The Perception-Driven Broadband Network Energy Efficiency Solution addresses one of the most critical challenges facing the broadband industry: how to reduce rapidly rising energy consumption and carbon emissions without compromising network performance, user experience, or computing power quality. Bringing together operators, equipment vendors, IDC service providers, and research institutions, the Catalyst delivers a coordinated, end-to-end approach to green and sustainable digital infrastructure. As 5G base station deployments scale and data centers and intelligent computing facilities become primary drivers of electricity demand, the industry faces an “energy consumption triangle”: surging energy costs, fragmented optimization across network and data center domains, and energy-saving measures that often degrade service quality. This Catalyst responds by introducing a new paradigm of experience-oriented energy efficiency, enabling energy conservation that is driven by real-time service perception and system-level optimization rather than isolated, component-level controls. The solution is built on one unified framework and three scenario-specific pillars. A network-computing-energy synergy framework establishes a global, end-to-end energy efficiency model spanning user terminals, access and transmission networks, cloud data centers, and intelligent computing centers. Energy management is deeply integrated with service and computing-power scheduling, allowing optimization decisions to be made at the system level rather than within individual silos. Three precision pillars address the industry’s highest energy-consumption scenarios. For 5G/4G base stations, AI-driven service awareness and traffic prediction dynamically adjust antenna power, carrier activation, and symbol-level shutdown strategies, achieving cell-, user-, and service-level energy savings while preserving high-value service experience. For traditional data centers, advanced cooling technologies combined with AI-based load-to-cooling coordination drive PUE toward best-in-class levels below 1.2. For intelligent computing data centers, elastic and green AIDC architectures optimize computing, cooling, and airflow, and schedule non-urgent training workloads during off-peak periods to maximize green energy usage and reduce energy consumption per computing task. Success is measured through tangible business and sustainability outcomes: 15–25% reduction in 5G base station energy consumption, data center PUE optimized to below 1.3, and 10–15% lower unit energy consumption in intelligent computing centers—all while ensuring high-value service KPIs degrade by less than 5%. At scale, the solution is expected to reduce annual carbon emissions by more than one million tons, directly supporting dual-carbon goals. By aligning energy efficiency with user experience and computing quality, this Catalyst establishes a scalable, perception-driven blueprint for the green and low-carbon transformation of broadband networks—enabling sustainable growth for the digital infrastructure of the future.