A YANG Data Model for Reporting Utilization Scores in ISAC

Introduction ISAC introduces a paradigm where network nodes perform both communication and sensing tasks. Several use cases and their requirements for ISAC have been defined in 3GPP and ETSI . A sensing task involves measurements of parameters such as Doppler, angle of arrival (AoA), radar cross section (RCS), micro-doppler, and range/delay , each consuming varying amounts of system resources such as memory, energy, and storage. To enable energy-efficient orchestration, it is important to measure and report the utilization impact of sensing measurements for a sensing task. This draft proposes a YANG data model that allows ISAC-enabled devices to report a utilization score (US) per sensing task, computed from weighted metrics such as CPU usage, memory, energy draw, storage load, and latency.

Model Scope The model is intended for systems that support ISAC and want to participate in energy-aware operations.

YANG Tree

YANG Module "; description "Reports a mandatory total utilization score for ISAC systems. Optionally exposes a per-parameter breakdown and per-measurement scores."; revision "2025-10-17" { description "Making per-parameter breakdown and per-measurement reporting; add timestamp/window."; } feature component-breakdown { description "Expose optional per-parameter components (compute, memory, energy, storage, latency) that contribute to the total score."; } feature per-measurement { description "Expose optional per-measurement (e.g., delay, Doppler) scores."; } typedef percent { type decimal64 { fraction-digits 2; } units "%"; description "Percentage 0..100 (not range-restricted here)."; } typedef duration-string { type string; description "Human-readable or ISO-8601 duration (e.g., PT60S)."; } container sensing-utilization { config false; description "Sensing utilization score."; leaf total-utilization-score { type uint8; mandatory true; description "Composite utilization score (0..100)."; } leaf timestamp { type yang:date-and-time; description "Time when the utilization score was computed."; } leaf window { type duration-string; description "Aggregation window (e.g., PT60S)."; } container parameters { if-feature component-breakdown; description "Optional per-parameter components."; leaf compute-score { type percent; } leaf memory-score { type percent; } leaf energy-score { type percent; } leaf storage-score { type percent; } leaf latency-score { type percent; } } list function-score { if-feature per-measurement; key "function"; description "Optional per-measurement utilization scores."; leaf function { type enumeration { enum delay; enum doppler; enum micro-doppler; enum aoa; enum rcs; } description "Type of sensing measurement."; } leaf total-utilization-score { type uint8; description "Per-measurement total (0..100)."; } leaf timestamp { type yang:date-and-time; } leaf window { type duration-string; } container parameters { if-feature component-breakdown; leaf compute-score { type percent; } leaf memory-score { type percent; } leaf energy-score { type percent; } leaf storage-score { type percent; } leaf latency-score { type percent; } } } } } ]]>

JSON Encoding example JSON encoding example for the total utilization score:

Security Considerations TBD

IANA Considerations TBD

Acknowledgements This work has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union's Horizon Europe research and innovation programme under Grant Agreement No 101192521 (MultiX). The following people have contributed to this draft: Sebastian Robitzsch