Table 2 Comparison of this study with other research papers in IoT healthcare systems.
From: Secure edge-based IoMT framework for ICU monitoring with TinyML and post-quantum cryptography
Aspect | Your research | Other research papers | Similarities | Improvements/novel contributions |
|---|---|---|---|---|
Focus area | ICU patient monitoring using TinyML, data fusion, and PQC21 | Both focus on integrating PQC into IoT systems for enhanced security | Specifically addresses healthcare IoT (ICU monitoring) | |
Post-quantum cryptography (PQC) | Kyber 512 used for encryption and decryption in OMNeT++ simulation of ESP32 nodes5 | Both emphasize the importance of PQC for securing IoT devices against quantum threats | Practical simulation of Kyber 512 on resource-constrained devices. | |
Edge computing | Real-time inference using TinyML models simulated on ESP32 nodes21 | Focus on edge computing for consumer electronics and IoT ecosystems1 | Both leverage edge computing to reduce latency and improve efficiency | Integration of TinyML with PQC for real-time inference in healthcare IoT |
Data Fusion | Data fusion techniques applied to enhance feature reliability and eliminate redundant/outlier data6,22 | No explicit mention of data fusion. | N/A | Introduction of advanced data fusion techniques tailored for ICU patient monitoring |
Machine learning models | Decision tree-based TinyML model for anomaly detection21 | No specific mention of machine learning models | Both aim to improve IoT functionality through intelligent systems | Explicit integration of TinyML for real-time inference on edge devices |
Synthetic dataset | Synthetic dataset simulating ICU patient scenarios validated by medical professionals14. | No mention of synthetic datasets | N/A | Meticulously generated and validated synthetic dataset for ICU scenarios |
Simulation framework | OMNeT++ simulation demonstrating zero observed packet loss and very low latency (mean: 18.7Â ms)19 | Emphasis on simulation and modeling for evaluating PQC in IoT ecosystems | Both use simulation frameworks to validate system performance | Detailed OMNeT++ results with real-world benchmarks (e.g., PhysioNet, Kaggle) |
Security challenges | Addresses sensor noise, network delays, and cyber threats using PQC and anomaly detection9,23 | Discusses general security challenges in IoT ecosystems, including quantum threats | Both address security challenges in IoT systems. | Specific solutions (e.g., Kyber 512, anomaly detection) for healthcare IoT |
Scalability | Framework supports scalability with multiple simulated ESP32 nodes19 | Emphasizes scalability in large-scale IoT deployments. | Both focus on scalability for IoT systems. | Demonstrates scalability in ICU settings with multiple patients and devices |
Energy efficiency | Optimized energy consumption on resource-constrained devices4 | Mentions energy efficiency as a challenge in IoT ecosystems | Both prioritize energy efficiency for edge devices | Integration of low-power hardware accelerators and efficient communication protocols |
Real-world validation | Validated synthetic dataset against real-world benchmarks (PhysioNet, Kaggle) and expert feedback21,23 | No mention of real-world validation | N/A | Extensive validation with real-world datasets and expert surveys |
Interoperability | Ensures compatibility with existing healthcare standards (e.g., HL7, FHIR)24 | Mentions interoperability as a challenge in IoT ecosystems | Both emphasize the importance of interoperability | Explicit compliance with healthcare standards for seamless integration |
