Conclusion and Future Directions

The thesis explores the effectiveness of spiking neural networks (SNNs) for edge devices, with a focus on resource efficiency, a key requirement for edge computing. The study primarily considers consumer radar as an application area, with a specific use case of gesture sensing and air-writing. To generalize the scope of this dissertation, the Fitbit tracker application is also utilized, covering a use case of time series forecasting for health data.