LyndHub
Jadwal Sholat

Memuat jadwal sholatโ€ฆ

โ† All journalsSearch index

Ilmu Komputer & AI editorial

Open AccessOA2026

E2-Conditioned Finite-Horizon Effective Capacity for Public-Safety MCX over Shared O-RAN

A finite-horizon service-assurance framework that replaces asymptotic effective capacity with E2-conditioned, confidence-calibrated capability profiles for mission-critical MCX orchestration at the Near-RT RIC.
Jingqing Wang; Wenchi Chengยท 2026ยท DOI 10.48550/arXiv.2608.25442

The core problem

Public-safety Mission Critical Services (MCX) must be assured over finite incident horizons, yet they are increasingly carried over shared Open radio access networks (O-RAN) where ordinary mobile traffic competes for the same radio and transport resources. The authors identify a structural mismatch between the assurance question and the tools used to answer it. Existing RAN key performance indicators (KPIs) are retrospective: they summarize what has already happened rather than what remains executable within the remaining mission window. Conventional effective capacity (EC), meanwhile, characterizes an asymptotic stationary regime. As the abstract states, this asymptotic view "suppresses both the initial E2-observed condition and the finite selection-to-actuation transient." In an O-RAN deployment, that transient is not a detail. Heterogeneous E2 domains expose different observations, control actions, and actuation latencies, so the state observed at selection time and the state at which control actually takes effect can differ materially over a short incident horizon. The paper therefore asks how to characterize the service capability that is actually executable within a finite mi

Innovation

The coupled MATLAB/ns-3 evaluations demonstrate the predicted short-horizon state and actuation effects. In other words, the finite-horizon conditioning is not merely a formal refinement: the simulations show behavior that the asymptotic view would suppress, namely sensitivity to the initial E2-observed condition and to the delay between selection and actuation. The evaluations further show that adaptive connectivity selection improves MCX supportability under O-DU degradation. This is the central operational result: when a distributed unit degrades, a static or retrospective selection policy loses MCX supportability, whereas an FH-EC-driven selection that accounts for the finite horizon and the actuation transient can maintain it. Importantly, the improvement is achieved while satisfying the configured multi-QoS and non-MCX protection requirements. That conjunction matters for a shared O-RAN: the framework does not buy MCX supportability by starving ordinary traffic or by violating the multiple QoS constraints configured for other services. The results are reported at the level of predicted short-horizon state and actuation effects, adaptive connectivity selection under O-DU degra
Public-safety Mission Critical Services (MCX) must be assured over finite incident horizons, yet they are increasingly carried over shared Open radio access networks (O-RAN) where ordinary mobile traffic competes for the same radio and transport resources. The authors identify a structural mismatch between the assurance question and the tools used to answer it. Existing RAN key performance indicators (KPIs) are retrospective: they summarize what has already happened rather than what remains executable within the remaining mission window. Conventional effective capacity (EC), meanwhile, characterizes an asymptotic stationary regime. As the abstract states, this asymptotic view "suppresses both the initial E2-observed condition and the finite selection-to-actuation transient." In an O-RAN deployment, that transient is not a detail. Heterogeneous E2 domains expose different observations, control actions, and actuation latencies, so the state observed at selection time and the state at which control actually takes effect can differ materially over a short incident horizon. The paper therefore asks how to characterize the service capability that is actually executable within a finite mission horizon, conditioned on what the E2 interface has observed, and how to use that characterization for orchestration at the Near-RT RAN Intelligent Controller (RIC). The contribution is an E2-conditioned finite-horizon effective capacity (FH-EC) framework for public-safety MCX over shared O-RAN, developed from a finite-horizon service model through an FH-EC formulation to confidence-calibrated capability profiles and an orchestration framework with contract certification, shared-resource protection, and FH-EC-driven profile selection.
The methodology proceeds in three linked stages. First, the authors establish a finite-horizon O-RAN MCX-based service model that incorporates three ingredients: E2-observed network states, control actuation latency, and correlation-aware connectivity diversity. The E2-observed state makes the model condition-dependent rather than stationary; actuation latency makes the selection-to-actuation transient explicit; and correlation-aware diversity prevents the model from treating multiple connectivity options as independent when their fading or congestion is correlated. Second, based on this model, the authors derive an FH-EC formulation that characterizes the executable service capability within a finite mission horizon. Conceptually, where asymptotic effective capacity takes a limit over an infinite horizon, FH-EC evaluates the capability over a horizon conditioned on an initial E2-observed state and an actuation delay :

Why it matters

The analysis reframes service assurance for public-safety MCX as a finite-horizon, condition-dependent problem rather than a retrospective or asymptotic one. Three implications follow. First, the E2 interface becomes a first-class input to assurance: because heterogeneous E2 domains expose different observations, control actions, and actuation latencies, the same nominal capability can correspond to different executable capabilities depending on which domain observed the state and how long actuation takes. Conditioning FH-EC on the E2-observed state and on actuation latency makes this explicit. Second, confidence calibration changes the orchestration problem from point prediction to risk-aware selection. A capability profile that carries a calibrated confidence allows the Near-RT RIC to select connectivity with an explicit notion of how much assurance is warranted, which is appropriate for mission-critical service where under-provisioning is costlier than over-provisioning. Third, the orchestration framework's contract certification and shared-resource protection address the multi-tenant reality of shared O-RAN. Contract certification ties selection to an enforceable commitment, while shared-resource protection encodes the non-MCX protection requirement so that MCX assurance does not come at the expense of ordinary traffic. The architecture can be summarized as follows:

The main limitation is scope: the evidence is simulation-based, using coupled MATLAB/ns-3 rather than a live shared O-RAN deployment, and the abstract reports no field trial. The taxonomy candidates for this work are Architecture, Network, and Cybersecurity, with Cryptography less directly implicated; the contract certification and protection mechanisms are the closest points of contact with security concerns. Overall, the framework's contribution is to make the finite mission horizon, the E2-observed initial condition, and the selection-to-actuation transient first-class quantities in MCX assurance over shared O-RAN.

Who should read this

CS practitioners and researchers

Opening member contentโ€ฆ