In the Fall of 2023, I immersed myself in the intricate field of signal processing, specifically focusing on motion and IMU data analysis. My journey began with an in-depth exploration of various techniques and algorithms crucial for interpreting signals from motion sensors like accelerometers and gyroscopes. I delved into data filtering, interpolation, normalization, offset correction, and time synchronization. These methods are pivotal in transforming raw data into meaningful insights about human movement.
Simultaneously, I dedicated substantial effort to mastering Python's Numpy and Pandas libraries, essential tools for data analysis. Through practical applications demonstrated in my notebooks "pandasTest.ipynb" and "numpyTest.ipynb," I honed my skills in handling and manipulating large datasets, a skill vital for my research.
As my understanding deepened, I embarked on a comparative analysis of motion data from AirPods and XSENS. XSENS's reputation for high precision in motion detection set a benchmark for my analysis. My objective was to assess AirPods' performance against this standard. To elucidate this comparison, I generated and shared detailed data visualizations, providing a clear depiction of the differences and similarities in the motion detection capabilities of these devices.
Recognizing the importance of thorough background knowledge, I referred to the documentation and specifications from AirPods and XSENS manufacturers. This approach helped me understand the nuances of their motion detection technologies and metrics. Additionally, I consulted various research papers and studies that compared these devices, enhancing my comprehension of their technological intricacies.
Taking a step beyond theoretical study, I visited NYU's therapy department to collect motion data firsthand. This experience was invaluable; it not only gave me a practical perspective on data collection but also enriched my understanding of the real-world applications and implications of this data. During this phase, I completed the normalization of the motion data and calculated the magnitude for each motion axis, which was crucial for my analysis.
In summary, my journey in Fall 2023 was a comprehensive exploration into the realms of signal processing for motion and IMU data. It was a blend of theoretical study and practical application, underscored by a deep dive into Python's Numpy and Pandas libraries. My comparative study of AirPods and XSENS motion data not only enhanced my analytical skills but also provided valuable insights into the capabilities of these technologies. The hands-on experience of data collection and preprocessing like normalization and magnitude calculation further solidified my understanding and expertise in motion data analysis. This endeavor was a significant step in my ongoing quest to excel in the field of data analysis and signal processing.
Here are three files for self-study on numpy and pandas:
These notebooks help me enhance my skills in handling and manipulating large datasets using Python's Numpy and Pandas libraries.
Here are the IMU real data for analysis