[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] GNU radio

From: Nick Foster
Subject: Re: [Discuss-gnuradio] GNU radio
Date: Wed, 21 Sep 2011 11:02:36 -0700


I've written a 1090MHz Mode S/ADS-B receiver for Gnuradio which can use Ettus hardware (N200/N210, E100/E110, USRP1). It works pretty well (up to >250nm range with line of sight). I'll try to answer your questions inline below as well.

On Wed, Sep 21, 2011 at 9:17 AM, Andrew Rich <address@hidden> wrote:
Maybe I can list my aims and you can tell me if GNU and N200 can do this ?

1. Receive on 1030 MHz - BW not sure yet
Yes, up to 25MHz bandwidth with N200/N210. You'll want to use WBX, SBX, or DBSRX2 as your daughterboard.
2. Receive on 1090 MHz - BW not sure yet
Yes, up to 25MHz bandwidth with N200/N210. You'll want to use WBX, SBX, or DBSRX2 as your daughterboard.

Keep in mind you will not be able to receive on 1030 and 1090 simultaneously with a single board. I'm not sure if your application requires this. A USRP1 plus two daughterboards, or two N200s in MIMO configuration would accomplish this.
3. Receive on 2700 - 2900 MHz - BW not sure yet
Yes, up to 25MHz bandwidth with N200/N210. You'll want to use SBX or DBSRX2 as your daughterboard.
4. Classify signals on these bands.
That's totally up to you. Remember that a USRP will just give you straight, basically unprocessed digital RF samples from wherever in the spectrum you ask it to. The processing of those signals into something intelligible is your responsibility, and Gnuradio is a toolkit to make this, if not easy, at least easier than doing all your DSP from scratch.

For example, for 1090MHz PPM signals, if you create a dead-simple Gnuradio flowgraph which just does AM demod (complex-to-magnitude) on the input samples, you'll have a PPM signal that looks just like the output of your 1090MHz front end. Further processing in Gnuradio can turn that into intelligible Mode S data.
5. Perform PPM decode - Pulse Position Decode on the 1090 MHz band
See the gr-air-modes package for an example of a Gnuradio receiver doing just this. Might give you a good idea of what goes into a Gnuradio flowgraph since it's a format you're already pretty familiar with.

Using a hardware device and GNU radio

The computer would be a MAC MINI running openSUSE LINUX


MacBook PRO Laptop running LINUX

with either USB or Gigabit ethernet.

Either is fine. Recommend using an N200 + GigE.

What would be the suggested software and hardware combinations ?

Can I use the N200 as a very basic spectrum analyser and a capture device ( I guess the capture device would be just continuous)
If your bandwidth of interest is under 25MHz it will give you a good spectrum analyzer display. You can "piece together" a wider spectrum by re-tuning and capturing one 25MHz swath at a time.

I undestand what I want to do and I have started decoding singals on 1090 MHz already

I just want to turbo charge the process and make it as fast as my Mode S 1090 MHz receiver I have now, which is a 1090 MHz front end and FPGA

Depends on what you mean by "fast". The N200 can get 25MHz of bandwidth down to the host, and since the Mode S waveform requires only 4Msps you're covered with room to spare.


- Andrew -

----- Original Message ----- From: "Marcus D. Leech" <address@hidden>
To: "Andrew Rich" <address@hidden>
Cc: <address@hidden>
Sent: Thursday, September 22, 2011 1:47 AM
Subject: Re: GNU radio

On 21/09/2011 11:34 AM, Andrew Rich wrote:
I was just looking at the N200

Do these hardware components have sensitivity figures ?
That depends entirely on the daughterboard you chose.  Although most of them have noise figures in the 4-5dB range at maximum gain.
 If you're just interested in RX in the 1090MHz range, I'd suggest the DBS_RX2.  Sensitivity is dominated by noise figure.  If you need
 lower noise figures you'll have to put a band-specific LNA in front, which is what I do for radio astronomy.

I am interested in passive RADAR

I have been using a 1090 MHz receiver and a cheap digital OSCilloscope commaned under LINUX as a capture device.

I guess that is sort of what the hardware and software of an SDR does ?

My system is very slow

In an SDR, nearly-all the processing is done on the host computer, so you need a fastish computer.  Overall compute requirements
 are roughly proportional to sample_rate * complexity-per-sample.

Discuss-gnuradio mailing list

reply via email to

[Prev in Thread] Current Thread [Next in Thread]