0
$\begingroup$

I've newly bought texas instruments Tiva C series TM4C123GXL processing board for developing DSP application project. Is that processing board is good for that purpose? Is that well work with Matlab? Or C2000 Piccolo TMS320F28075 32-bit microcontroller is good for my purpose ?

$\endgroup$

closed as off-topic by Matt L., Gilles, MBaz, Peter K. Sep 27 '16 at 0:34

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "This question does not appear to be about signal processing within the scope defined in the help center." – Matt L., Gilles, MBaz, Peter K.
If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ Can you please clarify what your question is exactly and what particular DSP aspects you require help with? $\endgroup$ – A_A Sep 26 '16 at 16:11
  • $\begingroup$ @A_A I want to use it for ground water leakage point detection and analyze the detected data. The first I have got the ground signal and analyze the data using DFT algorithm in matlab. $\endgroup$ – kaviranga Sep 27 '16 at 0:14
0
$\begingroup$

This question is apparently out of scope. But I would like to provide you a simplest guideline so that you can by yourself decide on whether a given microprocessor would be adequate for various DSP tasks.

1- The Floating Point Capabilities: Nowadays any processor claiming high quality DSP computation must have strong floating point hardware support. Look for its FLOPs (floating point operations) throughput. Also look for single precision (32 bits) or double precision (64 bits) FP capabilities. For most multimedia purposes (speech, music, video, image) 32 bit is more than enough. For more accurate scientific computations 64 bits is necessary. This being said, the experts can use integer capabilities of a microprocessor to implement most DSP algorithms. But that would require a lower level handling of the mpu. Depending on what kind of an algorithm you want to implement, you should first determine number of floating point operations per second and then check for the MPU if it can support that.

2- Parallel Execution Capabilities (SIMD) Many DSP algorithms (especially image and video prcessing types, or communication channel processing) will benefit from single instruction multiple data (SIMD) processing pipeline. It can be expensive to include in a simple MPU but if you have this capability, it's worth the money.

3- ADC / DAC capabilities are vital as you will eventually be inputting an analog signal, convert to digital, process it and cpnvert to analog before outputting. So look for for High Speed, High Accuracy ADC and DAC. Or if your application can tolerate low accuracy go with that one. And there must be enough number of channels so that if you have multiple signals being input or output you can use that.

4- Memory: in addition to limited program code memory for several applications, data has to be buffered inside RAM. this ram can be on-chip or out of chip. First determine how much ram you will need and then look for wheter it can be on chip (very fast) or out of chip.

$\endgroup$
  • 1
    $\begingroup$ This is great . It is a wide area because I'm a newbie to dsp . This answer includes the areas that I must cover before dig in to the problem . Very informative answer! $\endgroup$ – kaviranga Sep 27 '16 at 0:22
  • $\begingroup$ I wouldn't say that floating-point support is required for DSP applications nowadays. There are a lot of implementations even today that use fixed-point (integer) arithmetic, as it can be implemented much more cheaply in hardware. With that said, floating point is a lot easier in a lot of cases, if your platform supports it. $\endgroup$ – Jason R Sep 27 '16 at 0:48
  • $\begingroup$ The term "required" was in terms of ease of programming... $\endgroup$ – Fat32 Sep 27 '16 at 8:43

Not the answer you're looking for? Browse other questions tagged or ask your own question.