Lime Microsystems has announced a new revision of the LimeNET Micro 2.0 which adds compatibility with the newly-launched Raspberry Pi Compute Module 5 – delivering at least twice the performance and unlocking support for improved cellular capabilities.
“We are delighted to announce that LimeNET Micro 2.0, the eminently compact 5G/4G network-in-a-box, has been updated to be compatible with [Raspberry Pi’s] CM5 and to take advantage of the marked increase in performance that it provides, while also retaining compatibility with Compute Module 4,” Lime Micro’s Andrew Back explains.
Designed, as the name suggests, as the successor to the Raspberry Pi Compute Module 4, the Compute Module 5 is based on the same Broadcom BCM2712 system-on-chip as the Raspberry Pi 5 single-board computer. This features four Arm Cortex-A72 cores running at up to 2.4GHz, a high-performance Videocore-VII graphics processor, and up to 8GB of RAM – with a 16GB variant due to launch next year.
“With CM4 the LimeNET Micro 2.0 is able to support 5G NR TDD SISO operation with 10MHz RF bandwidth,” Andrew says. “However, with CM5 it is now able to support FDD 2×2 MIMO with 20MHz bandwidth. Which is to say a significant increase in performance and all while still being able to run a 5G Core network also on the Compute Module, for fully self-contained operation.”
All new orders placed for a LimeNET Micro 2.0, as well as the Developer Edition and the LimePSB RPCM, will feature CM5 compatibility, Lime Micro has confirmed. “While we expect that orders placed for LimeNET Micro 2.0 from January 2025 onwards will ship with CM5 fitted as standard,” Andrew adds, “please contact us first to confirm where this is a requirement.”
More information is available on the Lime Micro website.
Charles Lohr has been experimenting with stretching the meaning of “software-defined radio” by experimenting with using microcontrollers themselves as LoRa transmitters – declaring, tongue-in-cheek, that “microcontrollers are just radios in disguise.”
“Normally, to send RF signals, you would use radio chips, or microcontrollers with radios in them,” Charles explains, “or maybe get creative with your electrical engineering degree. But, really, any time there is a change in electrical potential along a conductor, an RF field is made. It can even be as simple as turning a light switch on or off.”
Charles is known for his work with the WCH Electronics CH32V003, an ultra-low-cost 32-bit microcontroller built around the free and open source RISC-V architecture. “By noodling one of its pins around at a couple of megahertz juuuust right,” Charles says, “it’s sending LoRaWAN messages to this commercial LoRa gateway at 904.5MHz, and that gateway is forwarding those packets to The Things Network where they can be accessed from around the world.”
Typically, radio frequency emissions from transceiverless microcontrollers are unintended – and, if strong enough, cause for a part to fail emissions tests. In Charles’ case, though, the emissions are being deliberately induced – and, unexpectedly, are powerful enough to be picked up by an off-the-shelf receiver at a distance of over 400 feet.
Charles has a full explanation in a YouTube video, with source code for the CH32V003 plus Espressif’s ESP32-S2 and ESP8266 microcontrollers on GitHub under a combination of the MIT licence and a custom license dubbed the “MIT-like-non-AI-licence.” A talk Charles gave during the Hackaday Supercon 2024 last month is also available on YouTube.
Pseudonymous developer “XQTR” has released a spectrum analyser and signal processor with a difference: written in Python, it’s designed to be used at a text console or within a terminal emulator – no graphical user interface required.
“After making the changes at the retrogram-rtlsdr project, I got the ‘virus’ to make something even more better and in a language I know better, as C is not my thing,” XQTR explains of the project’s origins. “So Python it was, as it may not be the faster, but it has great support and anyone can contribute, as it is a well known programming language.”
The PySpecSDR tool delivers a real-time spectrum analysis and waterfall display, rendered as text characters directly on the terminal. The signal processor handles FM, AM, and SSB demodulation, including outputting audio in real time, frequency scanning, signal classification, and there’s a bookmark system for highlighting frequencies of interest. Both RF signals and demodulated audio can be recorded, there are band presets for common frequency ranges, and multiple visualisation modes.
“For sure, it needs some refinement and work some quirks,” XQTR admits, “but it’s functional and it has more than any other text based SDR project… at least, I know of. Just grab a copy and ‘play’ with it.”
More information is available on XQTR’s blog, with the PySpecSDR source code published to GitHub under the reciprocal GNU General Public Licence 3.
Developer Steven C. Hageman has released a tool to make it easier to keep track of your favourite frequencies for use with the cross-platform SDR++ software: a Python script which converts spreadsheets into an importable JSON file.
“[This is] an easy way to build frequency lists in a [Microsoft Excel] XLSX Spreadsheet and convert them to a ‘frequency_manager_config.json’ file for direct use in SDR++,” Steven explains of his tool. “The supported Spreadsheet formatting is [Microsoft Excel] 2007 to 365 ‘.xlsx’ format only. Compatible with LibreOffice- and Excel-generated spreadsheets (perhaps others also, but untested).”
The idea is simple: you can use a spreadsheet to store your bookmarked frequencies, then run the Python script to convert it into a JSON file. Once SDR++ has been shut down, this file can be used to replace the existing JSON configuration for the Frequency Manager tool – automatically importing all your favourite frequencies.
“You can add ‘tabs’ to the spreadsheet workbook to organise frequency bookmarks together,” Steven notes. “Each ‘tab’ can contain grouped ‘bookmarks’ that are made up of names, frequencies, IF bandwidths, and demodulation modes.”
The Python source code, example spreadsheet, and user guide are available on GitHub under a public domain licence.
Radio amateur Umar “2E0UMR” Munir has released an initial design for a low-cost Bluetooth connected data modem for use with digital modes, with work in progress on turning the current modular layout into a single-board design.
“On [the] market there are lots of options for soundcard-based data modems, like Signalink, Digirig, and lots of DIY and cheap modems,” Umar explains of the inspiration behind the Blue Dmod project. “I haven’t seen any Bluetooth based audio data modem apart from Mobilinkd which is unfortunately Bluetooth serial port TNC [Terminal Node Controller]. In order to keep the cost to minimum I got together components I already had, [I] just had to order the microcontroller as I didn’t have any small form factor microcontrollers.”
Umar’s design uses a low-cost BT002 Bluetooth audio module, which connects to the radio’s microphone input and audio output plus an Arduino-compatible microcontroller. A relay delivers push-to-talk compatibility, with Umar choosing that route over a transistor or MOSFET for improved compatibility.
“To keep [it] all simple I used [an] RJ45 port to connect [the] radio to [the] modem,” Umar adds. “It [is] easy to make RJ45-based cables [rather] than using any special connector. I have tested the ‘Blue Dmod’ with Wouxun KG-UV9D, Wouxun KG-UV9K, and TCA/PRC 152 [radios]. On TCA/PRC 152 I had to add a ferrite core on high power.”
More information is available on Umar’s website, with source code and a schematic available on GitHub under the reciprocal GNU General Public Licence 3.
Researchers at Columbia University, North Carolina State University, and the City University of New York have made a step towards turning clothing into functional antennas – with the lead researcher proposing a future in which you can wear “a sweater that can double as a Wi-Fi signal booster.”
“The float-jacquard knitting technique used for making our textile metasurfaces is exactly the same technique that my mother used to make sweaters for me,” project lead and corresponding author Nanfang Yu explains of the team’s work. “I still remember a purple sweater I wore as a kid that had a row of white cats across the chest; I remember that when I inspected the inner side of the sweater, I saw white parallel yarns – the floats.”
The float-jacquard knitting technique, used in Fair Isle jumpers, has been adapted by the team to produce functional metasurfaces which can act as a vortex-beam generating device or a metasurface lens to focus radio signals. In both cases, the prototype devices were soft, foldable, and continued to operate even after repeated washing cycles.
“It’s important to stress that these devices were fabricated using commercially available off-the-shelf yarns and leveraging established fabrication techniques,” Nanfang adds. “I am almost certain that communities of knitters can come up with ingenious ways to integrate aesthetics and functionality into a sweater – a sweater that can double as a Wi-Fi signal booster.”
The team’s paper has been published in the journal Advanced Materials, though under closed-access terms.
A separate project carried out by researchers from the Johns Hopkins University Applied Physics Laboratory has also focused on novel antennas, working with 3D printing and shape-memory alloys to develop antennas capable of dynamically shifting their shape – inspired by sci-fi series The Expanse.
“I have spent my career working with antennas and wrestling with the constraints imposed by their fixed shape,” explains electrical engineer Jennifer Hollenbeck, who took inspiration for the project from the organic technology employed by the aliens in The Expanse. “I knew APL had the expertise to create something different.”
The team’s work focuses on 3D-printing nitinol, a shape-memory alloy made from nickel and titanium – and difficult to work with at the sizes and shapes required for a antenna suitable for high-power transmission. “We made shrapnel in the printer a few times because the antenna is trying to change shape as you’re printing it, due to the heat,” co-author Mary Daffron explains. “It wants to peel apart.”
“The shape-shifting antenna capability that has been demonstrated by this APL team will be a game-changing enabler for many applications and missions requiring RF adaptability in a low-size and -weight configuration,” claims APL’s chief engineer Conrad Grant of the work. “This is yet another powerful example of the innovation that occurs at the Laboratory through motivated, highly capable, multidisciplinary teams.”
The team’s paper has been published in Applied Engineering Materials under open-access terms; a supporting video is available on YouTube.
Radio amateur Sjef “PE5PVB” Verhoeven has written of his experience putting together a kit designed to adapt the NXP Semiconductors TEF6686, a chip originally designed for car radios, for use in surprisingly flexible yet low-cost receive-only SDR systems.
“As an active radio amateur (PE5PVB) in the Netherlands, I became intrigued by the enthusiastic reviews I started seeing of the TEF6686,” Sjef writes. “During the COVID lockdown of 2020, I started designing a completely open-source tuner that would wring the highest possible performance out of the chip for FM DXers. My enthusiasm grew when I found TEF6686 tuner modules on AliExpress. These contain a TEF6686 chip in a DIY-friendly package, suitable for through-hole soldering (the TEF6686 itself is a surface-mounted chip), and with radio-frequency shielding to help minimise interference. These modules are cheap—they can generally be found for around US $25.
Existing kits, though, proved awkward to use, so Sjef set about building his own – swapping out the commonly-used Arduino microcontrollers for a more powerful Espressif ESP32, adding a display board for local control, and making other improvements.
“In early 2021, I released on GitHub an initial version of the firmware and schematics for other DIYers,” Sjef writes of his design. “In the fall of 2021, I created a second version with a so-called human-machine interface (HMI) display from Nextion. This display has a built-in processor, so I could hand off more user-interface tasks from the ESP32. This sped up the radio considerably, and also opened up some new graphical possibilities.”
The full article is available on IEEE Spectrum; Sjef’s designs are available on GitHub under the reciprocal GNU General Public Licence 3, with additional information on his blog.
Giuseppe “IZ0GZW” Morlè, meanwhile, has been working on updating a compact SW antenna design – extending it to also cover MW as well.
“Do you remember the DICA 2 antenna? It was a small, shortened antenna with three ferrite cores inside, capable of tuning across all shortwave bands,” Guiseppe writes in a letter to the SWLing Post. “Unfortunately, it no longer exists in its original form.
“I have completely reworked the windings to enable the ferrites to also function for medium wave (MW) reception. Now, the antenna features: 35 turns for medium waves; four turns for shortwaves; and the same variable capacitor of over 1000pF remains in use.
“For medium waves,” Guiseppe notes, “the antenna works wonderfully by induction. You simply place the ferrite core near the tube and turn the variable capacitor to achieve perfect tuning across the entire MW range. I was genuinely amazed by its performance on shortwaves. Despite its small size, the antenna provides excellent gain, especially when a ground cable is connected to the variable capacitor’s casing.”
Guiseppe’s full write-up, including videos explaining the design, is available on the SWLing Post.
A team of radio amateurs has successfully received signals from NASA’s Voyager 1 probe, despite its 25 billion kilometre distance from the Earth – using a radio telescope built in the 1950s.
“We have used the historic Dwingeloo radio telescope to receive signals from the Voyager 1 spacecraft,” the team explains. “Only a few telescopes in the world have received these signals, which are very faint due to the distance of Voyager 1: almost 25 billion kilometers, more than four times the distance to Pluto.
“Since the Dwingeloo telescope was designed for observing at lower frequencies than the 8.4GHz telemetry transmitted by Voyager 1, a new antenna had to be mounted. At these higher frequencies, the mesh of the dish is less reflective, making it extra challenging to receive faint signals.
“To find the very weak carrier signal in the noise, we used orbital predictions of Voyager 1 to correct for the Doppler shift in frequency caused by motion of Earth and Voyager 1,” the team explains. “By doing so, the signal could be seen live in the telescope observation room. Later analysis confirmed that the Doppler shift corresponds to that of Voyager 1.
The full write-up is available on the C.A. Muller Radioastronomie Station website.
Finally, mononymous YouTuber Derek has published a video looking into the oldest satellite still transmitting: Transit 5B-5, launched on a Thor-Ablestar rocket in December 1964.
“It is a surviving relic of the age of the Cold War and the space race,” Derek explains of the satellite. “The very same spacecraft – built in the US by the [Johns Hopkins University] APL [Applied Physics Laboratory] during the Lydon B. Johnson administration, launched just seven years after Sputnik-1 and five years before the first crewed moon landing – can still be heard going strong.
“There are even older defunct artificial satellites, and dead rocket bodies, in Earth orbit, but as far as we know this is the oldest satellite still actively broadcasting a signal. Transit, also referred to as the NNSS or Navy Navigation Satellite System, acted as an early predecessor of today’s GNSS or Global Navigation Satellite System. But not all Transit satellites fulfilled this role. The initial four series of satellites – Transit 1, 2, 3, and 4 – acted as prototypes that demonstrated the feasibility of the concept.
“The Transit 5A series was the first with the specific goal of providing useful navigation services for the Navy,” Derek continues. “Of the three satellites launched between 1962 and 1963, only the last – Transit 5A-3 – managed to do so, as 5A-1 suffered an in-orbit malfunction while %a-2 never made it to orbit to begin with. Following the 5A series would be two successive ones: Transit 5BN and Transit 5E, launched together in three pairs of one 5BN and one 5E satellite each. Between the end of the 5BN series and the start of the [later] Oscar series, a single Transit 5C satellite was also launched.”
Transit 5B, though, doesn’t quite fit in to any of these families – not even the Transit 5BN family, its closest-named. The reason: a seemingly retroactive renaming – though Derek suggests there may be more to the story.
“This launch, which happened on December 13 1964, carried Transit-5E-5 as well as Transit Oscar 2, revealing that at some point in time the name ‘Transit 5B-5’ was retroactively, and possibly mistakenly, assigned to Oscar 2,” Derek explains. “[Its] radio signal, its frequency, modulation, and contents appear to be more in-line with a Transit 5E satellite rather than a Transit Oscar. When I look at all of this, I am personally inclined to believe the satellite we are observing today, that we know as Transit-5B, is in fact Transit 5E-5.”
The full video digs deeper into the satellite’s origin, naming confusion, and transmission, and is available on Derek’s YouTube channel.