Hello everybody, I have been following this forum for a long time, but I subscribed just now. I am into GNSSs and, currently, I am developing a Galileo software receiver. The problem I will tell you about is by the way encountered also in the gps. i used the search function, but i've only found a topic mentioning only what i know. i hope you will find some spare time to help me :) let us consider a receiver that is in tracking. I manage to get this ;) then I also manage to measure arrival times, both considering only the transmission time (i would refer to this case as the "common transmission time") and only the state of the registers (i would refer to this case as the "fixed reception time"). now, what i do is to compute misalignments between channels, waiting for a certain bit or considering the registers, and this is good if i decide to use a least squares pvt computer. by the way, i want to implement a kalman. the kalman itself is almost a piece of cake, since i can make it work with the pseudoranges given by other receivers. the problem is my set of pseudorange measurements. in fact, if i use the pvt, at each pvt sampling instant i just set the first front (or elder register) to 68.8ms (traveltime) and iterations give me a correct estimate of the pvt solution (actually, at the moment, i implemented only the pt solution, but i have already implemented pvt before) by the way, if you use a Kalman filter, you do not iterate it, but you run it just once, so it is no good to use a fixed travel time, since it won't get correct and it will provide a biased solution. i tried to use the clock bias estimated by the least squares block at the former step, but it didn't work. do you have any suggestion/idea/truth? :) synthetically: how do you compute the pseudoranges/ranges to be fed to the kalman filter? cheers