Annual Flagship Event of IEEE SPS NITK Student Chapter: IMPULSE 2023

You are here

Inside Signal Processing Newsletter Home Page

Top Reasons to Join SPS Today!

1. IEEE Signal Processing Magazine
2. Signal Processing Digital Library*
3. Inside Signal Processing Newsletter
4. SPS Resource Center
5. Career advancement & recognition
6. Discounts on conferences and publications
7. Professional networking
8. Communities for students, young professionals, and women
9. Volunteer opportunities
10. Coming soon! PDH/CEU credits
Click here to learn more.

News and Resources for Members of the IEEE Signal Processing Society

Annual Flagship Event of IEEE SPS NITK Student Chapter: IMPULSE 2023

IEEE SPS National Institute of Technology Karnataka Student Chapter
Annual Flagship Event: IMPULSE 2023

Impulse is the annual flagship event of the IEEE Signal Processing Society of IEEE-NITK (National Institute of Technology Karnataka, India) Student Chapter. It is a 24-hour signal processing hackathon, and 2023 was the third edition of this one-of-a-kind event. The event aims to foster interest in signal processing among first and second-year undergraduate students, by introducing them to open problems in the field, which are usually outside the classroom curriculum. The theme of the 2023 event was Multirate Signal Processing which was conducted over a period of five days, from 11-15 January 2023.

 

 

Registrations

With the event open to first and second-year undergraduates from all colleges across the country, Impulse-2023 saw 75 registrations and 132 participants, a considerable improvement over its predecessor. IEEE student members and NITK students had free entry to the competition, while other participants had to pay a registration fee of INR 150. Participants were supposed to register through the Impulse website.

Event

The entire event was conducted in three phases:

  • The preliminary round
  • Expert talk and technical workshop
  • Final hackathon (24 hrs.)

 

 

Preliminary Round

The preliminary round was completed on HackerEarth on 11 January 2023. The round consisted of 20 questions from signals and systems, Python programming, aptitude, and basic image processing.

Event Talk

The expert talk was organized to introduce to the participants the domain of Multirate Signal Processing and provide them with a glimpse of open research problems in the domain. The expert talk was conducted on 12 January 2023. Prof. Shayan Srinivasa Garani (IISc Bangalore) was invited to deliver a webinar on ‘Introduction to Multirate Signal Processing’. The event, being completely online, was conducted on Google Meet.

 

 

Workshop

The workshop was organized to equip the participants with the necessary tools to tackle the questions of the main hackathon. A workshop on ‘Multiresolution Pyramids’ was organized by M Shreyas Rao (SPS Chair, IEEE NITK) on 13 January 2023. Owing to other ongoing events at NITK, the workshop was recorded and shared among the participants.

 

 

Final Hackathon

The final hackathon was participated by 20 shortlisted teams from the preliminary round. The 24-hour hackathon was held from 21:00 hrs, 14 January 2023 to 20:59 hrs, 15 January 2023. The participants were allowed to attend the hackathon at the Central Computing Centre, NITK. The main problem statement involved deploying the finer directional wavelet transform and scattering transform as a low-level feature extractor in place of convolutional and pooling layers in the initial layers of CNNs. The participants were required to build a filter bank of directional filters using which they were then asked to perform the scattering transform on a given image as proof of concept. Finally, to put it all together, they were asked to feed the outputs of the scattering transform to a CNN(in this case, a trained model with loaded weights) for evaluation on a classification task.

Three review meets were conducted over the 24 hours to clarify the doubts of the participants. The event concluded with three successful submissions, and prizes were awarded to the best two teams. Prizes and refreshments for the event were completely funded by the IEEE Signal Processing Society.

 

 

SPS Social Media

IEEE SPS Educational Resources

IEEE SPS Resource Center

IEEE SPS YouTube Channel