age of information

Existing studies on version-aware age of information (VAoI) focus solely on its time-average, neglecting the full distribution—a critical limitation for characterizing content freshness and timeliness. This paper bridges that gap by systematically characterizing the steady-state distribution of VAoI in both single-hop and multi-hop networks. Leveraging queueing theory and stochastic processes, we derive closed-form expressions for the steady-state distribution and mean of VAoI under randomized, uniform, and threshold-based scheduling policies. We further propose the first analytically tractable method for optimal threshold design and rigorously prove the optimality of threshold policies in minimizing the VAoI distribution in the stochastic dominance sense. Our framework enables joint, fine-grained modeling of content novelty and timeliness, establishing a novel theoretical foundation and practical design principles for communication network scheduling aimed at optimizing data freshness.

Traditional communication paradigms—focused on accuracy, throughput, and latency—fail to capture the intrinsic value of information in real-time networked control systems. Method: This paper proposes a task-semantic-value-driven communication framework. It introduces, for the first time, a systematic multi-dimensional semantic-awareness metric encompassing content, version, contextual relevance, and historical dependency, and establishes its mapping to goal-oriented communication design principles. Integrating Markov decision processes, Lyapunov optimization, semantic distortion modeling, and task-relevance quantification, the framework achieves a paradigm shift from Age of Information (AoI) to task-semantic value. Contribution/Results: The approach significantly improves communication efficiency, task reliability, and closed-loop control performance. It provides an analyzable, optimizable, and deployable theoretical foundation and design methodology for 6G semantic communication.

This paper addresses the challenge of minimizing long-term average Age of Information (AoI) in multi-source–destination wireless networks where base stations lack real-time AoI feedback and suffer from unreliable links. To tackle this, we first derive a fundamental lower bound on achievable AoI performance. We then propose an optimal randomized scheduling policy that operates without any AoI observations. Furthermore, we design a joint MMSE-based estimator for AoI and transmission delay, and construct the first Max-Weight freshness-aware scheduling framework leveraging estimated AoI values. We rigorously prove its asymptotic optimality. Simulations demonstrate that, even with zero AoI knowledge, our framework significantly outperforms existing state-of-the-art randomized policies—reducing average AoI by up to 32%. The methodology integrates stochastic optimization, renewal process modeling, minimum mean-square-error (MMSE) estimation, and AoI-theoretic analysis.

In vehicular networks (IoV), conventional Age of Information (AoI) modeling suffers from inaccuracy due to dynamic channel conditions and high vehicle mobility, particularly under Doppler-induced channel errors and deterministic service delays. Method: This paper proposes the first joint D/M/1 and M/M/1 queueing model to jointly capture error-prone channel behavior (caused by Doppler shift) and deterministic service latency; it further introduces an environment-aware online data sampling rate adaptation algorithm that relaxes the ideal-channel assumption. Contribution/Results: Theoretical analysis and simulations demonstrate that the proposed hybrid model better reflects real-world vehicular communications, with D/M/1 significantly outperforming M/M/1 in AoI characterization. Under typical high-speed scenarios, the mechanism reduces average system AoI by up to 18.7%, validating the critical role of deterministic service modeling and adaptive transmission rate control in enhancing information freshness.

Semantic communication for 6G must transcend Shannon’s bit-level fidelity paradigm to jointly model multidimensional semantic metrics—such as information value (VoI), age of information (AoI), and age of incorrect information (AoII)—while balancing correctness, timeliness, and utility. Method: We propose the first tensorized Goal-of-Transmission (GoT) model, enabling unified representation and composable optimization of heterogeneous semantic metrics. Integrating semantic information theory, tensor algebra, and deep learning, we design an end-to-end, goal-oriented architecture that tightly couples perception, transmission, and control. Contribution/Results: This work establishes the first systematic, scalable framework for goal-oriented semantic communication. It rigorously formalizes key challenges—including semantic abstraction, dynamic metric composition, and cross-layer optimization—and provides a theoretically grounded, implementation-ready pathway toward both theoretical advancement and practical deployment of semantic communication in 6G systems.

This work addresses the limitation of traditional Age of Information (AoI) metrics, which neglect the carbon footprint associated with status updates and thus fall short in supporting green communication design. For the first time, carbon emissions are integrated into the AoI optimization framework, yielding a carbon-aware real-time status updating model that jointly optimizes scheduling policies and signal-to-noise ratio (SNR) under dynamic carbon intensity constraints. Leveraging M/M/1 and M/M/1* queueing models, closed-form expressions for average AoI are derived under both fixed and time-varying carbon intensity scenarios. The analysis reveals a non-trivial trade-off between AoI and carbon emissions: minimizing AoI does not necessarily reduce carbon footprint, and dynamic carbon intensity significantly impacts achievable information freshness. These insights provide a theoretical foundation and practical guidance for designing low-carbon real-time communication systems.

This work addresses the limitations of existing information freshness metrics, such as Age of Information (AoI), which neglect physical dynamics and semantic accuracy, thereby inadequately capturing the timeliness of digital twin states. To overcome this, the paper introduces a novel metric termed Age of Staleness (AoS), which jointly incorporates semantic accuracy and temporal freshness to quantify how recently a digital twin last accurately reflected its physical counterpart. Leveraging a Markovian source model, the authors derive a closed-form expression for AoS in the single-source case and prove its monotonic decrease with increasing sampling rate. For multi-source scenarios under a total sampling-rate constraint, they formulate a multivariate non-convex optimization problem and employ monotonicity-based algorithms—such as the polyblock method—to achieve near-optimal sampling allocation. This approach substantially enhances both timeliness and fidelity in digital twin monitoring, effectively overcoming the semantic blindness inherent in conventional AoI-based frameworks.

This study addresses the optimization of status updates for remote navigation agents in integrated sensing and communication (ISAC) systems, aiming to balance information freshness against system overhead. By formulating a long-term cost model based on the Age of Information (AoI), the work incorporates the stochastic success probabilities and associated costs of both sensing and communication into a sequential decision-making framework defined over a two-dimensional AoI state space. The problem is modeled as a discounted infinite-horizon Markov decision process, and theoretical analysis reveals that the optimal stationary policy exhibits a monotone threshold structure, characterized by a non-decreasing switching curve—offering both interpretability and implementability. Numerical experiments validate the structural properties of the value function and optimal policy, demonstrating that an AoI-driven optimization objective can effectively guide ISAC system design.

This paper addresses the Age-of-Information (AoI) minimization problem in goal-oriented communication systems, jointly optimizing source sampling/processing costs and action execution error (CAE) constraints. To tackle the coupled challenges of dynamically evolving environmental states, semantic distortion costs, and unreliable transmission, we propose the first joint AoI–CAE optimization framework. We model the environment as a discrete-time Markov chain and integrate reliability-aware signal processing with unreliable channel transmission. Theoretical analysis reveals fundamental trade-offs among AoI, processing overhead, and CAE, leading to a static randomized policy with provable performance guarantees. Numerical experiments demonstrate that the proposed policy achieves near-optimal AoI across diverse parameter regimes, explicitly characterizing system feasibility boundaries and the structural properties of near-optimal policies.

This study addresses the lack of timeliness metrics sensitive to the evolution of information content in multi-hop IoT networks and the insufficient characterization of the distributional properties of Version Age of Information (VAoI) under feedback mechanisms. The authors propose a two-layer optimization framework that jointly designs a rate-constrained update policy at the source and a feedback-aware forwarding mechanism at intermediate nodes. For the first time in multi-hop settings, they derive the steady-state VAoI distribution and its analytical relationship with the update rate, leading to a closed-form optimal threshold policy. Both theoretical analysis and simulations demonstrate that the proposed approach significantly reduces communication overhead while effectively preserving data freshness and informational validity at the destination.