The magnetoelectric (ME) coupling effect is a more effective approach for reducing antenna size. Nevertheless, at low frequencies, a single ME structure generates a weak signal strength and a low signal-to-noise ratio (SNR). This study utilized an array strategy to improve the radiation and induced electromagnetic field performance of ME antennas. In this study, the array ME coupling structure, composed of Metglas and Pb (Zr1−xTix) O3 (PZT) bilayers, was operated through ME coupling resonance modes. The relationship between the sensitivity and SNR of the serially connected ME antenna array and the number of series connections was determined. On the transmission side, the impact of the multi-source power supply modes on the radiation intensity and directivity of the ME antennas was analyzed. The sensitivity, SNR, magnetic detection limit, directivity, and radiation range of the single ME and array ME antennas were tested for the key parameters. Finally, it was demonstrated that at a frequency of 31.75 kHz, the array strategy achieved a low-frequency signal transmission distance of up to 48.1 m, which is 1.7 times that of a single ME antenna. This array strategy significantly enhances the radiation intensity and magnetic field SNR of ME antennas in the low-frequency range, demonstrating its application prospects in the field of low-frequency communication.

Wireless communication technology has greatly reduced communication costs and significantly promoted the exchange of social information.1,2 However, when high-frequency electromagnetic waves propagate in highly conductive environments such as seawater and mines, they experience rapid attenuation. Consequently, this severely hinders the information transmission of information.3,4 In contrast, the low-frequency (LF) band (30–300 kHz) offers a promising solution owing to its extended wavelength (1–10 km), which ensures reduced signal attenuation and strong penetration, thereby exhibiting significant potential for underwater communication applications.5,6 However, traditional antennas need to match their size with low-frequency electromagnetic waves, and miniaturization of the antenna size has reduced their gain.7,8 This leads to bulky LF communication systems,9 which make it difficult to realize information transmission on ocean platforms. Attempts to miniaturize LF antennas have primarily focused on three strategies: rotation of permanent magnets, piezoelectric resonance, and magnetoelectric (ME) coupling. A surge in research dedicated to the design of LF mechanical transmission antennas based on rotating magnets10–13 and piezoelectric resonance13,14 has been reported. Nevertheless, these approaches tend to yield low power and slow communication rates, rendering them suitable only for one-way communication in transmission antennas.15,16

Unlike traditional antennas that operate using oscillating-induced electromagnetic waves, ME antennas consist of a magnetostrictive layer and a piezoelectric layer. During transmission, periodic vibrations of the piezoelectric material induce the magnetostrictive material to radiate electromagnetic waves via the converse ME coupling effect. At the receiving end, the magnetic component of the electromagnetic field is sensed by the magnetostrictive material, which undergoes deformation. This deformation was transmitted to the piezoelectric layer to generate an electrical signal. Therefore, the ME coupling structure emits and receives an LF magnetic field. Because the ME coupling structure operates through its electromechanical resonance (EMR) frequency, it has miniaturization potential.17–19 

This process has attracted considerable attention in recent years. Potential applications in low-frequency (LF) and very-low-frequency (VLF) bands are highly anticipated.20–22 In 2020, Dong validated the emission and reception performance of the ME antenna and achieved near-field antenna modulation at 23.95 kHz.23 In 2022, Niu developed a self-transmitting and self-receiving ME antenna communication system that improved the communication distance using integrated bias magnets.24 An ME antenna and modulation and demodulation circuits achieve a signal transmission of 5.7 m via ASK modulation by Zhu et al.25 Joseph provides an experimental validation of the operating principles of a near-field multiferroic antenna, achieving dynamic magnetic field generation through a piezoelectric lead–zirconate–titanate stack inducing strain in a magnetoelastic iron gallium (Fe80.77Ga19.23) rod.26 

However, the actual communication distance of an LF communication system is still limited to a short range, which severely limits its potential for LF applications. Owing to size limitations, the transmitting signal strength and the receiving sensitivity of ME antennas still need to be improved. The array method has been proven to significantly enhance both aspects, with Dong et al.27 improving the relative bandwidth to 12.5% using an ME antenna array composed of three units. Zhang verified the array gain of the ME antenna, boosting the radiated signal by 2.5 times and expanding the bandwidth of the ME receiving antenna by 1.86 times through a three-unit array.28 In brief, ME antennas demonstrate considerable promise in the domain of LF communication. However, the drawback of a single unit’s subpar performance impedes its advancement. By implementing an array approach, the radiation and magnetic induction capabilities of the ME antennas can be significantly enhanced.

In this study, we designed and fabricated an antenna and its array based on a sandwich-type ME coupling structure consisting of a layer containing double-layer Metglas and PZT5H. This study elucidates the ways in which array strategies enhance the radiation performance and electromagnetic field-sensing capabilities of the ME coupling structure. Implementing a dual power supply mode at the transmission end significantly enhances the signal strength of the LF electromagnetic field radiation, achieving an amplification of 1.9 times compared with a single ME antenna. At the receiving end, the evaluation of the ME antenna array with two and three antennas in series configuration showed an increase in signal strength of 1.7 times and 2.5 times, respectively, compared to a single ME antenna. Moreover, it was verified that the LF electromagnetic field radiation from the ME coupling structure was driven by a dual power supply and sensing of the LF electromagnetic field through two series-mode ME coupling structures. An LF magnetic signal transmission distance of 48 m was achieved, surpassing the 1.7 times distance of traditional single-ME antenna configurations. Demonstrated Amplitude Modulation (AM) communication at 100 Hz with a working frequency of 31.75 kHz confirms the communication capability of the ME antenna array. This study provides a potential solution for LF communications by theoretically analyzing and experimentally verifying the impact mechanisms of array strategies on the radiation performance and magnetic field sensing capabilities of the ME coupling structure.

The ME coupling effect enables the radiation and reception of electromagnetic signals, with the highest signal radiation strength and reception sensitivity occurring in the resonant state of the ME coupling structure. ME antennas achieve signal radiation and reception based on the resonance of the ME coupling structure. In ME antennas, the magnetic material is Metglas and the piezoelectric material is PZT5H. Magnetostrictive materials are magnetized along their length, and piezoelectric materials are polarized along the longitudinal direction. The operating mode of the ME antenna was the L–T mode.

The transmission mode of the ME antenna is based on the converse ME-coupling effect. An AC signal applied to the piezoelectric layer in the ME coupling structure causes the piezoelectric slab to generate periodic vibrations, owing to the reverse piezoelectric effect. These vibrations, which are transferred through the heterojunction to the magnetostrictive material layer, produce electromagnetic wave signals with the same frequency as those of the excitation source, thereby achieving electromagnetic wave emission. When electromagnetic waves are received, the magnetostrictive material senses changes in the spatial magnetic field based on the ME coupling effect, which is opposite to the transmission mode. The magnetostrictive effect generates strain that is transmitted through the heterojunction to the PZT 5H piezoelectric material. The piezoelectric layer produces a corresponding voltage signal owing to the piezoelectric effect, thus achieving a conversion from a magnetic field to an electric field and, hence, the reception of electromagnetic waves.29  Figure 1 illustrates the principle of a magneto-electric antenna that radiates and receives electromagnetic waves.

FIG. 1.

(a) The ME antenna radiating the electromagnetic waves outward through the inverse magnetoelectric coupling effect. (b) The ME antenna receiving the electromagnetic waves through the magnetoelectric coupling effect.

FIG. 1.

(a) The ME antenna radiating the electromagnetic waves outward through the inverse magnetoelectric coupling effect. (b) The ME antenna receiving the electromagnetic waves through the magnetoelectric coupling effect.

Close modal

The ME antenna boasts versatility in adapting to electromagnetic waves of various frequencies and wavelengths while enabling significant downsizing of the antenna. Despite its size reduction, it retains high sensitivity. Its compact form factor facilitates integration with multiple devices, displaying unrivaled potential for LF band applications. As a recipient, the ME antenna is a passive component with low sensitivity and diminished output signal. This restricts the detection limit because phase-locked amplification is often required to detect electrical signals. Furthermore, when deployed as an emitter, a single ME antenna is limited by its low efficiency, which reduces the strength of the emitted radiation signals. However, owing to its size, which is markedly smaller than the corresponding wavelength of electromagnetic waves in the LF range, the antenna can be arranged in an array. The charge signals received by the individual antennas can be superimposed by employing an ME antenna array at the receiving end, thereby enhancing the sensitivity of the output signal. Similarly, at the transmitting end, the superposition of signals between the ME antenna array significantly increases the strength of the radiated signal.28  Figure 2 illustrates the law of influence of the ME antenna array on the transmitted and received signals. An ME antenna array can enhance the strength of the radiated signal, the sensitivity of reception, and the limit of detection in LF communication systems, ultimately improving the communication range of the system.

FIG. 2.

An ME antenna array with enhanced performance at the transmitter and receiver ends.

FIG. 2.

An ME antenna array with enhanced performance at the transmitter and receiver ends.

Close modal

The ME antenna, depicted in Fig. 3, is an ME coupling consisting of a magnetostrictive Metglas film (Shenzhen Yongsheng ME Ltd.), thin-film electrode, and piezoelectric material PZT5H (Zhejiang Shenlei Ultrasonic Material Ltd.), where PZT5H is polarized longitudinally and Metglas is magnetized along its length. Initially, a Metglas strip of 30 µm thickness was divided into 16 sections, with each section measuring 58 × 1 mm2. Subsequently, a c-axis-polarized PZT5H piezoelectric sheet was sectioned into pieces measuring 50 × 10 × 4 mm3. Simultaneously, two polyimide substrates measuring 58 × 10 × 0.06 mm3 were prepared, on which copper electrodes with a thickness of 20 µm were printed. Using Devcon 14 250 epoxy adhesive, the Metglas films were bonded to the electrodes and then symmetrically adhered to the piezoceramic. A pressure of 0.1 MPa was applied to ensure the bonding quality. The setup was left untouched for 24 h to completely cure the epoxy resin. The electrode pads were then extended to the exterior surface of the piezoelectric material using the backlead method to expose both the upper and lower electrodes. The lower electrode was wired to the front end through the printed circuit board (PCB), where the positive and negative terminals of the ME antenna were located. The final product, an ME antenna, is presented in Fig. 3 with the dimensions of 58 m × 10 × 4.9–4.92 (mm), volume of 2.842–2.853 cm3, and weight of 17.55–17.62 g.

FIG. 3.

Overall structure of the ME antenna.

FIG. 3.

Overall structure of the ME antenna.

Close modal

BNC coaxial lines were used as connectivity wires to mitigate environmental noise interference by utilizing a Helmholtz coil placed within a magnetic shielding barrel (magnetic disturbance below 30 fT). The Helmholtz coil, ME antenna, and magnetic field shielding barrel were aligned coaxially, and their geometric centers were ensured to coincide. A precision current source (Keithley-6221) energizes the Helmholtz coil, eliciting an alternating magnetic field. Subsequently, the ME antenna was connected to a phase-locked amplifier (Stanford Research Systems SR830) to read the induced voltage at the receiving end. This allows for the identification of the frequency response curve of the ME antenna and the correlation between the ME intensity and the output signal.

With the assistance of a tailored PCB board, the ME antenna established serial connections via the board’s leads. Using the receiving test platform, the spectral response and commensurate magnetic noise of both the ME antenna unit and the array were evaluated. The setup of the ME antenna receiving-mode test platform is illustrated in Fig. 4.

FIG. 4.

Performance testing platform for the reception mode of the ME antenna.

FIG. 4.

Performance testing platform for the reception mode of the ME antenna.

Close modal

Figure 5 illustrates the principal design of the test platform for the emission mode of the ME antenna. The Tektronix AFG3022C signal generator created a sinusoidal signal excitation, which was amplified using an Aigtek ATA-2082 high-voltage power amplifier. This amplified signal is then applied to the ME antenna, resulting in electromagnetic wave radiation. The performance of these radiated waves was tested using an LF electromagnetic signal analyzer (NF-5035), loop antenna, and ME antenna, all of which were manufactured by Aaronia AG. Tests for the intensity and communication distance were conducted by adjusting the strengths of both the input signal and output electromagnetic field. Given that the transmitter requires an independent power supply, the ME antenna was affixed to the PCB board. It outputs a dual-channel signal (max 800 Vpp, with dual-channel) via a high-voltage power amplifier, connects to the drive signal through a BNC lead wire, adjusts the alignment of the ME antenna, and carries out directionality testing of the antenna unit and array antenna using a calibrated coil positioned 1 m away.

FIG. 5.

Performance testing platform for the emission mode of the ME antenna.

FIG. 5.

Performance testing platform for the emission mode of the ME antenna.

Close modal

A waveform generator creates a modulated signal, which, upon amplification by the power amplifier, is imparted upon the ME antenna or its array for signal radiation. Subsequently, the signal is intercepted by the ME antenna and its array. Following bandpass filtering (20–40 kHz) of the signal, the waveform was visualized using an oscilloscope, and the strength of the output signal was gauged using a phase-locked amplifier (SR830). Figure 6 demonstrates the ME antenna signal radiation and reception of the ME antenna signal.

FIG. 6.

Testing platform for the emission and reception of the electromagnetic waves by the ME antennas array.

FIG. 6.

Testing platform for the emission and reception of the electromagnetic waves by the ME antennas array.

Close modal

The prepared ME antennas and the ME antenna array (connected in serial mode) are named 1, 2, 3, 1 + 2 (ME antennas 1 and 2 connected in serial mode), and 1 + 2 + 3 (ME antennas 1, 2, and 3 connected in serial mode). Tests and analyses of the impedance of the fabricated ME antennae and array were conducted using a network analyzer (Keysight E50618).

The spectral characteristics and detection limits of the ME antennae were determined using the test platform described in Sec. II C. The results are presented in Table I.

TABLE I.

ME antenna and serial mode array performance.

ME antenna1231 + 21 + 2 + 3
EMR (kHz) 31.76 31.75 31.76 31.75 31.75 
3 dB bandwidth (Hz) 400 420 440 470 500 
Sensitivity (mV/nT) 0.74 0.74 0.72 1.31 1.76 
aME (V/cm/Oe) 248 245 243 433 586 
Noise floor (nV/Hz81 79 77 114 139 
LOD (nV/Hz108.6 107.2 106.9 87.7 79.0 
ME antenna1231 + 21 + 2 + 3
EMR (kHz) 31.76 31.75 31.76 31.75 31.75 
3 dB bandwidth (Hz) 400 420 440 470 500 
Sensitivity (mV/nT) 0.74 0.74 0.72 1.31 1.76 
aME (V/cm/Oe) 248 245 243 433 586 
Noise floor (nV/Hz81 79 77 114 139 
LOD (nV/Hz108.6 107.2 106.9 87.7 79.0 
Magnetic fields sensitivity of ME antennas:
SME=UMEM,
(1)
Where UME is the voltage output at resonance of the ME antenna when the magnetic field strength is M. The ME coupling coefficient was calculated using the following equation:
αME=SMEdME,
(2)
where dME is the thickness of the ME antenna.
The limit of magnetic field detection of an ME antenna:
LOD=UNSME,
(3)
where UN is the noise floor of the ME antenna, measured using a dynamic signal analyzer (SR785) in a shielded barrel. The spectral characteristics and detection limits of the ME antenna were characterized using the ME antenna test platform described above. The test results are listed in Table I.
Consider m ME antenna units connected in serial mode. For simplicity, it is also assumed that the signal response of all these ME antenna units is generated in response to the applied magnetic field M and that the ME antenna performance is identical to the total voltage response of
Vs=mSMEM.
(4)
Similarly, the voltage noise level is calculated using the root mean square, which leads to
Vns=Vn12+Vn22+Vnm2=mVn.
(5)
Therefore, the SNR of m ME antennas in series is
SNR=VsVns=mSMMmVn=mSMMVn.
(6)

The structuring of an ME antenna into an array is conducive to enhancing its performance. However, its high impedance limits the effect of the parallel-array mode. The series mode array of the ME antenna effectively improves the sensitivity of the output signal.

Where aME is the ME coupling coefficient in the frequency domain, M is the measured magnetic field, m is the unit count of the series ME composite structure, Vs is the total output voltage, and Vns is the equivalent noise current, respectively.

From formulas (5) and (7), it can be inferred that the signal voltage of the ME antenna array in serial mode increases by a factor of m, thereby increasing the sensitivity by a factor of m, and the SNR increased with the m.30 ME Antennas 1, 2, and 3, which were prepared in the same batch, had frequencies of 31.76, 31.75, and 31.76 kHz, respectively. However, those connected in serial mode as 1 + 2 and 1 + 2 + 3 ME antenna arrays demonstrated a frequency of 31.75 kHz. This discrepancy might arise from manufacturing errors that cause inconsistencies in resonant peaks. The spectral response, magnetic field detection limit, and sensitivity of ME antenna array 1 and the 1 + 2 and 1 + 2 + 3 arrays connected in serial mode are shown in Figs. 7 and 8.

FIG. 7.

Relationship between the output voltage and the input voltage frequency of 1, (1 + 2), and (1 + 2 + 3) ME antenna.

FIG. 7.

Relationship between the output voltage and the input voltage frequency of 1, (1 + 2), and (1 + 2 + 3) ME antenna.

Close modal
FIG. 8.

Sensitivity and magnetic field detection limits of 1, (1 + 2), and (1 + 2 + 3) ME antennas.

FIG. 8.

Sensitivity and magnetic field detection limits of 1, (1 + 2), and (1 + 2 + 3) ME antennas.

Close modal

The background noise for ME antenna 1 as well as for the (1 + 2) and (1 + 2 + 3) arrays connected in series were found to be 81, 114, and 139 nV/Hz, respectively. This indicates that the theoretically projected noise level increases by √2 and 3 times, suggesting that the sensitivity should be two and three times that of a single ME antenna, respectively. However, the sensitivities of the (1 + 2) and (1 + 2 + 3) ME antenna arrays were only 1.75 and 2.35 times the sensitivity of a single ME antenna. These disparities probably resulted from manufacturing errors, wire losses, and noise in the test system. Despite these minor deviations from the theoretical values, the array structure displayed significant improvements in sensitivity and detection limits, thereby proving the efficacy of the series array mode. The theoretical LOD values for ME antenna 1 and ME antenna arrays (1 + 2 and 1 + 2 + 3) are listed in Table I. ME antenna 1 and the (1 + 2) and (1 + 2 + 3) ME antenna arrays connected in series yielded minimum measured magnetic field values of 170, 135, and 110 fT, respectively, corroborating the assertion that the ME antenna array enhances the detection limit.

Furthermore, the sensitivity of the test system was considerably improved using a phase-locked amplifier for readings with a series-connected ME antenna array. Given that the base noise of the phase-locked amplifier is assumed to be constant, the increased sensitivity assists in achieving lower magnetic-field detection limits during testing.

ME antennas 4 and 5 were fabricated as transmitting antennas with the same dimensions as those prepared previously. The ME antenna Tx mode frequency is very close to the Rx mode frequency. The Tx mode frequencies of the ME antennas 4 and 5 are 31.75 and 31.77 kHz, respectively. The impedance information of the ME antennas 4 and 5 is shown in Fig. 9.

FIG. 9.

Impedance information of ME antennas 4 and 5.

FIG. 9.

Impedance information of ME antennas 4 and 5.

Close modal

The mission performance testing platform described in Sec. III A was employed to characterize the relationship between the drive voltage and radiated signal strength. The signal generated by the signal source was input to a power amplifier (Aigtek ATA2802), leading to the signal input of ME antennas 4 and 5. Using a dual-channel power amplifier, identical signal inputs were implemented simultaneously on antennas 4 and 5. Using the loop antenna and LF electromagnet signal analyzer, the effects of different driving voltages on the magnetic field signals at a distance of 1 m were analyzed, yielding the results between the input signal strength and radiated signal strength. As illustrated in Fig. 10, the output signal strength peaks when the peak-to-peak voltage (Vpp) value is 350 Vpp, after which the signal strength slightly diminishes. As the voltage continued to increase, the signal radiated by the ME antenna no longer increased, presumably because the maximum withstand voltage of the piezoelectric crystal was achieved.

FIG. 10.

Driving voltage vs generated magnetic field for ME antenna monoliths and array.

FIG. 10.

Driving voltage vs generated magnetic field for ME antenna monoliths and array.

Close modal

The impedances of antennas 4 and 5 were measured to be 1967ej39.79° and 1928ej39.1°, respectively. The transmission power can be calculated as P=Vrms2zcosθ, resulting in transmission power levels of 23.92 and 24.65 W for antennas 4 and 5, respectively. Therefore, a 350 Vpp signal was used for the output in all transmission tests.

After defining the input signal strength, we characterized the directivity of the single ME antenna and its array. By employing a loop antenna 1 m away to characterize and normalize the emitted magnetic field strength of the antenna, we obtained the coordinate data, as depicted in Fig. 11(a). In this setup, the z-axis is along the length of the ME antenna, with the x-axis perpendicular to the thickness direction of the ME antenna and the y-axis perpendicular to the length direction of the ME antenna. Working at a frequency of 31.75 kHz, corresponding to a wavelength of 9448.82 m, the antennas were spaced 5 cm apart, which was significantly smaller than the corresponding wavelength. The signals from antenna 4 were tested individually and together with antenna 5. The radiation patterns in the xOz and yOz planes were similar to those of a magnetic dipole antenna, with maximum values occurring at 0° and 180° and minimum values at 90° and 270°. However, the voltage was not zero at 90° and 270°, which could be due to the magnetic signals generated by the wires or the reflected magnetic signals. This result was similar to that reported by Niu and Ren.24 The radiated voltages were produced in the xOy plane by the piezoelectric layer in the ME antenna. It was observed that the radiated signal strength on the xOy plane was ∼1/7.2 times that on the xOz and yOz planes, validating the hypothesis that the magnetic layer signal radiation is the principal radiation mode.23 Furthermore, the sensitivity of the test system was considerably improved using a phase-locked amplifier for readings with a series-connected ME antenna array. Given that the base noise of the phase-locked amplifier is assumed to be constant, the increased sensitivity assists in achieving lower magnetic-field detection limits during testing.

FIG. 11.

(a) Schematic diagram of the directional test. (b) xOz plane radiating signal direction. (c) yOz plane radiating signal direction. (d) xOy plane radiating signal direction. (e) ME antennas array (4 + 5) and single ME antenna (4) radiating signals on the xOz plane.

FIG. 11.

(a) Schematic diagram of the directional test. (b) xOz plane radiating signal direction. (c) yOz plane radiating signal direction. (d) xOy plane radiating signal direction. (e) ME antennas array (4 + 5) and single ME antenna (4) radiating signals on the xOz plane.

Close modal

An Amplitude-Modulated (AM) signal with a modulation depth of 100%, a frequency of 31.75 kHz, a modulation frequency of 100 Hz, and a central strength of 350 Vpp (Voltage Peak-Peak) was generated using a signal source. This signal was successfully used to drive a single ME antenna and array using a dual-channel power amplifier. At the receiving end, both the single ME antenna and its array functioned as transceivers. Signal analysis within a distance of 1 m was conducted using an SR785 and an oscilloscope. After envelope detection of the output signal of the antenna, it was discovered that the intensity of the temporal signal sent via the array was ∼3.1 times stronger than that of a single antenna, indicating an improvement in the radiation intensity and receiving sensitivity of the array ME antenna. This clearly indicates that the ME antenna array has superior communication capability. The experimental principles of the ME antenna signal radiation and reception, time domain, and frequency domain analysis are shown in Fig. 12.

FIG. 12.

(a) Principle of modulation and demodulation of the LF communication system, where SG is a signal generator, PA is a power amplifier, OS represents an oscilloscope, and SP is a spectrum analyzer. (b) Time–domain signal characteristics of the LF communication system after envelope detection. (c) Spectral characteristics of AM modulation.

FIG. 12.

(a) Principle of modulation and demodulation of the LF communication system, where SG is a signal generator, PA is a power amplifier, OS represents an oscilloscope, and SP is a spectrum analyzer. (b) Time–domain signal characteristics of the LF communication system after envelope detection. (c) Spectral characteristics of AM modulation.

Close modal
Assume that there are “m” identically designed transmitting antennae owing to the cubic power decay of the magnetic signal.
Bmr=mB0rm3,
(7)
B1r=B1r13,
(8)
rm=m3r0.
(9)

If “r” is utilized to quantify the distance, the magnetic field strength of a single ME antenna would be equal to that of an ME antenna array at a distance of m3 times greater. Compared with a single ME antenna, the radiated signal strength of an ME antenna array increases by a factor of “m” and the SNR ratio improves by a factor of m. Therefore, it is evident that the communication distance of an array composed of “m” identical ME antennae is m56 times that of a single ME antenna.

Tests were conducted on a single ME antenna (with antenna 4 transmitting and 1 receiving) and an ME antenna array (antennas 4 and 5 transmitting, with 1 and 2 receiving antennas, respectively), producing the attenuation curves illustrated in Fig. 13. The signal attenuation when the array is transmitting is proportional to r−2.66 and is proportional to r−2.69 when a single body is transmitting. The magnetic field amplitude decays to r−2.66 and r−2.69, which are slightly slower than the ideal r−3 decay law. This suggests that there may be other potential electromagnetic wave propagation paths in addition to the direct path. Examples include radiation from the current in the wire and reflection of the magnetic field in space.24 

FIG. 13.

Signal attenuation curve with distance during emission from the (4) ME antenna and reception from the (1) ME antenna and emission from the (4 + 5) ME antennas and reception from the (1 + 2) ME antennas.

FIG. 13.

Signal attenuation curve with distance during emission from the (4) ME antenna and reception from the (1) ME antenna and emission from the (4 + 5) ME antennas and reception from the (1 + 2) ME antennas.

Close modal

Within a 0.9 m radius, the signal from the ME antenna array falls below 1.9 times due to signal overlap. Beyond this distance, the signal from both ends was ∼1.9 times. A single ME antenna loses signal detectability at 28.2 m (1.55 µV), while the ME antennas array loses signal reception at 48.1 m (1.37 µV).

In prior tests conducted in a shielded environment, the integration time of the phase-locked amplifier was extended, thereby enabling a minimum detection voltage of as low as 0.1 µV. However, during tests under real-world conditions, factors, such as environmental noise and faint signals, extend the integration time. This addition led to an increase in the uncertainty in the signal detection, which resulted in discrepancies with the theoretical simulations. Despite such conditions, it was verified that the self-transmitting and self-receiving distance of the ME antenna array significantly surpassed that of a single ME antenna, with a theoretical transmission distance 1.78 times greater and an actual measurement of 1.7 times. When opting for a more sensitive antenna and setting the detection limit to 1 fT, the distance between the two pairs of ME antenna arrays can reach 800 m. Therefore, it is demonstrated experimentally and theoretically that increasing the number of ME antennas can significantly enhance the radiated signal strength and magnetic field induction SNR, thus improving the LF communication distance.

In conclusion, the newly introduced antenna array with a sandwich-type ME coupling structure represents a significant step forward in LF communication technology. Comprised of a dual layer of Metglas and Pb (Zr1−xTixO3) (PZT), the antenna operates at the resonance frequency of ME couplings, enhancing radiation intensity and the SNR for magnetic field sensing. For the transmitter, the dual-source power supply method effectively doubled the magnetic signal strength, increased the sensitivity to a single 2.5-fold by means of a three series-connected ME antenna array, and directly detected a magnetic field of 110 fT in the resonant mode. The achievement of 100 Hz communication using amplitude modulation is noteworthy. The relationship between the radiated signal strength, LF electromagnetic field sensitivity, and SNR of an array of “m” ME antennas to the number of arrays “m” is given. The array strategy achieved a low-frequency signal transmission distance of up to 48.1 m. The transmit–receive distance of a single ME antenna is 28.2 m. Ultimately, the proposed arrayed ME antenna significantly enhances the radiation proficiency of the LF communication system and lowers the detection limitation. In the future, by increasing the number of arrayed ME antennas, the LF communication range can be improved, providing a miniature, low-cost solution for two-way LF communication in mines and oceans.

This work was supported by the National Key Research and Development Program of China (No.2023YFB3208500) and the National Natural Science Foundation of China (Grant No. 92060112).

The authors declare that they have no known competing financial interests or personal relationships that could influence the work reported in this study.

Yinan Wang formulated an array strategy and orchestrated the design of comprehensive experiments. Guohao Zi prepared the ME antenna and its array, whereas Enzhong Song conducted reception tests on the ME antenna. Yuanhang Wang oversaw the transmission tests of the magnetron antenna, and Shanlin Zhao managed its ME antenna modulation. Zhibo Ma supplied funding and testing apparatus and provided overall supervision for experimental efforts.

Yinan Wang: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Writing – original draft (lead); Writing – review & editing (lead). Guohao Zi: Formal analysis (equal); Methodology (equal); Supervision (equal); Visualization (equal). Enzhong Song: Data curation (equal); Funding acquisition (equal); Methodology (equal); Resources (equal); Supervision (equal). Yuanhang Wang: Methodology (equal); Software (equal); Visualization (equal). Shanlin Zhao: Investigation (equal); Software (equal); Validation (equal); Visualization (equal). Zhibo Ma: Project administration (equal); Supervision (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
M. F.
Ali
,
D. N. K.
Jayakody
,
Y. A.
Chursin
,
S.
Affes
, and
S.
Dmitry
, “
Recent advances and future directions on underwater wireless communications
,”
Arch. Comput. Methods Eng.
27
(
5
),
1379
1412
(
2020
).
2.
A.
Pal
,
F.
Campagnaro
,
K.
Ashraf
,
M. R.
Rahman
,
A.
Ashok
, and
H.
Guo
, “
Communication for underwater sensor networks: A comprehensive summary
,”
ACM Trans. Sens. Networks
19
(
1
),
1
44
(
2023
).
3.
J. J.
Simpson
,
R. P.
Heikes
, and
A.
Taflove
, “
FDTD modeling of a novel ELF radar for major oil deposits using a three-dimensional geodesic grid of the earth-ionosphere waveguide
,”
IEEE Trans. Antennas Propag.
54
(
6
),
1734
1741
(
2006
).
4.
P.
Lunkenheimer
et al, “
Electromagnetic-radiation absorption by water
,”
Phys. Rev. E
96
(
6
),
062607
(
2017
).
5.
M. A.
Ali
,
S.
Kaja Mohideen
, and
N.
Vedachalam
, “
Current status of underwater wireless communication techniques: A review
,” in
2022 Second International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies (ICAECT)
(
IEEE
,
Bhilai, India
,
2022
), pp.
1
9
.
6.
B.
Majlesein
,
V.
Guerra
,
J.
Rabadán
,
J.
Rufo
, and
R.
Pérez-Jiménez
, “
Evaluation of communication link performance and charging speed in self-powered internet of underwater things devices
,”
IEEE Access
10
,
100566
100575
(
2022
).
7.
D. E.
Anagnostou
,
J.
Papapolymerou
,
M. M.
Tentzeris
, and
C. G.
Christodoulou
, “
A printed log-periodic Koch-dipole array (LPKDA)
,”
Antennas Wireless Propag. Lett.
7
,
456
460
(
2008
).
8.
M. N.
Hasan
,
O. J.
Babarinde
,
S.
Das
, and
K. V.
Babu
, “
Dispersion characterization of a UWB Vivaldi antenna in time and frequency domain
,” in
2018 IEEE Indian Conference on Antennas and Propogation (InCAP)
(
IEEE
,
Hyderabad, India
,
2018
), pp.
1
4
.
9.
C.
Pfeiffer
, “
Fundamental efficiency limits for small metallic antennas
,”
IEEE Trans. Antennas Propag.
65
(
4
),
1642
1650
(
2017
).
10.
O. C.
Fawole
and
M.
Tabib-Azar
, “
An electromechanically modulated permanent magnet antenna for wireless communication in harsh electromagnetic environments
,”
IEEE Trans. Antennas Propag.
65
(
12
),
6927
6936
(
2017
).
11.
F.
Fereidoony
,
S. P. M.
Nagaraja
,
J. P. D.
Santos
, and
Y. E.
Wang
, “
Efficient ULF transmission utilizing stacked magnetic pendulum array
,”
IEEE Trans. Antennas Propag.
70
(
1
),
585
597
(
2022
).
12.
H. C.
Burch
,
A.
Garraud
,
M. F.
Mitchell
,
R. C.
Moore
, and
D. P.
Arnold
, “
Experimental generation of ELF radio signals using a rotating magnet
,”
IEEE Trans. Antennas Propag.
66
(
11
),
6265
6272
(
2018
).
13.
Y.
Cui
,
C.
Wang
,
X.
Song
,
L. Y.
Li
,
Z.
Yuan
,
Z.
Li
, and
J.
Zhong
, “
A portable transmitter based on permanent magnets for low-frequency communications in multi-scenario with complex media
,”
Sens. Actuators, A
358
,
114428
(
2023
).
14.
J.
Xu
et al, “
Metamaterial mechanical antenna for very low frequency wireless communication
,”
Adv. Compos. Hybrid Mater.
4
(
3
),
761
767
(
2021
).
15.
M. A.
Kemp
,
M. A.
Franzi
,
A.
Haase
, and
E.
Jongewaard
, “
Analysis of a length extensional piezoelectric transformer for compact and efficient particle acceleration
,”
IEEE Trans. Plasma Sci.
47
(
1
),
313
323
(
2019
).
16.
M. A.
Kemp
et al, “
A high Q piezoelectric resonator as a portable VLF transmitter
,”
Nat. Commun.
10
(
1
),
1715
(
2019
).
17.
D. A.
Bas
,
P. J.
Shah
,
M. E.
McConney
, and
M. R.
Page
, “
Optimization of acoustically-driven ferromagnetic resonance devices
,”
J. Appl. Phys.
126
(
11
),
114501
(
2019
).
18.
B.
Lei
,
Z.
You
, and
Y.
Shi
, “
Analytical solutions for resonant radiation performance of bending-mode magnetoelectric antennas
,”
J. Appl. Phys.
134
(
12
),
124502
(
2023
).
19.
T.
Nan
et al, “
Acoustically actuated ultra-compact NEMS ME antennas
,”
Nat. Commun.
8
(
1
),
296
(
2017
).
20.
M.
Zaeimbashi
et al, “
Ultra-compact dual-band smart NEMS ME antennas for simultaneous wireless energy harvesting and magnetic field sensing
,”
Nat. Commun.
12
(
1
),
3141
(
2021
).
21.
E. A.
Burnside
,
S.
Tiwari
,
S. R.
Burnside
,
R. N.
Candler
,
R.
Henderson
,
S.
Grimm
, and
G. P.
Carman
, “
An axial mode magnetoelectric antenna: Radiation predictions via multiphysics modeling with experimental validations
,”
J. Appl. Phys.
134
(
14
),
144102
(
2023
).
22.
J. P.
Domann
and
G. P.
Carman
, “
Strain powered antennas
,”
J. Appl. Phys.
121
(
4
),
044905
(
2017
).
23.
C.
Dong
et al, “
A portable very low frequency (VLF) communication system based on acoustically actuated ME antennas
,”
Antennas Wireless Propag. Lett.
19
(
3
),
398
402
(
2022
).
24.
Y.
Niu
and
H.
Ren
, “
Transceiving signals by mechanical resonance: A miniaturized standalone low frequency (LF) ME mechanical antenna pair with integrated DC magnetic bias
,”
IEEE Sens. J.
22
(
14
),
14008
14017
(
2022
).
25.
M.
Zhu
et al, “
A fully packed ME VLF communication system based on self-designed circuits and wireless transmission into a metallic enclosure
,”
IEEE Trans. Antennas Propag.
71
(
10
),
8218
8822
(
2023
).
26.
J. D.
Schneider
,
J. P.
Domann
,
M. K.
Panduranga
,
S.
Tiwari
,
P.
Shirazi
,
Z. J.
Yao
,
C.
Sennott
,
D.
Shahan
,
S.
Selvin
,
G.
McKnight
,
W.
Wall
,
R. N.
Candler
,
Y. E.
Wang
, and
G. P.
Carman
, “
Experimental demonstration and operating principles of a multiferroic antenna
,”
J. Appl. Phys.
126
(
22
),
224104
(
2019
).
27.
B.
Dong
,
Z.
Yan
,
Y.
Zhang
,
T.
Han
,
H.
Zhou
, and
Y.
Wang
, “
Analysis of near field mutual coupling in wideband ME antennas array
,”
J. Appl. Phys.
134
(
11
),
114103
(
2023
).
28.
Y.
Zhang
et al, “
Research on a miniaturized VLF antenna array based on a ME heterojunction
,”
J. Mater. Sci.: Mater. Electron.
33
(
7
),
4211
4224
(
2022
).
29.
Z.
Wu
,
Z.
Xiang
,
Y.
Jia
,
Y.
Zhang
, and
H.
Luo
, “
Electrical impedance dependence on the direct and converse ME resonances in magnetostrictive/piezoelectric laminated composites
,”
J. Appl. Phys.
112
(
10
),
106102
(
2012
).
30.
Z.
Xing
,
J.
Zhai
,
J.
Gao
,
J.
Li
, and
D.
Viehland
, “
Magnetic-field sensitivity enhancement by ME sensor arrays
,”
IEEE Electron Device Lett.
30
(
5
),
445
447
(
2009
).