CS756: Mobile Computing and Sensor Networks (Fall 2024)
This course presents a comprehensive view of mobile and ubiquitous computing, from design choices for various mobile/IoT/robotic platforms, their applicability in smart cities, smart homes and digital health, and pertinent performance trade-offs in the context of mobile sensing. Topics will introduce fundamentals of sensing and analytics as applicable to smart phones, personal wearables, edge processors, AR/VR systems and mobile robots, and will cover a range of techniques relevant to applications, systems, networking, localization, activity classification, and deep learning under resource constraints. The course will also introduce students to emerging paradigms such as intermittent computing and human-robot teaming. Students will participate in instructor-led weekly discussions, acquire practical programming skills for various mobile platforms through in-class labs, critique mobile systems research papers through in-class presentations, and work on a semester-long project.
Course Prerequisites
While there are no formal prerequisites, prior familiarity with Unix-based operating systems, deep learning libraries in Python (e.g., Keras), and Java/Kotlin will be useful.Grading Policy
The course is designed such that there’ll be opportunities for evaluation and feedback throughout the semester. The course grade is allocated as follows:- Group Labs (45%) – Three lab assignments each worth 15% consisting of (a) instructed, hands-on programming of different mobile platforms (e.g., smartphone using Android Studio, Raspberry Pi/Jetson Nano, and educational mobile robots) and (b) answering design questions related to the platforms. Students will work in teams of 3-4.
- Individual Project (35%) – Students will work on a semester-long individual project related to mobile computing. Interim updates (i.e., abstract, 1-page update) will also contribute towards the component grade. Final write-up in a workshop-style format will be due a week after the final presentations. The Fall 2024 offering will involve students using the RealEye.io platform for gaze-based sensing applications.
- Paper critique (20%) – Each student will pick one paper from a list of papers provided by the instructor to critique and present in class (worth 10%). 10% of the total grade will be allocated for submitting summaries for at least three other papers presented by the peers.
Sample Schedule
Contact Me. If you're interested, drop me an email at kasthuri.jayarajah@njit.edu for any further clarifications.