DV4mini RPi Image Updated to Raspbian Stretch

The K2DLS DV4mini image for the Raspberry Pi 2 and 3 has been updated to Raspbian Stretch.  The DV4mini client and BMXTG 1.3 start automatically upon boot.  VNC has been updated to RealVNC.  Both VNC and ssh are started by default.

The installation fits nicely on an 8 GB SD card and it looks great on my 7″ touchscreen display.

The default password for the pi user is “raspberry” — please change it immediately!

The image may be found here.

Solar Eclipse Data Collection Project

Between 1600 – 2015 UTC on August 21, 2017, as the solar eclipse swept across the nation, I captured much of the lower 2 MHz of the radio frequency spectrum.  I used a Microtelecom Perseus SDR, a 130′ inverted L with four radials, and lots of disk space.  In doing so, I have created a permanent record of this portion of the RF spectrum during the solar eclipse.

I am making the spectrum capture files available for your analysis and research.  Each file contains a 5 minute segment.  If you download a group of files, they will play in succession.

You can use the demo version of the Perseus software or any other software that can read the Perseus data, such as Linrad or HDSDR.  You cannot use just an audio player to play the files, even though they have .wav extentions.

Should you perform any analysis or otherwise make use of the files, I’d like to hear from you in the comments below.

Special thanks to Jav, K4JH, for donating the bandwidth for this effort.

BMXTG 1.3 released on Github

I’ve added BMXTG to Github and updated it to version 1.3.  File locations have been formalized for consistency and I’ve created a .deb package for easier install.

Brandmeister XTG Dialer v1.3 by K2DLS

If you have a running DV4mini build and want to add BMXTG, grab the .deb file from the link in the README.

To install:

sudo dpkg -i bmxtg-1.3.deb

After installing, copy the desktop file to your Desktop directory.  This will make the desktop icon available.

cp /usr/local/share/bmxtg/bmxtg.desktop ~/Desktop/

 

aprx and weather reporting

I’ve been sending weather reports via APRS-IS and RF for some years and have recently re-architected the way it works.  My APRS station uses aprx software running on a Raspberry Pi Model B.

My weather station is a Davis Vantage Vue with WeatherLinkIP module. The module allows the weather station to plug directly into my ethernet network. I formerly used WeatherLinkIP to feed the data to CWOP which would in turn show up on APRS-IS.  I would then gateway my reports back to RF using an aprx filter.  This seemed convoluted to me, so I wanted to improve things.

I now use a program called Weather Display, a very capable weather program, which has the capability to generate an APRS WXNOW.TXT file.  Weather Display can directly poll the Davis station via IP.  Through the steps documented below, I now send my weather reports via RF and APRS-IS in the same step.

Every 10 minutes, via crond, I perform some preprocessing on the WXNOW.TXT file to embed the weather report in the APRS “Complete Weather Report Format — with Lat/Long position and Timestamp” and then secure copy the file over to my Raspberry Pi APRS system:

#!/bin/bash
#
if [ ! -f ~/tmp/WXNOW.TXT ]; then
   exit 1
fi
# Change to the Lat/Long of your weather station
LATLONG="4023.75N/07412.53W"
line=0
while read FILE; do
   if [ $line = 0 ]; then
      # Change America/New_York to your timezone
      TIME=`TZ=UTC date --date="TZ=\"America/New_York\" $FILE" +%d%H%M`
      line=$((line+1))
   fi
WX=$FILE
done < ~/tmp/WXNOW.TXT # echo /$TIME\z$LATLONG\_$WX > ~/tmp/wxnow.tmp
# 
echo `cat ~/tmp/wxnow.tmp | tr -d '\r'`XDsIP > ~/tmp/wxnow.txt
# You must setup ssh key based authentication for this to work
# Another method could be a file copy via NFS or CIFS
scp ~/tmp/wxnow.txt pi@aprs:/dev/shm/wxnow.txt
# 
rm ~/tmp/wxnow.txt ~/tmp/wxnow.tmp

On the APRS host, I defined the following beacon section in /etc/aprx.conf:

beaconmode both
cycle-size 10m

beacon via WIDE2-1 \
srccall N0CALL-13 \
exec /usr/local/bin/aprx-wxnow.sh

Finally, the beacon exec script, is installed in /usr/local/bin/aprx-wxow.sh:

#!/bin/bash
#
TIME=$(printf `date -u "+%d%H%M"`)
if [ -f /dev/shm/wxnow.txt ]; then
   if [ -f /dev/shm/wxold.txt ]; then
      FULLWXOLD=`cat /dev/shm/wxold.txt`
   else
      FULLWXOLD=""
   fi
   FULLWXNOW=`cat /dev/shm/wxnow.txt`
   if [ "$FULLWXOLD" == "$FULLWXNOW" ]; then
      # Convert date/times to minutes for date arithmetic
      CURDAY=`echo $TIME | cut -b 1-2`
      OLDDAY=`echo $FULLWXOLD | cut -b 2-3`
      CURHR=`echo $TIME | cut -b 3-4`
      OLDHR=`echo $FULLWXOLD | cut -b 4-5`
      CURMIN=`echo $TIME | cut -b 5-6`
      OLDMIN=`echo $FULLWXOLD | cut -b 6-7`
      CURTIME=$((10#$CURDAY * 1440 + 10#$CURHR * 60 + 10#$CURMIN))
      OLDTIME=$((10#$OLDDAY * 1440 + 10#$OLDHR * 60 + 10#$OLDMIN))
      # If report older than 20 minutes then not updating
      if (( $(($CURTIME-$OLDTIME)) > 20 )) ; then
         echo -n \>$TIME\z
         echo " WX rpt not updating"
         exit 0
      fi
      # Unchanged report but <= 20 min old then don't transmit
      echo
      exit 0
   else
      OLDRPT=`echo $FULLWXOLD | cut -b 9-`
      NEWRPT=`echo $FULLWXNOW | cut -b 9-`
      #if new and old report are same then don't transmit
      if [ "$OLDRPT" == "$NEWRPT" ]; then
         echo
         exit 0
      fi
      # Transmit report and copy wxnow.txt to wxold.txt
      echo $FULLWXNOW cp /dev/shm/wxnow.txt /dev/shm/wxold.txt
   fi
else
   echo -n \>$TIME\z
   echo " WX rpt not found"
fi

While debugging, I noticed that the wxnow.txt file would sometimes disapper from /dev/shm.  This turned out to be systemd cleaning up interprocess communication whenever the pi user logged out.  I fixed this by adding the following line to /etc/systemd/logind.conf:

RemoveIPC=no

Following this change, you must restart systemd-logind.service:

sudo systemctl restart systemd-logind.service

If you implement this on your aprx system, please leave a comment.

Radiation reporting to APRS

A large scale public health/public safety initiative when I worked for the City of New York was NYPD’s radiation monitoring program. Designed to detect a dirty bomb, it makes use of wireless real-time alerting.

Now you can perform your own monitoring as part of your amateur radio station and send the results in real-time to the Automatic Packet Reporting System.

Amateur radio operator “Adminck” (S55MA) in Slovenia has published this blog post. It explains how a Raspberry Pi Zero, a tiny geiger counter, and some scripts are sending his radiation measurements live to the world via the internet.

You can see the actual reports here.

Decode HD Radio on RTL-SDR!

During my 2017 SWL Fest presentation, I mentioned that there is no way to decode either HD Radio or ATSC HD Television using software designed for the RTL2832U dongles. The explanation I provided is that both protocols are covered by patents and that the holders have not been forthcoming on providing necessary details to the open source community.

A cybersecurity researcher, Theori, has cracked the codec used by the NRSC-5-C standard for US based terrestrial digital radio. I am now listening to HD Radio via an RTL SDR dongle. It takes a decent signal, so I’m not getting too many stations using an inside whip antenna, but there are enough to experiment with. It also takes a better dongle with good frequency stability. An older dongle without the TCXO was not up to the task, even on an i7 based system.

The discovery is summarized on the RTL SDR Blog. You’ll need some familiarity with building packages under Linux to grab the source from github and to compile it on your system. So far, I’ve compiled under Debian x86_64, Fedora x86_64, and Raspbian! Next, I want to get it running under Cygwin so that I can use it on the Windows 10 computer in the radio room.

It is a blast to be able to decode the alternate program streams. Audio quality is better than Sirius XM.

Thanks Theori!

DV4mini RPi Image Update

Note: This image has been replaced by this one.

I’ve created an updated image for use with the DV4mini and the Raspberry Pi 2 and 3. It incorporates changes through dv4mini software version date 20170517. It also includes my Brandmeister XTG Dialer for use with extended routing.

Changes include enhacements for YSF reflectors, DMR, DStar as well as the introduction of a text chat window.

Thanks to Ulrich Prinz (DC3AX) for his hard work in support of the DV4 community!

For default passwords, see this post.

Wanna Cry Patches for XP

In case you still have an XP machine running somewhere that you just cannot upgrade right away, Microsoft has released a patch for the Wanna Cry vulnerability. This is the vulnerability that was exploited in recent days to hold up the United Kingdom’s National Health Service and many other organizations for for ransom.

Although Microsoft stopped official support for Windows XP some time ago, the release of this patch for an unsupported product underscores the severity of the matter. The tool used is said to have come out of the United States National Security Agency.

Here is a link to the official Microsoft download. Patch away!

Google Home Goes Multiuser

Google Home has rolled out multiuser functionality. Despite an initial glitch, it seems to work well.

Each member of your household records their voice saying “Hey Google” and “OK Google” by kicking off a dialog on their Android mobile device. They then get access to their personal calendar, music content, and other personalized services to follow. Once enabled, each user can voice query, “Hey Google, what is my name?” and Google Home knows!

I was the initial owner of Google Home when we set up our two devices. After Tina enabled her profile, we found a glitch where Home did not know who I was any longer. It kept referring me to my Google Home app to enable multiuser functionality. But, it took a few days for Home to offer me that option within the app.

Yesterday, when I tried to listen to some content, Google Home complained that it didn’t recognize me and again asked that I enable multiuser. This time, the option was present and I was able to complete the setup.

Running BMXTG on Windows 10

I’ve been able to run the Brandmeister XTG Dialer on Windows 10 by following this procedure.

Visit the PyGTK FAQ page and fully read 21.1.

I installed the following items referred to in the FAQ:

Python 2.7.13
PyGTK for Microsoft Windows using the all-in-one installer

Run a Windows command prompt as Administrator.

cd \Python27
python -m pip install -U pip setuptools
python -m pip install -U requests

Exit the Administrator command prompt and then open a command prompt as a regular user.

cd %AppData%
mkdir Python

Copy bmxtg-v1.2.zip into your %AppData%\Python directory and unzip.

Create a desktop shortcut to bmxtg.py. The “Target” should be “C:\Python27\pythonw.exe bmxtg.py” and the “Start in” directory should be “%AppData%\Python\bmxtg-v1.2”.

Be certain to review and follow the configuration notes in the README. Then, you’re ready to fire it up!