CSCE 5013: Hot Topics in Mobile and Pervasive Computing (Fall 2009)

Course Description

Advances in wireless and sensor systems has fueled the growth and evolution of a new class of computing: mobile and pervasive computing. Mobile and pervasive computing is a conglomeration of several diverse areas of networking (mobile and sensor), and embedded systems. Hence, it suffers from unique complications including disruptive connectivity, variable bandwidth, energy constraints and easily sniff-able communications. Moreover, it spans a wide spectrum of applications including sensors networks, location-aware services, social networking, and wearable computing. This seminar will critically examine these topics from a systems perspective. It will cover a wide spectrum of applications under the umbrella of mobile and pervasive computing.

This course would be a seminar style course where students would be required to critically review papers from leading mobile, ubiquitous, and operating systems conferences such as Mobicom, SOSP, OSDI, MobiSys, Ubicomp, and Sensys.

Course Evaluation

The evaluation for the course would be based on class presentations and participation (15%), written paper reviews (25%), a take-home midterm (20%), and an independent research project related to mobile and pervasive computing (groups of at most two student are allowed) (40%). I have discussed each of these components below in detail.

Requirements

Knowledge of Operating systems and Computer Networking concepts would be required to understand the topics covered in the course.

Research Project

Paper Reviews

Writing critical reviews for research papers is an extremely important and rewarding exercise. Just because a paper is published in a renowned conference/journal does not necessarily imply that it is flawless. Analyzing loop holes and potential areas of improvement in research papers can help you understand what the authors have really done and how you could use it for your own research. Therefore, for a seminar course like this one your analysis of the papers is crucial to your understanding of the topics discussed.

I would want you to write a short review (less than 250 words) of the papers assigned. It should *not* be a summary of the paper and it should not be copied from the abstract of the paper. Here is my take on writing a good review. (1) Summarize what the authors have done or proposed in a couple of sentences (2) Bring out the potential loopholes and areas of improvement (remember: there is nothing like a perfect research!) (3) how you could use the ideas and/or techniques in your project/research (note that you might not be able to directly use the ideas in all the papers but you can use the methodologies and techniques used in the paper (such as the evaluation techniques)).

Reading a research paper is an art---something everyone can not master easily. Moreover, there is no single magical potion that you can drink and be able to read and understand research papers. It is mostly practice and following the correct techniques. Take a look at GA-6 for advice on reading papers and GA-7 on writing reviews for systems research papers.

All reviews are due at/before midnight before the class. This will give me time to go through your reviews and discuss them in class.

Class Presentation

Course Information

Instructor: Nilanjan Banerjee

Class lectures: M, W, Fr 9:30-10:20
Office Hours: Tue 3:45-4:45, Thur 3:45-4:45 or by appointment, Room 522, JBHT
Credits: 3
Mailing List: mobperv AT mailman.csce.uark.edu

Course News

Course Schedule

External References

Survey

Papers