| dc.contributor.author | Kaweesa, Wycliff | |
| dc.date.accessioned | 2023-01-20T06:45:38Z | |
| dc.date.available | 2023-01-20T06:45:38Z | |
| dc.date.issued | 2022-09-21 | |
| dc.identifier.citation | Kaweesa, Wycliff. (2022) Data compression for an edge-based machine learning model in IOT devices.(Unpublished undergraduate dissertation) Makerere University; Kampala, Uganda | en_US |
| dc.identifier.uri | http://hdl.handle.net/20.500.12281/14543 | |
| dc.description | A final year project report submitted to Makerere University in partial fulfilment of the requirements for the award of Bachelor of science in Electrical Engineering. | en_US |
| dc.description.abstract | In this project, an IoT device, comprising an AD8232 ECG sensor kit and ESP32 microcontroller, was designed and developed to collect ECG data. Compression using the wavelet transform was then applied on the collected data prior to transmission giving a compression ratio of 12.8. The transmitted data was then reconstructed on the edge and processed using a deep learning sequential model. The model had training and validation accuracies of 97.1% and 96.5% respectively. In the aforementioned approach, we considered the context of driving behaviour monitoring in where ECG signals data are collected from the driver using the developed IoT device and sent to an edge node for stress level detection. The results showed that the transmission time of the 5s segments of the electrocardiogram data was reduced on average by 1.931s when compression was used and this had no effect on the model’s accuracy. | en_US |
| dc.language.iso | en | en_US |
| dc.publisher | Makerere University | en_US |
| dc.subject | Data Compression | en_US |
| dc.subject | Machine Learning | en_US |
| dc.title | Data compression for an edge-based machine learning model in IOT devices. | en_US |
| dc.type | Thesis | en_US |
