Lime Microsystems chief executive officer Ebrahim Bushehri rounded out 2023 with an appearance on the December Crowd Supply Teardown Sessions, talking through the history of the company and the LimeSDR – as well as the upcoming LimeSDR XTRX and a new chip dubbed the LMS8001.
“Quite a few people have shown interest in [the LimeSDR] XTRX for embedded applications,” Ebrahim said during the discussion with crowdfunding platform Crowd Supply. “We work very closely with Vodafone here in Europe and one application is to actually do a 5G Network in a Box solution [that] uses a Raspberry Pi for baseband, and so this [LimeSDR] XTRX has been coupled together with Raspberry Pi to create a totally programmable solution for a 5G network in a box.”
The LimeSDR XTRX is the continuation of a Fairwaves project to create a flexible software-defined radio platform suitable for embedded applications. Following the discontinuation of the original XTRX, the design has been revamped and adopted into the LimeSDR family – complete with software support in Lime Suite.
Ebrahim also discussed Lime Micro’s LMS8001 chip. “The idea here is to extend the frequency up to 10GHz,” he said. “So LMS7002 and LMS6001 goes up to 3.8GHz, [and] LMS8001 is an up/down converter that takes an intermediate frequency and up-converts it to any frequency up to 10GHz and vice-versa. It’s an up/down converter that essentially extends the frequency range.”
The full discussion is available on the Crowd Supply YouTube channel now.
Andrew Back has reported on the LibreCellular project’s latest milestone: the creation of a new modular continuous integration (CI) radio access network (RAN) subsystem and the development of new duplexer and clock distribution boards.
“The original CI RAN subsystem, CIRAN1, was based on an original LimeSDR board plus an Intel NUC and two RF cavity duplexers,” Andrew explains. “However, the new CIRAN2 subsystem takes a more modular approach and can accommodate two base stations on removable sleds. The idea being that these sleds can be swapped out over time, in order to permit testing of new base station designs. The first sled is designed to accommodate an UP 4000 SBC [Single Board Computer] and a LimeSDR Mini 2.0, plus a new LTE Band 3 Low Power Duplexer (B3LPD) board.”
That board, B3LPD, is designed to replace the cavity duplexers – “physically large,” Andrew explains, “and overkill for applications such as this, with very low RF power levels” – in the original CI RAN subsystem. A new clock shaping and distribution board, CSAD10, has also been developed for the project. “CSAD10 utilises an LTC6957IMS-3 low phase noise buffer/driver at its input, which is able to translate a sine wave signal to logic levels. This is followed by an LMK00101 ultra-low jitter fanout buffer, which provides ten CMOS level 50 ohm outputs.”
The full status report is available in Andrew’s MyriadRF post, with links to documentation for the CIRAN2 and the two new boards.
Finnish technology news outlet Tivi has published a look back at the “mountain phone,” a “0G” precursor to the mobile phone operating as far back as 1946 – developed by a nomadic Sami, Aslak Partapuoli.
“Sweden’s first call on a mobile phone is said to have been made by engineer Sture Lauhrén on December 3, 1950. He worked at the state television agency Televerket, and the call was made from his car, in the back of which was a 40-kilogram line-connected radio,” Tivi’s Rosa Lampela writes, in translation. “But Göta Partapuoli, Aslak Partapuoli’s wife, called her husband with a ‘mountain phone’ already in 1946, Teknik Historia reports. Göta used a radio transmitter in the mountains and Aslak received the call with a radio in the village of Ammarnäs.”
Partapuoli had been planning such a device since at least 1939, but was delayed by restrictions in radio use during the war. Contemporaneous newspaper coverage of the invention reveals a range of 4-6 Scandinavian miles under normal conditions and up to 15 Scandinavian miles under “favourable conditions” – roughly equal to 25-37 miles and up to 93 miles, converted to Imperial measurements. Compared to a modern mobile, though, the devices were a little bulky, weighing in at over 30kg.
“The connection worked well,” Rosa writes, “during day and evenings on 37 of 50 test days. Televerket continued trials in winter and spring of 1947. When temperatures dropped below -35 degrees the connection noticeably weakened.”
Developer Sebastian Reichel has published the files required to simulate a Microchip MRF89XA multi-channel frequency-shift keying/on-off keying (FSK/OOF) transceiver using a LimeSDR radio and a copy of GNU Radio.
“This is a GNU Radio description to simulate a MRF89XA with a software defined radio,” Sebastian writes of the project. “It has been tested with a LimeSDR, but can easily be adopted to any other hardware supported by Soapy[SDR]. Additionally ‘cff3000.py’ implements the protocol used for the equally named door lock remote control, but is missing the encryption code. By adding the encryption code this can be used to (un)lock CFA3000 doors.”
The Microchip MRF89XA is a low-power single-chip embedded transceiver for FSK/OOK operation on the 863-970, 902-928, and 950-960Mhz frequency bands. It’s commonly used in electronic door lock systems – which led to Sebastian giving a talk at the 37th Chaos Communication Congress late last year on how the devices could be opened using software-defined radios.
Pseudonymous radio engineer “Leaning Tower” has designed a “pure analogue” Global Navigation Satellite System (GNSS) receiver – pairing a TinyFPGA BX development board with some Microchip ATtiny85 microcontrollers.
“[This is] a pure analogue Global Positioning System (GPS) data receiver that can receive the Navigation (NAV) message from the GPS satellite,” the pseudonymous Tower explains of the project. “The receiver is a dual conversion receiver that includes five homemade PCB boards. A TinyFPGA is used for generating the Coarse Acquisition (C/A) code which retrieve the direct-sequence spread spectrum (DSSS) GPS signal submerged below the thermal noise level.”
Those five boards are: a mixer board, a local oscillator, a second mixer and correlation board, a signal processing board, and an interface to the FPGA which runs gateware for code tracking and locking and for recovering the NAV message data.
“The final success [took] several weekends for troubleshooting,” the maker admits. “The most significant problem is cause by the interference of the 24.552MHz system clock generator. Originally I was using a Si5351’s programmed output as the system clock, however it generates a bunch of tones within the bandwidth of sensitive GPS antenna. Since it is hard to find any crystal works at this frequency, a final solution is done by the decapping [of] a 24.576MHz crystal and modifying the frequency by a Sharpie pen. It is also frequency stabilized through external temperature controller.”
The full project write-up, along with circuit diagrams and source code, is available on GitHub.
There are fewer components in Jon Dawson’s ham radio transmitter, meanwhile: its heart is a simple Raspberry Pi Pico microcontroller development board.
“Capable of outputting SSB [Single-Sideband], AM [Amplitude Modulation], and FM [Frequency Modulation] signals, this versatile transmitter allows you to cover frequencies from 0.5 to 30MHz, including the ham bands from 160m to 10m,” Jon explains of the simple circuit. “The transmitter is based on a Raspberry Pi Pico, which uses a powerful PIO [Programmable Input/Output] feature to output an RF oscillator with precisely controlled phase and frequency, reducing the part count and keeping the cost down. The transmitter also employs a PWM [Pulse-Width Modulation] output to generate an RF envelope for amplitude modulation.
“The hardware is pretty simple, the Pi Pico has 3 outputs. The RF oscillator is output on an IO pin, the phase and the frequency are precisely controlled by software. For constant amplitude modes like CW and FM, this is all we need, but for modes that modulate the amplitude like AM and SSB, we use a complimentary pair of PWM outputs to generate an RF envelope. The PWM outputs are passed through a simple RC low-pass filter with a cut-off frequency of around 3kHz. I run the PWM as fast as I can, to reduce ripple, we can get 8-bits of resolution at around 500kHz, at this frequency the filter attenuation is very strong.”
The full project write-up is available on Jon’s website, including schematics and source code.
Jon Mackey has built a miniaturised version of the NIST’s WWVB radio time signal station – delivering a compatible signal from an STMicro STM32 microcontroller over a range of around 10cm.
“I live in northern New Hampshire, and according to the NIST coverage map, I should be able to reliably receive the WWVB signal at night,” Jon says of his proximity to the National Institute for Standards and Technology’s radio station transmitting atomic-clock-derived time signals to clocks and watches. “I own several WWVB ‘Atomic’ clocks. Generally these clocks work quite well, but for more than two weeks in December 2023 my clocks were not updating.”
To solve the problem, Jon built his own “WWVB” – using a GNSS receiver for an accurate time source and an STM32 microcontroller to generate a compatible, though short-range, signal. “On startup, and on the half hour, the WWVB simulator’s time is updated from the serial stream coming from the GPS module,” he explains. “One of the STM32’s timers outputs a ~60kHz 50% duty cycle PWM [Pulse Width Modulation] signal that serves as the AM [Amplitude Modulation] carrier, the signal that drives the antenna. The antenna radiates, at most, about 10cm. It’s incredibly weak, but that’s OK.”
Jon has published a full guide to the project on Instructables.
A research team from Stanford University and the American University of Beirut have developed a new foldable antenna which is designed for rapid deployment in disaster zones – and which can shift between terrestrial and satellite communication at-will.
“The state-of-the-art solutions typically employed in [disaster] areas are heavy, metallic dishes. They’re not easy to move around, they require a lot of power to operate, and they’re not particularly cost-effective,” claims assistant professor of aeronautics and astronautics Maria Sakovsky of the problem her team set out to solve. “Our antenna is lightweight, low-power, and can switch between two operating states. It’s able to do more with as little as possible in these areas where communications are lacking.”
The resulting antenna looks like a giant version of a finger-trap toy, which Maria calls “a very untraditional antenna design” based on “shapes that have never been used on helical antennas before.” When folded, the antenna forms a ring around 2.5cm tall and a little under 13cm across. In this configuration, it acts as a directional antenna suitable for satellite communication; when stretched out to its full length, it turns into an omnidirectional antenna for terrestrial use.
“The frequency you want to operate at will dictate how large the antenna needs to be,” Maria explains, “but we’ve been able to show that no matter what frequency you operate at, you can scale this design principle to achieve the same performance.”
Pseudonymous maker “t0nito” has designed a portable antenna, too – this one targeting high-resolution picture transmission (HRPT) weather satellite signal reception and designed for 3D printing.
“This is a portable light weight handheld HRPT receiving antenna [for] weather satellite imagery,” t0nito explains of the design. “It can also be used on a motorised satellite tracker. Materials needed: a metallic conductive mesh, I used aluminium mosquito net; about 1.5m of 6mm² copper wire; sheet metal to cut a 133mm circular ground plane; 2mm self tapping flat head screws; a flange mount SMA male connector; 2 M8 bolts; 2 M8 washers; 4 M8 nuts.”
With those, and a suitable 3D printer, it’s possible to create t0nito’s antenna yourself: just download the files from Thingiverse, where they’re published under the Creative Commons Attribution-ShareAlike licence.