It's been more than 20 years since I submitted my EE master thesis about implementing a "feedforward" active noise cancellation algorithm (nlms) for headphones on a TI c50 fixed point DSP. It was a fun project and audio noise cancellation was a hot topic back in the days. Since then, active noise cancellation headphones have become mainstream.
I also remember that voice compression was a very active research field with researchers trying to get their algorithms in mobile phones. The University of Sherbrooke in Quebec/Canada where I was doing my master degree had a very successful research group on this very topic and the ACELP technology they developed can today be found inside virtually every mobile phone on the planet.
Fast forward to 2018, I would be curious to read your insights as to which DSP topics are currently trending. If the discussion gains traction, I think this thread could become a very interesting and useful read for students and young EEs looking to orient their DSP studies and careers.
I am working on my PhD in field of Signal Processing (to be more specific Signal Processing for communications). Based on what people do in research community here is one list:
1. Compressed and Finite rate of Innovation Sampling.
To get more insights you can read:
2. Sparse sensor selection and signals reconstruction
(Related to 1.)
3. Signal processing for biomedical applications
4. Signal processing for radars
5. Graph signal processing (signal processing on graphs) - still theoretical field
6. Distributed signal processing
7. Signal processing and deep learning ( I personally do not like way people do research related to this topic, but it is really treandy)
Fun to hear about your experience with noise cancellation. I worked for more than 30 years on adaptive filtering techniques for audio and data communications cancellation and equalization for speakerphones, network echo cancelers and modems.
I'm having fun these days with analysis of LTE radio signals in an effort to triangulate a position from the known position of the detectable cell towers.
I believe that there is still a large on-going effort in perfecting cellular communications with MIMO, beam-forming, OFDM modulation and other fun ways to improve user experience and network capacity.
The book "Trends in Digital Signal Processing: A Festschrift in Honour of A.G. Constantinides" presents a thorough review of the recent studies on different fields of digital signal processing.
You can find the content of the book here.
Computer Vision is burning hot nowadays.
This is a very good point to raise up nowadays, since signal processing has branched into so many fields of science and engineering. Of course, DSP is about the implementation of signal processing algorithms, but I believe that DSP will reach those topics shortly. From my reading in the IEEE Signal processing magazine, I can remember a few topics that are in tend now:
1- Massive MIMO.
2- DSP for mmWaves.
3- Artificial Intelligence.
4- Compressive sensing (not as new as the above).