Simulating a Doppler shifted signal

[paulbeetee]

Hello,

I am trying to generate a sampled time domain signal of a single tone being produced by a source moving past a stationary receiver .

I can calculate the expected received frequency a s a function of time. I use this expression to generate a sinusoid whose frequency changes with time to represent what I expect the receiver to receive.

However, when I chop the time domain sgnal into chunks to fft and then plot in a 2-D figure so I can see the generated signal change with time I get a strange excursion in the indicated frequency to a much lower frequency than expected.

I've attached the example Mathcad sheet but I can't see what I'm doing wrong. Anyone got any ideas?

Regards
Paul Beattie.

[ptc-1368288]

>I am trying to generate a sampled time domain signal of a single tone being produced by a source moving past a stationary receiver.<<br> _________________

That sounds so much "Fizeau-Doppler" !

Your work sheet didn't arrive to destination.

jmG

[RichardJ]

On 7/2/2009 8:44:15 AM, paulbeetee wrote:
>
>I've attached the example
>Mathcad sheet but I can't see
>what I'm doing wrong.

There is no worksheet attached

Richard

[TomGutman]

There is no work sheet. But if you are working with frequencies that vary with time, you might find the discussion at http://collab.mathsoft.com/read?24444,10e#24444 useful.
__________________
� � � � Tom Gutman

[ptc-1368288]

In reference to the thread:

http://collab.mathsoft.com/read?24444,10e#24444

the Rk solution may suffer from the "Rk syndrome". By quick inspection, the given DE has symbolic solution. It is possible to isolate the homogeneous coefficients as well as the forcing coefficients, i.e: pilot the function externally. Surprising that you didn't attach the work sheet as reminded by 3 collabs.

jmG