Lutz's blog post first describes and shows his RTL-SDR GOES reception setup. Then, it explains how he used Edge Impulse on his Raspberry Pi 4 to create an AI model that automatically detects the clouds in the image.
The process begins by importing 100 images into Edge Impulse, manually labelling the clouds in each image, training the model, and testing it. The result was an average detection accuracy of 90%.
NOAA satellite weather image APT transmissions have long been a staple of RTL-SDR hobbyists as they are among the easiest and cheapest satellite transmissions to receive. A simple V-dipole, RTL-SDR, and a computer are all that is required.
Sadly, it was recently announced that the NOAA satellites will be classified as end-of-life (EOL) on June 16 2025. However, the good news is that APT and HRPT data transmissions are expected to continue as usual. The end-of-life status indicates that NOAA will cease its own collection of data from these satellites and, hence, will no longer be distributing the datasets online. It also means that the satellites should no longer be used for mission-critical or scientific purposes and that no attempt will be made to fix a failing satellite.
These satellites have been operating well past their operational life, and given their advanced age, failure could come at any time. So the Open-Weather nowcast on June 15 2025 remains a nice way to honor their service to humanity.
Recently, software developer M. Khanfar has been releasing a lot of new SDR and RTL-SDR compatible software, which you can see by searching for Khanfar on our blog.
Recently M. Khanfar has released a new program called "Khanfar RTL-SDR CFAR - Mask Analyzer" which is software designed for RF signal detection and analysis. Khanfar writes:
This powerful tool combines advanced Constant False Alarm Rate (CFAR) algorithms with innovative Frequency Mask Trigger (FMT) techniques to provide reliable signal detection even in challenging RF environments.
CFAR (Constant False Alarm Rate) is an advanced adaptive threshold technique developed originally for radar systems but extremely valuable for any signal detection application.
The Frequency Mask Trigger (FMT) functionality allows you to define signal-free areas in the spectrum where you expect no signals or want to ignore existing signals. The software automatically adapts to these masks, preventing false detections in known signal areas while maintaining high sensitivity in areas of interest.
What You'll Learn: How CFAR adapts to changing noise environments Impact of different algorithm parameters Mask creation and application techniques Optimal settings for different signal types
Perfect For: Students learning signal processing Radar enthusiasts and researchers SDR hobbyists exploring advanced features Engineers developing detection systems
The Khanfar RTL-SDR CFAR - Mask Analyzer brings many features traditionally found only in expensive commercial spectrum analyzers and radar systems to the affordable RTL-SDR platform and provides an incredible value for education, hobbyists, and even many professional applications.
The software was also soon after updated to support Airspy devices, and add Telegram integration. The Telegram integration allows users to set up remote monitoring, and have alerts and data delivered directly to your mobile device via the Telegram messenger chat software.
As mentioned in previous posts, antivirus software can flag M. Khanfar's software as suspicious. We suspect these are false positive results due to the PyInstaller system used to create the exe, but please take your own precautions just in case.
Over on LinkedIn Khalil A. has uploaded a short post highlighting how critical monitoring radio communications was during the mass power outage in Spain. This week Spain and Portugal experienced country wide blackouts, leaving more than 55 million people without power for more than half a day.
During the blackout, news was difficult to obtain as mobile services and internet connections failed. Khalil used a charged laptop and an RTL-SDR to monitor FM, AM, amateur, and emergency services, providing up-to-date information to his neighbourhood.
RTL-SDR provides news during Spain and Portugal blackouts.
Over on the USradioguy.com blog, we've seen news from Carl Reinmann noting that NOAA 15, 18, and 19 will be classed as end-of-life on June 16, 2025. These NOAA satellites are ones commonly used by RTL-SDR hobbyists to download weather satellite images, either via APT on 137 MHz with a V-dipole antenna, or via HRPT on 1.7 GHz with a tracking dish antenna.
Initially, it was thought that this meant that transmissions would cease. However, Carl Reinmann has now clarified with NOAA that transmissions of the APT and HRPT signals will continue as usual. Importantly, NOAA urges that these transmissions will only be "data of opportunity" and should no longer be used for operational purposes (not for anything safety-critical, for example). The transmissions will be fine for everyday hobbyist use.
However, this does mean that should the sensors on these satellites start failing, no attempt will be made to repair them from the ground, and in case of critical failures, the satellites will be decommissioned. In the past, we've seen NOAA 15's scan motor fail multiple times before coming back to life. It's not clear if the satellite received commands from the ground that helped recover it or if the motor just recovered by itself.
The NOAA satellites have lived well past their operational life.
On YouTube, user MatdoFM has uploaded a video demonstrating what lightning sounds like over the radio and how to receive lightning pulse noise using an RTL-SDR.
When lightning strikes it releases a pulse of electromagnetic radiation, which shows up as a short wideband noise pulse over the radio spectrum. Lightning detection and mapping services like Blitzortung use a network of volunteer run VLF receivers spread out across the globe to determine the location of lightning pulses using time of arrival radio direction finding techniques.
In the video, MatdoFM uses a frequency of 124 MHz. Because the lightning pulse noise is so wideband, it extends from VLF to VHF and sometimes even UHF frequencies. At the end of his video, MatdoFM shows a lightning strike captured with his security camera and the corresponding radio sound produced by that strike.
DragonOS creator Aaron recently uploaded a video on YouTube showing how to capture IMSI data from an LTE-enabled phone by using the open-source LTE sniffer tool and Ettus X310 software-defined radio.
In the video, Aaron uses a simulated environment involving a Signal SDR Pro to simulate the LTE cell phone, a B205 Mini operating as the eNodeB (base station), and an Ettus X310 SDR for the actual LTE sniffing. The SRSRAN software running on DragonOS is used to simulate the LTE network environment.
Aaron goes on to show how the LTE sniffer software passively decodes the physical downlink control channels and captures IMSI numbers from user cell phones.
An IMSI is a unique identifier associated with a cell phone user's SIM card. IMSI sniffing cannot be used to listen to or decode voice, text, or data as they are all encrypted. However, bad actors can use IMSI sniffing to track the movement of devices/people.
Typically, a satellite dish is used to receive Elektro L3. As an example, our 70cm diameter Discovery Dish with linear feed can do this easily, and achieve an SNR of about 5-6 dB. However, as Meti shows, it is possible to receive this satellite even without a dish, and as he shows, an SNR of 1.5 dB is sufficient for decoding a perfect image.
Meti's antenna is an 11-turn RHCP helix made of copper wire, with a 17 x 17cm ground plane. In his post, he also notes a few interesting findings, noting that the height of the antenna off the ground is critical, rotating the helix can help, interference from cell towers can cause issues, and bending the corners of the ground plane can help.
In the rest of the post, Meti also shows how well the helix antenna works at receiving weather satellite signals from polar orbiting L-Band satellites like Meteor M2-3.
