Using Codar chirps to map the ionosphere in real-time

[RFSPACE, Inc.]

I've been playing with sudden ionospheric disturbance (SID) receivers. Initially I was looking at the signals around 24 KHz and plotting power over time. That got a little boring after a while so I decided to try something more exciting.
It turns out that there is a network of ocean radars around the world called Codar. Codar stands for Coastal Ocean Dynamics Applications Radar. The radar transmits chirp signals between 4 and 50 MHz. Most chirps are repeated every second with bandwidths of 18-100 KHz or so. These radar pulses scatter off the ocean waves and can be used to plot wave height and currents.

http://cordc.ucsd.edu/projects/mapping/maps/

This map shows the results of the Codar system.

It turns out that these radar chirps also illuminate the ionosphere and get refracted back to earth. The chirp signals make it really easy to measure the height of the ionosphere and measure other things like back-scatter and absorption.
The first thing that had to be done was write some software to "pulse compress" (aka de-chirp) the radar pulse. It turns out that the SpectraVue software already had a de-chirping algorithm to compress Pulsar transmissions. After about a week or hard work, the code was optimized (thanks to AE4JY) to properly compress the 1 second signal into a short pulse.

Initial tests were run at 4543 KHz. This transmitter is on the beach close to Destin, Florida. It is close enough to Atlanta, GA that it has groundwave most of the day. This makes it easy to calibrate the height of the ionosphere since you get the "zero" time reference signal. The 4543 KHz transmitter chirps at around -18378 Hz/second. This was calculated empirically by adjusting the chirp bandwidth to get the most narrow compressed pulse.

The hardware used is an RFSPACE SDR-IP receiver synchronized in frequency and phase to GPS using a Trimble Thunderbolt receiver. The SDR-IP samples at 32 KHz and it pulse compresses the pulse in the frequency domain using fast correlation inside the SpectraVue software. The following link explains some of the math involved.

http://www.ittc.ku.edu/workshops/Summer2004Lectures/Radar_Pulse_Compression.pdf

Results:

The graph below shows the 4660 KHz Codar transmitter observed from Atlanta, GA. The image covers 2 seconds per pixel from left to right. It covers 1500km from bottom to top (5ms). The total duration is 12.66 hours starting at 0033UTC on May 28th. It shows the ground wave all the way on the bottom fading in and out. The different refracting and back-scattering regions are shown all the way to 1500 km (top). Right around sunrise, everything is lost for a couple of hours.

http://www.rfspace.com/experiments/4660MHz_1500km_high_45576seconds_long_0033UTCstart_28May2011.png

The previous night, similar behavior can be observed around sunrise:

http://www.rfspace.com/experiments/CODAR4660_may273UTC.png

It will be interesting to observe what effects high plasma from solar flares will have on the different regions in real-time. If someone would like more details, please email me.

Pieter --
N4 IP

[Al Aburto]

Fascinating! The Gulf Stream comes in loud and clear off Georgia and Florida!

But what you're doing is even better! Wow.  Excellent work!

Also, pulse compression, what does that mean in terms of long range, low power, communication and detection I wonder?  Thinking about SETI actually...
Al

--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en

[Robert Fritzius]

Awesome! Any idea as to why the bold activity centered on 400 km seems to taper away apparently backwards in time at the sunrise interruption region? --- On Sat, 5/28/11, RFSPACE, Inc. <rfs...@gmail.com> wrote: