Advanced Topics in Next Generation Wireless Networks Qian

Advanced Topics in Next. Generation Wireless Networks Qian Zhang Department of Computer Science HKUST

Ice-Breaking

Course Information • Instructor: Qian Zhang, Lionel M. Ni, S. C. Cheung • Course homepage: – http: //www. cs. ust. hk/~qianzh/CSIT 600 K. htm • Selected conference and journal papers – Mobi. Com, SIGCOMM, INFOCOM, ICDCS, Dy. Span proceedings – IEEE Transactions, journals

Course Workload • Grading – Participation (30%) – Paper review and class presentation of papers (30%) – Project Report (40%)

Course Workload (cont. ) • Paper review and presentation – Everyone reviews and presents 2 papers – Email me ids of 3 papers that you’d like to present by July 15, 2008 – Submit a review for each paper of your choice before your presentation (1 page) – Review form coming soon • Classroom participation – Interactive!!!

Course Workload (cont. ) • Course project – Goal: obtain hands-on experience in pervasive computing and communication research – Work in a group of 2 – You may choose your own topic – Final report • In the form of a technical paper (just like the ones you read) --- no more than 6 pages -- IEEE 2 column format • The term paper needs to be formal -- abstract, intro, your proposed work etc.

1 page Paper Review • For the paper to be presented that week • Outline – Summary of main idea, contributions • Must be in your own words – Itemized list of Strengths – Itemized list of Weaknesses – Things that you have learned • A new area, novel approach, analysis, new evaluation methodology? – Moving on: ideas on how to extend/improve the work • Could be a paragraph or two, OR an itemized list

Class Presentation • Must use power-point (prepare 15~20 slides) • Read at least 2 related papers cited • Presentation outline (20 -25 minutes) – Significance of the problem – Why is solving the problem challenging? – Previous approaches (related work) • Also mention the ones that you have read – Proposed approach and motivation behind it – Analytical results/proofs (if any) – Simulation/experimental results – Summary • Lead classroom discussion

4 Steps in Doing Research 1. 2. 3. 4. Choosing the problem Solving the problem Evaluating solutions Selling your work

Choosing the Problem • Finding the right problem is important but hard – Relevance, significance • Several types of research – Tier 1: Point out a new direction – Tier 2: Under an interesting direction, figure out a new topic – Tier 3: Solve a well defined problem with new approach – Tier 3. 5: Find a good paper, do some enhancement – Tier 4: Duplicate other people’s work (X)

Choosing the Problem (Cont. ) • Read widely – Read conference papers and journals – Keep track of technology trends (tech news, industrial applications) and theoretical advances • Talk to others – People in different areas, in different organizations • Self-critical + patient – Try to avoid Tier 3. 5 like research – Problem statement should spend your ½ time • Willing to take a risk

Solving the Problem • Survey the related work (google!) – Avoid reinventing the wheel – Be skeptical • Acquire necessary knowledge (on demand) • Always think from different aspects – Diversity bring new research opportunities – Leverage different expertise • Ask for feedback

Evaluating Your Ideas • Be objective • Use simple scenarios to gain insights • Systematically explore the parameter space • Create realistic evaluation scenarios • Real system means a lot for wireless networking

Selling Your Work • Three ways of selling your work – Publish papers – Deliver presentation – Spontaneous situations • • Be clear about contribution and motivation Learn from good/bad examples Practice, practice, and practice Many helpful information on the Web to improve your presentation/communication skill