Filtering by
- All Subjects: wireless
- All Subjects: Educational AI
- Creators: Gupta, Sandeep
- Member of: Theses and Dissertations
In this work, I present novel approaches for the recognition of location, movement and handshape that are components of American Sign Language (ASL) using both wrist-worn sensors as well as webcams. Finally, I present Learn2Sign(L2S), a chat- bot based AI tutor that can provide fine-grained conceptual feedback to learners of ASL using the modular recognition approaches. L2S is designed to provide feedback directly relating to the fundamental concepts of ASL using an explainable AI. I present the system performance results in terms of Precision, Recall and F-1 scores as well as validation results towards the learning outcomes of users. Both retention and execution tests for 26 participants for 14 different ASL words learned using learn2sign is presented. Finally, I also present the results of a post-usage usability survey for all the participants. In this work, I found that learners who received live feedback on their executions improved their execution as well as retention performances. The average increase in execution performance was 28% points and that for retention was 4% points.
Wardriving is when prospective malicious hackers drive with a portable computer to sniff out and map potentially vulnerable networks. With the advent of smart homes and other Internet of Things devices, this poses the possibility of more unsecure targets. The hardware available to the public has also miniaturized and gotten more powerful. One no longer needs to carry a complete laptop to carry out network mapping. With this miniaturization and greater popularity of quadcopter technology, the two can be combined to create a more efficient wardriving setup in a potentially more target-rich environment. Thus, we set out to create a prototype as a proof of concept of this combination. By creating a bracket for a Raspberry Pi to be mounted to a drone with other wireless sniffing equipment, we demonstrate that one can use various off the shelf components to create a powerful network detection device. In this write up, we also outline some of the challenges encountered by combining these two technologies, as well as the solutions to those challenges. Adding payload weight to drones that are not initially designed for it causes detrimental effects to various characteristics such as flight behavior and power consumption. Less computing power is available due to the miniaturization that must take place for a drone-mounted solution. Communication between the miniature computer and a ground control computer is also essential in overall system operation. Below, we highlight solutions to these various problems as well as improvements that can be implemented for maximum system effectiveness.