Sensing movement automagically
At Dozee, our mission is to improve the quality of life through continuous health and well-being monitor. Our sensor tracks the quality of sleep and body vitals while a user is sleeping.
People move during their sleep. These movements are very important to track muscle activity which is a crucial parameter for sleep analysis. It is also an indication of restlessness during sleep. Your mission, should you chose to accept it, is to write code to identify whether each second of data has movement or not. In the test data folder you can find 5 files of 4 minutes of vibration data each with some movement instances in them.
The image below represents normal time series data without movement.
The image below represents time series data with movement.
A document explaining the approach along with the code (any programming language) is required for evaluation and should be mailed to gaurav@dozee.io.
For any clarifiactions please reach out to us at the same email address.
This repository will not self-destruct