原创 超越波浪理论：从维数方面理解DSP发展

Beyond the Third Wave: Thinking in Terms of Dimensions
By Gene Frantz
TI Principal Fellow and Business Development Manager, DSP

Anyone who watched the old Dr. Who show, from our friends at the BBC over the pond, knows that science fiction and quantum theory postulate that there are multiple dimensions in the universe. I’d like to take that theory to some degree and apply it to DSP innovation.

In my last two posts, we discussed waves of innovation. Those waves have been dominated by applications that harness video, imaging and audio breakthroughs. It is likely that future innovation won’t fall too far from this tree. But how do we track the evolution of these applications and create some sort of predictor for future innovation? This is where my dimension theory comes in, the main driver of which is performance increases.

In the beginning, the performance of DSPs limited their application to audio. Audio can be considered a one dimensional problem, so this was the first dimension. In my previous posts I’ve talked about the evolution of audio from the communication algorithms to the CD format to today’s MP3 applications.

Imaging is a two dimensional problem. So as DSP performance increased, we could move from one dimensional problems to two dimensional problems. With it, we added to audio second dimension - pictures. As performance allowed those pictures to move, we had a three dimensional problem. Finally, we added audio to moving pictures giving us the fourth dimension. Each time we added a dimension to the problem we required more performance from the DSP. Or was it every time we increased the performance of the DSP we could handle more dimensions?

Everyone with me?

The question I would like to pose to all of you, that has direct bearing on what will be possible beyond the Third Wave of Innovation, is: What will the fifth dimension be?

Every time we have added performance, we have added a dimension. Have we run out of dimensions? (We are still adding performance so I suspect there is a 5th dimension.) But what does the 5th dimension “look” like, “sound” like, or “feel” like?

Certainly, we will improve upon the elements of the first four dimensions. Better audio and video are a definite, and most of the future opportunities in signal processing lay upon those two paths.

But can we add to the other senses? Can we add smell? Can we add feel? Are these necessary additions that add to the quality of a product. Hopefully, some of you can give me a taste of what will be next. So?