ZTE Corporation; a major international provider of telecommunications, enterprise and consumer technology solutions for the mobile internet; together with the Zhejiang Branch of China Telecom, has jointly built a self-adaptive spatiotemporal cognitive network based on ZTE’s Radio Composer, improving dynamic user experiences in high capacity scenarios.
Under the collaboration with intelligent user navigation, the network solution matches network resources with traffic distribution more precisely and efficiently through on-demand elastic coverage of two-layer network, adapting to user group flow in space over different time periods.
Traffic distribution of different periods in one area
The spatiotemporal cognitive network intelligently predicts traffic distribution in the first place. According to location change of user groups in different time periods within base station coverage, the network solution, by virtue of long short-term memory (LSTM) algorithms, performs in-depth study and prediction of traffic distribution on physical grid level, and analyzes the traffic space distribution trend in different periods.
Based on the traffic distribution trend in both time and frequency, the spatiotemporal cognitive network implements the intelligent carrier power scaling function through power sharing, to achieve flexible coverage adjustment.
When the traffic loads within coverage of the two carriers are both high, the solution balances the two carriers with same coverage to guarantee capacity. When the traffic loads within coverage of the two carriers differentiates obviously, the solution adjusts the coverage mode. It adopts high power to cover the high-load area and decreases power in the low-load area, therefore precisely matching radio resources to ensure user experiences.
The spatiotemporal cognitive network focuses on intelligent experience collaboration and establishes AI logic grid knowledge base of base stations, in order to further balance network efficiency and user experiences.
With the model of intelligent carrier power scaling, real-time user experiences are evaluated according to the blocking rate. When user experiences cannot match service requirements well, with intelligent experience prediction based on the grid knowledge base, the users can be rapidly guided to the targeted grids, ensuring the optimal resource allocation.
The traditional fixed network coverage model changes with network intelligence and self-adaptation, from "user searching for network" to "network self-adapting to user."
Moving forward, the Zhejiang Branch of China Telecom and ZTE will keep innovating together to provide superb network performance and boost digital transformation.