Coordinated Energy Saving between RAN and Transport Network

This document addresses the usecases of RAN and transport network identified in through an coordinated energy saving mechanism. As described in Section 2.7 of , radio base stations are typically connected to the backbone network by means of fiber technologies. The capability of radio link can be adjusted accordingly to the traffic observed. Some of the power-related parameters are dynamically configured to adjust the power to the observed traffic levels with some coarse granularity to avoid the underutilized of energy. In addition, as described in describes the energy saving in transport network. However, under current energy-saving methods, energy saving of RAN and TN are done separately and without coordination. This leads to RAN adjusting its wireless link according to its own traffic conditions, but the transport network is unaware of this and still reserves resources based on the maximum traffic, especially the internal forwarding resources, optical module resources, and TDM slot resources, resulting in resource waste. Therefore, it is necessary to study an energy-saving coordination mechanism between the RAN and the transport network.

This section describes the detailed coordination mechanisms that may exist between the Radio Access Network (RAN) and the Transport Network (TN).

Energy Saving Coordination between RAN and Transport Network Access <-------+ +-------> Access <-------+ | | | | Node 1 | | Node 2 | | | | +----+----+ +----+----+ | | | | | | | | | | | | | +-----------------+--------------------+ | | | | | | | | | | +--------+-------+ | | | | | | Aggregation | | | | | | | Node | | | | | | +--------+-------+ | | | | | | | | | | +-----------------+--------------------+ | | | | | | | +----------+-------+ | | | | TN Controller <-------------------------------------------------+ | +------------------+ | | | +--------------------------------------------------------------------------------+ ]]> As show in Figure 1, the upper part represents the Radio Access Network(RAN), which consists of six RAN access nodes, labeled RAN1 through RAN6. These nodes provide radio coverage and user-plane termination for mobile terminals. All six RAN nodes are under the control of a single RAN controller, which is responsible for radio resource management, traffic scheduling, and RAN-side energy-saving decisions. The lower part represents the Transport Network (TN), which provides the connectivity between the RAN nodes and the core network. This part comprises three transport nodes arranged into a two-tier hierarchy, two access nodes, referred to as Access Node 1 and Access Node 2, and one aggregation node. Access Node 1 and Access Node 2 are directly interact with the RAN nodes, while the Aggregation Node provides aggregated streams toward the core network. All transport nodes in the TN are under the control of TN controller (e.g. for path computation and resource allocation), which is responsible for TN side energy-saving decisions such as port sleep, switching board sleep, and TDM timeslot deactivation, etc. The energy-saving coordination between the RAN and the TN may include:

Energy-related information advertisement: energy consumption prediction, link power status;

Energy-saving intent decomposition or policy: translating an operator-level energy-saving goal into concrete actions for both the RAN and the TN;

Action alignment: aligning wake-up latencies and co-sleep windows to avoid one domain sleeping while the other related resources remains active;

Traffic/load coordination: jointly migrating services and rerouting backhaul paths before shutting down a transport link;

This coordination can be realized through four approaches:

Controller-to-Controller (C2C) coordination is best suited for policy-level and long-horizon tasks such as intent decomposition, global state aggregation, and joint energy-budget management, where minutes-to-hours decision cycles are acceptable. This approach may exist between peer controllers or orchestrators across RAN and TN for unified energy saving.

Node-to-Node (N2N) coordination is best suited for data-plane and short-horizon tasks such as real-time action synchronization, link adaptation, and rapid wake-up upon unexpected traffic bursts, where controller involvement would introduce latency.

Hybrid coordination combining both C2C and N2N is also possible.

Depending on the specific coordination approach, different extensions may be required. For C2C coordination, the interaction typically relies on CPIs and may necessitate extensions to existing YANG data models and intent objects within the IETF GREEN framework or network management agent coordinations. For N2N coordination, nodes may leverage existing protocols, such as LLDP, IPv6, or some OAM protocol, without introducing new protocols. This part requires further study within the GREEN framework and its YANG modules.