Towards Net-Zero Carbon Emissions in Network AI for 6G and Beyond

Peng Zhang,Yong Xiao,Yingyu Li,Xiaohu Ge,Guangming Shi,Yang Yang

DOI: https://doi.org/10.1109/MCOM.003.2300175

2023-09-18

Abstract:A global effort has been initiated to reduce the worldwide greenhouse gas (GHG) emissions, primarily carbon emissions, by half by 2030 and reach net-zero by 2050. The development of 6G must also be compliant with this goal. Unfortunately, developing a sustainable and net-zero emission systems to meet the users' fast growing demands on mobile services, especially smart services and applications, may be much more challenging than expected. Particularly, despite the energy efficiency improvement in both hardware and software designs, the overall energy consumption and carbon emission of mobile networks are still increasing at a tremendous speed. The growing penetration of resource-demanding AI algorithms and solutions further exacerbate this challenge. In this article, we identify the major emission sources and introduce an evaluation framework for analyzing the lifecycle of network AI implementations. A novel joint dynamic energy trading and task allocation optimization framework, called DETA, has been introduced to reduce the overall carbon emissions. We consider a federated edge intelligence-based network AI system as a case study to verify the effectiveness of our proposed solution. Experimental results based on a hardware prototype suggest that our proposed solution can reduce carbon emissions of network AI systems by up to 74.9%. Finally, open problems and future directions are discussed.

Networking and Internet Architecture,Artificial Intelligence,Distributed, Parallel, and Cluster Computing

What problem does this paper attempt to address?

The paper aims to address the following issues:

1. **Reducing Carbon Emissions of Network AI Systems**: With the development of 6G technology, the overall energy consumption and carbon emissions of mobile networks are still rapidly increasing, especially after the introduction of resource-intensive AI algorithms. The paper proposes a new joint Dynamic Energy Trading and Task Allocation optimization framework (DETA) to reduce the carbon emissions of network AI systems throughout their lifecycle.

2. **Evaluating the Lifecycle of Network AI**: The paper introduces an evaluation framework for analyzing the lifecycle of network AI implementation and identifies the main sources of carbon emissions in network AI systems. In this way, researchers can better understand the energy consumption at each stage and its impact on the environment.

3. **Validating the Effectiveness of the Solution**: The paper uses a network AI system based on Federated Edge Intelligence (FEI) as a case study to experimentally validate the effectiveness of the proposed optimization solution. The results show that the solution can significantly reduce the carbon emissions of network AI systems, by up to 74.9%.

In summary, this paper mainly focuses on how to achieve the goal of net-zero carbon emissions in 6G and future communication technologies, particularly in the application of network AI technology, by effectively managing and optimizing energy consumption to reduce environmental impact.