🚀 go-pugleaf

Hello,

	I am looking for an algorithm that while generate the orbital positions 
of  a GPS satellite.  I am trying to generate an ephemeris file.  I have one 
algorithm but I can't not get the output data to match the values distributed by 
the U.S. Coast Guard.  The information I have is as follows:
the Epoch, the semi-major axis, the eccentricity, the right ascension, the 
argument of perigee, and the mean anomaly.

The algorithm I am using is coming from the book ""Satellite 
Communications" byTimothy Pratt and Charles Bostian  copyright 1986 John 
Wiley and Sons.  

1. calculate the angular_velc
	angular_velc = 1/semi-major_axis * 
	sqrt(MU/semi_major_axis)*Radians_to_Degrees.
	MU = 398601.352

2. find the time of perigee
	time_of_perigee = Epoch_date_converted_to_seconds - 
	mean_anomaly/angular_velc

3. find the mean anomaly at the analysis date.
	mean_anomaly = angular_velc*(analysis_date - time_of_perigee)

4. find E - eccentric anomaly
	mean_anomaly = E - sin(E);
	solved in an iterative approach and is around mean_anomaly + or - 1

5. find r0 - radius from the origin of the coor-system to the satellite.
	r0 = semi-major_axis*(1.0 - eccentricity*cos(E))

6. find the true_anomaly
	I solved this equation for the true_anomaly

	temp = ((semi-major_axis)/(r0*eccentricity))*(1.0 - 		
	eccentricity*eccentricity)) - 1.0/eccentricity

	true_anomaly = acos(temp)

7. find the x0 and y0 values.  There is no z component because the equations 
are derived in the z plane
	x0 = r0*cos(phi)
	y0 = r0*sin(phi)

the last part solves for the xi,yi,zi triplet.  This is the triplet I am using to 
compare my results from the program called GSS.  I am getting x and z values 
to be very close to the output from the program and I am not getting the y 
values close at all.  The next part is complex because I had to solve the matrix 
to get the equations.  I am going to use some shorthand to simplify the 
equations.
	a = cos(right_ascension)	b = sin(right_ascension)
	c = cos(arg_of_perigee)	d = sin(arg_of_perigee)
	e = cos(inclination)	f = sin(inclination)

	a11 = a*c - b*e*d
	b11 = -a*d - b*e*c			the matrix
	c11 = -b*d + b*e*c	|xi|	| a11	b11	na | |x0    |
	d11 = a*d + b*e*c	|yi|  =	| c11	d11	na | |y0    |
	e11 = d*f		|zi|	| e11	f11	na | |z0 = 0|
	f11 = c*f

	xi = a11*xo + b11*y0
	yi = c11*xo + d11*y0
	zi = e11*x0 + f11*y0

after implementing the algorithm I ran the program I was getting out data that 
looked like the following
	x0 = 15254.499(km)	y0 = -21551.16(km)
	xi = -17367.91923	yi = -829.09		zi = 15610.593
From the GSS program
	xi = -17207.79		yi = 12394.955		zi = 15732.192
I have compared this results with the  U.S. Coast Guard and they seem close, so 
I can assume my algorithm has the problem.

I solved for the c11 and d11 coefficient and should be getting results for c11 to 
be -.0057 and d11 to be-.57.  After doing some manipulation of changing the 
equations the best I could do was to come up with a c11 = -.0047 and a d11 = -
.642 and this brought my yi to be 13764.811 still far away from the expected.  I 
have run with different data at different time and still get the same type of 
results.  The x and z are good but the y is wrong. 

I need some help in figuring about a different algorithm that will work and give 
me close data are find out what I am doing wrong in this algorithm.

Also,  in the Yuma file format I need to know how to convert the Right Ascen 
at TOA which is from the GM(Greenwich meridian) to the right ascen from the 
Vernal Equinox.  And how to convert the Mean anom in yuma format to the 
almanac format.  In the case of the right ascension value it is 26.7 degrees 
different from the almanac file I am using and the Mean anon is about 59.9 
degrees different.  This causes a problem because the algorithms assume the 
angles from the VE not the GM.

Any help these areas would be very useful.  I am starting to run into deadlines 
and need all the help I can get.  

If you post the response to the net for the others to gain information, please 
email me also.  I don't get the time to read much news and would not like the 
response to get lost.

thank you
shane gustafson

korin@eng.umd.edu