matched filter of chirp signals

Question

In simulating linear frequency modulated (LFM) chirp signals, with the parameters: Pulse width, bandwidth.

I find the matched filter response to be this:

In particular, there is a increase in the energy over the right side of the plot.

My question is, is this normal, or should the tapering be going down?

think about this: since your pulse shaper is constant envelope (right?), your MF should be, too. The fact that you're seeing intensity variations at all is probably because of your analysis methods. — Marcus Müller, Jul 9, 2018 at 6:12

matthew · Accepted Answer · 2018-07-12 02:15:28Z

0

I found that the reason for the spike on the right is due to the inclusion of pulse width of time along the time axis.

answered Jul 12, 2018 at 2:15

matthew

955 bronze badges

$\begingroup$ where is this spike on the right? it looks like it should be just touching the bump on the right. where is Nyquist in this frequency-domain plot? this chirp signal must've been windowed with a window of some sort. it looks a little Hannish or Hammish. And it looks like the window closed before the frequency changed a lot (or is the spike at DC)? Turns out that if you're doing linearly-swept frequency chirps, windowing them with a Gaussian window allows for some mathematical properties that can be made use of. $\endgroup$
– robert bristow-johnson
Sep 10, 2018 at 7:17
$\begingroup$ a while ago i did another answer that discusses the ideal continuous-time model of a linearly-swept chirp with a gaussian window. $\endgroup$
– robert bristow-johnson
Sep 10, 2018 at 7:21
$\begingroup$ thanks! the "spike" refers to the gradually elevating response on the right side of the plot. Also, the plot x-axis is in time domain, and i simulated the chirps with rectangular windowing $\endgroup$
– matthew
Sep 17, 2018 at 13:06

Add a comment |

1 Answer 1