How to make 5G more resilient and responsive

Commercial context

To meet rising and increasingly diverse demands for 5G connectivity, CSPs are starting to integrate artificial intelligence (AI) deeply into their 5G networks. The ultimate goal is to embed AI into all aspects of 5G network operation and maintenance, thereby reducing costs and improving efficiency through network self-configuration, self-optimization and self-healing.

At the same time, CSPs are trying to collect and analyze user perception data to help them predict changes in demand and user experience in advance. This enables them to proactively configure network supply capacity to deliver the desired user experience. Ideally, 5G networks should also provide differentiated services that meet the specific needs of each industry automatically, improving service to end customers and reducing environmental impact.

The solution

Supported by China Mobile, China Unicom, AsiaInfo, Huawei and ZTE, the Smart 5G Catalyst is building a user perception evaluation map for an entire mobile network, based on user behavior and perception characteristics. This will help CSPs to identify areas where the network is not performing well, and develop solutions to improve the user experience. The map will be supplemented by a scoring system of user business model and perception indicators. These tools can inform the scheduling of the operations and coverage capabilities of 5G base stations to further improve a network’s efficiency.

The Catalyst is also developing a compensation mechanism to maintain services in the event of a network fault, to ensure that users are not adversely affected. “This Catalyst combines research on user behavior, perception data and beamforming to develop a solution for the intelligent ability of NR (New Radio) base stations based on synaesthesia, which can compensate for outages and provide businesses with network scheduling and coverage specific to their needs,” explains Yang Ranran, an Engineer at China Mobile Anhui.

Applications

The Catalyst’s participants have tested AI-based solutions to support smart 5G connectivity in parks, for policing, transportation and other business scenarios. In total, they have piloted eight specific use cases. For example, to help CSPs anticipate the movements of mobile network users, the project developed a ‘spatio-temporal recurrent’ neural network algorithm, which analyzes users’ location information over a week to achieve an accurate prediction of the overall distribution of users. It also employs a machine vision algorithm and multi-cell joint optimization model to quickly determine the optimal configuration of antenna beam parameters. Applied in large supermarkets, universities, and office buildings, these tools achieved an 11% increase in user perception, an 8.4% increase in 5G revenues, and a 5.2% decrease in 5G single station maintenance costs.

The Catalyst has also developed a solution to quickly locate and compensate for network outages. In the event of a network fault, it applies an ‘iterative adjustment convergence algorithm’ to determine the optimum compensation adjustment range, and immediately activates nearby network elements accordingly, ensuring the overall stability of the network. The Catalyst found this approach can serve 92% of the traffic that would otherwise have been lost, and reduce the number of user complaints by 36%. The solutions developed by the Catalyst have also been used to cut the cost of inspecting the east-west Chizhou ultra-high voltage power transmission line, which runs through a complex mountainous area, by 2.2 million yuan over one year. This was supported by partnership between China Mobile Anhui and ZTE to develop a 5G- and AI-based solution to enable transmission line inspection by uncrewed vehicles and drones.

Wider value

The results of the Smart 5G Catalyst point to how 5G and AI can work together to provide individual users and entire industries with more reliable and responsive connectivity. As they support high bandwidth and low latency connectivity, 5G large-scale antenna arrays enable greater integration of sensory data and communication technologies, helping to achieve better coordination between a network's constituent parts and the customers who use them.

Smart 5G solutions can also help to reduce a network’s energy consumption and lower costs. “We are implementing the Smart 5G business philosophy of being smarter, more agile and greener,” concludes Qiao Jun, Deputy Manager of Network Department at China Mobile Anhui. “Our Catalyst project will solve the needs of the transformation from traditional communication services to new information services, improve the supply capacity of information services, promote the coordinated development of the upstream and downstream of the 5G industry chain, and help realize the value goal of the digital intelligence transformation of the whole society.”