# Coded long pulse test with Ch

# 25oct09

Links to plots:

The
average bandpass shape for 10 seconds of data (.ps) (.pdf):

Dyanmic
spectra for 10 seconds of averaged data (.gif):

A test was done with the rdev
spectrometer on 25oct09 using the ch 430 transmitter and
the coded long pulse program.

### The setup:

- ch tx,receiver using crr2 centered at 430 Mhz.
- coded long pulse program run with 440 usec pulse, 10 millisecond
ipp.

- Rdev setup:
- 160 Mhz clock, decimate by 5: 32 Mhz Bw, 16 bits output,
data rate 32 Mhz i,q.

- txSamples: numsamples:20000
- hghtSamples: start: (20000+30000)/32e6=1562.5 usecs or 234.775
km range.
- compute 1591 heights at 1 usec height step (234km-473 km) range

- Set rms in counts at digitizer to about 90 counts on the data.
It was higher than this since the rms included the rcvr blanking.

- Data file: used
'/share/tdevs0/sasdr.20091025.b0a.00000.pdev -> 00011.pdev .. 12
files.

### The processing:

The data
processing consisted of:

- Input 1000 ipps (10 seconds of data).
- for each ipp

- for each height of an ipp
- Decode, fourier transform, compute spectral density, and
accumulate in height buffer

- Do this for 1000 ipps.
- Compute the median bandpass for all heights.
- Fit a 1591 heights at each ipp, compute power, and accumulate
each heights power separately.

The display
processing consisted of:

- Compute the average bandpass over all heigths to show the
bandpass shape.
- To compute the dynamic spectra:
- use 422.5 to 438.5 . This is 8192 data points.

- for each height, fit a 1st order polynomial and a 23rd order
harmonic to 423 - 437 Mhz.
- divide this height by the fit then subtract 1.

- reduce the image to displayable dimensions:
- from 8192 to 1024 channels using a peak hold on groups of 8
channels.
- average adjacent heights to go from 1591 to 795 heights.

The average bandpass shape for 10 seconds of
data (.ps) (.pdf):

- The vertical scale is db above Tsys (defined as the median value
424-436 Mhz).

Dyanmic spectra for 10 seconds of
averaged data (.gif):

- The plasmaline gets lost at the peak because of the bandwidth of
the ch system.

processing:
x101/rdev/25oct09/test_25oct09.pro

<-
page
up

home_~phil