{"id":4231,"date":"2021-09-02T07:36:05","date_gmt":"2021-09-02T11:36:05","guid":{"rendered":"https:\/\/jeremyclark.ca\/wp\/?p=4231"},"modified":"2026-01-02T14:33:30","modified_gmt":"2026-01-02T19:33:30","slug":"rtl-sdr-for-satellite-weather-on-goes16-signal-decoding","status":"publish","type":"post","link":"https:\/\/jeremyclark.ca\/wp\/telecom\/rtl-sdr-for-satellite-weather-on-goes16-signal-decoding\/","title":{"rendered":"RTL-SDR for Satellite Weather on GOES16 – Signal Decoding"},"content":{"rendered":"\n

Introduction<\/h2>\n\n\n\n

In the previous two posts we looked at planning for reception of GOES16 weather signals and we did a signal capture as shown above. Improved performance resulted when the LMR400 coax was inserted between the SAWBIRD Filter\/LNA at the back of the antenna and the RTL-SDR . From the first planning post, the average noise level was -100dBm. The BPSK spectrum is about 8dB above the noise, so say -92dBm. Our planning was for a signal at -84dBm based on a guess of 60dBmi tx power. Thus a better guesstimate is for a lower satellite Tx power of 52dBmi. <\/p>\n\n\n\n

Receive Signal Simulation<\/h2>\n\n\n
\n
\"\"
Fig.1 BPSK Scicos Simulation<\/figcaption><\/figure>\n<\/div>\n\n\n
\"\"
Fig.2 PRBS Data & BPSK 180deg Phase Shift<\/figcaption><\/figure>\n\n\n
\n
\"\"
FIg.3 Rx Data – LPF Output – Mixer Output<\/figcaption><\/figure>\n<\/div>\n\n\n
SNR=20*log10[Sp2p\/Np2p]<\/strong><\/td>BER<\/strong><\/td><\/tr>
20*log10[2Vp2p\/1.0vp2p]=6dB <\/td>[<1\/1000]=<0.001 <\/td><\/tr>
20*log10[2Vp2p\/2vp2p]=0dB <\/td>[6\/1000]=0.006 <\/td><\/tr>
20*log10[2Vp2p\/4vp2p]=-6dB <\/td>[41\/1000]=0.041<\/td><\/tr><\/tbody><\/table>
Table 1 SNR vs. BER for BPSK<\/figcaption><\/figure>\n\n\n\n

Figure 1 shows a Scicos channel model for BPSK transmission. A PRBS data signal directly multiplies a carrier frequency and White Gaussian Noise is added in the channel. The receiver reinserts the carrier and the data is recovered with a LPF and comparator. Table 1 lists the BER vs SNR. Note that this is just for BPSK with no Viterbi or RS coding, just to get a rough idea of performance. At 6dB we have better than 10^-3BER, so with 8dB and Viterbi\/RS coding we should have adequate performance to decode pictures successfully. <\/p>\n\n\n\n

Decoding Software – goestools on RPi4<\/h2>\n\n\n\n

In order to decode the GOES16 signal, I used Ref.3<\/a> & 4<\/a> (from rtl-sdr.com & Alexy) that describe the installation and setup of goestools on a RaspberryPi. The articles are very well written, so not too much to add. The first step was to download an image file for Raspbian Lite OS<\/a>. I had several old 8Gbyte SD cards left over from earlier installations, so I cleaned off the various partitions from one of them using a DOS utility DISKPART running on Windows10, in a command window.

After that, now with an empty SD card, I used RaspberryPi Imager<\/a> to load and install the RPiLiteOS. I inserted the SD card into an RPi4 and booted up using my TV as an HDMI monitor. The default user\/password is pi\/raspberry. I ran <sudo raspi-config> and enabled SSH access. Then I downloaded Putty on my laptop and connected the RPi4 to my LAN. I was able to connect over the LAN to the RPi4 with 192.168.0.12 assigned to the RPi4 (Fig.4). Once connected over the LAN, I ran an update & upgrade <sudo apt update> & <sudo apt full-upgrade>. Then I followed all the steps from Ref.4. For the last step, I used nano to write the parameters into goesrecv.conf directly, rather than the cat command. Once goestools was installed, I inserted the RTL-SDR and ran a test to detect it <rtl_test> as shown in Fig.5.<\/p>\n\n\n

\n
\"\"
Fig.4 PUTTY Configuration Screen RPi4 LAN Connection<\/figcaption><\/figure>\n<\/div>\n\n
\n
\"\"
Fig.5 rtl_test to Detect RTL-SDR on USB Port<\/figcaption><\/figure>\n<\/div>\n\n\n

The final step was to run goesrecv, orient the dish antenna and see if I could get the Viterbi error rate to decrease to a point where no packets were lost. With no signal, vit(avg) > 2000, when the dish was pointed I could get the vit(avg) to <180 which resulted in no drops as in Fig.6. The YouTube video in Fig.7 shows the procedures used. In the next post we will use a permanent mechanical dish mount and download actual information from the satellite. <\/p>\n\n\n\n

\"\"
Fig.6 Shows the Console Output for goesrecv<\/figcaption><\/figure>\n\n\n\n
\n