Enabling High-Throughput and Cost-Efficient Web 3.0 Services for Intelligent Sensing Networks via GNN-Enhanced Masked Deep Reinforcement Learning in Multi-Chain Systems

Intelligent sensing networks generate heterogeneous application-level service requests from aggregated sensing measurements and events, raising critical requirements for data integrity, traceability, and trustworthy service interactions. Blockchain-enabled Web 3.0 infrastructures provide decentralized trust, tamper-evident records, and verifiable service execution to support these requirements. However, the heterogeneous characteristics of blockchain platforms, coupled with complex smart-contract dependencies and diverse sensing-service requirements, introduce significant challenges in achieving cost-efficient and low-latency service deployment. To address these challenges, this paper investigates dependency-aware smart-contract orchestration across heterogeneous multi-chain systems at the blockchain service layer of intelligent sensing networks. Specifically, we formulate the smart-contract deployment problem as a constrained optimization problem that minimizes the weighted financial and time costs while satisfying transaction throughput, execution resource, security, and application latency constraints. To solve this problem, we propose a focus-aware graph neural network (GNN)-enhanced Masked Proximal Policy Optimization (PPO) framework. The proposed Focus-GNN employs edge-aware multi-head graph attention to capture smart-contract dependencies and incorporates a focus-context fusion mechanism to integrate the target contract representation, global graph features, and blockchain resource states. Furthermore, action masking eliminates infeasible deployment choices, enabling constraint-aware sequential decision-making. We evaluate the proposed framework using heterogeneous blockchain configurations and synthetic workloads representing anomaly warning, sensing data archiving, and environmental monitoring services. Experimental results demonstrate that the proposed framework achieves improved training performance and lower deployment overhead, highlighting its potential for efficient smart-contract orchestration in blockchain-enabled intelligent sensing networks.

Authors

Institutions

Publication Details

Journal
Sensors
Published
2026-11-13
DOI
https://doi.org/10.3390/s26227174
Primary Topic
Blockchain Technology Applications and Security
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Enabling High-Throughput and Cost-Efficient Web 3.0 Services for Intelligent Sensing Networks via GNN-Enhanced Masked Deep Reinforcement Learning in Multi-Chain Systems

霍如, Tao Huang, Wei Chen, Xuepeng Shi et al.
Sensors
Blockchain Technology Applications and Security
article

Enabling High-Throughput and Cost-Efficient Web 3.0 Services for Intelligent Sensing Networks via GNN-Enhanced Masked Deep Reinforcement Learning in Multi-Chain Systems

霍如, Tao Huang, Wei Chen, Xuepeng Shi, Yang Liu
article en

Abstract

Intelligent sensing networks generate heterogeneous application-level service requests from aggregated sensing measurements and events, raising critical requirements for data integrity, traceability, and trustworthy service interactions. Blockchain-enabled Web 3.0 infrastructures provide decentralized trust, tamper-evident records, and verifiable service execution to support these requirements. However, the heterogeneous characteristics of blockchain platforms, coupled with complex smart-contract dependencies and diverse sensing-service requirements, introduce significant challenges in achieving cost-efficient and low-latency service deployment. To address these challenges, this paper investigates dependency-aware smart-contract orchestration across heterogeneous multi-chain systems at the blockchain service layer of intelligent sensing networks. Specifically, we formulate the smart-contract deployment problem as a constrained optimization problem that minimizes the weighted financial and time costs while satisfying transaction throughput, execution resource, security, and application latency constraints. To solve this problem, we propose a focus-aware graph neural network (GNN)-enhanced Masked Proximal Policy Optimization (PPO) framework. The proposed Focus-GNN employs edge-aware multi-head graph attention to capture smart-contract dependencies and incorporates a focus-context fusion mechanism to integrate the target contract representation, global graph features, and blockchain resource states. Furthermore, action masking eliminates infeasible deployment choices, enabling constraint-aware sequential decision-making. We evaluate the proposed framework using heterogeneous blockchain configurations and synthetic workloads representing anomaly warning, sensing data archiving, and environmental monitoring services. Experimental results demonstrate that the proposed framework achieves improved training performance and lower deployment overhead, highlighting its potential for efficient smart-contract orchestration in blockchain-enabled intelligent sensing networks.

SensorsVol. 26(22)
Beijing University of Posts and Telecommunications (CN), Beijing University of Technology (CN), China Academy of Information and Communications Technology (CN), Purple Mountain Laboratories (CN), State Key Laboratory of Networking and Switching Technology (CN), Beihang University (CN)
National Key Research and Development Program of China
Industry, innovation and infrastructure
Blockchain Technology Applications and Security
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon