Re: [Discuss-gnuradio] A problem to the FFT Block

[Tom Rondeau]

On Wed, Nov 14, 2012 at 5:08 AM, Zhiwen He <address@hidden> wrote:

Thank you very much. I have solved the problem, I find that, I didn't use the multiply_const block. But there is another question, when I see the data in Matlab with plot function, the figure before fft and after ifft is not same, there are something wrong on the position, where the valueis 0 in the square source. why?

Thanks

Zhiwen He

I feel like there's a representational issue here. Why is your scale x10^9 (before the FFT) when the amplitude of the signal generator is 1? I'm also not clear what I'm looking at in the plotted figures. Can you label them?

Tom

 

On Wed, Nov 14, 2012 at 4:49 AM, Marcus D. Leech <address@hidden> wrote:

On 11/13/2012 10:38 PM, Tom Rondeau wrote:

On Tue, Nov 13, 2012 at 10:43 AM, Zhiwen He <address@hidden> wrote:

hello all,


When I use the fft blcok, it seems not right.

I build the test system like this:   Square source -> IFFT -> FFT -> File Sink,  when the input is 1, the output in the file sink is 1, and when the intput is 0, the output is random, sometimes 0, sometimes between -1 and 1.

Like this:

input:

						0	0	0	0	0	0	0							0	0	0	0	0	0	0							0	0	0	0	0	0	0
						0						0							0						0							0						0
						0						0							0						0							0						0
						0						0							0						0							0						0
						0						0							0						0							0						0
						0						0							0						0							0						0
0	0	0	0	0	0	0						0	0	0	0	0	0	0	0						0	0	0	0	0	0	0	0						0

output:

						0	0	0	0	0	0	0							0	0	0	0	0	0	0							0	0	0	0	0	0	0
						0						0							0						0							0						0
	0					0						0							0						0							0						0
0						0						0			0				0						0							0						0
			0		0	0						0		0		0			0						0		0					0						0
		0		0		0						0	0				0	0	0						0	0		0	0	0		0						0
						0						0							0						0						0	0						0

How should I do, so that the source data is the same as the sink data?

Thanks.

Heyutu

Hi Heyutu,

I just tried this right now and don't have a problem. This is my full graph:

sig_source (square wave) -> throttle -> stream_to_vector(fftsize) -> fft(fftsize, reverse) -> multiply_const(fftsize*[1/fftsize], fftsize) -> fft(fftsize, forward) -> vector_to_stream -> time sink

Tom

_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

I did the attached.

Note that I've turned off the default window function in both the forward and reverse transform

-- 
Marcus Leech
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org

_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio