Your shopping cart is empty! Please select the type which you want.
Artificial intelligence (AI) platform will undoubtedly open another wave of 5G killer applications. The evolution process from 4G to 5G is not only the complexity of the network, but also the equipment and types that must be managed. Therefore, more intelligent artificial intelligence (AI) platform is needed to help the telecom operators to improve their operating efficiency and create a new mode of commercial application.
Yao Dan, deputy general manager and technology manager of Ericsson, in Taiwan, said that the 4G era usually monitors the actual quality of service statistics through the Key Performance Indicators (KPI) to monitor the real quality of the services provided, but in the 5G phase, it is more guided than the network traffic. In different technologies such as network slicing, software defined network (SDN) or network function Virtualization (NFV), only using existing traditional KPI network behavior analysis, it seems unable to fully meet the support of large data analysis and new technology needed by 5G, so AI platform and machine learning are needed to assist automatic analysis and build more intelligence. Able network to improve user experience.
Based on this, Ericsson actively set up a dedicated AI expert analysis system --EEA (Ericsson Expert Analytics) for telecommunications. Yao Dan said that Ericsson has a different focus on the development of AI platforms such as Google, IBM, Microsoft and other companies in the industry, which can be applied to any industry; Ericsson is focused on developing AI and machine learning technology that is most suitable for communications.
When 5G goes to a more high-frequency development, Yao says it needs more base station coverage. In Taiwan, it needs nearly 10~20 million base stations, and the data that the base station reads without the AI platform will be useless. However, with the introduction of AI technology, it can meet three application modes, including more intelligent optimization and management of the network, the improvement of customer experience, and precision marketing operations due to different customer characteristics, and help the telecom operators to build better maintenance and marketing applications.
For example, China Mobile's big Sima data, through the mobile network data collected by the telecom operators, analyzes population characteristics, mobile terminal features, communication social features, outbound travel features, spatial location distribution and APP usage behavior, and establishes a large data system. If a store is ready to open a new store, in search of the exact location that meets the shopkeepers, we can enter Sima's big data, input some conditions, and then compare the user data and action services by Sima's big data, and lock out the location of the guest group to provide a reference for the location of the store.
It is worth the problem that, Yao said, the future telecom business should move towards the direction of building the AI platform, because the communication will be the same as electricity in the next few years, and it is inseparable from the human life. Therefore, the flow of data generated by the users in the next generation will be astonishing, and if the telecom business itself has no AI capability, it will not be able to meet the huge data analysis and traffic demand in the future.
