{"title":"Intrusion detection in internet of things using differential privacy: A hybrid machine learning approach","authors":"Ankit Manderna , Upasana Dohare , Sushil Kumar , Balak Ram","doi":"10.1016/j.adhoc.2025.103818","DOIUrl":null,"url":null,"abstract":"<div><div>With the rising integration of the Internet of Things (IoT) into our daily lives, ensuring the security and privacy of these interconnected systems has become paramount. Traditional cybersecurity approaches often fall short in addressing the dual challenges of protecting IoT systems from intrusions while preserving user privacy, particularly given the complexity of IoT data and increased concerns about data privacy. Therefore, a Machine Learning (ML) based model is implemented in the proposed system to accurately classify intrusions, tailored for IoT security. In this context, this paper proposes a hybrid machine learning model: Average Orthogonal Probabilistic Random Forest-Extreme Gradient Boosting (AOPRF-XGBoost) to classify the presence and absence of intrusions in the datasets while considering different data privacy budgets. The AOPRF-XGBoost model makes use of a proposed enhanced version of Random Forest: Average Orthogonal Probabilistic Random Forest and Extreme Gradient Boosting models. A key aspect of this work is the incorporation of differential privacy mechanisms to safeguard sensitive data during model training. Differential Privacy Datasets are created by adding Gaussian noise to the existing datasets: AFDA-IDS and UNR-IDD for the AOPRF-XGBoost model, ensuring privacy protection. The experimental results of the AOPRF-XGBoost model show that the model’s accuracy improves by 3% and 1% in AFDA-IDS dataset and UNR-IDD dataset respectively, over state-of-the-art existing machine learning models, achieving a balance between Security and Privacy.</div></div>","PeriodicalId":55555,"journal":{"name":"Ad Hoc Networks","volume":"174 ","pages":"Article 103818"},"PeriodicalIF":4.4000,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ad Hoc Networks","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1570870525000666","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}
引用次数: 0
Abstract
With the rising integration of the Internet of Things (IoT) into our daily lives, ensuring the security and privacy of these interconnected systems has become paramount. Traditional cybersecurity approaches often fall short in addressing the dual challenges of protecting IoT systems from intrusions while preserving user privacy, particularly given the complexity of IoT data and increased concerns about data privacy. Therefore, a Machine Learning (ML) based model is implemented in the proposed system to accurately classify intrusions, tailored for IoT security. In this context, this paper proposes a hybrid machine learning model: Average Orthogonal Probabilistic Random Forest-Extreme Gradient Boosting (AOPRF-XGBoost) to classify the presence and absence of intrusions in the datasets while considering different data privacy budgets. The AOPRF-XGBoost model makes use of a proposed enhanced version of Random Forest: Average Orthogonal Probabilistic Random Forest and Extreme Gradient Boosting models. A key aspect of this work is the incorporation of differential privacy mechanisms to safeguard sensitive data during model training. Differential Privacy Datasets are created by adding Gaussian noise to the existing datasets: AFDA-IDS and UNR-IDD for the AOPRF-XGBoost model, ensuring privacy protection. The experimental results of the AOPRF-XGBoost model show that the model’s accuracy improves by 3% and 1% in AFDA-IDS dataset and UNR-IDD dataset respectively, over state-of-the-art existing machine learning models, achieving a balance between Security and Privacy.
期刊介绍:
The Ad Hoc Networks is an international and archival journal providing a publication vehicle for complete coverage of all topics of interest to those involved in ad hoc and sensor networking areas. The Ad Hoc Networks considers original, high quality and unpublished contributions addressing all aspects of ad hoc and sensor networks. Specific areas of interest include, but are not limited to:
Mobile and Wireless Ad Hoc Networks
Sensor Networks
Wireless Local and Personal Area Networks
Home Networks
Ad Hoc Networks of Autonomous Intelligent Systems
Novel Architectures for Ad Hoc and Sensor Networks
Self-organizing Network Architectures and Protocols
Transport Layer Protocols
Routing protocols (unicast, multicast, geocast, etc.)
Media Access Control Techniques
Error Control Schemes
Power-Aware, Low-Power and Energy-Efficient Designs
Synchronization and Scheduling Issues
Mobility Management
Mobility-Tolerant Communication Protocols
Location Tracking and Location-based Services
Resource and Information Management
Security and Fault-Tolerance Issues
Hardware and Software Platforms, Systems, and Testbeds
Experimental and Prototype Results
Quality-of-Service Issues
Cross-Layer Interactions
Scalability Issues
Performance Analysis and Simulation of Protocols.