en.Wedoany.com Reported - At a recent telecommunications industry exhibition, Li Tiejun, General Manager of the Customer Service Center at China CICT Mobile, gave an exclusive interview to C114, elaborating on how operators can accelerate their transition from traffic management to Token management in the era of artificial intelligence, and how CICT Mobile utilizes its full-stack capabilities of digital-intelligent integration to help the communications industry move toward a new phase of efficiency, intelligence, and convergence.
This year marks the seventh anniversary of 5G commercial use, and the network has shifted from large-scale construction to refined operations. As of the end of April 2026, China had over 5.009 million 5G base stations, with 1.262 billion 5G mobile phone users, achieving a penetration rate of 68.7%. Li Tiejun believes that the industry has entered a phase of stock management, where traffic and user growth have peaked, making the traditional model of revenue growth through traffic unsustainable. Against this backdrop, operators are reshaping their businesses through AI Token management, accelerating the transition from "traffic management" to "Token management," with all three major operators having launched Token packages. Li Tiejun stated that operators have shifted their focus to precise investment and deep value mining, upgrading their business models from indicator management to traffic management and AI Token management. To this end, CICT Mobile has introduced three solutions: revitalizing existing networks through the HiNet2.0 platform; building modular intelligent agents for rapid deployment and quantifiable results; and leveraging the EagleSee intelligent monitoring platform to link network and business data, delivering targeted solutions to boost traffic revenue and Token management.
In the era of stock management, converting network data into commercial value is key. CICT Mobile's EagleSee platform integrates multi-source big data to achieve a closed loop from network-business data linkage to operational decision-making. The platform features three progressive capabilities: first, wireless holographic perception, collecting grid-level data based on high-precision digital twins; second, network-business data linkage, correlating network quality with data such as ARPU and churn rate to identify low-efficiency, high-value areas; and third, automatically generating optimization and marketing deployment lists, connecting business systems to form a closed loop. For example, the platform can precisely identify areas with limited traffic, such as during events or concerts, and through intelligent site selection, 3D simulation, and AI optimization, release traffic and enhance user experience, helping operators achieve dual improvements in user perception and business revenue. For Token management, CICT Mobile is exploring the construction of a new three-tier computing network architecture—"intelligent computing center, edge computing room, base station computing power"—leveraging distributed computing network scheduling agents to integrate idle base station computing power, reducing Token production costs, and using automated planning, construction, maintenance, and optimization agents to cut operational costs, assisting operators in building an ecosystem of intelligent agent products for both B-end and C-end markets, driving the shift from traffic operations to integrated computing-network value management.
Autonomous network optimization is crucial for enhancing user experience. CICT Mobile's upgraded HiNet2.0 autonomous network architecture includes three core optimizations: first, building an AI agent matrix to achieve fully automated closed loops for root cause localization of quality degradation, performance optimization, and experience assurance; second, upgrading the NetTwins digital twin engine, using multi-layer grid-level wireless simulation models to preview solutions and reduce trial-and-error costs; and third, adding an agent orchestration layer that decomposes operational capabilities into standardized APIs, enabling end-to-end automated processing, with the entire capability set deployable on edge servers. Li Tiejun noted that CICT Mobile adopts a "three-layer decoupling, scenario-driven" agent architecture: the knowledge layer uses lightweight RAG combined with a communications-specific knowledge graph for low inference costs; the decision layer integrates rule engines with AI reasoning to improve decision effectiveness and availability; and the execution layer decomposes operational capabilities into atomic APIs, with multiple agents collaborating to achieve full-process automation, effectively reducing existing network operational labor costs.
In the satellite internet field, CICT Mobile holds a first-mover advantage. Li Tiejun stated that the company leads the development of international satellite internet standards and possesses full-stack self-developed products, including satellite-borne equipment, ground gateways, core networks, and terminals. It offers three types of service solutions for operators: satellite-terrestrial integrated managed services to enhance satellite link stability; the AirLink Intelligent Test drone testing system for rapid air interface fault localization; and full-stack end-to-end delivery for one-stop deployment of scenarios such as low-orbit networking and direct satellite connectivity for mobile phones. Among these, satellite NTN can extend AI computing power and new services to ground network blind spots, enabling uninterrupted operation of devices across all areas. Li Tiejun emphasized that CICT Mobile provides a complete paradigm for the digital-intelligent transformation of communication networks, and these innovations are driving the communications industry toward a smarter and more efficient stage of development.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
