Usage of CC6678 Digital Signal Processor (DSP)

Aguilera

Usage of CC6678 Digital Signal Processor (DSP) [Copy link]

Like one of the inventors of the DSP, Gene Frantz , there is nothing more fun for me than to see how customers put our processors to use in real-world situations. However, I find the KeyStone based TMS320C6678 , an 8-core C66x DSP, particularly interesting because it is one of the first high-performance multicore DSPs on the market. While over 100 customers have already adopted the C6678 DSP in their products, I can safely say that there are at least 50 different ways to use the fixed-point and floating-point C6678 DSP! But don’t worry, I won’t throw that long list at you, I just want to share with you 5 of the most common applications.

Speeding up backhaul —The Keystone family of wireless base station systems-on-chip (SoCs) has already flooded the wireless market, but the C6678 DSP has also found its way into wireless backhaul. The Sub10 system uses the C6678 DSP as the foundation of its wireless backhaul network synchronization platform, relying on the Keystone-based DSP for real-time algorithm processing and network connectivity.

Take a new scan —Since image processing is the forte of digital signal processing, a large number of TI DSPs play a very important role in medical imaging equipment, and the C6678 DSP is no exception. Take a look at how to apply this Keystone multi-core DSP to your next medical imaging application.

Stay on the bus —or go wireless when using Synthetic Aperture Radar (SAR) products. SAR technology is popular in commercial and military radar-based applications, and because SAR relies heavily on prior signal processing and computationally intensive processing, the C6678 multicore DSP is the best core for this function.

Without getting into too much detail - we are all consuming more and more streaming video on our smartphones, tablets, etc., which means we have less time for face-to-face conversations these days. All these video streams are seriously taxing servers and content distribution devices in the network, which is why the implementation of video compression/decompression standards such as H.265 is critical for real-time performance. The C6678 is highly efficient at performing video encoding and is a software programmable chip with very low power consumption. In terms of how the C6678 DSP is an ideal H.265 processing engine.

Fast Fourier Transform (FFT) is key - As avionics and defense applications rely heavily on real-time signal processing, it is no wonder that these applications are closely related to Fast Fourier Transform (FFT); in many cases, very large FFTs are used. View this FFT white paper to learn how the C6678 DSP with FFT coprocessor is ideal for FFT processing and avionics and defense applications, and how to maintain optimal performance at low power and small footprint.