[Discuss-gnuradio] Decoding CCSDS

[Michael Sabino]

Hi,

I figured out that adding a descrambling block with mask 0x81, seed
0x7f, and len=4 enabled me to get some interesting strings related to
the AIM spacecraft. I went from recording with a 3rd party tool to
demodulate BPSK -> [viterbi CCSDS27 decoder in the gnuradio flowgraph]
-> [descrambler with mask=0x81,seed=0x7f,len=4] -> file sink. I'm not
sure if it's actually scrambled or not, or I did something wrong in
the recording that left two peaks on the histogram in the demodulator
output near 0x7f and 0x81. Not sure. Either way, this seems to fix
most of it.

I'm trying to output the frames in a format readable by wireshark so I
understand that I need to run this through a Reed-Solomon decoder
first. I'm unfamiliar with Reed-Solomon, how much variety is there in
the choice of parameters for CCSDS spacecraft? it should be 223,225
correct?

Are there any blocks I can use for Reed-Solomon decoding that will
enable me to output the frames to a file so I can open in wireshark?

PS: Please disregard the previous message. My IE browser locked up and
accidentally hit send.

Michael Sabino

On 2/1/16, Michael Sabino <address@hidden> wrote:
> On 1/30/16, Michael Sabino <address@hidden> wrote:
>> I'm going to post my flowgraph. I have added a CCSDS decoder block.
>> Please, if you can, provide input on how to improve the amount of
>> successfully decoded data with a short wave file and a doppler
>> shifting BPSK carrier.
>>
>> My .wav file: http://1drv.ms/1bS3KxX
>>
>> and here is my gnuradio flowgraph: http://1drv.ms/1m3o3N6
>>
>> Using several different decoding methods, I have obtained strings
>> found in these two documents:
>>
>> http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19880002723.pdf
>> http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19800022222.pdf
>>
>> (for the Aeronomy of Ice in the Mesosphere spacecraft). One is old
>> NASA teleoperator system reference, if that helps.
>>
>> However, the gnuradio method (I was also using a seperate program
>> written for an SDR analysis/display utility called SDR#/SDRSharp), is
>> presently suboptimal.
>>
>> I have some times where I see a BPSK constellation, but other times
>> (some in between the good constellations) I just see a ring where it
>> loses synchronization. Please provide help on how to improve my BPSK
>> demodulator.
>>
>> Thanks
>>
>> Michael Sabino
>>
>> On 1/27/16, Achilleas Anastasopoulos <address@hidden> wrote:
>>> The problem is not the length 2048 but the way you use the fsm
>>> constructor
>>>
>>> trellis.fsm(2,7,[79,109])
>>>
>>> You are trying to create an FSM for a (7,2) CC.
>>>
>>> This means you need to provide a vector of 2x7 integers, each
>>>
>>> representing the generator polynomial between the i-th input (i=0,1)
>>>
>>> to the j-th output (j=0,1,...,6)
>>>
>>>
>>> However you only provide 2 of them.
>>>
>>>
>>> Are you sure this is what you want to do?
>>>
>>> ------------
>>>
>>> this is a snippet from fsm.cc
>>>
>>>
>>> //######################################################################
>>>     //# Automatically generate the FSM from the generator matrix
>>>     //# of a (n,k) binary convolutional code
>>>
>>> //######################################################################
>>>     fsm::fsm(int k, int n, const std::vector<int> &G)
>>>
>>
>