GPS L2C Implementation on E100 FPGA

GNU Radio
1

Hi all,

I am generating my masters thesis proposal and I have a few questions
about
the capabilities of the E100 and the feasibility of my project. I have
read
other posts to gather information, but correct me if anything I’ve
determined is wrong. My goal is to implement an 8 channel GPS L2C
receiver
on an FPGA. I am interested in doing this with the E100 since I may be
able
to take advantage of the ARM processor as well. Can anyone with
experience
in this area provide advice?

I would like to use the RFX1200 to receive the L2C signal since it has
better performance characteristics than the DBSRX. This would prevent
me
from first acquiring L1 signals to aid in acquisition, so I propose
using
preloaded almanac data to estimate SV position/Doppler frequency. This
would reduce the number computations when acquiring L2 CM code. The end
result only has to go as far as to decode the navigation bits: I will
not
need a position solution.

The root of my question is: does the FPGA on the E100 have enough space
to
handle this project?

Does it sound like I’m on the right track?

Thank you in advance!
Ryan Wolfarth

2

On Thu, Sep 1, 2011 at 8:42 AM, Wolfarth, Ryan [email protected]
wrote:

I would like to use the RFX1200 to receive the L2C signal since it has
Does it sound like I’m on the right track?

Sounds like a very interesting project. I have two suggestions:

  1. The SBX daughterboard has a better noise figure (5dB vs. 8dB typ.),
    with
    better linearity, and would be better suited for GPS work, especially if
    you
    aren’t using front-end filters. SBX will also receive both the L1 and L2
    frequencies.
  2. The E110 is identical to the E100 except that it uses the XC3SD3400A
    FPGA
    instead of XC3SD1800A, and so has additional FPGA resources.

Best,
–n

3

On 01/09/2011 1:08 PM, Nick F. wrote:

interested in doing this with the E100 since I may be able to take
position solution.

with better linearity, and would be better suited for GPS work,
especially if you aren’t using front-end filters. SBX will also
receive both the L1 and L2 frequencies.
In the specific case of GPS, most active GPS antennas (which is what
most people use) have GPS-optimized LNAs with very low noise figure,
and easily enough gain to swamp the noise figure of whatever
receiving hardware is downstream, so the noise figure of the receiver
board is largely irrelevant in most cases with GPS active antennae.

In my radio astronomy work (where Tsys is absolutely critical), as long
as the downstream receiver has a noise figure below about 15dB or
so, I generally don’t care what it is, since I use a tuned vLNA in
front of it. It’s a very similar situation with GPS.

4

Thanks for the input! After studying the problem further we’ve decided
that
this project will be too much for one person to complete in the given
time
span. I’ll be moving to a post processing software implementation of
L2C
and L5 signals instead.

Thanks again,
Ryan