Project desert tortoise - Revision history

NebarnixWikiSysop at 18:21, 5 May 2016

2016-05-05T18:21:04Z

NebarnixWikiSysop: /* Result 1 Nov 2015 */

2016-05-05T17:58:52Z

Result 1 Nov 2015

NebarnixWikiSysop at 17:57, 5 May 2016

2016-05-05T17:57:50Z

NebarnixWikiSysop: /* Work */

2016-05-01T03:44:49Z

Work

NebarnixWikiSysop: /* Result 1 Nov 2015 */

2016-04-30T03:32:39Z

Result 1 Nov 2015

NebarnixWikiSysop: Added time sections

2016-04-28T14:10:40Z

Added time sections

NebarnixWikiSysop: added WFF TX site

2016-02-26T14:39:33Z

added WFF TX site

NebarnixWikiSysop at 14:31, 26 February 2016

2016-02-26T14:31:07Z

NebarnixWikiSysop: ass known transmitter list

2016-02-26T14:30:45Z

ass known transmitter list

NebarnixWikiSysop: Added links

2016-02-26T14:20:06Z

Added links

@@ Line 164: / Line 164: @@
   [[File:WFF_DCS5.jpg|250px]] [[File:WFF_DCS3.jpg|250px]] [[File:WFF_DCS4.jpg|250px]] [[File:WFF_DCS2.jpg|250px]] [[File:WFF_DCS1.jpg|150px]]
-== List of Discovered Transmitters==
+== List of Discovered Transmitters ==
-68961013 - NASA Goddard, Maryland, or in the area.
+68961013 - NASA Goddard, Maryland, or in the area.  <br>
 B90000 - NASA Goldstone, around (35.335948, -116.806606) <br>
 FB3AD000 - Telonics, Mesa, AZ (33.384355, -111.811219)<br>
 D81000 - NASA Wallops NESDIS (37.94648, -75.46059) <br>

@@ Line 146: / Line 146: @@
 === Result 1 Nov 2015 ===
-Results are shown below, the plotted points represent the solutions from two passes 20 minutes apart (NOAA15 and NOAA18). The distantly spaced models uses the 3 DOF model which assume a doppler scaling of 1.0 (just trust the raw frequency data) whereas the more closely spaced data uses the 4 DOF model that solves for slight calibration errors of the spacecraft. Both models solve for frequency offsets of the transmitter.
+Results are shown below, the plotted points represent the solutions from two passes 20 minutes apart (NOAA15 and NOAA18). The distantly spaced models uses the 3 DOF model which assume a doppler scaling of 1.0 (just trust the raw frequency data) whereas the more closely spaced data uses the 4 DOF model that solves for slight errors of the curves (but where are the errors coming from? Ionisphere? spacecraft calibration?). Both models solve for frequency offsets of the transmitter.
 <br>
 [[File:DopplerCurveFit.jpg|250px]] [[File:DopplerModelResultsVisualization2.jpg|250px]] [[File:DopplerModelResultsVisualization1.jpg|250px]] [[File:DopplerModelResults.jpg|250px]] [[File:DopplerModelResults3vs4.jpg|250px]]

@@ Line 166: / Line 166: @@
 == List of Discovered Transmitters==
-B90000 - NASA Goldstone <br>
+68961013 - NASA Goddard, Maryland, or in the area.
-FB3AD000 - Somewhere in Phoenix <br>
+B90000 - NASA Goldstone, around (35.335948, -116.806606) <br>
-D81000 - NASA Wallops <br>
+FB3AD000 - Telonics, Mesa, AZ (33.384355, -111.811219)<br>

← Older revision		Revision as of 03:44, 1 May 2016
Line 3:		Line 3:

	==Work==		==Work==
		+	===Doppler Model===
		+	The model is complex enough to need its own page... [http://wiki.nebarnix.com/wiki/Doppler_models]

	===Time===		===Time===

@@ Line 147: / Line 147: @@
 <br>
 [[File:DopplerCurveFit.jpg|250px]] [[File:DopplerModelResultsVisualization2.jpg|250px]] [[File:DopplerModelResultsVisualization1.jpg|250px]] [[File:DopplerModelResults.jpg|250px]] [[File:DopplerModelResults3vs4.jpg|250px]]
 === Result 2 Feb 2016 ===

@@ Line 3: / Line 3: @@
 ==Work==
 ===Orbital Determination===
@@ Line 90: / Line 131: @@
 ===Doppler Data===
-Doppler shifts produce sigmoid functions, so we should see a very specific shape with an inflection point where the satellite passes overhead and relative velocity reaches zero, then reverses. The last three bytes hold the key. The most data I have so far comes from a Baltic sea area transmitter (fishing ship? sea float?) with the ID 0x24000530 (in hex). There is probably enough data here to determine rough location, but more data to characterize the receiver range (units of frequency shift?) would be nice. Also, all [[Doppler_models | doppler calcs]] will show two identical points for every solution -- multiple passes must be captured in order to eliminate the incorrect solution.
+Doppler shifts produce sigmoid functions, so we should see a very specific shape with an inflection point where the satellite passes overhead and relative velocity reaches zero, then reverses. The last three bytes hold the key. Some good data I have comes from a Baltic sea area transmitter (fishing ship? sea float?) with the ID 0x24000530. By analyzing the bytes, we can figure out how to combine these three bytes to produce the frequency as received by the spacecraft.
 [[Image:DCS_0x24000530_suffix1.PNG|thumb|left|Three separate bytes, doesn't look right]]
@@ Line 98: / Line 139: @@
 === Latest Doppler Shift Model for Geolocating ARGOS Transmitters ===
-A matlab script that interfaces with an updated version of the above python script has been created that does pretty much all of the hard work for you. You must provide Spacecraft ID (15, 18, 19) T0 time and date of the time axis (xdata vs ydata provided) and there is a small calibration factor to the orbital period that you can use to run a second pass if required after the first pass. This script will attempt to find the best position and spit out the latitude and longitude but please be aware that there are always TWO possible solutions, and they become more and more fuzzy the fewer the data points you feed it. Multiple satellite passes are required to discern which location is the correct one!
+A matlab script that interfaces with an updated version of the above python script has been created that does pretty much all of the hard work for you. You must provide Spacecraft ID (15, 18, 19) T0 time and date of the time axis, and the time and frequency data. There is a function to fine tune the orbital period and and satellite radius to minimize the error directly at the frequency inflection point by providing a fudge factor for each (which is different for every solution set). This script will attempt to find the best position and spit out the latitude and longitude. Be aware that there are always TWO possible solutions, and they become more and more fuzzy the fewer the data points you feed it. Multiple satellite passes are required to discern which location is the correct one!
 http://wiki.nebarnix.com/wiki/File:Doppler3DRotationInclination3FitandWebephem.m

← Older revision		Revision as of 14:30, 26 February 2016
Line 105:		Line 105:

	[[File:DopplerCurveFit.jpg\|250px]] [[File:DopplerModelResultsVisualization2.jpg\|250px]] [[File:DopplerModelResultsVisualization1.jpg\|250px]] [[File:DopplerModelResults.jpg\|250px]] [[File:DopplerModelResults3vs4.jpg\|250px]]		[[File:DopplerCurveFit.jpg\|250px]] [[File:DopplerModelResultsVisualization2.jpg\|250px]] [[File:DopplerModelResultsVisualization1.jpg\|250px]] [[File:DopplerModelResults.jpg\|250px]] [[File:DopplerModelResults3vs4.jpg\|250px]]
		+
		+	==Discovered Transmitters==
		+	01B90000 - NASA Goldstone
		+	FB3AD000 - Somewhere in Phoenix
		+	01D81000 - NASA Wallops

@@ Line 1: / Line 1: @@
 ==Problem Statement==
-I found an embedded data stream in the NOAA DSB broadcasts that contains doppler data from ground-based transmitters. These messages contain short bits of data (up to 64 bytes) and the time vs frequency data required to geolocate them to a km or so (in theory). My friends at the hackerspace kept making jokes about using to find a desert tortoise with a huge transmitter glued to its back, and thus the project name was born...
+I found an embedded data stream in the [[NOAA_POES_TIP_Demodulation|NOAA DSB broadcasts]] that contains [[Doppler_models |doppler data]] from ground-based transmitters. These messages contain short bits of data (up to 64 bytes) and the time vs frequency data required to geolocate them to a km or so (in theory). My friends at the hackerspace kept making jokes about using to find a desert tortoise with a huge transmitter glued to its back, and thus the project name was born...
 ==Work==