Unconsidered but influencing interference in unmanned aerial vehicle cabling system

ABSTRACT


INTRODUCTION
The development of information technology and electronics in unmanned aerial vehicles (UAVs) has led to increasingly complex electrical and electronic systems [1].However, the limited compartment space in UAVs poses a potential source of unwanted electromagnetic interference (EMI) [2]- [4].Previous research has shown that 60% of EMI occurrences on aircraft are caused by electromagnetic interference coupled to UAV wiring systems [5].Cable length has been identified as a major factor influencing the electromagnetic susceptibility of devices to EMI, both from internal and external sources [2], [6]- [9].The interference problem in UAVs is increasingly complex due to the structure of the UAV body, which is generally made of lightweight and strong composites, providing a fairly good level of flexibility compared to metal.This UAV structure offers flexibility but is less effective at withstanding electromagnetic interference compared to metal materials [7], [10].
Several research studies have investigated methods for predicting radiated emission from cables and reducing EMI.One such approach involves utilizing the simplified simulation program with an integrated Int J Elec & Comp Eng ISSN: 2088-8708  Unconsidered but influencing interference in unmanned aerial vehicle cabling system (Imas Tri Setyadewi)

23
circuit emphasis (SPICE) model [11].The result estimates the amount of emission (current) received by cables starting from the frequency range of 60 MHz to 1 GHz in the vertical antenna polarization direction [12].Another method focuses on reducing the radiation absorbed by printed circuit boards (PCBs) in the farfield at a distance of 1 meter, achieved by incorporating a smart patch loop on the PCB, resulting in an average emission reduction of 83% [13].However, this method can only be applied to integrated circuit (IC) circuits.
Investigations have also been conducted to explore the impact of radiation on different types of cables at varying distances from the source [14].Using a radiation source from a supply cable with an input signal of 10 V at a frequency of 2 MHz, the amount of crosstalk in the receiving cable was analyzed by adjusting the distance from 1 to 10 cm.The results show that distances exceeding 4 cm from the source can reduce the amount of EMI received by up to 60% [14].Additionally, cables with copper shields demonstrated a 50% reduction in EMI at a distance of 1 cm compared to cables without shields [15].
A measurement of radiated electromagnetic emission using a shielded cable by varying the height of the cable to the ground has been conducted.Parameters that affect emissions, such as current distribution, dielectric properties of the ground (reflection coefficient), and reflected current on the cable, were considered [3].As a result, this measurement model can be used to predict emissions from shielded cables on certain ground materials, but the distance factor of the cable to the source has yet to be determined.
Similarly, Park et al. [16] developed a measurement method by simplifying the estimation of common mode radiation from a cable connected to a real device, which is a mobile device, by measuring the common mode current.The simulation is carried out by modeling a mobile device like a conducting box.The equivalent source between the conducting box and the power cable can be determined by measuring the common mode current in the cable, which contains all the noise coupled to the body of the mobile device [16].This measurement can only be used at frequencies below 300 MHz to estimate EMI at the radiation peak.
Although previous studies have explored methods to mitigate interference in UAV systems, the effects of cable distance and polarization angle have not been adequately addressed.Therefore, this paper proposes an alternative approach to investigate and analyze the effects of these two factors, comparing the results obtained from measurements and simulations.Understanding the influence of cable distance and polarization angle will provide valuable insights for designing UAV cable line systems and configuring onboard equipment to effectively mitigate interference effects on UAV system operation.

METHODS 2.1. Measurement setup
Figure 1 illustrates the setup used for measuring electromagnetic radiation emission.The measurements were conducted in a semi-anechoic chamber, where the antenna was placed at various distances (L) based on the characteristics of the wave impedance.These distances included the near-field at 0.1 m, Fresnel at 0.33 m, and far-field at 1 m away from the cable under test (CUT).The calculation of the near-field, Fresnel, and far-field was based on the wavelength (λ) and length of the antenna [17].The VERT dipole antenna was used, with an operating working frequency of 824 to 960 MHz.For this study, a frequency of 905 MHz was chosen as it is widely used for UAV working frequencies.To measure the variation in polarization angle, the antenna was positioned at different angles from the cable [18].This included 90°, which means perpendicular to the cable, at 45° and 0° in a horizontally laid position with the cable.The cable used was a single wire cable pair, 0.7 m long, and had a resistance of 50 Ohms at each end.The cable was placed horizontally on a foam table with a height of (h) 1 m above the ground chamber.The FieldFox microwave spectrum analyzer N9917A was connected to the other end of the cable as an instrument reader output.The transmit power of the EXG Keysight N5173B analog signal generator to the antenna was varied from -19 to 20 dBm in increments of 5 dB on both vertical and horizontal polarization antennas.
The magnitude of the radiation intensity received in the far-field was measured in the form of a power receive (Pr).The amount of power received was calculated as stated in (1).
is the radiation efficiency of the receiving antenna, where   is the directivity receiving antenna in the direction   , ∅  , and power density (  ) is calculated using (2).
where   is the transmit power,   is the gain of the transmitting antenna in the direction   ,   , and R is the distance between the transmitting and receiving antenna [19].The power received represented in (1) assumes that the transmitting and receiving antennas were matched, and the polarization of the receiving antenna was the same as the transmitted polarization.

Cable simulation
Cable simulation using the computer simulation technology (CST) Cable Studio has been conducted to compare with the value obtained by laboratory measurements.The simulation was carried out at a far-field distance by using a plane wave as a radiation source.The magnitude of the electric field intensity (V/m) was calculated using (3).
The magnitude of the electric field (E) was obtained by multiplying the power density (  ) parameter received based on the calculation with the air impedance in the field area.In the far-field, the air impedance is equivalent to 377 ohms or the value of the impedance with space impedance is the same ( 0 =   ).In the near-field, the impedance (  ) is a function of distance, as shown in Figure 2, so (4) can be used [17], [20], [21].Using (4) and refers to Figure 2, a   value of each field can be obtained.
The simulation used schematic circuits that were identical to the measurement configuration in the chamber, as shown in Figure 3.A plane wave was used as a radiation source by providing   input according to the field area.The simulation results were expressed in the form of power induction received by the cable in dBW.

Measurement radiated susceptibility of cable
The electromagnetic interference was measured by varying the transmitting power of the dipole antenna source at polarization angles of 0°, 45°, and 90° over a 0.7 m length of cable that was laid on a table in the horizontal direction, as shown in Figure 1.The frequency source was set to 905 MHz, and the measurement results were monitored by the spectrum analyzer and analyzed by comparing them with the simulation.The measurement results are shown in Figure 4.
Figure 4(a) shows the values of the power radiation received by the cable when the distance between the cable and the source was 0.1 m or in the near-field area.Meanwhile, the radiation received value by the cable at a distance of 0.33 m or in the Fresnel zone area was shown in Figure 4(b), and the radiation in the far-field area was shown in Figure 4(c).The Y-axis represents the amount of radiation received by the cable, while the X-axis represents the power transmitted to the antenna ranging from -20 to 19 dBm with an increment interval of 5 dBm.From the overall results obtained, it was observed that there was a linear relationship between the transmitted power of the source and the radiation received by the cable.An increase in transmit power led to an increase in the amount of radiation received by the cable, which was in accordance with (1) and (2).Therefore,   was linear to   .For every 5 dBm increase given on the cable, an average increase of 4.8 dBm was obtained.
The influence of distance on electromagnetic interference received by the cable was found to be significant, particularly at distances above 0.33 m.The maximum radiation values of -19.36 and -20.9 dBm were observed at the Fresnel and far-field distances, respectively, in the polarization direction of 0° (square symbol).Conversely, the highest radiation was recorded at the near-field distance with a polarization angle of 90° (triangle symbol), reaching -12.5 dBm.It was observed that increasing the distance between the cable and the source resulted in a linear negative relationship, causing a decrease in the radiation received by the cable.This relationship was expressed by (1), which demonstrated that in the far-field, the radiation received decreased quadratically as a function of the distance from the source.The calculations revealed that the magnitude of radiation received was within acceptable limits, decreasing quadratically in proportion to the emitted power, as indicated by the equation for a radiation magnitude of 19 dBm in Figure 4(c).However, this trend did not apply to the polarization direction of 90°, where the highest radiation value was obtained at the near-field distance.Furthermore, the radiation value at a polarization angle of 90° was found to be significantly lower in the far-field region, with a difference of up to 15 dBm compared to the near-field distance.The findings highlight the importance of considering cable distance and polarization angle in EMI analysis.The influence of antenna polarization can be analyzed based on the polarization loss factor (PLF) equation, where the polarization mismatch causes the power received by the receiving antenna to be less than the maximum.Polarization losses can be defined as the magnitude of the cosine vector angle between the wave vector and its polarization vector.The PLF equation is given by ( 5) [19].
where  ̂ is a wave unit vector,  ̂ is a polarizing vector and   is the angle between the two vector units.If the angles of the two vectors meet, which means they form an angle of 0°, then the magnitude of the PLF in decibels (dB) will be zero (7).This is because the electric field of the antenna will be oriented in the same direction as the conductors in the cable, resulting in a stronger coupling between the antenna and the cable.
() = 10  10  = 0 (7) If the polarization angle is 45° (circle symbol), then the loss factor is 3 dB.Figures 4(b) and 4(c) demonstrate the connection between polarization and the amount of energy received, which was smaller at 45° polarization than at 0° polarization.The radiation received was smaller by an average of 5 dBm.When the distance to the source was 0.3 m in the Fresnel zone, the radiation output at a 45° angle was closer to the same result as the 0° angle, but when the distance was 1 m, it shifted closer to the same polarization value as the 90° angle.Based on ( 6) and ( 7), if the PLF angle was 90°, then the loss factor became infinite.This is because the electric field of the antenna will be oriented perpendicular to the conductors in the cable, resulting in a weaker coupling between the antenna and the cable [22].

27
This behavior can be explained by a decrease in the dominance of one of the fields, either magnetic (H) or electric (E) field, as shown in Figure 5.As the distance approaches the far-field, the E and H values will converge to the same value, forming a plane wave.The field in the Fresnel zone has a large attribute where the electric field and magnetic field almost resemble plane waves, but there is still some influence of field reflection from the near-field.In contrast, when compared to radiation received in the near-field, the greatest polarization is received at an angle of 90° or perpendicular to the cable.In the near-field, one of the field components dominates, and the waveform is not in phase and does not form a plane wave.Therefore, 90° produces a larger value which could be due to the higher magnetic field (H) in that area compared to the electric field (E) [20].

Simulating the polarization effect
To establish a relationship between the measurement parameters and ideal simulation values, simulations were conducted to estimate the radiation received by the cable.The electric field intensity obtained from the power density calculation in (1) was used to determine the magnitude of the intensity in the far-field.This value served as an input for the field intensity in the simulations, which were performed using plane waves at an antenna polarization angle of 0°.This configuration ensured that the electric field direction was parallel to the cable.The simulations were conducted at a far-field distance of 1 meter, as (1) specifically applies to the far-field region.Therefore, the electric field intensity value assigned to the plane wave was only valid for the far-field distance [19].
The simulations conducted in this study focused on a transmit power range of 0 to 19 dBm.This range was chosen due to the linear relationship observed in the simulation values, enabling reliable predictions.Thus, within this range, it was expected that a correlation could be established between the measured values and the simulation results.Figure 5(a) illustrates that the simulations exhibited similar trends to the measurements, indicating that higher transmit power resulted in increased radiation received by the cable at polarization angles 0° and 45°.However, at a polarization angle of 90°, there was a notable difference in the power received by the cable between the simulation and measurement results, as depicted in Figure 5 The simulation results aligned with the PLF equation, which suggests that the power received by the cable is reduced due to weaker coupling between the antenna and the cable caused by polarization mismatch.Conversely, the measurement results also indicated that power coupling still occurred from the antenna to the cable.It is important to acknowledge that the PLF equation assumes idealized conditions and may not fully capture all the factors influencing polarization losses in real-world scenarios [19].Antenna polarization and its impact on electromagnetic interference can be influenced by various factors, such as the environment, multipath propagation, and antenna characteristics.Therefore, a practical analysis of polarization losses often requires more complex models and measurements.
Furthermore, when analyzing electromagnetic interference and polarization effects in practical scenarios, additional factors and techniques need to be considered.The simulations accounted for cable connectors and reflections, which can affect the overall measurement value [23].Additionally, the simulations utilized plane waves as a source without considering correction factors for the antenna's radiation pattern.As a result, reflections, whether phased or unphased, can either enhance or weaken the power received by the cable.Some reflections may interfere with the original wave, leading to constructive or destructive interference.When a plane wave encounters the cable, reflections can occur, influencing the cable's received power based on the phase and amplitude of the reflected wave.
To address these factors, correction factors may be incorporated into the antenna's radiation pattern through measurements or simulations that consider the reflections and environmental geometry.These correction factors improve the accuracy of the simulations and account for the impact of reflections on the received power [24]- [27].By integrating correction factors into the simulations, inaccuracies can be mitigated, leading to enhanced overall accuracy of the results.When simulating the performance of a cable receiving a plane wave from a real antenna, it is crucial to consider the possibility of reflections and incorporate appropriate correction factors to ensure more reliable and accurate outcomes [28].

CONCLUSION
In this study, a comprehensive analysis of the electromagnetic radiation effects on UAV cable systems was presented, considering transmit power, polarization angle, and distance from the interference source.The measurement results demonstrated a linear relationship between the transmitted power of the source and the radiation received by the cable.The results revealed that multiple variables significantly influenced the amount of electromagnetic radiation received by the cable.Moreover, the impact of the polarization angle varied between the near-field and far-field regions.In the near-field region, the greatest radiation was received when the polarization angle was perpendicular to the cable, whereas, in the far-field, the greatest radiation was received when the polarization angle matched the cable.
The findings highlighted the importance of considering cable distance and polarization angle in EMI analysis.The PLF equation provided insights into the impact of antenna polarization on the power received by the cable.The simulations and measurements showed that the PLF equation was applicable, with polarization mismatch resulting in a reduction in the power received by the cable.However, the measurements also indicated that power coupling still occurred from the antenna to the cable, suggesting additional factors influencing polarization losses in real-world scenarios.To improve the accuracy of simulations, correction factors are suggested to be applied to the antenna's radiation pattern, taking into account the possibility of reflections.
Overall, understanding the effects of cable distance, polarization angle, and reflection phenomena is essential for designing UAV cable line systems and configuring onboard equipment to effectively mitigate EMI and ensure optimal system operation.Future research should consider more complex models and measurements to capture the diverse factors influencing polarization losses in practical scenarios.

Figure 2 . 25 𝑧
Figure 2. Field impedance as a function of distance

Figure 3 .
Figure 3. Schematic diagram in CST cable studio