1. Introduction
The past two decades have seen the spread of software-defined radio (SDR) devices into the nonprofessional area (see [
1] for an overview of SDR technologies, with schematics, concept of usage, and more; for an example of a practical application, such as an ionospheric sounding station, see [
2]). This was facilitated by a reduction in hardware prices, as well as the evolution of the field-programmable gate array (FPGA) technology, which streamlined the structure of high-frequency (HF) SDR devices, leaving aside application-specific integrated circuits (ASICs). Also, last but not least, another reason is the fact that the main suppliers granted free access to the FPGA builder tools for noncommercial usage [
3]. Nowadays, analog-to-digital converters (ADCs) [
4,
5] and digital-to-analog converters (DACs) [
6] with a sampling rate of more than 80 MHz capture all high-frequency (HF) bands (3 to 30 MHz) for FPGA processing. All this makes SDR devices suitable for monitoring short-wave transmissions through the ionosphere and, thus, for the implementation of the linear frequency modulation (LFM—a sinusoidal wave that increases in frequency linearly over time) of the ionosonde [
7,
8]. To monitor the current ionospheric situation, vertical ionospheric sounding using short coded pulses with a carrier frequency from 1 to 20 MHz is widespread. At present, it is difficult to purchase a turnkey solution for vertical and oblique soundings of the ionosphere. The few offers available are high-priced. Meanwhile, such devices are necessary to solve the problems of diagnosing the ionosphere and HF channels of long-range radio-wave propagation. A spatially distributed network of several such devices is extremely useful in supporting active experiments in the ionosphere and allows one to obtain data on the horizontal movements of large ionospheric irregularities. The present paper describes the implementation of the available SDR devices and hardware components in the vertical sounding ionosonde prototype developed at the NIRFI and considers the results of the tests of its continuous operation at the “SURA” facility (including the results of supporting the optic experiment in August 2022). Also, we show its extension to a chirp ionosonde receiver and discuss the prospects for its further modernization in order to turn it into a multifunctional ionospheric diagnostic platform.
The vertical sounding ionosonde is a radio-locating station operating in the HF band (for more details, refer to [
9]). While running, a typical ionosonde radiates short radio-wave pulses with a carrier frequency from 1 to 20 MHz. In general, amplitude- or phase-shift keying is applied to the transmitted pulses. The receiver captures and handles pulses reflected from the ionosphere. For each reflected probe pulse, its delay relative to the moment of its emission is recorded. The product of the delay times half the speed of light is called the effective height of reflection. The dependence of the effective height on the frequency of sounding pulses is called the altitude–frequency characteristic of the ionosphere or ionogram. By analyzing ionograms, it is possible to obtain the dependence of the electron concentration on the height (electron concentration profile for an altitude range from 80 to 700 km)—the most important characteristic of the ionosphere.
Because of its technical features, the prototype of the device described below, if widely used, can contribute to an effective solution to the problem of diagnosing the upper ionosphere.
3. Operation Test Results
At the present time, the developed prototype is in the stage of trial operation at “Vasilsursk” experimental base.
Figure 6 shows the locations of the main elements of the transmitting and receiving parts of the ION-FAST ionosonde. The receiving and transmitting antennas are separated by a distance of about 400 m. The hardware of the transmitting part is placed in the instrument room of the “SURA” facility. The receiver operates in the laboratory building.
Currently, the ionosonde uses two compact 53 m antennas (T2FD—Tilted Terminated Folded Dipole, [
16]) as the transmitting and receiving antennas (see
Figure 7), so they are regarded as regular components of the future device. The T2FD is a general-purpose shortwave antenna that performs reasonably well over a broad frequency range, without marked dead spots in terms of frequency, direction, or angle of radiation above the horizon. The design properties of the antenna make it ideal for use in small spaces at long wavelengths. Other advantages are fast installation and a low cost for the available commercial versions.
The ION-FAST has two main operating modes. The first one is a daily monitoring mode. The Ionosonde starts every 5 min, except on the 0th, 15th, 30th and 45th minutes of each hour. The prototype makes 60 full 1 s cycles and then performs averaging over 60 ionograms (total 1 min averaging). Thus, the ION-FAST has the same effective time for obtaining an ionogram as the CADI ionosonde. The exception of the 0th, 15th, 30th and 45th minutes is due to the schedule for starting the CADI ionosonde. This is necessary to avoid interference. Using both the ION-FAST and the CADI ionosondes, we can simply compare the registered ionograms and verify the obtained results. This approach provides more reliable information about the state of the ionosphere.
Figure 8 shows a comparison of ionograms obtained in the daily monitoring mode with the same registration time for both of the ionosondes. The presented images clearly demonstrate that the behavior of the dependence of the effective height of reflection on the frequency of the radio pulse in both of the pictures is similar. Furthermore, the values of the effective reflection height obtained by the two ionosondes coincide, taking into account the time difference of 5 min. For example, f
OF2, determined from the ION-FAST ionogram, was 5.3 MHz, and from the CADI ionogram, 5.4 MHz. Although this comparison is made to check the results of ION-FAST, it is noticeable that the trace on the left graph is sharper and cleaner, and the dynamic range is greater. You can see additional examples of 60 s ION-FAST’s ionograms in
Figures S1 and S2.
The second mode is fast registration. The presence of this mode partly determined the name of the prototype. In this operating mode, ION-FAST allows obtaining ionograms with a time resolution of about 1 s, which is required to monitor the fast processes in the ionosphere during heating experiments, because the characteristic times of the processes of the redistribution of the electron concentration over height during such experiments are also of about 1 s. Such an experiment was performed in August 2022. The combination of two 53 m T2FD (transmitting and receiving) antennas provided a signal-to-noise ratio of about 32–34 dB (at the radiated power of 600 W) that affected the quality (noisiness) of the obtained ionograms in comparison with the 1 min averaging, though the altitude–frequency profile information could be recovered (see
Figure 9). In [
13], we presented some results with 1 s ionograms, but they were obtained on a set with more sensitive antennas that had better characteristics and gain.
4. Chirp-Ionosonde Receiver Trial
The ION-FAST is not only a vertical ionosonde. The diagnostic platform provides the basement for a set of different solutions for ionosphere diagnosis. The developed vertical ionosonde is currently a more mature and reliable instrument that has been tested for a long time in various modes, but recently, we became interested in our own implementation of an oblique chirp-sounding station, based on the developed hardware platform.
In the wake of increasing interest in short-wave radio communication, the task of developing new diagnostic equipment for monitoring the characteristics of radio links is becoming increasingly urgent. The ionosondes of oblique sounding using a chirp signal (chirp ionosondes) can effectively solve this problem. A comprehensive overview elucidating the need for a chirp ionosonde and its role in radio communication diagnostics can be seen in [
17]. For an adequate response to modern challenges, the developed equipment must meet the following criteria: multifunctionality, compliance with world analogues in terms of technical characteristics, and accessibility. SDRs have been successfully used for oblique ionospheric sounding since the late 2000s [
18]. The existing solutions can be divided into two categories: professional and DIY. Moreover, the latter are often in no way inferior to the former. Professional solutions use versatile, high-speed SDR devices (like the Ettus Research USRP N210 [
19]) and a suite of proven software [
20] that uses major software packages such as GNU Radio [
21]. DIY solutions use self-made SDRs (e.g., [
22]), usually combining ADCs and FPGAs that have an interface for communication with a PC (as in professional solutions). A distinctive feature of all of the noted solutions is that the PC receives a digital stream of samples with a bandwidth of ≥25 MHz, which provides coverage of the entire HF range (3–30 MHz). Processing such a high-speed stream of samples on a PC requires considerable computing resources and optimized algorithms. Another type of existing DIY chirp receiver uses software control of the HF transceivers (e.g., [
23]), which ensures that a received frequency changes synchronously with the chirp transmitter. In this case, signal registration on the PC is performed in a narrow band (usually 16–24 kHz) [
24]. In a number of solutions, a specialized direct digital synthesis (DDS) chip is used instead of an HF transceiver. In this case, a narrow reception bandwidth is achieved by implementing the digital down converter (DDC) function [
25,
26]. At this moment, we have a prototype of the receiving part of this kind of a station implemented on the existing ION-FAST platform. In the current implementation, our solution is close to solutions using DDS and DDC. The difference is that DDS and DDC in our solution are implemented not with the help of separate chips, but on the FPGA. In addition, in the future, our solution will not require a separate PC for DSP.
As mentioned before, the developed platform provides full control over the transmitting and receiving parameters of the signal, as well as over the entire cycle of digital signal processing. The prototype allows the expansion of functionality due to the implementation of certain transceiver functions on the FPGA. This section describes the implementation of the chirp ionosonde function based on a previously developed layout, and presents the first test results.
To carry out an experiment with a real chirp signal, we used another prototype of a fast vertical ionosonde, which was already available at the moment we started the experiment, at the “Vasilsursk” experimental base with minor changes in the analog and digital parts. At the moment, we are focused on exploiting just the receiving part of the ION-FAST device, considering dealing with the existing sources of the chirp-sounding signals.
The current version of the chirp ionosonde receiver consists of (1) a low-noise amplifier with a gain of approximately 20 dB; (2) a low-pass filter with a pass band of 20 MHz; (3) a Red Pitaya SDRlab 122-16 SDR device; (4) a Leo Bodnar precision GPS reference clock; and (5) an analog mixer. Unlike the vertical ionosonde prototype, in the chirp-sounding receiver, we used a combination of analog and digital down conversion. The external analog mixer and the low-pass filter, as a pair, handle conversion of the input chirp signal to a frequency of 1 MHz using a Red Pitaya device as the source of a local oscillator signal with linear frequency modulation. It is necessary to meet the Red Pitaya analog input requirements determined by a high-pass filter. Then, we perform the digital down conversion after the ADC on the FPGA using DDS with a fixed output frequency of 1 MHz. As in the vertical ionosonde, we used a Leo Bodnar Precision GPS reference clock as the source of the reference clock signal for the FPGA and the source of the PPS signal for synchronization of the sweep between the transmitter and the receiver. Another difference concerns the antenna we used. In the chirp-sounding ionosonde receiver prototype, we employed a CT-HF-FD antenna.
Figure 10 illustrates a block diagram of the newly developed chirp-sounding ionosonde receiver prototype.
In order to provide a chirp-ionosonde function, we had to make some changes to the firmware. First, we added an additional DDS to obtain an analog local oscillator signal. The phase increment is streamed to DDS from the FM_Modulator in order to provide sweeping (linear frequency modulation). After the subsequent digital-to-analog conversion, a cosine component inputs the analog mixer. To provide the intermediate frequency of 1 MHz, sweeping is upshifted to 1 MHz in comparison with the transmitter’s local oscillator. Second, we changed the operating mode of the digital down-converter. Now, the corresponding DDS that generates cosine and sine components has a fixed output frequency of 1 MHz. The other DSP on the FPGA remains the same (as realized in the vertical ionosonde). We envision using the Xilinx Zynq7020 Processing System in the future to perform spectrum analysis, normalization, framing, etc., but currently, in the first stage of trial operation, we are utilizing a PC for postprocessing.
Figure 11 illustrates a flowchart of the receiver firmware on the SDR 122-16 board providing the chirp-ionosonde function. We still use the advantages of the Processing System to start ionogram recording at any time, and we prepare schedules for chirp-sounding in accordance with the transmitter’s working time.
During trial operation of the receiving part of the chirp-sounding ionosonde, we decided to use the well-known transmitter located on Cyprus. That allowed us to compare the results obtained by our developed prototype and a commercial chirp-ionosonde receiver implementing an HF receiver control technique that we have in use. The transmitter we worked with has the following radiating parameters: a frequency range of 8–30 MHz, sweep rate of 100 kHz/s, and a broadcast time with a periodicity of 5 min, starting from the time of 00:00:20.
Figure 12 shows the oblique-sounding ionogram of the Cyprus-Vasilsursk short-wave channel, recorded using the SDRLab 122-16 board.
Figure 13 shows the same ionogram registered by the commercial chirp receiver and the inverted-V antenna. More examples of ION-FAST’s oblique ionograms are shown in
Figures S3 and S4.
It should be noted that the commercial ionosonde uses proprietary software, which implements ionogram filtering for stationary interference. This function has not yet been implemented in our ionosonde. Otherwise, all of the characteristic features of wave propagation along the HF channel, which can be emphasized from one ionogram, are also visible on the other. In
Figure 13, in the frequency range of 22–28 MHz, a reflected signal with delays of <9.5 ms is visible. This is a well-known artifact associated with the reception of strong signals and the auto gain control operation in a commercial ionosonde. Because of the fact that we have full control over the DSP in ION-FAST, such problems are excluded.
5. Discussion
While being field-tested, the ionosonde prototype displayed efficiency as a multipurpose ionosphere sounding tool that allows obtaining contrast ionograms in a monitoring mode, as well as 1 s ionograms for the fast diagnosis of ionospheric movements. We know that there is at least one more ionosonde with a declared ionogram registration time of 1 s [
27]. Its published technical characteristics significantly exceed the characteristics of our developed prototype. Distinctive features of our universal diagnostic platform, in comparison, for example, with the most famous commercial ionosonde DPS-4D [
28], are structural simplicity and low cost, which make it easy to replicate our development to equip ionosphere monitoring networks without significant financial investments.
In addition, the possibility of oblique sounding ionogram registration on the same hardware platform was demonstrated. Both vertical and chirp ionosondes provide ionograms comparable to those obtained using commercial ionosondes. The following can be noted as the distinctive features of the ION-FAST: low cost and availability of the components used in the development; extremely compact (even at the prototype stage); no need for additional computational resources; versatility and full control over DSP. We hope that these properties will ensure a high rate of adoption of ION-FAST by all testing sites of the Radiophysical Institute. This can become the basis for the geophysical monitoring network being developed, which, in a broad sense, is a continuation of the modernization of the unique “SURA” scientific installation, which began in 2020–2021 [
29]. The use of several spaced ionosondes in combination with a flexible control system will make it possible to obtain operational data on the speeds and directions of movement of traveling ionospheric disturbances. The selected platform (SDRlab 122-16) provides the possibility of combining and synchronizing several boards [
30]. Therefore, multichannel registration of ionograms using phased antenna arrays can be implemented, which will allow determining the polarization and the angle of arrival of signals reflected from the ionosphere [
31,
32]. Since the ION-FAST ionosonde has two receiving channels, it is possible to measure the frequency dependences of the Doppler frequency shifts of each ionospheric mode [
33] without additional hardware upgrades. This functionality is not new and has already been implemented in commercial devices [
27,
28]. However, we consider our implementation to be competitive since it involves combining all functions in a much simpler (in a computational sense) and cheaper device.
At this stage of development of the prototype, there is a limitation due to the technical characteristics of the development board we have used (SDRlab 122-16). For example, we cannot change the ADC parameters that directly affect the sensitivity and dynamic range of the analog path of the prototype. In the future, we plan to develop and manufacture our own version of the board, covering all our needs for recording, including low-power signals radiated from remote chirp transmitters.
The earliest plans for ION-FAST’s further modernization are to implement the transmitting part of the chirp ionosonde and to register ionograms on a completely controlled long-range HF path. Developing our own transmitting part will allow us to choose the start time of sounding, synchronizing it with the operating time of the SURA facility. This can provide more accurate information about the characteristics of radio links during heating experiments. In combination with the active influence on the HF radio-waves’ propagation, which can be provided by the SURA facility, this will allow organizing a comprehensive study of effects such as aspect scattering of radio waves from artificial ionospheric inhomogeneities extended along the geomagnetic field [
34]. It is also planned to implement a function for filtering ionograms from stationary interference.
During the active experiments at the “SURA” heating facility related to the ionospheric artificial airglow (557.7 nm atomic oxygen line) registration in August 2022, the ION-FAST prototype was used for a permanent control of the sporadic E layer and frequency with no necessity to put a technological pause into the facility operation scheme. The techniques of such experiments and the results of the analysis of data obtained are described in [
35,
36]. It was shown in [
13] that recording ionograms at a rate of 1 ionogram per second makes it possible to reveal that the characteristics of the Es layer can change quite quickly (in a time of the order of several seconds). Thus, provided that the radiative lifetime of the O1s level (corresponding to the 557.7 nm line) is 0.7 s, monitoring the Es layer with a time resolution of 1 s is necessary to unambiguously establish the connection between the presence of the Es layer and the generation of luminescence. In 2022, registration of artificial glow was carried out by several scientific groups at three points spaced from each other by 100–150 km. Each participant in the experiment had the opportunity to observe one-second ionograms in real time using video conferencing software (Zoom Cloud Meetings 5.11.10). The results of the 2022 experiments will be published soon. Thus, continuous monitoring of ionospheric conditions was carried out. Currently, no commercial ionosonde is able to provide such functionality.
6. Conclusions
In conclusion, the ION-FAST platform stands as an innovative and cost-efficient solution for ionosphere diagnostics, merging both vertical and oblique sensing capabilities. Through rigorous testing, this multifunctional device has showcased promising results. The platform’s versatility and low production cost herald a new era in ionospheric research and technological innovation. Looking ahead, our focus remains on enhancing the platform’s capabilities. Future modernization efforts will aim to broaden its diagnostic scope. Notably, ongoing plans involve registration of both polarizations (O and X modes), measuring Doppler frequency shifts for each ionospheric mode, multichannel registration and so on. These strides will not only bolster the device’s efficiency but also pave the way for expanded applications in scientific experiments and ionosphere monitoring.
Moreover, the platform’s successful integration and performance during active experiments in the ionosphere, notably in active experiments at the “SURA” heating facility related to the ionospheric artificial airglow (557.7 nm atomic oxygen line) registration in August 2022, signify its potential in enabling real-time monitoring and yielding invaluable insights into ionospheric behaviors. In essence, the ION-FAST platform, with its current accomplishments and envisioned advancements, holds the promise of significantly contributing to ionosphere research and scientific investigations, underscoring its relevance in the realm of multifunctional diagnostic tools.