MZI-based Method to Design Microwave-Photonics Processor for Ultra-Wideband Reactive Blocking of Any Unauthorized Wireless Channels

4.1 Traditional method using standard digital radio channel

Following the information from the previous section, all worldwide manufacturers produce purely electronic jamming devices, the operating principle of which is to introduce noise into individual working RF channels of UAVs, significantly increasing their circuitry’s complexity, as well as their response time, weight and size characteristics, and cost. The analysis conducted within the framework of patent research allows us to draw the following common conclusions regarding the current level of electronic blocker development.

1. All models, both [7, 14–15] serially produced and under development, are built based on a traditional, purely radio-electronic approach.

2. The following can be attributed to the general fundamental disadvantages of existing radio-electronic communication devices:

• The need for frequency scanning in the ultra-wideband frequency range to find, receive, and decrypt the command code leads to a relatively low response speed.

• The need for computer control for storing and additionally introducing into the received signals a masking noise interference from the optimal interference library, which also reduces the response speed.

• The complexity of the digital radio-electronic processor circuit leads to a deterioration in the weight and size characteristics, as well as an increase in the cost of the product.

We will conduct a specific analysis based on the most modern industrially produced UAVs of the FPV model. An FPV (first person view) drone is an UAV that allows the operator to control it from the pilot’s point of view, thanks to real-time video. An FPV drone has a camera that transmits images to special glasses, creating a sense of presence. FPV drones are often used for racing, aerial photography, and other activities where speed and maneuverability are important. In addition, the uncontrolled use of FPV kamikaze drones can pose a safety risk, especially when flying in restricted areas or near airports. Operating frequencies play an important role in controlling and transmitting video for FPV drones. They determine the connection between the RC and the UAV, as well as the quality and stability of the video stream. Common frequency ranges for controlling FPV drones include 2.4 GHz and 5.8 GHz, each with its own advantages and disadvantages. For video transmission, the most common ranges are 5.8 GHz, 2.4 GHz, and 1.2 GHz, which also have their own characteristics and limitations. In particular, the most common operating ranges of control and video channels for modern FPV drone models are provided in Table 1.

Conclusion: An obvious disadvantage of the traditional purely electronic suppression means is the need to select the current operating frequency of the suppressed channel, which is done by manual switching, whereas in modern FPV UAVs, this operation is performed automatically and randomly by the “No connection” signal on the control panel. This also significantly increases the complexity of their circuitry, response time, weight and size characteristics, and cost.

4.2 Newer method using the LoRaWAN protocol

For today, a new class of communication network titled “Long Range Wide Area Network” (LoRaWAN) is being deployed on a large scale in several countries worldwide. On the other hand, this widespread adoption has led to the need for a reliable and effective solution to such an important and complex task related to blocking an unfriendly radio channel, for example, when using LoRaWAN to activate RC explosive devices planted at a ground object, as well as blocking the radio control channels of modern UAVs. As is known, LoRa technology is a powerful and versatile wireless communication protocol widely used in Internet of Things (IoT) networks. Now, it has found application in different industries such as utilities, medicine, agro-industry, security, urban infrastructure management, and environmental quality monitoring. Below are the results of an analytical study, which showed nine advantages of its deployment [16–18].

1. Long range: As its name suggests, the key feature of LoRa technology is the ability of devices to operate over long distances. In particular, they are capable of transmitting data over several kilometers in open space, ensuring uninterrupted communication even in remote areas. For example, using the LoRa protocol, information can be transmitted over a distance of up to 15 km in open areas. Such a long range is essential for the implementation of smart agriculture, smart cities, and industrial monitoring, where devices can be dispersed over vast areas.

2. Low power consumption: LoRa technology is optimized for devices with low power consumption. This advantage has made it popular for battery-powered IoT devices. Thanks to the technology protocol, the battery life reaches five years, which reduces the cost of regular maintenance and increases the overall efficiency of the entire system, where hundreds or even thousands of sensors are able to operate.

3. Cost-effective solution: The cost of devices and infrastructure deployment is one of the key factors in any IoT network. LoRa technology offers a cost-effective solution: the infrastructure required to create and configure the network is relatively simple and inexpensive compared to other wireless technologies. LoRa-enabled devices have a low cost of production and purchase. All this makes them accessible for various IoT projects, including large-scale deployments.

4. Scalability: Scalability is an important aspect of IoT networks. When creating such projects, investors expect that, over time, the network will support more devices or expand its coverage. LoRa technology can offer fast and universal scalability. For example, the LoRaWAN protocol, built on LoRa technology, is able to connect an unlimited number of sensors, meters, and other devices. A system deployed based on this protocol allows for easy expansion of the infrastructure and effective management of devices in large-scale IoT projects.

5. Security: LoRa includes robust security features to protect sensitive information. The AES-128 protocol is used to encrypt information and secure communications. It protects data from unauthorized access and tampering. In addition, LoRaWAN uses device authentication mechanisms, which prevent unauthorized devices from connecting to the IoT network.

6. Bidirectional communication: LoRa provides continuous two-way communication between devices and gateways. This feature allows sensors to not only receive commands but also transmit data back to the network – for example, simultaneously receiving instructions and transmitting data. This is very important for projects that require real-time control and monitoring.

7. Geolocation services: LoRa technology offers geolocation capabilities, eliminating the need for additional GPS hardware to track and position devices. Utilizing the time difference of arrival (TDoA) method, LoRaWAN networks accurately estimate the location of devices. This feature finds application in projects such as asset tracking, logistics management, and geomagnetic sounding.

8. Interoperability: LoRa protocols have been developed following open standards, ensuring efficient interoperability between different devices and LoRaWAN networks. In particular, the receiving and transmitting equipment can operate at one of five speeds, including 0.1, 1, 10, 100 кбит/с, and 1 Mbit/s, in nine frequency bands, including 170, 230, 433, 470, 868, 915, 2,400, 5,200, and 5,800 MHz.

9. High protection against jamming the radio channel: LoRaWAN is protected from interferences that degrade the quality of signal transmission by taking the following measures: i) the low intended transmission rate of LoRa devices limits the probability of collisions, ii) channel hopping spreads messages across channels, reducing the probability of collisions, and iii) LoRa devices can trade data rate for sensitivity to punch through noise [19].

As a result, this new class of communication network is being deployed on a large scale in several countries worldwide. On the other hand, the above-mentioned advantages have led to the need for a reliable and effective solution to such an important and complex task related to blocking an unfriendly radio channel – for example, when using LoRaWAN to activate RC explosive devices planted at a ground object, as well as blocking the radio control channels of modern UAVs. Currently, this problem is solved using various hardware approaches and protection tools. However, judging by the literature, the most acceptable method widely used for protection against UAVs remains the introduction of noise into the radio control channel. Of course, this approach has limited effectiveness for such a highly protected network as LoRaWAN, which normally operates at a signal-to-noise ratio of −30 dB.

Taking the above into account, in this paper we propose and preliminarily analyze another approach related to the introduction of false signals into the commands transmitted via the control channel from the operator console, which we previously developed for protection against UAVs [7, 14]. A fundamentally important advantage of the proposed approach is the implementation of the blocker circuit based on the microwave-photonics principle, which ensures its simultaneous operation at all of the above radio frequencies and digital information transmission rates.

5.1 The prototype

To confirm the correctness and efficiency of the proposed approach and to conduct confirmatory tests, the layout of the MPJ prototype was designed and built based on the circuits in Figures 3 and 4. Its appearance, with the RF-transparent cap removed, is shown in Figure 5 [10].

As it follows from the Figure, the layout consists of a rectangular case, inside which are located a battery of Virtustec (21 V), boards for converting the supply voltage to 12, 5, and 3.3 V, a battery charging board, input and output RF amplifier modules, and an MPP module, including laser and photodetector control boards, laser and photodiode modules, a low-noise amplifier module, and an FO-MZI module. On the rear wall is located the antenna block, consisting of a receiving (on the left) and transmitting (on the right) Vivaldi antennas (the width of the radiation diagram is about 60°, and the gain is near 7 dBi). In addition, the antennas are positioned at an angle of 90° and are separated by a metal shield, which increases the isolation of the receiving part. The prototype also has a front panel that includes a power switch and a battery level indicator, and it runs on a built-in battery for at least 1.5 hours. There are also two optical connectors on the front panel, the main function of which is to quickly select the optimal length of the L-shaped FO-MZI bracket (see Figure 4). This is achieved by connecting an external SMF-28 optical fiber of a certain length. The external dimensions of the prototype body are 330x300x80 mm, with a weight of 2.8 kg.

The main parameters of the prototype can be justified as follows. As is known, the loss of a radio signal in free space at a distance of 1 km (taking into account the antenna gain) is about 90 dB at the upper operating frequency of 6 GHz. Therefore, with an output signal power of 1 W, the minimum input power level should be near −60 dBm. According to the manufacturer of the semiconductor laser used (DFB-1550-EAM-12, Optilab, USA), the level of the modulating signal should be in the range of −5.3 dBm. In addition, the calculation of the transmission coefficient of the MPP (see Figure 4) taking into account the insertion losses in two single-mode optical splitters included in the FO-MZI and the sensitivity of the photodetector (PQW20A-L, Albis, Switzerland), is about −35–40 dB. Taking into account the above data, the gain of the input amplifier is about 55 dB. This is implemented using two cascaded MSLA2-20,180-4.0 M amplifiers (manufactured by JSC Microwave Systems, Moscow). In addition, a pre-RF amplifier (gain factor of 40 dB), included in the MPP, is implemented in-house. The RFLUPA02M06GA power amplifier, manufactured by RF-Lambda, USA, is used as the output amplifier.

6.1 Computer simulation for LoRaWAN jammer

The general diagram of the proposed LoRaWAN jammer corresponds, in general, to Figure 3, except for the FO-MZI diagram. Our preliminary research has shown that the most likely and simple way to increase the level of RF channel blocking effect in a highly jam-resistant network such as LoRaWAN is to increase the number of false signals, which can be easily achieved by increasing the number of FO-MZI arms. The structural diagram of the four-arm-based FO-MZI, proposed for further research, is shown in Figure 6, where the following abbreviations are adopted: OS – optical splitter, OA – optical amplifier, SLE – semiconductor laser emitter, PD – photodetector. It should be noted that the OAs introduced into each branch are intended to compensate for losses, which should lead to an increase in the jamming effect.

6.1.2 Approach

For the preliminary analysis of the method of suppressing radio control using LoRa technology, a mathematical model of a transmitter and receiver utilizing the LoRa modulation and coding scheme is employed. The model is implemented in the MATLAB environment (Engage version). A LoRa frame consists of the following sequence of symbols:

x(t)=∑n=1NmSn(t−nTs)E2

where Nm is the total number of symbols in the frame. That is, the simulation considers the transmission and reception of data packets, which are compared with the checksum at the receiving end during verification. An illustration of a LoRa frame, including the length, spreading factor, and coding rate used to encode the symbols, is shown in Figure 7.

In the Figure, the preamble consists of padded symbols that do not represent any information but are used for synchronization purposes. The payload is where the message is encoded and can be up to 255 bytes long.

The proposed LoRa receiver emulator uses a two-step process in which the received signal is first compressed and then demodulated based on multifrequency shift keying. To perform the compression, the received signal is mixed with a sequence of inverted signals without frequency shifting. The compressed signal is represented as follows:

s*(t)=∑n=1Nmexp(jπBt−j2πβt2)⊗δ(t−nTs)E3

The compression signal is then mixed with the received signal:

y(t)=[g{x(t)}+i(t)+n(t)]s*(t)=r(t)+i(t)+n(t),E4

here, g(x) – channel function; i(t) – radio spectrum interference; n(t) – additive white Gaussian noise.

In the model used, a noncoherent demodulation is achieved by implementing the energy determination in the standard minimal frequency shift keying signal using a fast Fourier transform (FFT), which extracts the dominant component of the power spectral density (PSD), that is, its peak. This dominant frequency component represents the original shift, which encodes the symbol value. Also, the model implements error-correcting Hamming coding. LoRa uses a Hamming code of 4/(4 + CR), where the data word length is 4 and is transformed into a code word length of 4 + CR. The payload can be encoded using any of the Hamming code combinations, while the header is always encoded using a Hamming (4,8) code. A sign of signal suppression is the appearance of errors in the received message after its detection and decoding.

6.2 The results

First of all, the simulation was carried out at the nominal power of the radio signal and the maximum frequency deviation. The generated spectrum of the signal at the output of the LoRa transmitter is shown in Figure 8. The spectrum shown in Figure 9 was obtained at the receiver output when the radio signal power was reduced to –130 dBm.

Further reduction of the radio signal power led to errors in the data structure and in complete loss of the transmitted packet. For example, with a radio signal power of −131 dBm, the received message is decrypted as: “Microwave photonic radio suppresSIMl”3 = 74ai.” Thus, the receiver sensitivity threshold is configured in the model.

In the subsequent studies, numerical simulations were performed to determine the presence of the LoRaWAN signal’s blocking effect and to calculate the optimal delay time from the point of view of blocking at all operating frequencies in the arms of the FO-MZI for three of its configurations, including two-arms, three-arms, and four-arms. For this purpose, the model of the proposed blocking device (see Figure 6) was introduced between the LoRa transmitter and receiver models. The results of calculations for the above FO-MZI option are presented in Table 3, where the relative values of the time delays between the arms are considered, designated as: τ₂₁ for 2-arm, τ₃₂ for 3-arm, or τ₄₃ for 4-arm FO-MZI configuration.

The following conclusions can be drawn from the table.

1. The optimal values of the relative time delay in the two and three-arm FO-MZIs turned out to be completely identical, but they almost doubled in the 4-arm FO-MZI.

2. In all 4-arm FO-MZI configurations, the value of the optimal relative time delay increases approximately threefold when transmitting a digital signal at the lowest required speed.

3. In the result, it was shown that the optimal principle for designing a device for blocking an RF signal transmitted in such a highly noise-immune network is the use of a microwave-photonic processor based on a 3-arm fiber MZI.

6.3 Laboratory studies

To prove a synchronous blocking effect on all carrier RF of a modern UAV and to determine the requirements for the optimal delay difference in the 2-arm FO-MZI, laboratory studies of the RFB-1 prototype (see Section 5) were conducted. In order to exclude the influence of local CNSS and WiFi channels operating at close frequencies, the tests were carried out without antennas. Their influence was modeled by introducing appropriate attenuators into the measuring setup. Both tests of the prototype under study, the main components of which are described in detail in sub Section 5.1, were carried out using standard methods described in the operating manuals of the measuring instruments specified below.

First, the overall transmission coefficient of the model was measured using the C1209 Vector Network Analyzer from PLANAR LLC, Chelyabinsk. The measurement results, in the form of a screenshot from the analyzer, are presented in Figure 10, where mark 1 shows the transmission coefficient at a frequency of 0.9 GHz, mark 2 at a frequency of 2.4 GHz, and mark 3 at a frequency of 5.8 GHz.

Thus, the unevenness was no more than 5 dB across the entire measurement band of 0.9–5.8 GHz.

Furthermore, a number of laboratory tests were carried out in the time domain using a real-time oscilloscope, KEYSIGHT UXR0402AP. For testing, the remote controller of a modern Autel EVO Max 4 T UAV (0.9, 2.4, 5.2, 5.8 GHz) was applied to the input of our prototype (see Figure 5). An example of measurement results at a frequency of 0.9 GHz is shown in Figure 11 in the form of screenshots from the oscilloscope. The comparison of oscillograms for the direct connection of a UAV controller (Fig. a) and through the tested MPJ (Fig. b) shows that the signal processing in the MPJ led to a chaotic appearance of their shape, processing the pulse and pause, which argues the presence of a blocking effect on the signal during transmission.

As is known, the quality of transmission of RF signals with digital modulation is standardly determined by a standard parameter known as error vector magnitude (EVM), which was measured in the course of laboratory studies using a Keysight UXR0402AP real-time digital oscilloscope and the corresponding RF signal from a Keysight M9384 vector signal generator that is applied to the input of the RFB-1 prototype. As a result, Figure 12 shows the graphs of the dependence of the EVM values on the L and L0 FO-MZI arm’s time delay difference (TDD) for the standard BPSK communication format, which is widely used in UAVs, at transmission rates from 10 kbps to 2 Mbps and for all RF carriers of the UAV under test. It should be noted that the horizontal dotted lines in the graphs show the standard EVM threshold of the digital RF signal retransmitted by the FO-MPJ under test, equal to 17.5% for a BPSK RF signal.

As it follows from the Figure, the same dependence of the transmitted signal quality on TTD is observed at all speeds and all RF carriers, namely, acceptable transmission quality for practically identical TDD arms of the FO-MZI-based MPP and its degradation to unacceptably high values, when their time delay exceeds the above threshold.

Thus, from the obtained experimental results, one can make an unambiguous conclusion that the proposed RFB-1 scheme allows reliable and effective blocking of any RF signals from the UAV’s operator console, with a minimum delay difference in the MPP arms of 270 ps, which corresponds to an inequality in their arm lengths of only 5.5 cm.

6.4 Preliminary field tests

The above laboratory test results were qualitatively confirmed in two field tests using UAV models of the Autel EVO Max 4 T and DJI Mavic Pro 2. The test results are presented in Table 4.

As follows from the Table, the preliminary full-scale tests conducted showed the general operability of the developed principle of designing the device for blocking radio channels of modern UAVs. The deterioration in the reliability of blocking with increasing distance from the UAV to the MPJ sample is due to its insufficient output power and the non-optimality of the Vivaldi antennas used.