I have owned and used radio scanners for many years, and loved them as my posts before December 2011 will testify.
In that month I became the proud possessor of a FUNCube Dongle Plus and discovered the joys of software defined radio, since then I purchased a FUNCube Dongle Pro+ and extended my SDR adventures in to the realms of HF and I have several of the insanely cheap RTL2832 based dongles.
As much as loved my scanners there was a major flaw with them, which has been brought into sharp focus now that I have used SDR.
No matter how fast or as sensitive as the scanner is are you are still playing a game of chance. You are limited by the frequency steps, demodulation modes and scanning rate of the receiver and you could zip through the band all day and still miss those elusive signals.
SDR and the waterfall display is a revolution, you can view a portion of the spectrum in real time and actually see the signals, they may be short lived bursts of data and voice, or continuous data transmissions.
The RTL-SDR dongles excel in this respect with their wide sampling rate you can view up to 2MHz of the spectrum at once, the following images show typical waterfalls captured this morning using one of my RTL-SDR dongles.
The first one, shows the cluster of data channels (was the old Vodaphone Paknet system) around 164.2 – 164.4 MHz, a trunking control channel and various data bursts, which are mostly Taxi Mobile Data Terminal (MDT) transmissions.
A little lower down the spectrum and another trunking control channel, a speech conversation, more data bursts and a faint digital channel.
Further up the spectrum into the UHF, a cluster of data transmissions.
Simply moving the cursor on the display and you can hear the transmission, if necessary change the demodulation type, widen or narrow the filter bandwidth, save the frequency. If you capture the IQ file you can then replay it endlessly tweaking and refining until you extract the information you want.
During the weekend I was experimenting and noticed there was lots of data bursts in the 163-168MHz range, I confess that I already knew what most of them were as I have experimented before with a scanner (with a discriminator tap) and Ian Wraith’s Java based Taxi MDT decoder. I decided to reinvestigate them using the RTL-SDR as the receiver.
While many taxi companies still use voice transmissions, many have adopted automated data terminal systems, where the dispatcher sends information about jobs to terminals in the cars, the drivers then can accept jobs, get information and send information back to the dispatcher.
Ian’s decoder which requires the Java runtime environment decodes systems that use the same physical layer as MPT1327 i.e 1200 Hz and 1800 Hz tones transmitted at 1200 bps. The two main systems used in the UK, are the Autocab and Auriga. The Taxi MDT Decoder currenly decodes the Autocab, but the coding for the Auriga system is still an unknown, so just outputs the raw data.
More information about Taxi MDT Decoder can be found here I confess to having one slight niggle with it, often I couldn’t get it to accept sound from the selected input. A work around I found was to first open the Audacity sound editor which I had installed and select the input and start a recording, then opening the decoder seems to make it work!
Ian has also written the excellent DMRDecoder which allows analysis of the DMR digital mode which is becoming more widespread. I intend to post some details soon about decoding digital modes, keep watching.
I created a video showing the Taxi MDT Decoder in action, the quality is pretty dire but you can get the idea, I identify the Auriga as being encrypted, it might be but as nobody on the team knows the protocol yet!