george seifert Posted December 3, 2009 Report Share Posted December 3, 2009 Has anyone created an envelope detector? I have a 300 KHz signal that will be on for varying amounts of time (20 to 40 msec) with a gap between the pulses of about 10 msec. I'd need to figure out the actual on time of the signal. I suppose I could use a peak detector to find all the peaks in the burst and then determine when the peaks drop out, but that just seems kind of clumsy. Any better ideas? George Quote Link to comment
Gary Rubin Posted December 3, 2009 Report Share Posted December 3, 2009 Has anyone created an envelope detector? Envelopes are typically rectangular, and have little triangular flaps with a strip of nasty-tasting glue. You just need to pattern match on that Sorry. Couldn't resist. Quote Link to comment
BobHamburger Posted December 5, 2009 Report Share Posted December 5, 2009 Use your favorite low-pass filter; that's all that an envelope detector really is anyway. Quote Link to comment
Gary Rubin Posted December 5, 2009 Report Share Posted December 5, 2009 Use your favorite low-pass filter; that's all that an envelope detector really is anyway. It may have to be more complicated than that, depending on whether can or can't teal with the reduced time resolution associated with the LPF. Maybe a combination of LPF + peak detection to find coarse position, then peak detection on the raw signal to get finer position. Quote Link to comment
Norm Kirchner Posted December 5, 2009 Report Share Posted December 5, 2009 Has anyone created an envelope detector? I have a 300 KHz signal that will be on for varying amounts of time (20 to 40 msec) with a gap between the pulses of about 10 msec. I'd need to figure out the actual on time of the signal. I suppose I could use a peak detector to find all the peaks in the burst and then determine when the peaks drop out, but that just seems kind of clumsy. Any better ideas? George What kind of Eb/No and signal are you working with? Digital/Analog, constant envelope, Peak/Average ratio. I would imagine you would do a power vs time analysis and just look for the power edge. Quote Link to comment
rpursley Posted December 7, 2009 Report Share Posted December 7, 2009 Has anyone created an envelope detector? I have a 300 KHz signal that will be on for varying amounts of time (20 to 40 msec) with a gap between the pulses of about 10 msec. I'd need to figure out the actual on time of the signal. I suppose I could use a peak detector to find all the peaks in the burst and then determine when the peaks drop out, but that just seems kind of clumsy. Any better ideas? George Take a look at using a Hilbert Transform on your pulses. Create an analytic signal (original signal + j (Hilbert transform of original signal)) and the magnitude of this signal will be the envelope. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.