A Haptic Data Repository Framework

The concept of a data repository is not a new idea; almost any scientific approach to a statistical-based problem requires a data storage effort for the purpose of post-processing and post-examination of acquired data sets. When faced with haptic-data, however, we run into several problematic issues, such as high bandwidth requirements (sampling rates equal to or higher than 600 Hz), data accuracy and relevance, data filtration and presentation. In this paper, we propose a haptic-data repository proof-of-concept capable of retrieving, filtering, and storing the compacted data sets in a local database. We introduce an Intelligent Filtering Algorithm, a Thread Management Algorithm, as well as an I/O optimization algorithm. By combining these algorithms, and as our performance evaluation demonstrates, we have successfully been able to save accurate haptic data at sampling rates equals to or higher than 1 KHz.