Travel Report from University of South Australia, Adelaide, Fall 2002

Elisabeth Uhlemann

Introduction

In the fall 2002 I spent two months at the Institute for Telecommunications Research (ITR) at University of South Australia (UniSA) in Adelaide, Australia supported by travel grants mainly from ARTES but also from Ericsson.

ITR is a research institute at UniSA in Adelaide, specializing in technology for digital wireless communications, including both fixed and mobile satellite and terrestrial radio services. ITR is located in the Signal Processing Research Institute building at The Mawson Lakes Campus.  This campus is located at Mawson Lakes, 14 km north of the center of Adelaide. The 80 staff and postgraduate students in ITR engage in research programs designed to be of direct benefit to industry. 

Purpose of Visit

The ITR has a recognized reputation for both theoretical research and practical implementation in the area of concatenated coding and iterative decoding a currently very hot topic. Iterative decoding is a sub-optimal decoding method with good accuracy and significantly lower complexity than the optimal method. Generally it is known that most block codes are good, provided that they are long enough. However, the complexity of optimal decoding is NP-hard and exponentially increasing with code length. Recently, a sub-optimal coding and decoding strategy has been discovered where, provided that the code is concatenated, the output of a sub-optimal decoder can be fed back again in order to be refined iteratively. This procedure, termed iterative decoding or turbo-decoding, has performance close to the optimal one with manageable complexity increase.

For real-time communication purposes iterative decoding and concatenated codes are very interesting topics. We want to benefit from the high quality that the long concatenated codes provide, but at the same time keep the decoding complexity, and hence the time to decode, at a minimum. This deadline dependent coding and decoding procedure is what I am currently looking at in my research.

Dr. Alex Grant is the leader of the Coding and Information Theory group at ITR and I was invited to work closely together with him and the members of his group during my visit. Dr. Grant is a recognized expert in iterative signal processing with a particularly emphasis on analytical tools for performance evaluation. Currently, Dr. Grant has four students working on various aspects in iterative signal processing. In addition, my main supervisor Prof. Lars K. Rasmussen was visiting the ITR during most of 2002 and could, consequently, provide me with necessary supervision.

Results

The work resulted in two conference papers; one is accepted [1] and will be presented at the IEEE Wireless Communications and Networking Conference. The other is submitted and decisions are due in March. A poster presentation of the work in [1] will also be presented at the Australian Communications Theory Workshop in Melbourne [2]. Furthermore, I took a course in Multi-user detection given by Dr. Grant and Prof. Rasmussen. Finally, on the way back to Sweden I stopped in San Jose, California, USA to attend and present a paper at the Real-Time and Embedded Technology and Applications Symposium, RTAS 2002.

Graduate Studies in Australia

The Ph.D. program in Australia is typically only three years. No coursework is mandatory but is often encouraged by the supervisor. Since the ITR is a research institute they do research for both academia and industry, but provide no teaching of undergraduate students. This means that neither professors nor Ph.D. students have any mandatory teaching duties. If you want to become a Ph.D. student you have to apply for funding yourself, usually through government-funded scholarships. Upon doing this you should also state your research topic and suggest an appropriate supervisor. This means that lots of responsibility is put on the student from the very beginning.

Adelaide

Adelaide is the capital of South Australia hosting 1.2 million people originating from more than 100 different countries around the world. Many migrants arrived from Europe (especially Italy) after World War II and introduced the café culture that lends Adelaide its relaxed atmosphere. The city was named after Queen Adelaide, the wife of the British King William IV. Adelaide was unusual in that it was settled by free people - the city has no convict history. Adelaide has a reputation of being the festival capital of Australia and also the food and wine capital of Australia.

Superfluous details

Σor Australia versus Sweden [3]:

If you want to cross the street you better be fast. If you walk normally you almost make it half way across the street before the lights become red again.

No stores have the opening hours listed on their doors! It is very annoying. There is a little supermarket very close to where I live, but no one knows its opening hours.

The water tastes like a mix of chlorine and seaweed. Adelaide is located in the driest part of the driest continent, which makes it hard to access clean water. Most people in Adelaide buy bottled water or install purifying filters in their homes.

When at a restaurant you seldom order your food at the table. You notice this by the fact that nobody in the staff seems to care about you. You are then supposed to go to the counter and place your order. The weird thing is that you cannot order food and drinks at the same counter. You pay for food at one counter and then go over to the next and order drinks. This is regardless if you want water, coke or wine. It may even be the same guy that took your food order that makes his way over to the bar to take your drink order. This is especially annoying if you wish to pay for the entire dinner with credit card and are not sure upfront how much you will eat or drink.

The Australian milk does not contain 3% fat, but is 97% fat free! Why dont you write how much fat there is in the milk instead of writing how much fat there isnt in the milk?

All buss stops are request stops, which means you wave your hand to the bus with the appropriate number. This is troublesome because often three buses with different numbers arrive at the same time. How do you stop the last bus? My bus is always the last one so I never see the number until it has already passed usually by overtaking the other two others that have already stopped.

Buying candy is hard, especially so called lösgodis (loose weight candy?). In Sweden there is one price for all sorts. In Australia it isnt every sort has an individual price stated on the respective boxΣ You have to take one bag for each sort and since I usually buy 20 different sorts (obviously none of them with the same price), I gave up.

Conclusions

My visit was of great benefit to my research. Having to describe real-time communication problems to people with non-real-time backgrounds, but with excellent communication knowledge is very instructive and gives you new insight to your problem. In addition, it is always valuable and interesting to visit a different research institute than the one you are used to and see different solutions to similar problems.

References

[1]   Elisabeth Uhlemann, Tor M. Aulin, Lars K. Rasmussen and Per-Arne Wiberg, Packet combining and doping in concatenated hybrid ARQ schemes using iterative decoding, accepted for publication at the IEEE Wireless Commun. and Networking Conf., New Orleans, LA, USA, March 2003.

[2]   Elisabeth Uhlemann, Tor M. Aulin, Lars K. Rasmussen and Per-Arne Wiberg, Concatenated hybrid ARQ using iterative decoding for real-time communication, accepted for publication at the Australian Communications Theory Workshop 2003, Melbourne, Australia, February 2003.

[3]   F. Brännström, Personal Conversation, Chalmers University of Technology, 2002.