Blockchain-Based Smart Monitoring Framework for Defense Industry

被引:1
作者
Alqahtani, Abdullah [1 ]
Alsubai, Shtwai [1 ]
Alanazi, Abed [1 ]
Bhatia, Munish [2 ]
机构
[1] Prince Sattam Bin Abdulaziz Univ, Coll Comp Engn & Sci, Dept Comp Sci, Al Kharj 16278, Saudi Arabia
[2] Natl Inst Technol Kurukshetra, Dept Comp Applicat, Kurukshetra 136119, India
来源
IEEE ACCESS | 2024年 / 12卷
关键词
Monitoring; Personnel; Data models; Real-time systems; Internet of Things; Cloud computing; Sensors; Blockchains; US Department of Defense; Digital twins; Surveillance; Smart cameras; Blockchain; defense industry; digital twin; smart surveillance; DIGITAL TWIN; SURVEILLANCE; WORKOUTS;
D O I
10.1109/ACCESS.2024.3421573
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Internet of Things (IoT) technology has been widely adopted across various industries for remote decision-making, monitoring, and surveillance. The proliferation of IoT applications in sensitive sectors, such as national defense and security, has been driven by the ability to obtain in-depth information on ubiquitous occurrences. Conspicuously, the current research presents a comprehensive framework based on the IoT-empowered Digital Twin technology for assessing the national integrity of defense personnel. The primary objective is to identify the integral behavior of military personnel by securely tracking everyday activities. The proposed method demonstrated the ability to accurately analyze an individual's anomalous occurrences in activities using a hybrid Convolution Neural Network with Gated Recurrent Units. Moreover, each personnel is mapped using a secure blockchain platform for acquiring social interactions and activities to identify potential threats to national security. The proposed model has been validated using challenging date sets obtained from public repositories. The computed results indicate that the proposed solution is successful in facilitating the development of high-quality defense services. The effectiveness of the suggested solution is evaluated using statistical metrics including vulnerable activity recognition (Precision 95.24%), model training and testing (Precision 95.24%, Recall (95.00%), and F-Measure 94.11%), latency rate (7.45 seconds), and data processing cost( $\theta $ ((n-1) logn)).
引用
收藏
页码:91316 / 91330
页数:15
相关论文
共 55 条