Home > Communication Service Provider Solutions > Enabling Telecom Transformation > Network-Based Energy Saving Opportunities in RAN for Improved TCO > Energy-efficient E2E approach
Focusing more deeply on O-RU architectural aspects, coupled with AI/ML inference engines that can leverage telemetry can significantly improve efficiency. Novel AI/ML approaches can distill intelligence from the vast quantity of measurement data that modern-day RANs generate, allowing the network to direct policy to O-DU and O‑RU network elements that are optimized from a global network operations perspective, resulting in greater overall network energy efficiency.
To illustrate the potential of O-RAN-based opportunities for efficiency and savings, this section discusses several features in O-RAN’s NES Use Case Report [4]. Through an analysis of these use cases, Table 1 and Figure 1 model the energy savings potential for each use case.
The calculations done here are based on using daily average usage (DAU) traffic loads modeled with the O-RAN use cases for a particular scenario. In each case, savings are realizable while ESF is activated on an individual stand-alone basis. Table 1 describes the energy saving scenarios.
Scenario | O-RAN NES use case | Description | Energy savings methods based on DAU | Energy savings relative to scenario 2 |
Scenario 1 | Laboratory Benchmark | 100% traffic load O-RU Benchmark: maximum power consumption | No energy savings. Laboratory use case that establishes an O-RU 100% traffic load energy consumption. | – |
Scenario 2 | Operational Benchmark | O-RU 30% traffic load DAU Benchmark: typical power consumption | No energy savings. Baseline energy consumption from which savings are calculated. | 0% |
Scenario 3 | PA Dynamic Voltage Bias Adaptation | Power Amplifier (PA)-efficiency optimization based on traffic load | Energy savings through adaptive rebiasing to optimize efficiency of key components. | Up to 9% |
Scenario 4 | Symbol TRx Shutdown and User Packing | Energy-efficient scheduling with user packing and symbol blanking | Energy savings though shutting down components when symbols are blank – no traffic to transmit. | Up to 25% |
Scenarios 5–7 | RF Reconfiguration and Carrier Switch OFF | Multiple Input – Multiple Output (MIMO) reconfiguration and antenna branch shutdown | Energy savings through shutting down O-RU MIMO antenna paths from 4 paths to 3, to 2, to SISO when traffic patterns allow. | 11–86% |
Scenario 8 | Deep Sleep | Traffic steering sector or cell switch off | Energy savings from shutting down entire O‑RUs. | 97–100% |
O-RAN does not mandate the use of ESFs, so the specific implementation of O‑RAN‑compliant network equipment will determine how many and which ESFs are supported. O‑RU and O-DU vendors must explicitly incorporate these ESFs in their architectures at the design stage. Even when adhering to O-RAN, having functional innovative features in next-generation network equipment requires the right set of skills, innovation, and network equipment design expertise. O-RU designs must be able to leverage cross-layer innovation for long-term reduced TCO to ensure increased benefits over the lifetime of a network.
While each ESF scenario is distinct and independent, there is opportunity for combining features that are not mutually exclusive. Invoking multiple ESFs simultaneously, however, does not guarantee linear aggregation of benefits and may not provide a cumulative gain.