[OpenSource_GPS] GPS-determination velocity with carrier phase

[Filipe Santos]

Hi Tak,

the time span between to two consecutive phase measurements is aprox. 1 
second, and i get the observed range rate from derivative of observed  
integrated  carrier phase, but i think that the problem is not here 
because i use the derivative of pseudorange but there have the same 
offset, only with expected more noise. I use the superstar II receiver 
with carrier option at 1Hz...

for calculation of velocity of satellite i use the derivative of 
equations of ephemeris, and i get the same values that is equal to 
derivative of position at the time transmitted  less and more dt/2... 
Could be the relativist effect?
I have made the correction, but is the same. I do like this:
Xvels=Xvel-[X(2) -X(1) 0]'*Omegae_dot;
and at the end, the result:

    pos(5:8)= inv(B'*B)*B'*omc_b;

and i subtract
    pos(5)=pos(5)+pos(1)*Omegae_dot;
    pos(6)=pos(6)-pos(2)*Omegae_dot;

and B and omc_c is obtains like this:

  for kl=1:length( tabela(:,1))
               if( tabela(kl,1)== sats(i) & tabela(kl,4)==0 & el>20)

                   h=[ (-(Rot_X(1)-pos(1)))/obs(i)...
                     (-(Rot_X(2)-pos(2)))/obs(i) ...
                     (-(Rot_X(3)-pos(3)))/obs(i)];
                   B=[B;h 1];
                  
                   %Xvels=Xvel-[X(2) -X(1) 0]'*Omegae_dot;
                   %Xvels=Xvel-[pos(1) -pos(2) 0]'*Omegae_dot;
                   Xvels=Xvel;
                   %Xvels = e_r_corr(traveltime,Xvel);
                   %lambda1=1;
                   vel_dl=(tabela(kl,3))*lambda1+dtcorr*v_light+h*Xvels;
                  
                   omc_b=[omc_b; vel_dl ];

               end
           end

You have say:

If you take an "average" range rate using two phase measurements, i.e. 
[v(t+dt)+v(t)]/2, the "observed" range rate is not instantaneous but in 
the middle of the time span, dt, by assuming a constant acceleration 
during that time. This might give you those small offsets in your 
velocity estimates.

do you think this offset can be this magnitude?

Another question, can anyone help-me with documentation with this basic 
problem of formulation on velocity of gps  and the correct corrections?
 

thanks,
Filipe Santos



tebinuma wrote:
> Hi Filipe,
>
> How much is the time span between two consecutive
> phase measurements and how do you get "observed"
> range rate from them?
>
> The GPS satellites are accelerating at <1G. That 
> gives you a few m/s difference in range rate every
> sec.
> If you take an "average" range rate using two phase
> measurements, i.e. [v(t+dt)+v(t)]/2, the "observed"
> range rate is not instantaneous but in the middle of
> the time span, dt, by assuming a constant acceleration
> during that time. This might give you those small
> offsets in your velocity estimates.
>
> Hope this helps,
>
> Tak
>
> --- Filipe Santos <fbnsantos at gmail.com> wrote:
>
>   
>> Hello,
>>
>> Sorry for my email, but I'm looking for help on GPS
>> for determination 
>> velocity...
>> I'm working on GPS data, and I'm using the carrier
>> phase for 
>> determination of velocity, but in static test the
>> results have a bias  
>> (like offset -1m/s , 1.4m /s and 0.4 m/s) , what
>> could be the problem? 
>> can anyone help-me?
>>
>> I use
>>
>> obs = Hv
>>
>> and H is the direction cossenos
>> obs= derived( phase) - dtcorr*v_light + Hv_{velocity
>> of satellite, given 
>> by derivation of equations of ICD  }
>>
>> were dtcorr is the derive clock drift = Eph(2,k)*dt
>> + Eph(20,k)
>>
>> and solve this by means of least squares..
>> should I subtract something else? the relativist
>> term should be used?
>>
>> the ionosphere and troposphere could be neglected  ?
>>
>> I have attached one graf witch the red is velocity
>> derived by position 
>> and blue is the estimation by using carrier
>> phase....
>>
>> many thanks,
>> Filipe Santos
>>     
>>> _______________________________________________
>>>       
>> opensource_gps mailing list
>> opensource_gps at lists.psas.pdx.edu
>>
>>     
> http://lists.psas.pdx.edu/cgi-bin/mailman/listinfo/opensource_gps
>   
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com 
>
>