Abstract

The Kalman filter is a signal processing algorithm that plays an essential role in real-time mobile applications as nano drones navigation, spacecraft orbit control, GPS positioning, image recognition, and sensor data fusion for wearable health monitoring. However, the Kalman filter is a compute-intensive kernel since it implements a high amount of complex matrix operations like multiplications and inversions. The most timing-consuming tasks of the Kalman filter is the Kalman gain. Therefore, this work presents four architectural solutions to explore real-time and energy-efficient Kalman gain processing.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call