Privacy and convergence analysis for the internet of medical things using massive MIMO
Machine learning is the analysis based on data that gives strategic decisions to cultivate an accurate and stable framework for different applications. Access to medical data with the utmost privacy and high data rates is still a challenging problem. To accomplish the above-mentioned features, the p...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-06-01
|
| Series: | e-Prime: Advances in Electrical Engineering, Electronics and Energy |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772671124001049 |
| Summary: | Machine learning is the analysis based on data that gives strategic decisions to cultivate an accurate and stable framework for different applications. Access to medical data with the utmost privacy and high data rates is still a challenging problem. To accomplish the above-mentioned features, the performance of federated learning (FL) with 5G massive multiple-input-multiple-output (MIMO) is investigated for IoMT systems. This provides an energy-efficient and privacy-preserving solution with high throughput for digital health system. In the proposed model, the uplink scenario is investigated using different detection techniques. The performance are evaluated at the central server and edge devices for different signal-to-noise ratios (SNRs) and different fading channels. The ML bit error rate (BER) is better than MRC but with higher complexity. The accuracy obtained is approximately 90% with an improvement of around 8% to 9% as compared to the baseline approach. |
|---|---|
| ISSN: | 2772-6711 |