From RevSpace
Jump to navigation Jump to search
Event Foundations Friday: Software Defined Radio workshop
Name Foundations Friday: Software Defined Radio workshop

Duration 2019-06-28 20:00 - 2019-06-28 23:00

* Location: Overgoo 1, Leidschendam
* Open from: 20:00

Software Defined Radio workshop: receiving a wireless microphone

1. Goal of the workshop:

At the end of the workshop (ca two hours), participants will be able to

  • Receive a simple wireless microphone signal from the presenter
  • Using their own laptop
  • Using their own RTL-SDR dongle or other software-defined radio receiver
  • Understand a few basic concept of radio signals, signal processing, and software-based radio signalprocessing in particular.

2. Prerequisites:

Please bring your own radio-receiver:

  • A RTL-SDR hardware (very cheap, $20-$40 or less)
  • A funcube dongle
  • Other SDR hardware (USRP or other GNURadio supported SDR)
  • A laptop with GnuRadio and Gnuradio-companion installed (version 3.7.5 ... 3.7.11 is ok)

There might be a few available @ revspace coin-op machine by the time the workshop starts.

  • Optional: a gqrx installation or other software-defined radio/waterfall receiver with WBFM support

3. Quick introduction

Quick introduction into GNURadio companion 3.7.x user interface and concepts and keyboard shortcuts (have it preinstalled and git client to download the samplecode)

4. Theory of signal processing

This takes about half an hour, it is an interactive session so that we can adapt the speed a bit depending on prior knowledge of the participants. Questions in between are very welcome, we will be with a small group.

5. Radio receivers and/or detectors

  • Build a simple mono FM receiver using standard gnuradio blocks
  • Decomposing the function of the gnuradio blocks

The transmitter in question will be an ordinary wireless presentation microphone for stage-usage, that the presenter will bring and setup to transmit at a fixed frequency.

It will generate a mono, wideband FM signal in the 473-494 MHz band. The transmitter will also feature a tone-code of 35 kHz that will be inaudible to the public and normal receivers. The tone is used to detect nearby presence of the transmitter, and estimate the signal level.

The goal is to - find the exact frequency the presenter is using - demodulate the FM signal using basic gnuradio blocks - play back the audio from a local speaker or headphone

The latter might also introduce one into the problem of samplerate-conversions and free-running clocks between the radio-receiver, audio device sample clock, and a PC's local software clock

6. Practice of signal processing

The examples will be provided via git and just need to be started, so the focus is understanding the signal flow, making modifications predict what will happen, and then test the result in practice. This takes about another half hour

6.1 Real signals

Real sinewaves

 * In the timedomain
 *  Displayed using time sink / oscillioscope
 *  Displayed uisng frequency sink / FFT / spectrogram

Multiplying real-sinewaves, in the time domain

* Displayed using time sink / oscillioscope
* Displayed uisng frequency sink / FFT / spectrogram
* Usage of the word 'mixing' in radioterms (multiplication or usage of other non-linearities)

6.2 Complex signals

Quadrature or complex (IQ) sinewave signal sources

* Displaying IQ signals in timedomain ('constellation' diagram)
* Displaying IQ signals in frequency domain (using FFT or frequency sinks)

Multiplying quadrature (IQ) sinewaves and spectrogram

* Effect of quadrature multiplication (phases of are added) 
* Usage of complex multiplication in signal processing

6.3 Filtering complex signals

Filtering IQ signals using FIR filters with complex coefficient/taps.

  • FIR or Finite Impulse Response filters.
  • IIR filters will be skipped.
  • Hilbert transform filters might be skipped too (generating complex analytical signals from real ones).

6.4 Calculating FIR filter coefficients or parameters

Filtering coefficients/taps, impulse response of filter

 * Calculating FIR filter coefficients/taps using gr-filter 

Performance limitations, tradeoffs between

 * Computational complexity of filter (number of multiply/add operations per processed sample)
 *   Steepness of cut-off frequency
 *   Flatness of passband of filter (integrity of signal that still should pass through the filter)
 *   Depth and ripple of stopband of filter (flatness of suppression of unwanted signal)
 *   (Using inverse FFT/inverse DFT and windowing functions to determine filter coefficients/taps will be skipped)


(Nick)Name Has (some) SDR Reciever
cmpxchg Yes
Pepman Yes
GNyers Yes
Aloha Yes
Edward H. Yes
Atoomnet Yes