low communication overhead through effectively coordinating the data and model interactions between devices is one of the most important research problems for 6G-oriented intelligent wireless communication systems.

This task will address the radio map construction problem, explore the key technologies of 6G intelligent networks through distributed learning methods, and guide the research direction.

Competition Task 3：

Recently, as an important topic of pre-research of 6G, AI-based wireless communication system attracted lots of attentions of academia and industry. However, in addition to obtaining the high-performance advantages brought by data-driven methods, AI-based solutions also face generalization problems in different scenarios with the high cost of data acquisition and model training, which bring the challenges for the landing of the technologies. Therefore, to solve these problems, the 6G AI competition will take “Design of wireless AI system for requirements of online updating with limited dataset” as the topic to explore possible solutions.

3、Competition Schedule

※Please keep an eye on the competition platform and the communication group for the specific schedule of the competition.

With upholding public welfare, adhering to the principles of fairness, justice and openness, the main objective of the competition is to attract the global wireless communication technology research enthusiasts, researchers from enterprises and institutions, students and teachers from universities and research institutions to participate in the competition, and to find outstanding talents to help 6G research.

Task 1,2:

https://developer.huaweicloud.com/contest/6g-wireless-ai.html

Task 3:

https://www.datafountain.cn/competitions/624