## Texas DSP 2812 and McBSP

Digital Signal Processing

### Next

• 1. Help on galois field arithmetic
Hello The following query though elementary, is elusive to me. Please bear with me for a posting, which I feel is trivial. I need to find the product of two GF(2^m) numbers, using generated lookup tables of Log and Exp. Essentially using the following Let Q=2^m -1, for example when m=3, Q=7; Let x and y be two numbers in GF(2^m); mul(x,y)=exp( (log(x) + log(y)) % Q) ....... Eq(1) where % in Eq(1) implies modular operation. I have generated Tables for GF(2^3), tell me if these are correct i Log(i) Exp(i) ---------------------- 0 0 1 1 0 2 2 1 4 3 3 3 4 2 6 5 6 7 6 4 5 7 5 1 To test the correctness of the above tables Let x=2;y=3; Q=7; Then,Eq(1) gives us 2*3 = exp( (log(2) + log(3)) % 7) 6 = exp((1+3)%7) from the table above = exp(4) = 6 . Which is equal to the left hand side. But, when the product goes more that 7,( I am missing something) Let x=4;y=5; Q=7; Now,Eq(1) gives us 4*5 = exp( (log(4) + log(5)) % 7) 20 = exp((2+6)%7); from the table above 20%7= exp(1) 6 = 2 which is absurd, What is that I am missing? Are my tables correct in the first place? If yes, where have I gone wrong? Give me some references,preferebly web links, which deals with Galois arithmetic. Thanks in advance KK
• 2. Harmonics of sample frequency
Hello all, I'm sorry if this is a duplicate. I think my news server choked on the previous one, though. I have a sound experiment where I generate a 125 Hz signal digitally and broadcast it through a power amp to a speaker. Sample frequency is 1 kHz. I've got a low-pass filter between the controller and the amp. If I disable the LP filter, there are audible harmonics of the sample frequency broadcast through the speaker. I understand that this is to be expected, and is one of the reasons that professional audio is sampled so high. I've looked back in my signal processing book (Proakis and Manolakis) and poked around the web, but can't find where anyone says WHY. There is a mention of it in my book (276, 3rd edition), saying "The process of sampling this (continuous time) signal with a sampling frequency Fs introduces replicas of the spectrum about multiples of Fs." That seems connected. I've drawn pictures and examined the FFTs, but I don't get why. This is one of these annoying things that is considered so self-evident that no one ever explains it, I guess. Would anyone like to give me a pointer or a hint? By the way, I am a student, but this is not a homework assignment. I took a signal processing course several years ago but don't have anyone I can ask right now. Thanks, Amanda
• 3. Classical IIR filter problem
Hi, I had designed a digital IIR filter from the existing analog prototype. The filter coefficients were obtained using MATLAB. The DSP which is used here is the TMS320F2812 processor. The problem is scaling of filter coefficients.The input data length to the filter is 12 bits. I had scaled the filter coeffiecients to 9 bits. But then with these available datalengths,its difficult to predict the no. of output bits as the filter is recursive. Two filters have been implemented. One is a second order filter amd the other one is a fourth order filter. I know that these quantization errors are non linear and its very difficult to predict the error. But whats the work around? Shd I use trial and error method to find out the exact scaling for my filter coefficients or any other better way exists? Any help in this regards would be greatly appreciated. Regards, Kalyani
• 4. Urgent:RTDX issue
Hi all I'm running a massive test harness on DSK6416, I'll send the input data from Matlab via RTDX to the target and I get back the data for analyzing in Matlab via RTDX. After a while , matlab reported a failure in input channel , at the same time the CCS has crashed and error in CCS is as following "stack overflow" Comments or suggestions would be very much appreciated. thanks , hamid
• 5. Odd frequency DFT with real inputs
A paper I am reading describes how a transform can be simplified by recognising it as "an odd DFT with real inputs". The equation is: X(k) = sum_(k=0 to N-1) x(n) exp(-j2.pi.n.(k-0.5)/N) He says that this can be realized by an FFT with N/2 complex valued inputs instead of N real valued inputs and gives the following reference: M. Bellanger, "Digital Processing of Signals", Wiley, 1989, p.83 Since I dont have acces to this reference, could someone please show me how to do the above trick! Cheers....

### Texas DSP 2812 and McBSP

```Hi all DSP experts :-)

I have a problem with the McBSP receive interrupt on the 2812 DSP from
Texas. Do anyone have working code on this DSP ?
I have done much the same as on the 5510 DSP where I did not have this
problem, so I do not understand why it is not working here ?

Torgeir

```

### Re: Texas DSP 2812 and McBSP

```There are more things you need to enable to get interrupts on the 2812 than
there are on the 5510.  Are you sure that you have enabled the PIE, set the
proper bit in the PIEIER for the McBSP interrupt, acknowledged the PIE
(PIEACK), set the corresponding IER bit, and cleared the INTM bit?

Also, are you sure that you have configured the McBSP to generate a receive
interrupt?  Are you seeing the RRDY flag which indicates there is data in

```

### Re: Texas DSP 2812 and McBSP

```Thank you Brad,

I have done the things you describe:
- Enabled PIE
- Set PIEIER for McBSP receive interrupt
- Acknowledged the PIE
- Set IER
- Cleared INTM.

First it looked like RRDY was not set, but now the RRDY is set. I have also
enabled the transmit interrupt to see if this was working better.  I can see
that RRDY and XRDY is set but the PIEIFR register (and PIEACK) does not flag
any interrupt, even if McBSP is set to generate interrupt on RRDY and XRDY
and the PIEIER is set to enable both interrupts.
I am using the sample rate generator, and the clock and frame signals are
generated correctly (measured with oscilloscope).

Any suggestions ?

than
the

```

```I have problems with the receive and transmit interrupts on this DSP. I have
set the McBSP to generate interrupts on RRDY and XRDY, and I can see that
these bits are set when I emulate the program on my hardware (using XDS510PP
Plus and CCS). This should also have activated the corresponding interrupt
flags in the PIEIFR6 regsiter (bit 5 and 6), but this does not happen.
I can see that my ISRs are working if I set the interrupt flags in PIEIFR6
manually, so the problem is the connection between the McBSP and the PIEIFR6
register. What do I have to do to make the McBSP RRDY and XRDY activate the
flags in PIEIFR6 ?
I use the internal sample rate generator, and I can see the frame and clock
signals with my oscilloscope.

```

```Hi everyone!
I am new in using dsp 2812.

I want to make an T3PINT interrupt.

I just read "TMS320x281x DSP System Control and Interrupts Reference
Guide"(spru078e) and
"TMS320x281x DSP Event Manager (EV) Reference Guide"(spru065e),

On page 84 in spru065e,and on page 122 in spru078e,
what's mentioned in the priority field seems different.
Why?

I am traceing the C of the example "sw_prioritized_interrupts"
mentioned in the document "C281x C/C++ Header Files and Peripheral
Examples".

It seems that cpu branchs to ISR to setup priority.Is this the way it
does? Or Is my understanding wrong?
(FILE:   DSP281x_SWPiroritizedPieVect.c)

If priority can be set in code,why priority are mentioned in these
document?

Dose someone know how to accomplish this task?

Bell
```

```Hello,
I work on  a 2812 dsp board, I make a c# application to test it and now
want to flash it.

On this board i have four Dsp (not interconnected), i use SDFlash wit
four com port and four sdp project.

I can create a process which will launch four SDFlash.
Does SdFlash can take arguments in a command line, like this ?
\SdFlash clientprjCom1.sdp
...
\SdFlash clientprjCom4.sdp

An other idea is to create my own window working like SDFlash.
In this case, do you know if it's possible to have SDFlash source code
or informations about c++ input methods that i can call to flash dsp.

A last question, does it exist an industrial method to flash more fastly
several dsp.

Laurence

```

```Hi everybody !!!

Can someone help me to solve the following problem ?
I need to use an external signal in my program in order to make a goo
enslavment of a signal.
So what, i want to get a very accurate value of my external signal.
I use the ADC module of the DSP 2812 but when I look a the value returne
there's a lot of noise on it.
Can someone explain me why this measure is so noisy and how i coul
improve it ?
Thanks a lot

Laurence

```

```Hi All,

I want to call SCI_BOOT in my program in order to use SDFlash to update
the flash. The resistors are set up in such a way that it boots from
flash and I don't want to change that. Or can the DSP-resetted in such
a way that it ignores the hardware and boots from SCI_BOOT anyway.

Kind regards, Frank

```