arXiv subjects

Dirk Pesch

Publications and source records attributed to Dirk Pesch.

At least 19 recordsLinked to original sources

arXiv2026

Scaling 5G-TSN Bridges: Operating Regimes, Scheduling, and Time Synchronisation Under Heterogeneous Industrial Traffic

3GPP Release 16 enables a 5G system to operate as a transparent IEEE 802.1 TSN bridge, but its scalability under heterogeneous industrial workloads remains insufficiently characterised. This paper uses the nascTime framework on OMNeT++/Simu5G to evaluate how many TSN endpoints a single 5G NR cell can bridge before per-flow QoS degrades. We model closed-loop control, machine vision, bulk telemetry, and IEEE 802.1AS traffic over a four-bearer SDAP architec- ture, varying the number of endpoints from 1 to 40, MAC scheduler, radio bandwidth (10 MHz and 20 MHz), and channel model. Results show three operating regimes. Below saturation, non-DRR schedulers perform similarly; near saturation, QoS- aware PF reduces critical-flow P99 latency by up to two or- ders of magnitude relative to channel-aware and fairness-based schedulers; and under overload, only QoS-PF maintains near- complete delivery for the highest-priority traffic. Across the two evaluated bandwidths, the saturation threshold approximately doubles when bandwidth doubles. We also show that isolating IEEE 802.1AS/gPTP traffic on a dedicated high-priority radio bearer reduces clock-servo instability, although endpoints carry- ing lower-priority data still experience elevated synchronisation delay under saturation because of reduced MAC scheduling frequency. Finally, the evaluated sub-6 GHz, 30 kHz-SCS con- figuration exhibits an effective latency floor of approximately 2.25 ms, indicating that sub-3 ms TSN deadlines may require radio-configuration changes such as configured grants or higher numerology

cs.NI↗

arXiv2026

EBGT: Epistemology-aided Bayesian Game Theory for Uplink Power Control in Stochastically Distributed IoT Tiers

Uplink power control in dense, heterogeneous Internet-of-Things (IoT) tiers is fundamentally limited by incomplete channel-state information (CSI) and mutual interference, while low size, weight, and power (SWaP) devices cannot afford the feedback and computation of conventional distributed schemes. This paper proposes EBGT, an epistemology-aided Bayesian game-theoretic framework for decentralized uplink power minimization in stochastically distributed IoT networks. Interfering users are modeled as spatially random through a Poisson point process (PPP), and each device reasons about its rivals through a two-layer belief hierarchy of inter-epistemic beliefs about opponents and intra-epistemic self-assessment, so that the transmit-power equilibrium is reached without repeated inter-node feedback. We derive a closed-form coverage-probability payoff via stochastic geometry and quantify belief convergence toward equilibrium using the Jensen--Shannon divergence (JSD) of the resulting power distributions. Monte-Carlo simulations validate the analytical coverage expressions and show that EBGT sustains the target coverage probability while reducing transmit power relative to fixed power control (FPC) and stochastic non-cooperative power control (SNCPC) baselines, particularly under stringent SINR and high-density regimes.

cs.GT↗

arXiv2026

Quality over Quantity: Value-Driven Distributed Congestion Control for the Collective Perception Service

While the Collective Perception Service (CPS) enables the exchange of sensor information among Intelligent Transport System Stations (ITS-S'), frequent transmission of Collective Perception Messages (CPMs), their highly variable size, and load from other vehicular services can cause severe channel congestion. Existing Distributed Congestion Control (DCC) Access layer mechanisms typically regulate channel load without considering the relative importance of the objects carried in CPMs. This limits their ability to preserve high-value information under constrained radio resources. More recently, Facilities layer DCC mechanisms attempt to prioritise high value objects within the specified radio resource limits but may not operate well in heterogeneous environments where the number of sensed objects and their importance can vary significantly over time or between ITS-S'. This paper proposes a value-based DCC Facilities layer 'quality' selector that couples a Value of Information (VoI) per bit rate controller with object-level selection. It is benchmarked against state of the art approaches from standards and the literature, with results showing that the proposed method maintains channel load near the target CBR while retaining more high-VoI objects than state of the art approaches, thereby improving the dissemination of perception-critical information.

cs.NI↗

arXiv2026

Behaviour-Conditioned Neural Processes for Adaptive Residential Short-Term Load Forecasting

Residential short-term load forecasting (STLF) is challenging because household demand is heterogeneous, temporally variable, and shaped by diverse behavioural routines. This work investigates whether inferred behavioural structure can be embedded within the forecasting mechanism of a Neural Process-based probabilistic model, rather than used only as an external grouping signal, for context-conditioned residential STLF. We propose a behaviour-conditioned Attentive Neural Process framework that treats each load profile as a forecasting task. Behavioural structure is represented by a discrete latent variable inferred from the available context and used for behaviour-conditioned decoder conditioning, while a continuous latent variable captures shared functional uncertainty across heterogeneous profiles. To enable conditioning without ground-truth behavioural labels, clustering-derived information provides weak supervision during training, whereas test-time conditioning relies only on context-inferred class distributions. Experiments on the Smart Grid, Smart City (SGSC) dataset use user-disjoint train/validation/test splits, variable context lengths, and multi-step forecast horizons, with comparisons against a label-agnostic ANP baseline and fixed-window deterministic STLF baselines. The proposed variants improve MAE and CRPS over ANP across horizons and context settings, with the largest gains under limited context. The best-performing variant achieves average reductions of 7.9% in MAE and 6.9% in CRPS relative to ANP. Compared with fixed-window baselines, this variant achieves lower RMSE across all evaluated horizons while maintaining competitive MAE, suggesting fewer large prediction deviations under heterogeneous consumption patterns. These results support single-model, uncertainty-aware forecasting across heterogeneous households, contexts, and horizons.

cs.LG↗

arXiv2026

Lyapunov Optimization based Queue-aware Traffic Shaping for 5G-TSN in Industrial Environments

Manufacturing companies look increasingly at Private 5G networks to manage Automated Guided Vehicles (AGVs). While 5G promises Ultra-Reliable Low Latency Communication (URLLC), its service quality is challenged by industrial environments characterized by dense metallic structures, which frequently cause line-of-sight (LOS) blockage events, causing deep fades in received signal levels that can degrade channel capacity to near-zero. Standard transport protocols and rate adaptation mechanisms fail to react sufficiently fast to these deep fades, resulting in bufferbloat and latency spikes that violate safety margins. In this paper, we propose a cross-layer rate control algorithm based on Lyapunov Drift-plus-Penalty theory. The proposed controller dynamically optimizes the trade-off between service utility and queue stability based on instantaneous buffer states, without requiring predictive channel models. We validate the approach using a trace-driven simulation framework that replicates the stochastic dynamics of 5G blockage using 3GPP-compliant capacity data. Numerical results demonstrate that while baseline scheduling schemes suffer from catastrophic queue accumulation, leading to excessive delays upon reconnection, the proposed Lyapunov controller effectively eliminates bufferbloat. By preventing congestion-induced backlog, the system ensures immediate low-latency operation as soon as the channel recovers, maintaining near-deterministic behavior.

cs.NI↗

arXiv2026

nascTime: A Full-Stack 5G-TSN Bridge Simulation Framework with SDAP-Based QoS Mapping and IEEE 802.1AS Transparent Clock

3GPP Release~16 specifies how a 5G system can operate as a transparent IEEE~802.1 TSN bridge, yet no existing simulation framework implements the complete bridge architecture with end-to-end QoS mapping through the SDAP layer, per-flow Data Radio Bearer selection, and IEEE~802.1AS transparent clock behaviour with measured residence time. Existing tools model either QoS mapping without time synchronisation, or time synchronisation without a data plane. This paper presents nascTime, a simulation framework built on OMNeT++~6.3, INET~4.6, and Simu5G that implements the full 3GPP 5G-TSN bridge model. The NW-TT and DS-TT are realised as modular compound modules that integrate with INET's \texttt{LayeredEthernetInterface} and streaming PHY. QoS mapping traverses the complete PCP\,$\rightarrow$\,DSCP\,$\rightarrow$\,QFI\,$\rightarrow$\,SDAP/DRB pipeline, and gPTP frames are transported through the simulated 5G radio path via L2-in-GTP-U encapsulation with per-message residence-time correction. We validate the framework with a three-endpoint factory topology under both ideal and fading channel conditions. In the ideal scenario, high-priority traffic achieves 99.9\% delivery with a mean end-to-end delay of 2.58\,ms, while the measured 5GS residence time exhibits a variance below 0.2\,$\mu$s. Under a fading channel, residence-time variance increases to 48\,$\mu$s, confirming that the framework captures radio-induced timing effects absent from abstract-delay simulators. nascTime is publicly available and constitutes the first full-stack 5G-TSN bridge simulation with SDAP-based QoS differentiation and measured IEEE~802.1AS transparent clock behaviour.

cs.NI↗

arXiv2026

A Framework for Hybrid Collective Inference in Distributed Sensor Networks

With the ever-increasing range of applications of Internet in Things (IoT) and sensor networks, challenges are emerging in various categories of classification tasks. Applications such as vehicular networking, UAV swarm coordination and cyber-physical systems require global classification over distributed sensors, with tight constraints on communication and computation resources. There has been much research in decentralized and distributed data-exchange for communication-efficient collective inference. Likewise, there has been considerable research involving the use of cloud and edge computing paradigms for efficient task allocation. To the best of our knowledge, there has been no research on the integration of these two concepts to create a hybrid cloud and distributed approach that makes dynamic runtime communication strategy decisions. In this paper, we focus on aspects of combining distributed and hierarchical communication and classification approaches for collective inference. We derive optimal policies for agents that implement this hybrid approach, and evaluate their performance under various scenarios of the distribution of underlying data. Our analysis shows that this approach can maintain a high level of classification accuracy (comparable to that of centralised joint inference over all data), at reduced theoretical communication cost. We expect there is potential for our approach to facilitate efficient collective inference for real-world applications, including instances that involves more complex underlying data distributions.

cs.DC↗

arXiv2025

Epistemology-Inspired Bayesian Games for Distributed IoT Uplink Power Control

Massive number of simultaneous Internet of Things (IoT) uplinks strain gateways with interference and energy limits, yet devices often lack neighbors' Channel State Information (CSI) and cannot sustain centralized Mobile Edge Computing (MEC) or heavy Machine Learning (ML) coordination. Classical Bayesian solvers help with uncertainty but become intractable as users and strategies grow, making lightweight, distributed control essential. In this paper, we introduce the first-ever, novel epistemic Bayesian game for uplink power control under incomplete CSI that operates while suppressing interference among multiple uplink channels from distributed IoT devices firing at the same time. Nodes run inter-/intra-epistemic belief updates over opponents' strategies, replacing exhaustive expected-utility tables with conditional belief hierarchies. Using an exponential-Gamma SINR model and higher-order utility moments (variance, skewness, kurtosis), the scheme remains computationally lean with a single-round upper bound of $O\!\left(N^{2} S^{2N}\right)$. Precise power control and stronger coverage amid realistic interference: with channel magnitude equal to $1$ and a signal-to-interference-plus-noise ratio (SINR) threshold of $-18$ dB, coverage reaches approximately $60\%$ at approximately $55\%$ of the maximum transmit power; mid-rate devices with a threshold of $-27$ dB achieve full coverage with less than $0.1\%$ of the maximum transmit power.Under $80\%$ interference, a fourth-moment policy cuts average power from approximately $52\%$ to approximately $20\%$ of the maximum transmit power with comparable outage, outperforming expectation-only baselines. These results highlight a principled, computationally lean path to optimal power allocation and higher network coverage under real-world uncertainty within dense, distributed IoT networks.

eess.SY↗

arXiv2025

QoS-Aware Proportional Fairness Scheduling for Multi-Flow 5G UEs: A Smart Factory Perspective

Private 5G networks are emerging as key enablers for smart factories, where a single device often handles multiple concurrent traffic flows with distinct Quality of Service (QoS) requirements. Existing simulation frameworks, however, lack the fidelity to model such multi-flow behavior at the QoS Flow Identifier (QFI) level. This paper addresses this gap by extending Simu5G to support per-QFI modeling and by introducing a novel QoS-aware Proportional Fairness (QoS-PF) scheduler. The scheduler dynamically balances delay, Guaranteed Bit Rate (GBR), and priority metrics to optimize resource allocation across heterogeneous flows. We evaluate the proposed approach in a realistic smart factory scenario featuring edge-hosted machine vision, real-time control loops, and bulk data transfer. Results show that QoS-PF improves deadline adherence and fairness without compromising throughput. All extensions are implemented in a modular and open-source manner to support future research. Our work provides both a methodological and architectural foundation for simulating and analyzing advanced QoS policies in industrial 5G deployments.

cs.NI↗

arXiv2025

SDAP-based QoS Flow Multiplexing Support in Simu5G for 5G NR Simulation

The Service Data Adaptation Protocol (SDAP) plays a central role in 5G New Radio (NR), acting as a bridge between the core and radio networks, by enabling QoS Flow multiplexing over shared Data Radio Bearers (DRBs). However, most 5G simulation frameworks, including the popular OMNet++-based Simu5G, lack SDAP support, limiting their ability to model realistic QoS behavior. This paper presents a modular, standardscompliant SDAP extension for Simu5G. The implementation includes core elements such as QoS Flow Identifer (QFI) flow tagging, SDAP header insertion/removal, and configurable logical DRB mapping. The proposed design supports multi-QFI simulation scenarios and enables researchers to model differentiated QoS flows and flowaware scheduling policies. Validation results confirm correct SDAP behavior and pave the way for advanced 5G simulations involving per-flow isolation, latency-sensitive traffic, and industrial QoS profiles.

cs.NI↗

arXiv2025

Resilient Time-Sensitive Networking for Industrial IoT: Configuration and Fault-Tolerance Evaluation

Time-Sensitive Networking (TSN) is increasingly adopted in industrial systems to meet strict latency, jitter, and reliability requirements. However, evaluating TSN's fault tolerance under realistic failure conditions remains challenging. This paper presents IN2C, a modular OMNeT++/INET-based simulation framework that models two synchronized production cells connected to centralized infrastructure. IN2C integrates core TSN features, including time synchronization, traffic shaping, per-stream filtering, and Frame Replication and Elimination for Redundancy (FRER), alongside XML-driven fault injection for link and node failures. Four fault scenarios are evaluated to compare TSN performance with and without redundancy. Results show that FRER eliminates packet loss and achieves submillisecond recovery, though with 2-3x higher link utilization. These findings offer practical guidance for deploying TSN in bandwidth-constrained industrial environments.

cs.NI↗

arXiv2025

Comparative Performance Evaluation of 5G-TSN Applications in Indoor Factory Environments

While Time-Sensitive Networking (TSN) enhances the determinism, real-time capabilities, and reliability of Ethernet, future industrial networks will not only use wired but increasingly wireless communications. Wireless networks enable mobility, have lower costs, and are easier to deploy. However, for many industrial applications, wired connections remain the preferred choice, particularly those requiring strict latency bounds and ultra-reliable data flows, such as for controlling machinery or managing power electronics. The emergence of 5G, with its Ultra-Reliable Low-Latency Communication (URLLC) promises to enable high data rates, ultra-low latency, and minimal jitter, presenting a new opportunity for wireless industrial networks. However, as 5G networks include wired links from the base station towards the core network, a combination of 5G with time-sensitive networking is needed to guarantee stringent QoS requirements. In this paper, we evaluate 5G-TSN performance for different indoor factory applications and environments through simulations. Our findings demonstrate that 5G-TSN can address latency-sensitive scenarios in indoor factory environments.

cs.NI↗

arXiv2025

Scalability Analysis of 5G-TSN Applications in Indoor Factory Settings

While technologies such as Time-Sensitive Networking (TSN) improve deterministic behaviour, real-time functionality, and robustness of Ethernet, future industrial networks aim to be increasingly wireless. While wireless networks facilitate mobility, reduce cost, and simplify deployment, they do not always provide stringent latency constraints and highly dependable data transmission as required by many manufacturing systems. The advent of 5G, with its Ultra-Reliable Low-Latency Communication (URLLC) capabilities, offers potential for wireless industrial networks. 5G offers elevated data throughput, very low latency, and negligible jitter. As 5G networks typically include wired connections from the base station to the core network, integration of 5G with time-sensitive networking is essential to provide rigorous QoS standards. This paper assesses the scalability of 5G-TSN for various indoor factory applications and conditions using OMNET++ simulation. Our research shows that 5G-TSN has the potential to provide bounded delay for latency-sensitive applications in scalable indoor factory settings.

cs.NI↗

arXiv2025

A Survey and Tutorial of Redundancy Mitigation for Vehicular Cooperative Perception: Standards, Strategies and Open Issues

This paper provides an in-depth review and discussion of the state of the art in redundancy mitigation for the vehicular Collective Perception Service (CPS). We focus on the evolutionary differences between the redundancy mitigation rules proposed in 2019 in ETSI TR 103 562 versus the 2023 technical specification ETSI TS 103 324, which uses a Value of Information (VoI) based mitigation approach. We also critically analyse the academic literature that has sought to quantify the communication challenges posed by the CPS and present a unique taxonomy of the redundancy mitigation approaches proposed using three distinct classifications: object inclusion filtering, data format optimisation, and frequency management. Finally, this paper identifies open research challenges that must be adequately investigated to satisfactorily deploy CPS redundancy mitigation measures. Our critical and comprehensive evaluation serves as a point of reference for those undertaking research in this area.

cs.NI↗

arXiv2024

Auction-based Adaptive Resource Allocation Optimization in Dense and Heterogeneous IoT Networks

Efficient and reliable resource allocation within densely-deployed massive IoT networks remains a key challenge due to resource constraints among low-size, weight, and power (SWaP) IoT devices and within the network and limitations of conventional centralized methods under incomplete information. We propose a novel auction-based framework for adaptive resource allocation, combining space-time-frequency spreading (STFS) techniques with Bayesian Game approaches. We introduce novel modified Simultaneous Ascending Auction (mSAA) mechanism tailored to densely-deployed and low-complexity IoT networks, enabling distributed computation and reduced power consumption. By incorporating Bayesian game-based bidding strategies and optimizing dispersion matrices for signal transmission, the proposed approach ensures enhanced channel throughput and energy efficiency. Comparative analysis against traditional auction types, including First-Price and Second-Price Sealed-Bid Auctions, as well as the Vickery-Clarke-Groves (VCG) mechanism, demonstrates the superiority of mSAA in terms of surplus maximization, revenue efficiency, and robustness in risk-prone bidding environments. Simulation results validate the model's adaptability to heterogeneous IoT nodes and its potential for dense deployment across different environments and verticals.

cs.GT↗

arXiv2024

Semantic Vehicle-to-Everything (V2X) Communications Towards 6G

Semantic Communication (SEM-COM) has emerged as one of the disruptive technologies facilitating the evolution towards sixth-generation (6G) wireless networks. This article presents the potential of SEM-COM to transform Vehicle-to-Everything (V2X) communications, with a particular emphasis on its ability to enhance communication efficiency and intelligence. We discuss the core components and metrics that characterize SEM-COM, providing insights into its operational framework within the context of V2X communications. We illustrate the applicability and practicality of SEM-COM through real-world vehicular use cases, demonstrate the potential of SEM-COM to enhance aspects of intelligent mobility, such as communication efficiency and decision-making. Finally, the article identifies key open research questions for SEM-COM V2X, pointing to areas that require further exploration and thus setting a foundation for future work in this evolving domain.

cs.NI↗

arXiv2024

Unmasking the Nuances of Loneliness: Using Digital Biomarkers to Understand Social and Emotional Loneliness in College Students

Background: Loneliness among students is increasing across the world, with potential consequences for mental health and academic success. To address this growing problem, accurate methods of detection are needed to identify loneliness and to differentiate social and emotional loneliness so that intervention can be personalized to individual need. Passive sensing technology provides a unique technique to capture behavioral patterns linked with distinct loneliness forms, allowing for more nuanced understanding and interventions for loneliness. Methods: To differentiate between social and emotional loneliness using digital biomarkers, our study included statistical tests, machine learning for predictive modeling, and SHAP values for feature importance analysis, revealing important factors in loneliness classification. Results: Our analysis revealed significant behavioral differences between socially and emotionally lonely groups, particularly in terms of phone usage and location-based features , with machine learning models demonstrating substantial predictive power in classifying loneliness levels. The XGBoost model, in particular, showed high accuracy and was effective in identifying key digital biomarkers, including phone usage duration and location-based features, as significant predictors of loneliness categories. Conclusion: This study underscores the potential of passive sensing data, combined with machine learning techniques, to provide insights into the behavioral manifestations of social and emotional loneliness among students. The identification of key digital biomarkers paves the way for targeted interventions aimed at mitigating loneliness in this population.

cs.HC↗

arXiv2024

Evolving AI for Wellness: Dynamic and Personalized Real-time Loneliness Detection Using Passive Sensing

Loneliness is a growing health concern as it can lead to depression and other associated mental health problems for people who experience feelings of loneliness over prolonged periods of time. Utilizing passive sensing methods that use smartphone and wearable sensor data to capture daily behavioural patterns offers a promising approach for the early detection of loneliness. Given the subjective nature of loneliness and people's varying daily routines, past detection approaches using machine learning models often face challenges with effectively detecting loneliness. This paper proposes a methodologically novel approach, particularly developing a loneliness detection system that evolves over time, adapts to new data, and provides real-time detection. Our study utilized the Globem dataset, a comprehensive collection of passive sensing data acquired over 10 weeks from university students. The base of our approach is the continuous identification and refinement of similar behavioural groups among students using an incremental clustering method. As we add new data, the model improves based on changing behavioural patterns. Parallel to this, we create and update classification models to detect loneliness among the evolving behavioural groups of students. When unique behavioural patterns are observed among student data, specialized classification models have been created. For predictions of loneliness, a collaborative effort between the generalized and specialized models is employed, treating each prediction as a vote. This study's findings reveal that group-based loneliness detection models exhibit superior performance compared to generic models, underscoring the necessity for more personalized approaches tailored to specific behavioural patterns. These results pave the way for future research, emphasizing the development of finely-tuned, individualized mental health interventions.

cs.HC↗