RL-ASL: A Dynamic Listening Optimization for TSCH Networks Using Reinforcement Learning

F. Fernando Jurado-Lasso; J. F. Jurado

cs.NIApr 8, 2026

RL-ASL: A Dynamic Listening Optimization for TSCH Networks Using Reinforcement Learning

Authors: F. Fernando Jurado-Lasso, J. F. Jurado

Organizations: independent researcher, based in Cali, Colombia · Department of Basic Science, Faculty of Engineering and Administration, Universidad Nacional de Colombia Sede Palmira, Palmira 763531, Colombia

Abstract

Time Slotted Channel Hopping (TSCH) is a widely adopted Media Access Control (MAC) protocol within the IEEE 802.15.4e standard, designed to provide reliable and energy-efficient communication in Industrial Internet of Things (IIoT) networks. However, state-of-the-art TSCH schedulers rely on static slot allocations, resulting in idle listening and unnecessary power consumption under dynamic traffic conditions. This paper introduces RL-ASL, a reinforcement learning-driven adaptive listening framework that dynamically decides whether to activate or skip a scheduled listening slot based on real-time network conditions. By integrating learning-based slot skipping with standard TSCH scheduling, RL-ASL reduces idle listening while preserving synchronization and delivery reliability. Experimental results on the FIT IoT-LAB testbed and Cooja network simulator show that RL-ASL achieves up to 46% lower power consumption than baseline scheduling protocols, while maintaining near-perfect reliability and reducing average latency by up to 96% compared to PRIL-M. Its link-based variant, RL-ASL-LB, further improves delay performance under high contention with similar energy efficiency. Importantly, RL-ASL performs inference on constrained motes with negligible overhead, as model training is fully performed offline. Overall, RL-ASL provides a practical, scalable, and energy-aware scheduling mechanism for next-generation low-power IIoT networks.

Figures & tables

Fig. 1: Network model for a TSCH network with five nodes.

Panel of Figure 2 — Fig. 2: Experimental network topologies from Strasbourg IoT-LAB site: (a) Simple 5-node topology used for training and validation, (b) Larger star topology, and (c) Deployment of the star topology in the FIT IoT-LAB.

1 [respect-blocks] \RowStyle Pattern	Jittered	Transmission intervals per node ID
High Traffic	✓	All nodes: 13 s
\Block 4-1 Heterogeneous Traffic	\Block 4-1✓	IDs 3, 11, 15, 19: 17 s
		IDs 4, 12, 16, 20: 30 s
		IDs 5, 6, 13, 17, 21: 50 s
		IDs 4, 18, 22: 73 s
Sparse Traffic	✓	Alternating IDs: 60 or 73 s

TABLE I: Traffic generation patterns and per-node Tx intervals.

1 [respect-blocks] \RowStyle Parameter	Value	Parameter	Value
Episode length ( $T_{e}$ )	500	Discount factor ( $\gamma$ )	0.9
Learning rate ( $\alpha$ )	0.15	Initial exploration ( $\varepsilon_{0}$ )	1.0
Min. exploration ( $\varepsilon_{\min}$ )	0.05	Decay factor ( $\eta_{\varepsilon}$ )	0.997
Reward success ( $R_{\mathrm{succ}}$ )	+1.0	Skip reward ( $R_{\mathrm{skip}}$ )	+0.5
Idle cost ( $C_{\mathrm{idle}}$ )	–0.5	Miss penalty ( $C_{\mathrm{miss}}$ )	–1.0
Terminal reward (success)	+5.0	Terminal penalty (failure)	–5.0

TABLE II: RL-ASL Q-learning configuration parameters.

Fig. 3: Convergence of average reward during Q-learning training in the relay topology for different learning rates

\alpha

1 [respect-blocks] \RowStyle $R_{\mathrm{skip}}$	Latency [ms]	PDR [%]	RDC [%]
0.25	202.37	99.91	1.35
0.5	196.38	99.97	1.32
0.75	198.70	99.97	1.33

TABLE III: Sensitivity of RL-ASL to the reward parameter

R_{\mathrm{skip}}

in the simple topology.

Panel of Figure 7 — Fig. 4: Comparison of PDR across different traffic patterns for the simple topology : (a-d) PDR under high, heterogeneous, sparse, and periodic traffic patterns.

Panel of Figure 8 — Fig. 5: Latency distributions across different traffic patterns for the simple topology : (a-d) Latency under high, heterogeneous, sparse, and periodic traffic patterns.

Panel of Figure 9 — Fig. 6: Power consumption breakdown across different traffic patterns for the simple topology : (a-d) Power consumption under high, heterogeneous, sparse, and periodic traffic patterns.

Panel of Figure 10 — Fig. 7: Trade-off comparison across different traffic patterns for the simple topology : (a-d) Overall performance under high, heterogeneous, sparse, and periodic traffic patterns. The metrics are normalized for comparison.

Panel of Figure 11 — Fig. 8: Comparison of PDR across different traffic patterns for the Star topology : (a-d) PDR under high, heterogeneous, sparse, and periodic traffic patterns.

Panel of Figure 12 — Fig. 9: Latency distributions across different traffic patterns for the Star topology : (a-d) Latency under high, heterogeneous, sparse, and periodic traffic patterns.

Panel of Figure 13 — Fig. 10: Power consumption across different traffic patterns for the Star topology : (a-d) Power consumption under high, heterogeneous, sparse, and periodic traffic patterns.

Panel of Figure 14 — Fig. 11: Trade-off comparison across different traffic patterns for the Star topology : (a-d) Overall performance under high, heterogeneous, sparse, and periodic traffic patterns. The metrics are normalized for comparison.

1 [respect-blocks] \RowStyle Protocol	Simple [mW]	LT [days]	Star [mW]	LT [days]
Orchestra	0.996	99	1.008	98
Orchestra-LB	1.419	70	1.453	68
PRIL-M	0.529	185	0.751	131
RL-ASL	0.561	174	0.647	152
RL-ASL-LB	0.789	124	0.856	114

TABLE IV: Average Power and Estimated Lifetime (3 V, 220 mAh Battery)

1 [respect-blocks] \RowStyle Protocol	PDR%	Radio Duty Cycle [%]	Latency [ms]
RL-ASL	93.1	2.8	186.97
RPL Baseline	96.7	3.4	184.70

TABLE V: Impact of Moderate Mobility on RL-ASL Performance

Industrial Internet of Things

See all 16 papers

DeFRiS: Silo-Cooperative IoT Applications Scheduling via Decentralized Federated Reinforcement Learning
Mar 16, 2026Zhiyu Wang, Mohammad Goudarzi, Mingming Gong +1Federated RL Non-IID Federated Learning
The Quantum Cloud Computing and Distributed Systems (qCLOUDS) Laboratory, School of Computing and Information Systems, The University of Melbourne, Australia · The Faculty of Information Technology, Monash University, Australia · School of Mathematics and Statistics, The University of Melbourne, Australia
Harnessing Large Language Models for Intelligent Resource Allocation in the Internet of Everything
Jul 29, 2026Haijun Zhang, Zhuojun Duan, Zijun Wu +2Large Language Model-Guided Optimization Internet of Things
Hebei Key Laboratory of Space-Air-Ground Intelligent Communication and Beijing Engineering and Technology Research Center for Convergence Networks and Ubiquitous Services, University of Science and Technology Beijing, Beijing, China
Auction-Based RIS Allocation With DRL: Controlling the Cost-Performance Trade-Off
Feb 18, 2026Martin Mark Zan, Stefan SchwarzWireless Resource Allocation Reconfigurable Intelligent Surfaces
Institute of Telecommunications, TU Wien, Vienna, Austria

Mar 16, 2026J/K move · Enter open · S save

DeFRiS: Silo-Cooperative IoT Applications Scheduling via Decentralized Federated Reinforcement Learning

Zhiyu Wang, Mohammad Goudarzi, Mingming Gong, Rajkumar Buyya

The Quantum Cloud Computing and Distributed Systems (qCLOUDS) Laboratory, School of Computing and Information Systems, The University of Melbourne, Australia · The Faculty of Information Technology, Monash University, Australia · School of Mathematics and Statistics, The University of Melbourne, Australia

Federated RL Non-IID Federated Learning

Open paper AskPDF arXiv

Next-generation IoT applications increasingly span across autonomous administrative entities, necessitating silo-cooperative scheduling to leverage diverse computational resources while preserving data privacy. However, realizing efficient cooperation faces significant challenges arising from infrastructure heterogeneity, Non-IID workload shifts, and the inherent risks of adversarial environments. Existing approaches, relying predominantly on centralized coordination or independent learning, fail to address the incompatibility of state-action spaces across heterogeneous silos and lack robustness against malicious attacks. This paper proposes DeFRiS, a Decentralized Federated Reinforcement Learning framework for robust and scalable Silo-cooperative IoT application scheduling. DeFRiS integrates three synergistic innovations: (i) an action-space-agnostic policy utilizing candidate resource scoring to enable seamless knowledge transfer across heterogeneous silos; (ii) a silo-optimized local learning mechanism combining Generalized Advantage Estimation (GAE) with clipped policy updates to resolve sparse delayed reward challenges; and (iii) a Dual-Track Non-IID robust decentralized aggregation protocol leveraging gradient fingerprints for similarity-aware knowledge transfer and anomaly detection, and gradient tracking for optimization momentum. Extensive experiments on a distributed testbed with 20 heterogeneous silos and realistic IoT workloads demonstrate that DeFRiS significantly outperforms state-of-the-art baselines, reducing average response time by 6.4% and energy consumption by 7.2%, while lowering tail latency risk (CVaR

_{0.95}

) by 10.4% and achieving near-zero deadline violations. Furthermore, DeFRiS achieves over 3 times better performance retention as the system scales and over 8 times better stability in adversarial environments compared to the best-performing baseline.

Abstract

Industrial Internet of Things

Explore similar work

DeFRiS: Silo-Cooperative IoT Applications Scheduling via Decentralized Federated Reinforcement Learning

Harnessing Large Language Models for Intelligent Resource Allocation in the Internet of Everything

Auction-Based RIS Allocation With DRL: Controlling the Cost-Performance Trade-Off