Valentina Palazzi, Ricardo Correia, Xiaoqiang Gu, Simon Hemour, Ke Wu, Alessandra Costanzo, Diego Masotti, Enrico Fazzini, Apostolos Georgiadis, Hooman Kazemi, Ricardo Pereira, Naoki Shinohara, Dominique Schreurs, Jung-Chih Chiao, Alexandru Takacs, Daniela Dragomirescu, Nuno Borges Carvalho
©SHUTTERSTOCK.COM/SERGEY NIVENS
Providing energy to Internet of Things (IoT) apparatuses is ever more challenging. Humans are increasingly invested in the revolutionary new scenario that experts call Industry 4.0, where billions of electronic devices are interconnected with one another. The operations of battery charging or battery replacement for these devices will soon become infeasible. As a consequence, radiative wireless power transfer (WPT) will soon become a leading technology as it is the only method available to support this technological revolution, and several information communication technology companies have already begun to actualize their interest in it.
The interest in ultralow-power, energy-autonomous wireless sensors has been further fueled by 5G communication systems, consisting of a massive number of interconnected devices communicating at low bit rates [1], and it is expected that the interest in energy-autonomous wireless sensing platforms will only continue into 6G systems and beyond.
This article aims to address radiative WPT technology, systems, and strategies for the future of this new electrical engineering paradigm.
The function of a radiative WPT system is to use RF signals to power electronic apparatuses, which are far from the RF energy source. The RF power propagating in free space is captured by a circuit called a rectenna. The main building components of rectennas are antennas, which are used to capture the RF power, and rectifiers, which are responsible for converting the RF power into dc power. Starting from the analysis of the basic components of radiative WPT systems, this article aims to bring together RF circuit and system designers with different backgrounds to
The article benefits from a wide network of experts from both academia and industry that address the existing and upcoming challenges in wireless power transmission scenarios in an interdisciplinary manner, paving the way for future generations of wireless power transmission solutions and the associated regulations.
At present, the power transmission limits and operating frequency bands for radiative WPT systems are not regulated. For WPT systems operating in the Industry, Science, Medicine (ISM) band of 2.4–2.5 GHz, power transmission must satisfy the effective isotropic radiated power (EIRP) limits (36 and 27 dBm in the United States and Europe, respectively [2]). The transmitted power is usually enough to supply ultralow-power IoT devices [3] at a distance of a few meters. Indeed, the power consumption of transponders based on low-cost transmission protocols [such as Bluetooth Low Energy (BLE) or semiactive RFIDs] with low-duty cycle operation is on the order of 1 ${\mu}$W–1 mW. In far field, assuming the signal propagates in free space, the power density decreases as the square of the distance, as shown in Figure 1. Therefore, the maximum distance between the transmitter and the tag is approximately 10 m. The studies on ultralong-range WPT systems (transmitting a few kilowatts) have been reported for power-beaming applications as well [4].
Figure 1. Received power as a function of the transmitter-to-receiver distance for different values of transmitted power EIRP. The receiver gain is equal to 3 dB.
Antennas are one of the key building blocks of radiative WPT systems. They determine how the RF power is transmitted through space, and how it is acquired by WPT receivers.
Antennas are also the most cumbersome parts of WPT receivers. Therefore, the manufacturing technologies and design approaches adopted for the antenna determine the characteristics of the WPT receiver in terms of area, weight, flexibility, and robustness.
Radiative WPT systems can be classified based on the type of available energy source, which can be intentional or unintentional.
In WPT systems with intentional sources, the RF power is transmitted by ad hoc RF transmitters. This means that the frequency of the RF signal that conveys the power is known. Therefore, narrow-band antennas can be used.
If the position of the receivers with respect to the transmitter is also known, high-gain antennas and arrays can be used both to transmit and receive RF power, if this is allowed by the spatial constraints associated with the application. On the transmitter side, beamforming techniques can be used to shape the radiation pattern so that the power is transmitted only in the direction of the receivers (closed-loop WPT systems have also been investigated to maximize the output dc power at the receiver side).
If the position of the receivers with respect to the transmitters is unknown, which is true in many applications, omnidirectional radiation patterns are needed. Additionally, circularly or dual-polarized antennas can be used to mitigate the polarization loss caused by an arbitrary receiver-to-transmitter orientation [5].
In WPT systems with unintentional sources, however, there are no dedicated RF power sources. The receivers harvest the RF power available in the environment. We usually talk of “energy-harvesting” (EH) or “energy-scavenging” systems, wherein the frequency of the available signals is not usually known in advance. Therefore, broadband and multiband antennas must be used to cover all the relevant bands used for telecommunications [6], [7]. This is essential to maximize the acquired power in a given scenario, and to improve system reliability. Omnidirectional and circularly polarized antennas are required as well [8]. Rectenna arrays can also be used (in applications with no restrictive spatial constraints) to increase the active area of the WPT receiver [3], [9], where each antenna is equipped with its own rectifier. This way, the received power is rectified by each element independently, and the obtained dc power is finally summed.
Antennas for WPT systems can be either matched to 50 Ω or directly to the input impedance of the rectifier. With the latter approach, the input matching network between the rectifier and the antenna can be avoided, thereby reducing the rectenna’s complexity, losses, and area [10]. Nevertheless, this makes the antenna design more challenging (the input impedance of the rectifier for low, available input powers has a significant capacitive component and a small, real part). Usually, dipole antennas with T-matching networks are used for this purpose. Additionally, the antenna and the rectifier cannot be tested separately. Therefore, this solution can be adopted only if the component models and the adopted manufacturing technologies are mature.
Planar microstrip antennas, such as rectangular patches, are mostly chosen for WPT receivers used in wearable applications [11], [12]. The presence of a ground plane reduces the impact of the human body on the antenna’s performance. Thick fabrics are used to improve system efficiency.
As WPT receivers can be used to power sensors placed on common items, flexible antennas can be used to conform the circuit to the object. In this case, wire and slot antennas are preferred as thin substrates usually do not allow for efficient microstrip antenna solutions in the sub-6-GHz spectrum.
In Figure 2, two flexible multiband rectennas are shown [7], [13]. The solution in Figure 2(a) and (b) is based on two nested annular slot antennas matched to 50 Ω in the bands of 790–960 MHz and 1.71–2.69 GHz, while the solution in Figure 2(c) is based on a Yagi–Uda antenna directly matched to the input impedance of the rectifier at two operating frequencies (915 MHz and 2.45 GHz).
Figure 2. Examples of flexible rectennas. (a) and (b) A multiband rectenna based on two nested annular slots [7] and (c) a dual-band rectenna based on a Yagi–Uda antenna [13].
3D-printed antennas and 3D antenna arrays have been also investigated [14] to allow WPT receivers to harvest energy from different directions (see, for instance, Figure 3).
Figure 3. An example of a 3D-printed antenna for RF EH [14].
RF-dc converters are RF circuits that convert RF signals back to dc voltage. They are normally composed of a matching network (to allow the maximum power transfer between the antenna and the nonlinear rectifier), a nonlinear rectifying device, and a low-pass filter (LPF), which is then connected to a load or a current source.
Usually, the nonlinear rectifying device is a Schottky diode due to its low-voltage threshold and low-junction capacitance, which results in a more efficient operation at low powers and a higher maximum operation frequency.
Figure 4 presents the RF-dc conversion efficiency of different rectifiers with different topologies from 450 MHz to 94 GHz, based on [15]. Besides the single-band converters presented in Figure 4, dual-band converters that are able to convert RF power from two different transmitter sources are presented in [16], and a triple-band converter is presented in [17].
Figure 4. The state of the art in RF-dc converters, with different topologies, from 450 MHz to 94 GHz.
Many traditional rectifiers can exhibit only reasonable RF-dc conversion efficiency with a narrow input power range. The efficiency declines severely when the input power deviates from the operating range, limiting wireless charging applications with considerable variations of input power. Thus, it is necessary to implement and design rectifiers with a wide operating input power range [18], [19], [20]. The authors in [21] designed an adaptive rectifier for WPT that can adapt the configuration of the rectifier to the input power level.
In Figure 5, different topologies of rectifiers are presented. Charge pumps (voltage multipliers) are used the most in RFID applications as they require a considerable dc voltage to power electronics. The following main conclusions can be drawn from the simulations and experimental work:
Figure 5. Different rectifier topologies. (a) A series diode rectifier. (b) A shunt diode rectifier. (c) A shunt diode with a m/4 stub. (d) An N-stage Dickson voltage multiplier.
Most of the circuits for WPT applications are based on diode solutions and are optimized for low-power RF signals [22], [23]. High-efficiency rectifiers for a few watts of input power have been studied as well [24], [25]. If the main objective is the use of these RF-dc converters in very high power configurations, then transistor-based RF-dc converters can be used.
The efficiencies of rectifiers are very low for lower input powers. Some authors have addressed this problem in diode-based solutions, including in [26] and [27], which showed that RF-dc conversion efficiency can be improved by selecting the appropriate excitation signal. In [28], the use of multitone signals was proposed to increase the efficiency in RF-dc converters with Schottky diodes. In [29], the use of chaotic signals was proposed to increase RF-dc conversion efficiency for EH systems as well as for WPT. The authors in [29] used the Schottky diode approach to implement the rectifier. The use of high peak-to-average power ratio (PAPR) waveforms [intermittent continuous-wave (CW) signals by varying the duty cycle, ultrawideband signals, power-optimized waveforms, chaotic signals, white noise, modulated signals, and multicarrier signals] to increase and improve the WPT efficiency has been demonstrated in [26], [27], [28], [29], [30], [31]. Some more recent work focuses on the model to enable an analysis of a multisine-based WPT system, focusing on the bandwidth of the signal and the rectifier [32]. The authors in [33] demonstrated the use of modulated signals to improve RF-dc converter efficiency. They proposed instantaneous power variance to describe, more accurately than PAPR, the variation of the instantaneous power and the occurrence of signal peaks that directly affect RF-dc converter efficiency.
As the difference in the diode nonlinear junction resistance under forward and reverse biases is not substantial for low input power, the Shockley diode model is adopted instead of the commonly used ON-OFF switch model to develop an efficiency chain for Schottky diode-based low-power rectifiers [34]. Figure 6(a) shows the equivalent circuit model of a typical low-power rectifier containing four parts: an RF input source, a matching network, diode, and resistive loading. The matching network removes the influence of diode packaging inductance ${L}_{p}$ and capacitance ${C}_{p}.$ Thus, the rectifier efficiency ${\eta}$ can be described as follows, focusing on the Schottky diode itself [35]: \[{\eta} = {\eta}_{{\text{RF}}\_{\text{dc}}}\,{\cdot}\,{\eta}_{p}\,{\cdot}\,{\eta}_{\text{tr}} \tag{1} \]
Figure 6. (a) The equivalent circuit of a typical Schottky diode-based low-power rectifier. The rectifier efficiency for (b) SMS7630 and (c) HSMS2860 against an input power range of −50—10 dBm, where load resistance is 5.4 kΩ and operating frequency is 0.9 GHz.
in which \begin{align*}{\text{diode}}\,{\text{RF}}{-}{\text{dc}}\,{\text{conversion}}\,{\text{efficiency}}{:}\,{\eta}_{{\text{RF}}\_{\text{dc}}} = \frac{{P}_{\text{in}}\cdot{\mathfrak{R}}_{I}^{2}\cdot{R}_{j}^{2}}{{R}_{l} + {R}_{s} + {R}_{j}} \tag{2a} \\ {\text{diode}}\,{\text{parasitic}}\,{\text{efficiency}}{:}\,{\eta}_{p} = {\left({\frac{1}{{1} + {\left({{\omega}\cdot{C}_{j}}\right)}^{2}\cdot{R}_{s}\cdot{R}_{j}}}\right)}^{2} \tag{2b} \\ {\text{source}}\,{-}\,{\text{load}}\,{\text{dc}}\,{\text{transfer}}\,{\text{efficiency}}{:}\,{\eta}_{\text{tr}} = \frac{{R}_{l}}{{R}_{l} + {R}_{s} + {R}_{j}} \tag{2c} \end{align*} where ${P}_{in}$ is RF input power; ${\mathfrak{R}}_{I}$ is current responsivity, a factor measuring a diode’s nonlinearity; and ${R}_{s},$ ${R}_{j},$ and ${C}_{j}$ are the series resistance, nonlinear junction resistance, and junction capacitance of the diode, respectively. ${\omega}$ and ${R}_{l}$ are angular frequency and load resistance, respectively.
Diode RF-dc conversion efficiency ${\eta}_{{\text{RF}}{-}{dc}}$ is only associated with nonlinear junction resistance ${R}_{j}$ as only ${R}_{j}$ supports a dc path, as shown in Figure 6(a). Also, (2a) indicates that higher RF input power ${P}_{in}$, stronger diode nonlinearity ${\mathfrak{R}}_{I},$ and larger diode junction resistance ${R}_{j}$ all drive up ${\eta}_{{\text{RF}}{-}{dc}}.$ Diode parasitic efficiency ${\eta}_{p}$ characterizes the amount of RF power going through ${C}_{j}$ and is eventually dissipated by ${R}_{s}.$ A smaller junction capacitance ${C}_{j}$ and series resistance ${R}_{s}$ can help leverage ${\eta}_{p},$ according to (2b). Finally, source-load dc transfer efficiency ${\eta}_{\text{tr}}$ quantifies how much dc power arrives at ${R}_{l},$ which can be calculated by the voltage divider rule in (2c). The optimum load resistance ${R}_{l}$ can be approximated as ${R}_{j} + {R}_{s}$ based on (2c).
This analysis is suitable for understanding the loss (efficiency) mechanism of Schottky diode-based low-power rectifiers with explicit formulae. Zero-bias values ${R}_{j0}$ and ${C}_{j0}$ are generally adopted in the analysis. However, ${R}_{j}$ and ${C}_{j}$ vary under different input power levels. Therefore, the aforementioned analysis is valid in a limited power range (<−35 dBm). An analytical model is then developed to extend the dynamic range of an accurate prediction for rectifiers’ power conversion efficiency (PCE). The PCE can be defined as [36] \[{\eta}_{\text{PCE}} = \frac{{P}_{\text{dc}}}{{P}_{\text{in}}}\times{100}{\%} = \frac{{P}_{\text{dc}}}{{P}_{{R}_{j}} + {P}_{{R}_{s}}}\times{100}{\%} \tag{3} \] where the dc output power ${P}_{dc}$ can be calculated by the generated dc current multiplying the load resistance ${I}_{DC}^{2}\cdot{R}_{l}$. The RF input power ${P}_{in}$ contains two parts: the power absorbed by ${R}_{j}$ $({P}_{{R}_{j}}),$ and the power consumed by ${R}_{s}$ $\left({{P}_{{R}_{s}}}\right){.}$ Considering the voltage-controlled current (I) − voltage (V) relation of ${R}_{j}$ and capacitance (C) − voltage (V) relation of ${C}_{j},$ the generated dc current ${I}_{DC}$ and RF input power ${P}_{in}$ can be calculated with better accuracy. Hence, the dynamic range of this model for PCE prediction can be extended to −10 dBm, which is enough for ambient RF energy rectifier analysis.
As one example, Figure 6(b) and (c) demonstrate the rectifier conversion efficiency and loss mechanism for two diodes: SMS7630 and HSMS2860, respectively. From the comparison, SMS7630 is clearly a better candidate for low-power rectifier design as it delivers higher efficiency throughout the entire power range. When the input power increases, parasitic loss tends to reduce due to a smaller effective ${R}_{j},$ while harmonic loss has an opposite tendency. Regarding the HSMS2860 diode, its large junction resistance ${R}_{j}$ (due to a small saturation current ${I}_{s})$ offers smaller RF-dc conversion loss and source-load dc transfer loss at a low power level. However, a large ${R}_{j}$ brings significant parasitic loss, as seen in Figure 6(c).
In far-field WPT, the power is exchanged by radiative electromagnetic (EM) waves, allowing us to cover long-distance links and typically involving low levels of power. The most important figure of merit in this scenario is the efficiency \[{\eta}_{\text{LINK}} = {\eta}_{\text{TX}}{\eta}_{\text{FS}}{\eta}_{\text{RX}} = \frac{{P}_{\text{TX}}}{{P}_{\text{BIAS}}}\frac{{P}_{\text{RX}}}{{P}_{\text{TX}}}\frac{{P}_{\text{dc}}}{{P}_{\text{RX}}}{.} \tag{4} \]
Within the three main efficiencies involved, as shown in Figure 7, the TX and the RX efficiencies $({\eta}_{TX}$ and ${\eta}_{RX},$ respectively) have almost reached their upper limits. The main bottleneck is represented by the transmission efficiency in free space, ${\eta}_{FS}, $ because only a small amount of the transmitted power actually reaches the receiver location, and this causes a drop in the overall system performance, mainly because the attenuation makes the power decrease rapidly after a few meters. The solution is offered by the so-called “Smart RF” or “mm-wave showers,” illuminators that are able to provide a high level of reconfigurability in real time under demanding power-control conditions [37]. Here, two technologies are reported that are aimed at improving ${\eta}_{FS},$ both of which leverage the concept of diversity in its two guises: frequency diversity applied by frequency-diverse arrays (FDAs), and time diversity applied by time-modulated arrays (TMAs).
Figure 7. A far-field WPT block schematic and the efficiency contribution.
In frequency diverse arrays, each radiating element radiates with a frequency slightly different from its neighbors. The result, given by the intermodulation of the different tones, allows us to dynamically compose an array factor that, for a fixed instant, is able to provide energy in real time not only in prescribed directions but also at a prescribed distance from the transmitting source. In this way, the transmitted power can be concentrated at desired spots, and can be drastically reduced in undesired ones, with the twofold advantage of increasing energy efficiency where it is needed and minimizing the power where it is not needed. After several studies on the characteristics of traditional linear FDAs, attention has moved to improving the focusing capabilities of these radiating architectures: planar bidimensional array arrangements have been identified answers to this requirement. The multifinger frequency diverse array belongs to this family, where the array is arranged as a tree, with M radiating elements aligned along the x-axis and N along the y-axis. An example of the layout and its astonishing properties are presented in Figure 8, which highlights the availability of a planar-array configuration, which allows us to improve the focusing capability with respect to a linear FDA, reducing the area of the spot in the ${\theta}$ direction, while the “S shape” of the beam pattern is still present, as shown in Figure 8(c). This shows the main limitation of FDAs: the displacement in time of the main beam. This means that as time passes, the power is not fixed and moves farther away from the source periodically with a period equal to the inverse of the frequency increment applied to two consecutive elements or groups of elements.
Figure 8. Normalized beam pattern (BP) of the standard FDA for t = 100 ns: (a) in the plane (r, q, f = 90°), (b) in the plane (r, q, f = 45°), (c) in the plane (r, q, f = 0°), (d) a normalized BP in the (t, r, i = 0°, f = 0°) space, and € a planar array in “tree” configuration operating at 24 GHz, with M = 8 “branches” and N = 16 series-fed patch antennas.
Several techniques have been proposed to counterbalance this undesired effect for WPT, and the most promising ones are called time-controlled FDAs (TCFDAs). Each input signal is modulated by a periodic pulse presenting two new design parameters: the first one is able to intercept the power at a desired distance from the source, and the second one is able to define the coverage area. As a consequence of the application of this time-based technique, the power can be fixed at a certain distance from the transmitting source and drastically reduced in undesired regions, as shown in the examples presented in Figure 9.
Figure 9. (a) Backprojection (BP) (accounting for isotropic attenuation) versus time when the TCFDA is designed to focus at (r, i, z)=(30 m, 0°, 0°). (b) BP (accounting for isotropic attenuation) versus time when the TCFDA is designed to focus at (r, i, z)=(10 m, 0°, 0°).
The proposed design highlights the great potential of FDAs for selective far-field WPT “on-the-move” applications.
TMAs, first introduced in the 1950s, have recently garnered increased interest due to exponential electronic development that is able to produce and sustain the time-diversity concept: in a linear array with high-performance RF switches driven by precise control sequences, a simultaneous multifrequency radiation can be exploited to allow it to point the many beams in different directions at the same time. The almost unlimited number of driving sequences leads to manifold benefits: sidelobe suppression relaxing the array design constraint, electronic beam scanning through the generation of multiple beams pointing at different angles without the use of phase shifters, and finally, an important reduction in system cost. The schematic representation and the multiharmonic radiation mechanism of a TMA are presented in Figure 10.
Figure 10. (a) The schematic representation of a linear, n-element time modulated array (TMA), with detailed diode switch bias networks, including dc-block capacitors, and (b) multifrequency radiation patterns of a 16-element linear TMA.
In WPT applications, this technology represents one of the most promising solutions for simultaneous energization of objects located in different positions [38]. The main concern is the synthesis of the optimum sequence for the driving of the RF switches to fully leverage the time-diversity capability. As an expression of this great potential, Figure 11 shows both the ON-OFF sequence of 16 switches of a 16-element linear array operating at f0 = 2.45 GHz and the corresponding simultaneous radiation patterns involving the fundamental carrier and its upper- and lower-sideband harmonics (f0 ! fM = 2.45 GHz ! 100 kHz, in this case).
Figure 11. (a) Switches control patterns optimized for WPT purposes for a 16-monopole array. (b) Radiation patterns at the fundamental and at the two first symmetrical sideband harmonics.
These promising results, obtained by means of the simple architecture implementation shown in Figure 10, indicate that TMAs are potential candidates for agile and reconfigurable WPT systems to be exploited in many civil and industrial scenarios.
Beam propagation is associated with one of the major problems in WPT, that is, spillover losses (see Figure 12). These happen whenever the emitted radiation misses its target, mainly due to the microwave’s divergence but also due to system misalignments.
Figure 12. A microwave beam in a traditional WPT system. The beam’s divergence reduces the amount of energy reaching the target.
Antennas’ far field and associated tools are traditionally used in WPT. However, these do not provide us with the means to reduce, or even account for, the spillover losses. To do that, a new approach to the radiation’s study can be pursued, one that remains practical but is capable of analyzing the microwave’s radiation in a more comprehensive manner so that the beam’s divergence is understood and controlled.
Quasioptics appears to be the theoretical framework of choice for the previously mentioned requirements by approximating the emitted EM radiation with Gaussian beams [39]. These flexible mathematical entities are the building blocks of the complete theory, providing information on beam propagation throughout any system (see Figure 13). The electric field distribution of a Gaussian beam in the fundamental mode, propagating in the $\hat{z}$ direction, is given by \[{E}\left({r,z}\right) = \sqrt{\frac{2}{{{\pi}\varpi}^{2}}}{\text{exp}}\left({{-}\frac{{r}^{2}}{{\varpi}^{2}}{-}{ikz}{-}\frac{{i}{\pi}{r}^{2}}{{\lambda}{R}} + {i}{\phi}_{0}}\right) \tag{5} \]
Figure 13. Different theoretical frameworks for analyzing EM radiation and radiative WPT, ordered according to the depth of their analysis. Quasioptics is proposed as the one that best balances implementation difficulty with analysis depth.
where $\varpi$ is the beam radius, the most important parameter for WPT as it provides information about the beam’s divergence. It is defined as the radial distance at which the power falls to 1/e of the on-axis value (e is Euler’s number). The point where the power is most concentrated is given by ${z}_{0},$ and the beam radius is at its minimum at this point; it is called the beam waist, ${\varpi}_{0}^{2}.$ The electric field varies with the distance to this point, $z,$ and the distance from the axis of propagation, $r.$ Finally, $R$ is the wavefront’s radius of curvature, ${\phi}_{0}$ is the phase shift, and ${\lambda}$ is the beam’s wavelength.
With it, the matrix formalism of optics can be used in contexts of significant divergence, as is the case of microwaves traveling distances of meters and kilometers to transform and optimize the beam to the system requirements (Figure 14). Although mostly used in RF and millimeter-wave (mm-wave) astronomy, its application in WPT systems may be paramount to reducing spillover losses, contributing to an overall increase in efficiency [40], [41], [42].
Figure 14. In this schematic, a quasi-optical system transforms an input beam to a thin ray, similar to an optical system. The beam’s divergence can therefore be accounted for, potentially reducing spillover losses. The distance from the input to the output beams in the quasi-optical system is given by ${d}_{in}$ and ${d}_{out},$ respectively.
EH is a very broad and diverse field, given that each EH technology defines a completely different field of research [43]. The physics behind EH technologies, including solar, mechanical, thermal, and RF, have been known for decades, even centuries. More than a hundred years ago, Nikola Tesla envisioned the wireless transmission of power. Advances in materials and fabrication techniques such as additive manufacturing have enabled the miniaturization, reduction of cost, and performance improvement of EH devices, making them suitable for low-power wireless sensors. With advances in electronic design and integrated circuit technologies, which have led to further reductions in operating power and cost of electronic circuits, wireless sensor platforms that are powered by EH are becoming increasingly feasible.
Table 1 presents some indicative performances from various types of energy harvesters suitable for micropower generation. There is a large variation among the size of the transducers and the amount of energy that can be generated. As a result, transducer selection depends on application requirements and the scenario, which makes the presented results of Table 1 only indicative of the potential of the various harvesting methods.
Table 1. The harvested power potential from different transducer types [44].
Each transducer technology has distinct advantages and disadvantages and is thus suitable for different application scenarios. One characteristic of ambient energy that makes the design and use of harvesting circuits particularly challenging is its variable and probabilistic nature. A particular challenge is the efficiency of the transducer device itself, which is typically limited and/or strongly dependent on the available input energy and output load. For example, the availability of light energy is reduced in indoor scenarios or at night; thermal energy harvesters are limited by a low maximum transducer efficiency, depending on the existing temperature gradient, while kinetic energy harvesters are sensitive to the natural vibration frequencies of the harvester and application settings. Finally, the available EM energy density is usually orders of magnitude below the corresponding values of the other energy sources, although measurements in urban settings have demonstrated the possibility of harvesting a useful amount of EM energy using wideband or multiband harvesters [45], [46], [47]. However, the highly variable nature of ambient energy availability gives rise to application scenarios where EM EH is justified. One characteristic example is the variation of solar light energy between daytime and nighttime. A second example is placement of sensors sufficiently near transmitting RF base station antennas where RF energy levels may exceed the typical values.
The different energy sources can be combined in a variety of ways. The signal combination can be done at the input or the output of the diode rectifier circuits (i.e., in dc or ac signal form) [43]. Furthermore, the dc signal combination can be done in series or in parallel, resulting in a different optimum load value corresponding to a maximum efficiency. For example, the various signals of RF energy harvesters may be combined in the RF stages before they are fed to the rectifier circuit, resulting in a directive antenna system. Alternatively, the dc outputs of different rectennas can be combined in series or in parallel [48]. Every topology has different advantages and requires a careful design to optimize the dc conversion efficiency for a given output load and for the desired input power ranges.
To address the aforementioned challenges associated with ambient energy availability, we highlight efforts toward hybrid multitechnology harvesters in the next section [43], [49].
Example multitechnology circuit implementations include a solar-thermal-EM energy harvester based on a patch antenna integrating a solar cell and a thermoelectric generator [50]. A preliminary prototype is shown in Figure 15 [50]. The concept of using solar power as an auxiliary power source to power up passive RFID tags is proposed in [51]. A different implementation of a solar energy-assisted passive RFID tag where the dc power from a solar cell is converted to RF using an oscillator and then fed to the RF pins of a commercial passive ultrahigh frequency RFID tag is shown in Figure 16 [52]. Finally, ambient backscattering has been proposed as an ultralow-power communication technique where information from an ultralow-power sensor tag is superimposed on existing ambient signals [53]. A low-cost ambient backscatter platform based on commercial off-the-shelf components based on ambient frequency-modulation station backscattering is depicted in Figure 17 [54]. The potential for combining ambient backscattering with multiple EH technologies may further expand the concept of batteryless, “zero-power” sensors.
Figure 15. The stacked configuration of a patch antenna, solar cell, and thermoelectric generator for EH [50]. TEG: thermoelectric generator.
Figure 16. A solar-power-enhanced passive ultrahigh frequency RFID tag [52]. (a) A read-rate performance measurement, (b) a tag without a solar cell, and (c) a tag with an integrated solar cell.
Figure 17. Ambient backscattering of frequency-modulation (FM) signals [54]. RTL: Realtek; SDR: software-defined radio.
Mm-wave frequencies provide key advantages in radiative WPT applications, such as compactness and focused energy, which can be delivered to a target at range with a smaller rectenna surface. Another major advantage of the mm-wave frequency range is fewer regulatory restrictions for transmission of signals in populated areas compared to the congested 2.45- and 5.8-GHz frequency bands.
Figure 18 represents the antenna beamwidth and spot size at a 1-km distance as a function of transmitter frequency for three transmitter aperture sizes. As the transmitter frequency is increased from the RF to the mm-wave frequency range, the antenna beamwidth narrows, leading to a smaller spot size on target. This allows for a focused beam of the radiated energy with less spreading loss, while the spot size is not as small as optical beams where pointing and tracking is required, thereby reducing complexity of the rectenna-based receivers.
Figure 18. An RF and mm-wave WPT system analysis, highlighting (a) antenna beamwidths and (b) spot sizes as a function of Tx frequencies for three Tx antenna aperture sizes.
Figure 19 represents the plot of the total RF and mm-wave power received at the receive antenna as a function of range. For a fixed-aperture comparison between the various operating frequencies, the mm-wave approach performs best due to its increased power received at all ranges. The power delivered on target due to the focused beam and higher power density can either manifest itself into a smaller size and weight of a power-beaming system or a reduced overall cost for watts delivered at range compared to RF and microwave-beaming systems. Advances in high-power monolithic microwave integrated circuit (MMIC) amplifiers in GaN technology as well as high power-density 3D packaging, together with commercial and defense application pulls, will promote the cost-effectiveness of mm-wave power systems in the near future.
Figure 19. Received dc power at the rectenna as a function of range for five RF and mm-wave WPT systems. All the Tx powers are at 100 kW, with transmit and receive aperture sizes being 2 and 1 m in diameter, respectively.
The high-power regime for WPT requires a high power-handling rectifier diode to handle the high incoming radiation at W/cm2 input power levels. The current device technologies used for wireless power rectification at mm-waves are GaAs and CMOS. The key to high-efficiency RF-dc rectification is the high switching speed of the diode (high cutoff frequency), which entails a low series resistance and low associated capacitance. For high-power operation, high forward current handling and reverse breakdown voltage are required. GaAs planar air-bridged diodes and CMOS transistors are characterized by high cutoff frequencies. However, their breakdown voltage is limited. For high power operation, a high breakdown voltage is required, moving the choice of technology toward GaN: much like in transistor-based power amplifier integrated circuits with high output power density. Various GaN Schottky diodes have been developed, highlighting the improvement in output rectified power. However, the higher frequency of operation has been limited by the device’s design and its fabrication process. Figure 20 illustrates a technology cross section of the nano-Schottky diode (NSD).
Figure 20. A technology cross section of the nano-Schottky diode. The anode metal contacts laterally with a 2D electron gas (2DEG) formed in an AlGaN/GaN high-electron mobility transfer structure. The cathode is made of a regrown n+GaN on the 2DEG.
The epitaxial structure consists of an AlGaN/GaN high-electron mobility transfer (HEMT) structure where a 2D electron gas (2DEG) is formed near the AlGaN/GaN interface. The sheet electron density and mobility of the 2DEG are 1.2 × 1,013 cm2 and 1,600 cm2/V · s, respectively. The 2DEG is confined in the GaN channel layer within a few nanometers below the interface, enabling a nanoscale Schottky contact between the anode metal and the 2DEG. The small Schottky contact area and a lateral depletion length of the 2DEG determine the junction capacitance (Cj). The cathode ohmic contact is formed by an n + GaN layer regrown on the 2DEG. The low contact resistance (equal to 0.1 W·mm) and low sheet resistance of the 2DEG (equal to 330 W/sq) contribute to a low series resistance (Rs) of the diode. The distance between the anode and cathode is 0.5 µm. The GaN NSD rectifier was used to design a variety of rectenna circuits for a performance evaluation of the developed technology. Figure 21 shows a voltage-doubler rectifier fabricated as an integrated circuit directly on the GaN/SiC wafer using two 2 × 2 × 10-µm diodes. The circuit includes an input LPF using a double-stub matching network designed for roll off at 96 GHz, which prevents any higher harmonics power generated by the rectifier to be radiated out of the input port. Similarly, a large, integrated thin-film capacitor connected through a substrate via has a low-loss RF short circuit, allowing only dc to be extracted at the output port. An integrated thin-film capacitor with a value of 350 fF was also used for the voltage doubler for charging and discharging operation during forward and reverse operation RF input swing across the rectifier diodes.
Figure 21. A GaN nano-Schottky rectenna MMIC integrated with a voltage-doubler diode topology, input LPF, and dc circuitry measuring 2.1 mm by 0.75 mm. CPWG: coplanar waveguide.
To characterize the rectifier circuit, a large signal characterization setup was configured (the details are reported in [55]). Figure 22(a) represents the dc voltage versus current generated from the circuit for each input power level across a 240-Ω load resistor. The data measured falls on a straight line, indicating that the diode rectifier is being driven in a forward and a reverse direction by the incoming RF signal and has not reached the maximum limits of forward current and reverse breakdown voltage. Figure 22(b) highlights the latter in detail, where the dc power generated by the circuit is shown as a function of the input power absorbed by the circuit. A fitted curve is also included to highlight which trajectory is expected for higher input power.
Figure 22. Performance of the GaN nano-Schottky rectenna MMIC. (a) dc voltage versus current measured under a large signal at 93 GHz. Each data point indicates an input power-level incident on the rectenna circuit. (b) dc power generated by the rectenna circuit versus incident power. The dotted line represents the higher power-level trajectory if more power were available on the test system.
The rectifier is connected to an antenna to form a rectenna. Figure 23 represents the conversion-efficiency plots for the measurements as a function of the W-band input power absorbed by the circuit, and the power density (the power absorbed per unit area of the MMIC). The data represent the highest efficiency rectenna circuit reported at the W-band frequency range. It can be seen that the efficiency has not peaked completely, and with more input power even higher values could be achieved. The input power density is also calculated as the ratio of the input power versus the area of the circuit MMIC. The value of 0.56 W/cm2 indicates the highest power-handling level of a W-band rectifier to date. A conversion efficiency of 61.5% was recorded during the high input power of 9.512 dBm, representing the highest reported mm-wave rectifier efficiency at this power level. Table 2 represents the latest published W-band rectenna circuit performance metrics compared to the GaN nano-Schottky rectenna circuit. Si CMOS circuits provide very high switching speeds but fall short on power handling due to the lower inherent breakdown voltage and current-carrying capability of the diode rectifier. However, CMOS presents a high degree of integration and can provide substantial advantages for lower power arrays due to its many layers of interconnects and capability of integrating the power conditioning circuits. GaAs diodes, a predominant mm-wave and sub-mm-wave device technology, also provide great switching speeds and moderate breakdown voltage and hence improve over the Si CMOS-based rectenna metrics. The GaN nano-Schottky rectenna circuits reported in this work show an improvement of 1.5 times in power handling over GaAs and 9.5-times over CMOS rectenna circuits while improving efficiency by 15.7% over the highest previously reported W-band rectenna circuits. Therefore, GaN nano-Schottky represents next-generation technology for high-power mm-wave wireless power beaming.
Figure 23. The plots of RF-dc efficiency at 93 GHz are shown versus the input power level and power density.
Table 2. A comparative performance of various published rectenna circuits at 94–95 GHz.
In the 1960s, W.C. Brown first used narrow-beam microwaves as energy carriers [61]. However at the time, commercial applications of narrow-beam WPT were limited except for space applications, in particular, space-based solar power (SBSP). P.E. Glaser proposed SBSP in 1968 [62], following Brown’s success in 1964 with the first narrow-beam WPT experiment in a drone aircraft. However, theoretically, narrow-beam WPT requires large antennas, even in the case of microwaves, to increase beam efficiency. Hence, narrow-beam WPT has historically been viewed as economically infeasible, and wired power transfer is still preferred. SBSP is the only exception to this rule.
Typical SBSP design parameters when used with microwaves are as follows (Figure 24):
Figure 24. A typical SBSP image with microwaves.
Some SBSP systems are designed with laser power transfer. The U.S. Naval Research Laboratory recently demonstrated laser power transfer. The U.S. Department of Defense also launched a US$100 million partnership with Northrop Grumman, the Space Solar Power Incremental Demonstrations and Research Project, which aims to launch an SBSP demonstration spacecraft called Arachne in 2024 [63]. Japan proposed a road map to commercial SBSP by 2050 by considering the required R&D steps and potential commercial spin-off WPT technologies. For example, in 2018, a narrow-beam WPT experiment was conducted on a drone with a phased-array antenna at 5.8 GHz in Japan. WPT-assisted drones are one of the expected spin-off technologies [63]. Recently, China launched a national project for SBSP [64]. China plans to use a new superheavy lift rocket, which is currently under development, to construct a massive SBSP in a geostationary orbit. It is supported by a Chinese rocket company. China hopes to develop a megawatt-level power generation facility around 2030 [65], [66], [67]. The United Kingdom also initiated a feasibility study on SBSP as a CO 2 -free source of power.
SBSP requires long distance and high-efficiency beam WPT. Recently, there have been some interesting beam WPT demonstrated in the world. In Japan, a high-efficiency and thin phased-array antenna with a newly developed GaN HEMT was developed [63]. In 2015, they carried out a 1.6-kW–50-m-distance narrow-beam WPT at 5.8 GHz and succeeded in receiving 330 W of dc. Now they are developing a sandwich module with solar cells and a phased-array antenna for space experiments in the near future. In the United States, a 91.2-kW–1-km-distance narrow-beam WPT at 10.5 GHz with a Cassegrain antenna succeeded in receiving 1.6-kW of dc in 2021 [4]. Chinese researchers carried out a 910-W–30-m-distance narrow-beam WPT at 5.8 GHz with a phased-array antenna and succeeded in receiving 36 W of dc, also in 2021. Building on their successful development and demonstration of long-distance and high-efficiency WPT, all three countries, Japan, the United States, and China, plan to carry out satellite experiments on SBSP.
There are other WPT space applications in addition to SBSP. In the 1990s, a research group at Kyoto University in Japan designed a power satellite using 35-GHz WPT, which could supply wireless power to a user satellite (see Figure 25) [68]. With such a power source, we can reduce the weight of a working satellite using lightweight rectennas instead of heavy solar cells and batteries.
Figure 25. A power satellite image.
WPT on the moon is another interesting WPT application. We could power a moon rover via microwave or mm-wave wireless energy transfer. We could also supply wireless power to nonsunlit areas of the moon’s surface from a 100-km lunar orbit. This would bring us a step closer to practical SBSP and achieving a new paradigm for space exploration.
Not only narrow-beam but also wide-beam and coupling WPT can be applied in space systems. The Internet of Space, based on wide-beam WPT, is an important new concept in space system engineering. In small satellites, we can reduce the weight of power cables using inductive WPT. Overall, WPT shows novel potential to expand human space activities.
In IoT networks, the challenge is to combine wireless powering with data communications, also called simultaneous wireless information and power transfer (SWIPT) (Figure 26). The requirements are often contradictory, and therefore an optimal balance must be found between efficient wireless powering on the one hand and reliable communications [i.e., low bit error rate (BER) and high data rate] on the other. As an example, the cutoff frequency of the LPF of the rectifier should be as low as possible for optimal power harvesting, thus avoiding a ripple, although the ripple is essential for data communication as it contains the modulated information [32]. Thus, research on SWIPT focuses on two main aspects: the IoT node design and base station signal design [69], [70].
Figure 26. In an IoT network, the base station would transmit both wireless power and data toward the IoT nodes and receive the uplink data communications [72].
Various IoT node architectures have been explored (Figure 27), and a promising one is the compact “integrated” configuration, whereby the split between EH and information retrieval is not achieved at RF but in the baseband. If the split were to happen at RF, an additional downconversion block would be required, and such an additional RF oscillator-mixer topology increases the power requirements of the IoT node.
Figure 27. The possible architectures for SWIPT IoT nodes. (a) Time switching, (b) power splitting, and (c) an integrated topology with a data/energy split in the baseband. PMU: power management unit.
Even though it is possible to harvest power from existing modulations adopted in IoT networks, such as BLE [71], the performance is suboptimal in terms of power harvesting. Therefore, research is being conducted on novel modulation techniques that improve PCE while minimizing the deterioration of communication performance metrics (i.e., data rate and BER), for example, in [72] and [73]. Besides the base station’s waveform design, another degree of freedom to improve SWIPT performance is by exploiting the base station’s antenna configuration, namely, by adopting distributed antenna settings [74].
Currently, most of the U.S. Food and Drug Administration (FDA)-approved medical implants that implement wireless power charging capability utilize inductive or resonance coupling by coils. They include neurostimulators that are implanted under the skin at the waist or chest, and cochlear implants behind the ear. The tissue distance is typically within a few millimeters, and the size of the implants are on the order of centimeters. As the tissue thickness is fixed, the dielectric characteristics between the transmitter and implant coils are known, and the variations for wireless power efficiency depend solely on spatial misalignment of coils. The systems are operated in low-megahertz frequency ranges due to the lower ac resistance.
For deeper tissues, the field divergence and misalignment issues become challenges. A miniature gastric stimulator attached to the stomach wall, which inevitably moves during digestion, can harvest sufficient power at 1.3 MHz through 7-cm thick tissues [75]. Field manipulation with spiral coils can give better power coverage to tolerate misalignment [76]. The field distributions cannot be modeled in an equivalent circuit. Finite-element simulations are needed to evaluate the available field strengths at a certain implant depth.
Midfield power transfer techniques at higher megahertz and lower gigahertz ranges have been proposed for millimeter-scale implants that are deeper in the body. Unwanted signals with higher frequencies attenuate faster in tissue, preventing interference; and a smaller antenna spatially reduces noise coupling into the electronics. However, the tradeoffs highlight the need for better focusing of energy onto the implant via field manipulation.
An electrical stimulator with a 2-mm diameter and 2.5-mm length can be powered at a depth of 5 cm with a power of 500 mW at 1.6 GHz [77]. The transmitter’s metal pattern includes four split rings with independent excitation ports to adjust phases, and a circular slot array to generate the desired current distributions. The fields converge on the implant with a measured power of 200 ${\mu}$W in phantoms. At a 10-cm depth, 10 ${\mu}$W of power can be received, which is comparable to the power consumption of a cardiac pacemaker.
A meandered-line patch antenna in a capsule-type implant receives 6.7 mW of power at a tissue thickness of 5 cm with 1 W of transmitting power [78]. The transmitter’s antenna contains four ports to control the phases for better radiation focusing. Dynamic phase adjustment gives operational flexibility to power the imaging capsule as it moves through the gastrointestinal tract.
Far-field power transfer has been demonstrated from a horn antenna with a 7.6-dBi gain to a subcutaneous implant with a 4 × 8-mm2 inverted-F antenna at 2.45 GHz [79]. The transmitter’s power is limited to 1 W by safety rules. This technique can remotely charge the implant with ambient wireless energy; however, efficiency may vary significantly due to the directivities between antennas.
For mid- and far-field power transfer systems, their implementation examples are illustrated in Figure 28, as compared to the near field, where coils are placed directly on top of the subcutaneous implant. For mid- and far-field systems, exposure limits to tissue generally follow the guidelines of IEEE Standard C95.1 [80] and International Commission on Non-Ionizing Radiation Protection (ICNIRP) Guidelines [81]. The absorption limits per unit weight of tissues are defined by specific absorption rate (SAR) and specific absorption (SA) to prevent thermal damage. IEEE Standard C95.1 limits the 10-g averaged SAR and SA for a 6-min time period to 1.6 W/kg and 576 J/kg, respectively. ICNIRP Guidelines limit the 10-g averaged SAR of the head to 2 W/kg below 10 GHz and 2-mJ/kg averaged SA for a single pulse. Power flux densities for implants at different depths for 915 MHz and 2.4 and 5.8 GHz have been studied [82]. The methodologies for assessing exposure and anatomical models for dosimetry in WPT systems are discussed in [83].
Figure 28. An illustration of near-, mid-, and far-field WPT to implants.
The efficient design of wireless sensor networks (WSNs) including wireless battery-free sensor nodes (BFSNs) involves several key points that have to be simultaneously addressed at both the network and the sensor levels, on hardware and software implementation. An example of a meshed wireless BFSN network topology designed for structural health monitoring applications in civil engineering is reported in [84], [85] and depicted in Figure 29.
Figure 29. An example of a network topology, including wireless BFSNs. BFSNs are wirelessly powered using a far-field radiative WPT technique based on RF sources driven by the communicating nodes (CNs).
The WSN represented in Figure 29 is composed of two kinds of nodes: communicating nodes (CNs) and BFSNs. BFSNs are critical elements designed to be embedded in concrete for performing physical measurements (e.g., temperature and humidity) to enable monitoring of civil engineering structures. To reduce the dc consumption of each BFSN, only a directional uplink communication (from BFSN to CN) was implemented. BFSNs are cold-start compatible and powered using a far-field radiative WPT technique from dedicated RF sources (operating at ISM 868 MHz) driven by a CN (and usually co-located with other CNs). As proof of concept, two BFSNs were prototyped: 1) a long-range (LoRa) BFSN, represented in Figure 30; and 2) a BLE BFSN, represented in Figure 31.
Figure 30. (a) The topology and (b) photo of the LoRa BFSN.
Figure 31. (a) The topology and (b) photo of a manufactured prototype for the BLE BFSN. The external rectenna providing the dc power at the input Vin of the BLE BFSN is not represented here. PMU: power management unit; CH: channel; SOC: system on chip; PIFA: planar inverted F-antenna; RST: reset.
As far as the topology, the wireless chipsets (Murata CMWX1ZZABZ-091 for the LoRa BFSN and NXP QN9080 for the BLE BFSN), sensors (Texas Instruments HDC 2080), and power management unit (Texas Instruments BQ25570) were selected; there are only a few ways to reduce the dc consumption of a BFSN. A typical dc power consumption for the LoRa BFSN is represented in Figure 32(a), while Figure 32(b) highlights the evolution of dc voltage of the storage capacitor (22 mF) as a function of the performed task.
Figure 32. (a) The measured dc consumption for an LoRa BFSN and (b) dc voltage at the port of the storage capacitor (22 mF) as a function of the performed task (experimental results obtained for a measured RF power of −5 dBm at the input of the RF rectifier). MPPT: maximum power point tracking; WAN: wide area network.
For the LoRa BFSN, the dc consumption can be minimized by reducing the RF transmitting (Tx) power (dc current is reduced) and/or by increasing the data rate (thus decreasing the duration of the frame transmission). Both the Tx power and data rate are software controlled. The experimental results demonstrated that by reducing Tx power from +14 dBm to +4 dBm, the dc power consumption is reduced from 150 to 60 µW, while the energy/data byte is reduced from 52 to 26.5 mJ. There is no practical interest in reducing the Tx power below +4 dBm because the communication range is highly impacted with a minimal gain in terms of energy/byte reduction. By increasing the data rate from 250 to 5,470 bps (Tx power: 4 dBm), the energy/data byte is reduced from 26.5 to 5 mJ.
For the BLE SN, the sensing data are sent to the CN over four start cycles (on each start, the same data frame is broadcast over BLE channels 37, 38, and 39). This high redundancy on the BLE SN side minimizes the probability of errors caused by potential interference, mainly with Wi-Fi transmitters operating in roughly the same frequency band. The measured dc power consumption (Tx power: 0 dBm) is reported in Figure 33(a). The measured first and second charging times of the storage capacitor (100 µF) as a function of the EIRP of the RF source is presented in Figure 33(b). The measurements were performed in an anechoic chamber to avoid interference and multipath effects. The distance between the RF source and the BLE BFSN was fixed at 2 m. The first physical measurement/sensing and transmission are performed after the first charge, and the periodicity of sensing and transmission is the same as the duration of the second charge. These durations can be roughly controlled by controlling the EIRP of the RF source. The BLE BFSN implementation is more efficient than the LoRa BFSN implementation in terms of energy per transmitted frame (and per transmitted byte). Sending the data frame by BLE BFSN during one start cycle requires approximately 48 µJ (Tx power: 0 dBm) or roughly 1.2 mJ for a complete process, while the LoRa BFSN requires approximately 20 mJ for sending the data frame (Tx power: +4 dBm, data rate: 5,470 bps). A better performance in terms of communication range in favor of the LoRa implementation justifies this significant difference.
Figure 33. (a) Measured dc consumption for the BLE BFSN (dc voltage: 3 V) for a complete frame transmission (the red line is the average dc current) and (b) duration of the first charge (including cold-start procedure) and of the second charge of the energy storage capacitor (100 µF) as a function of RF source EIRP. SN: sensing node.
Two different SWIPT approaches were implemented. For the LoRa BFSN, the same antenna operating at 868 MHz in combination with a circulator was used for energy reception (from an RF source) and data transmission (to CN). In the case of the BLE BFSN, two antennas were used: one operating at 868 MHz [86] for energy receiving and one at 2.45 GHz for data communication.
This article provided an overview of the current research progress on radiative WPT, describing where we are and where we want to go. Emerging research directions have been sketched, including hybrid EH, mm-wave EH, and energy showers, which can improve reliability and versatility of the WPT approach. Diverse applications of radiative WPT have been reported as well, from medical, to space, to structural health monitoring. All this shows the potential and positive impact that can be produced by radiative WPT on a growing number of application scenarios and its key role in the evolution of the IoT.
[1] F. Boccardi, R. W. Heath, A. Lozano, T. L. Marzetta, and P. Popovski, “Five disruptive technology directions for 5G,” IEEE Commun. Mag., vol. 52, no. 2, pp. 74–80, Feb. 2014, doi: 10.1109/MCOM.2014.6736746.
[2] “Relating to the use of Short Range Devices (SRD),” CEPT Electronic Communications Committee, Copenhagen, Denmark, ERC Recommendation 70-03, Feb. 2021. [Online] . Available: https://docdb.cept.org/download/25c41779-cd6e/Rec7003e.pdf
[3] C. Song, P. Lu, and S. Shen, “Highly efficient omnidirectional integrated multiband wireless energy harvesters for compact sensor nodes of Internet-of-Things,” IEEE Trans. Ind. Electron., vol. 68, no. 9, pp. 8128–8140, Sep. 2021, doi: 10.1109/TIE.2020.3009586.
[4] C. T. Rodenbeck et al., “Terrestrial microwave power beaming,” IEEE J. Microw., vol. 2, no. 1, pp. 28–43, Jan. 2022, doi: 10.1109/JMW.2021.3130765.
[5] M. Wagih, A. S. Weddell, and S. Beeby, “Omnidirectional dual-polarized low-profile textile rectenna with over 50% efficiency for sub-μW/cm wearable power harvesting,” IEEE Trans. Antennas Propag., vol. 69, no. 5, pp. 2522–2536, May 2021, doi: 10.1109/TAP.2020.3030992.
[6] V. Palazzi, M. Del Prete, and M. Fantuzzi, “Scavenging for energy: A rectenna design for wireless energy harvesting in UHF mobile telephony bands,” IEEE Microw. Mag., vol. 18, no. 1, pp. 91–99, Jan./Feb. 2017, doi: 10.1109/MMM.2016.2616189.
[7] V. Palazzi et al., “A novel ultra-lightweight multiband rectenna on paper for RF energy harvesting in the next generation LTE bands,” IEEE Trans. Microw. Theory Techn., vol. 66, no. 1, pp. 366–379, Jan. 2018, doi: 10.1109/TMTT.2017.2721399.
[8] P. D. Hilario Re, S. K. Podilchak, S. A. Rotenberg, G. Goussetis, and J. Lee, “Circularly polarized retrodirective antenna array for wireless power transmission,” IEEE Trans. Antennas Propag., vol. 68, no. 4, pp. 2743–2752, Apr. 2020, doi: 10.1109/TAP.2019.2952011.
[9] J. A. Hagerty, F. B. Helmbrecht, W. H. McCalpin, R. Zane, and Z. B. Popovic, “Recycling ambient microwave energy with broad-band rectenna arrays,” IEEE Trans. Microw. Theory Techn., vol. 52, no. 3, pp. 1014–1024, Mar. 2004, doi: 10.1109/TMTT.2004.823585.
[10] C. Liu, H. Lin, Z. He, and Z. Chen, “Compact patch rectennas without impedance matching network for wireless power transmission,” IEEE Trans. Microw. Theory Techn., vol. 70, no. 5, pp. 2882–2890, May 2022, doi: 10.1109/TMTT.2022.3156936.
[11] G. Monti, L. Corchia, and L. Tarricone, “UHF wearable rectenna on textile materials,” IEEE Trans. Antennas Propag., vol. 61, no. 7, pp. 3869–3873, Jul. 2013, doi: 10.1109/TAP.2013.2254693.
[12] D. Vital, S. Bhardwaj, and J. L. Volakis, “Textile-based large area RF-power harvesting system for wearable applications,” IEEE Trans. Antennas Propag., vol. 68, no. 3, pp. 2323–2331, Mar. 2020, doi: 10.1109/TAP.2019.2948521.
[13] R. Scheeler, S. Korhummel, and Z. Popovic, “A dual-frequency ultralow-power efficient 0.5-g rectenna,” IEEE Microw. Mag., vol. 15, no. 1, pp. 109–114, Jan./Feb. 2014, doi: 10.1109/MMM.2013.2288836.
[14] J. Kimionis, M. Isakov, B. S. Koh, A. Georgiadis, and M. M. Tentzeris, “3D-printed origami packaging with inkjet-printed antennas for RF harvesting sensors,” IEEE Trans. Microw. Theory Techn., vol. 63, no. 12, pp. 4521–4532, Dec. 2015, doi: 10.1109/TMTT.2015.2494580.
[15] R. Correia, “Passive backscatter wireless sensor with wireless power transmission for IoT applications,” Ph.D. thesis, DETI - Departamento de Eletrónica, Telecomunicações e Informática, Aveiro, Portugal, Apr. 2019. [Online] . Available: https://ria.ua.pt/handle/10773/29173
[16] Y. Suh and K. Chang, “A high-efficiency dual-frequency rectenna for 2.45- and 5.8-GHz wireless power transmission,” IEEE Trans. Microw. Theory Techn., vol. 50, no. 7, pp. 1784–1789, Jul. 2002, doi: 10.1109/TMTT.2002.800430.
[17] B. L. Pham and A.-V. Pham, “Triple bands antenna and high efficiency rectifier design for RF energy harvesting at 900, 1900 and 2400 MHz,” in Proc. IEEE MTT-S Int. Microw. Symp. Dig., Jun. 2013, pp. 1–3, doi: 10.1109/MWSYM.2013.6697364.
[18] Z. Zhong, H. Sun, and Y. Guo, “Design of rectifier with extended operating input power range,” Electron. Lett., vol. 49, no. 18, pp. 1175–1176, Aug. 2013, doi: 10.1049/el.2013.1988.
[19] R. Correia and N. B. Carvalho, “Single transistor passive backscatter sensor,” in Proc. IEEE MTT-S Int. Microw. Symp., Jun. 2017, pp. 820–823, doi: 10.1109/MWSYM.2017.8058705.
[20] C. Wang, B. Yang, and N. Shinohara, “Study and design of a 2.45-GHz rectifier achieving 91% efficiency at 5-W input power,” IEEE Microw. Wireless Compon. Lett., vol. 31, no. 1, pp. 76–79, Jan. 2021, doi: 10.1109/LMWC.2020.3032574.
[21] H. Sun, Z. Zhong, and Y.-X. Guo, “An adaptive reconfigurable rectifier for wireless power transmission,” IEEE Microw. Wireless Compon. Lett., vol. 23, no. 9, pp. 492–494, Sep. 2013, doi: 10.1109/LMWC.2013.2250272.
[22] C. R. Valenta and G. D. Durgin, “Harvesting wireless power: Survey of energy-harvester conversion efficiency in far-field, wireless power transfer systems,” IEEE Microw. Mag., vol. 15, no. 4, pp. 108–120, Jun. 2014, doi: 10.1109/MMM.2014.2309499.
[23] F. Zhao, D. Inserra, G. Gao, Y. Huang, J. Li, and G. Wen, “High-efficiency microwave rectifier with coupled transmission line for low-power energy harvesting and wireless power transmission,” IEEE Trans. Microw. Theory Techn., vol. 69, no. 1, pp. 916–925, Jan. 2021, doi: 10.1109/TMTT.2020.3027011.
[24] Z. He, J. Lan, and C. Liu, “Compact rectifiers with ultra-wide input power range based on nonlinear impedance characteristics of Schottky diodes,” IEEE Trans. Power Electron., vol. 36, no. 7, pp. 7407–7411, Jul. 2021, doi: 10.1109/TPEL.2020.3046083.
[25] N. Sakai, K. Noguchi, and K. Itoh, “A 5.8-GHz band highly efficient 1-W rectenna with short-stub-connected high-impedance dipole antenna,” IEEE Trans. Microw. Theory Techn., vol. 69, no. 7, pp. 3558–3566, Jul. 2021, doi: 10.1109/TMTT.2021.3074592.
[26] M. Trotter, J. Griffin, and G. Durgin, “Power-optimized waveforms for improving the range and reliability of RFID systems,” in Proc. IEEE Int. Conf. RFID, Apr. 2009, pp. 80–87, doi: 10.1109/RFID.2009.4911196.
[27] S. A. Rotenberg, S. K. Podilchak, P. D. H. Re, C. Mateo-Segura, G. Goussetis, and J. Lee, “Efficient rectifier for wireless power transmission systems,” IEEE Trans. Microw. Theory Techn., vol. 68, no. 5, pp. 1921–1932, May 2020, doi: 10.1109/TMTT.2020.2968055.
[28] A. S. Boaventura and N. B. Carvalho, “Maximizing DC power in energy harvesting circuits using multisine excitation,” in Proc. IEEE MTT-S Int. Microw. Symp., Jun. 2011, pp. 1–4, doi: 10.1109/MWSYM.2011.5972612.
[29] A. Collado and A. Georgiadis, “Improving wireless power transmission efficiency using chaotic waveforms,” in Proc. IEEE/MTT-S Int. Microw. Symp. Dig., Jun. 2012, pp. 1–3, doi: 10.1109/MWSYM.2012.6259595.
[30] A. Boaventura, D. Belo, R. Fernandes, A. Collado, A. Georgiadis, and N. B. Carvalho, “Boosting the efficiency: Unconventional waveform design for efficient wireless power transfer,” IEEE Microw. Mag., vol. 16, no. 3, pp. 87–96, Apr. 2015, doi: 10.1109/MMM.2014.2388332.
[31] C. R. Valenta and G. D. Durgin, “Rectenna performance under power-optimized waveform excitation,” in Proc. IEEE Int. Conf. RFID, Apr. 2013, pp. 237–244, doi: 10.1109/RFID.2013.6548160.
[32] N. Pan, D. Belo, M. Rajabi, D. Schreurs, N. B. Carvalho, and S. Pollin, “Bandwidth analysis of RF-DC converters under multisine excitation,” IEEE Trans. Microw. Theory Techn., vol. 66, no. 2, pp. 791–802, Feb. 2018, doi: 10.1109/TMTT.2017.2757473.
[33] J. Blanco, F. Bolos, and A. Georgiadis, “Instantaneous power variance and radio frequency to DC conversion efficiency of wireless power transfer systems,” IET Microw., Antennas Propag., vol. 10, no. 10, pp. 1065–1070, Jul. 2016, doi: 10.1049/iet-map.2015.0790.
[34] X. Gu, P. Burasa, S. Hemour, and K. Wu, “Recycling ambient RF energy: Far-field wireless power transfer and harmonic backscattering,” IEEE Microw. Mag., vol. 22, no. 9, pp. 60–78, Sep. 2021, doi: 10.1109/MMM.2021.3086335.
[35] S. Hemour et al., “Towards low-power high-efficiency RF and microwave energy harvesting,” IEEE Trans. Microw. Theory Techn., vol. 62, no. 4, pp. 965–976, Apr. 2014, doi: 10.1109/TMTT.2014.2305134.
[36] X. Gu, S. Hemour, and K. Wu, “Far-field wireless power harvesting: Nonlinear modeling, rectenna design, and emerging applications,” Proc. IEEE, vol. 110, no. 1, pp. 56–73, Jan. 2022, doi: 10.1109/JPROC.2021.3127930.
[37] E. Fazzini, A. Costanzo, and D. Masotti, “Range selective power focusing with time-controlled bi-dimensional frequency diverse arrays,” in Proc. 2021 IEEE Wireless Power Transfer Conf. (WPTC), pp. 1–4, doi: 10.1109/WPTC51349.2021.9458208.
[38] D. Masotti, A. Costanzo, M. Del Prete, and V. Rizzoli, “Time-modulation of linear arrays for real-time reconfigurable wireless power transmission,” IEEE Trans. Microw. Theory Techn., vol. 64, no. 2, pp. 331–342, Feb. 2016, doi: 10.1109/TMTT.2015.2512275.
[39] P. F. Goldsmith, Quasioptical Systems: Gaussian Beam Quasioptical Propogation and Applications. Piscataway, NJ, USA: IEEE Press, 1998.
[40] R. Pereira, N. Carvalho, and J. Da Cunha, “Quasi-optical analysis of a double reflector microwave antenna system,” Wireless Power Transfer, vol. 5, no. 2, pp. 75–86, Mar. 2018, doi: 10.1017/wpt.2017.19.
[41] R. A. M. Pereira, N. B. Carvalho, and A. Georgiadis, “Focus location measurement of a quasioptical double reflector system,” in Proc. 2021 IEEE Wireless Power Transfer Conf. (WPTC), pp. 1–4, doi: 10.1109/WPTC51349.2021.9457870.
[42] D. R. Smith et al., “An analysis of beamed wireless power transfer in the Fresnel zone using a dynamic, metasurface aperture,” J. Appl. Phys., vol. 121, no. 1, p. 014901, Jan. 2017, doi: 10.1063/1.4973345.
[43] A. Georgiadis, A. Collado, and M. M. Tentzeris, Energy Harvesting, Technologies, Systems and Challenges. Cambridge, U.K.: Cambridge Univ. Press, 2021.
[44] A. Georgiadis, “Energy harvesting for autonomous wireless sensors and RFID’s,” in Proc. 31st URSI General Assem. Sci. Symp. (URSI GASS), Beijing, China, 2014, pp. 1–5, doi: 10.1109/URSIGASS.2014.6928982.
[45] H. J. Visser, A. Reniers, and J. Theeuwes, “Ambient RF energy scavenging: GSM and WLAN power density measurements,” in Proc. Eur. Microw. Conf., Amsterdam, The Netherlands, 2008, pp. 721–724, doi: 10.1109/EUMC.2008.4751554.
[46] Y. Kawahara, K. Tsukada, and T. Asami, “Feasibility and potential application of power scavenging from environmental RF signals,” in Proc. IEEE AP-S Antennas Propag. Int. Symp. (APSURSI), Charleston, SC, USA, 2009, pp. 1–4, doi: 10.1109/APS.2009.5171785.
[47] R. J. Vyas, B. S. Cook, Y. Kawahara, and M. M. Tentzeris, “E-WEHP: A batteryless embedded sensor-platform wirelessly powered from ambient digital-TV signals,” IEEE Trans. Microw. Theory Techn., vol. 61, no. 6, pp. 2491–2505, Jun. 2013, doi: 10.1109/TMTT.2013.2258168.
[48] R. J. Gutmann and J. M. Borrego, “Power combining in an array of microwave power rectifiers,” IEEE Trans. Microw. Theory Techn., vol. 27, no. 12, pp. 958–968, Dec. 1979, doi: 10.1109/TMTT.1979.1129774.
[49] A. Georgiadis, “Energy harvesting for autonomous wireless sensor platforms,” in Proc. 34th URSI General Assem. Sci. Symp. (URSI GASS), Rome, Italy, 2021, pp. 1–3, doi: 10.23919/URSIGASS51995.2021.9560611.
[50] M. Virili, A. Georgiadis, A. Collado, P. Mezzanotte, and L. Roselli, “EM characterization of a patch antenna with thermo-electric generator and solar cell for hybrid energy harvesting,” in Proc. IEEE Radio Wireless Symp. (RWS), San Diego, CA, USA, 2015, pp. 44–46, doi: 10.1109/RWS.2015.7129773.
[51] A. P. Sample, J. Braun, A. Parks, and J. R. Smith, “Photovoltaic enhanced UHF RFID tag antennas for dual purpose energy harvesting,” in Proc. 2011 IEEE Int. Conf. RFID, pp. 146–153, doi: 10.1109/RFID.2011.5764615.
[52] A. Georgiadis and A. Collado, “Improving range of passive RFID tags utilizing energy harvesting and high efficiency class-E oscillators,” in Proc. 6th Eur. Conf. Antennas Propag. (EuCAP), Prague, Czech Republic, 2012, pp. 3455–3458, doi: 10.1109/EuCAP.2012.6206429.
[53] V. Liu, A. Parks, V. Talla, S. Gollakota, D. Wetherall, and J. R. Smith, “Ambient backscatter: Wireless communication out of thin air,” in Proc. ACM SIGCOMM Comput. Commun. Rev., 2013, pp. 39–50, doi: 10.1145/2534169.2486015.
[54] S. N. Daskalakis, J. Kimionis, A. Collado, G. Goussetis, M. M. Tentzeris, and A. Georgiadis, “Ambient backscatterers using FM broadcasting for low cost and low power wireless applications,” IEEE Trans. Microw. Theory Techn., vol. 65, no. 12, pp. 5251–5262, Dec. 2017, doi: 10.1109/TMTT.2017.2765635.
[55] H. Kazemi, “61.5% efficiency and 3.6 kW/m power handling rectenna circuit demonstration for radiative millimeter wave wireless power transmission,” IEEE Trans. Microw. Theory Techn., vol. 70, no. 1, pp. 650–658, Jan. 2022, doi: 10.1109/TMTT.2021.3110966.
[56] S. Hemour, C. H. P. Lorenz, and K. Wu, “Small-footprint wideband 94GHz rectifier for swarm micro-robotics,” in Proc. IEEE MTT-S Int. Microw. Symp. Dig., May 2015, pp. 1–4, doi: 10.1109/MWSYM.2015.7167106.
[57] H.-K. Chiou and I.-S. Chen, “High-efficiency dual-band on-chip rectenna for 35- and 94-GHz wireless power transmission in 0.13-μm CMOS technology,” IEEE Trans. Microw. Theory Techn., vol. 58, no. 12, pp. 3598–3606, Dec. 2010, doi: 10.1109/TMTT.2010.2086350.
[58] N. Weissman, S. Jameson, and E. Socher, “W-band CMOS on-chip energy harvester and rectenna,” in Proc. IEEE MTT-S Int. Microw. Symp. Dig., Tampa, FL, USA, Jun. 2014, pp. 1–6, doi: 10.1109/MWSYM.2014.6848243.
[59] K. Matsui et al., “Microstrip antenna and rectifier for wireless power transfer at 94 GHz,” in Proc. IEEE Wireless Power Transf. Conf. (WPTC), Taipei, Taiwan, May 2017, pp. 1–3, doi: 10.1109/WPT.2017.7953902.
[60] P. He, J. Xu, and D. Zhao, “A W-band rectenna using on-chip CMOS switching rectifier and on-PCB tapered slot antenna achieving 25% effective-power-conversion efficiency for wireless power transfer,” in Proc. IEEE MTT-S Int. Microw. Symp. Dig., Aug. 2020, pp. 1055–1058, doi: 10.1109/IMS30576.2020.9223846.
[61] W. C. Brown, “The history of power transmission by radio waves,” IEEE Trans. Microw. Theory Techn., vol. 32, no. 9, pp. 1230–1242, Sep. 1984, doi: 10.1109/TMTT.1984.1132833.
[62] P. E. Glaser, “Power from the sun: Its future,” Science, vol. 162, no. 3856, pp. 857–886, Nov. 1968, doi: 10.1126/science.162.3856.857.
[63] C. T. Rodenbeck et al., “Microwave and millimeter wave power beaming,” IEEE J. Microw., vol. 1, no. 1, pp. 229–259, Jan. 2021, doi: 10.1109/JMW.2020.3033992.
[64] N. Shinohara, San Diego, CA, USA. 6 Recent Advance of Beam Wireless Power Transfer for Solar Power Satellite in Japan, IEEE Wireless Power Week 2021 School. (Jun. 1, 2021). [Online Video] . Available: https://www.youtube.com/watch?v=JLVx56c9_30
[65] “China to build space-based solar power station by 2035.” English.GOV.CN. [Online] . Available: https://english.www.gov.cn/news/topnews/201912/02/content_WS5de47b72c6d0bcf8c4c182a1.html
[66] A. Jones. “China’s super heavy rocket to construct space-based solar power station.” Space News. [Online] . Available: https://spacenews.com/chinas-super-heavy-rocket-to-construct-space-based-solar-power-station
[67] H. Xinbin and W. Li, “Study on multi-rotary joints space power satellite concept,” Aerosp. China, vol. 19, no. 1, pp. 19–26, 2018.
[68] H. Matsumoto, N. Kaya, S. Kinai, T. Fujiwara, and J. Kochiyama, “A feasibility study of power supplying satellite (PSS),” Space Power, vol. 12, pp. 1–6, Jan. 1993.
[69] B. Clerckx, R. Zhang, R. Schober, D. W. K. Ng, D. I. Kim, and H. V. Poor, “Fundamentals of wireless information and power transfer: From RF energy harvester models to signal and system designs,” IEEE J. Sel. Areas Commun., vol. 37, no. 1, pp. 4–33, Jan. 2019, doi: 10.1109/JSAC.2018.2872615.
[70] A. Costanzo, D. Masotti, G. Paolini, and D. Schreurs, “Evolution of SWIPT for the IoT world: Near- and far-field solutions for simultaneous wireless information and power transfer,” IEEE Microw. Mag., vol. 22, no. 12, pp. 48–59, Dec. 2021, doi: 10.1109/MMM.2021.3109554.
[71] G. Paolini et al., “RF energy harvesting from GFSK-modulated BLE signals,” in Proc. IEEE Topical Conf. Wireless Sensors Sensor Netw. (WiSNeT), 2021, pp. 27–29, doi: 10.1109/WiSNeT51848.2021.9413780.
[72] M. Rajabi, N. Pan, S. Claessens, S. Pollin, and D. Schreurs, “Modulation techniques for simultaneous wireless information and power transfer with an integrated rectifier–receiver,” IEEE Trans. Microw. Theory Techn., vol. 66, no. 5, pp. 2373–2385, May 2018, doi: 10.1109/TMTT.2018.2811491.
[73] S. Claessens, N. Pan, D. Schreurs, and S. Pollin, “Multitone FSK modulation for SWIPT,” IEEE Trans. Microw. Theory Techn., vol. 67, no. 5, pp. 1665–1674, May 2019, doi: 10.1109/TMTT.2019.2908645.
[74] S. Shen, J. Kim, C. Song, and B. Clerckx, “Wireless power transfer with distributed antennas: System design, prototype, and experiments,” IEEE Trans. Ind. Electron., vol. 68, no. 11, pp. 10,868–10,878, Nov. 2021, doi: 10.1109/TIE.2020.3036238.
[75] S. Deb et al., “An endoscopic wireless gastrostimulator,” Gastrointest Endosc., vol. 75, no. 2, pp. 411–415, Feb. 2012, doi: 10.1016/j.gie.2011.09.052.
[76] M. Q. Nguyen, Z. Hughes, P. Woods, Y. Seo, S. Rao, and J. Chiao, “Field distribution models of spiral coil for misalignment analysis in wireless power transfer systems,” IEEE Trans. Microw. Theory Techn., vol. 62, no. 4, pp. 920–930, Apr. 2014, doi: 10.1109/TMTT.2014.2302738.
[77] J. S. Ho et al., “Wireless power transfer to deep-tissue microimplants,” Proc. Nat. Acad. Sci. USA, vol. 111, no. 22, pp. 7974–7979, Jun. 2014, doi: 10.1073/pnas.1403002111.
[78] A. Basir and H. Yoo, “Efficient wireless power transfer system with a miniaturized quad-band implantable antenna for deep-body multitasking implants,” IEEE Trans. Microw. Theory Techn., vol. 68, no. 5, pp. 1943–1953, May 2020, doi: 10.1109/TMTT.2020.2965938.
[79] C. Liu, Y. Guo, H. Sun, and S. Xiao, “Design and safety considerations of an implantable rectenna for far-field wireless power transfer,” IEEE Trans. Antennas Propag., vol. 62, no. 11, pp. 5798–5806, Nov. 2014, doi: 10.1109/TAP.2014.2352363.
[80] IEEE Standard for Safety Levels with Respect to Human Exposure to Electric, Magnetic, and Electromagnetic Fields, 0 Hz to 300 GHz, IEEE Standard IEEE C95. 1, 2019.
[81] “ICNIRP guidelines on limiting exposure to time-varying electric, magnetic and electromagnetic fields (100 kHz to 300 GHz),” International Commission on Non-Ionizing Radiation Protection, 2020. [Online] . Available: https://www.icnirp.org/cms/upload/publications/ICNIRPemfgdl.pdf
[82] R. A. Bercich, D. R. Duffy, and P. P. Irazoqui, “Far-field RF powering of implantable devices: Safety considerations,” IEEE Trans. Biomed. Eng., vol. 60, no. 8, pp. 2107–2112, Aug. 2013, doi: 10.1109/TBME.2013.2246787.
[83] A. Christ, M. Douglas, J. Nadakuduti, and N. Kuster, “Assessing human exposure to electromagnetic fields from wireless power transmission systems,” Proc. IEEE, vol. 101, no. 6, pp. 1482–1493, Jun. 2013, doi: 10.1109/JPROC.2013.2245851.
[84] G. Loubet, A. Takacs, and D. Dragomirescu, “Implementation of a battery-free wireless sensor for cyber-physical systems dedicated to structural health monitoring applications,” IEEE Access, vol. 7, pp. 24,679–24,690, Feb. 2019, doi: 10.1109/ACCESS.2019.2900161.
[85] G. Loubet, “Autonomous wireless sensor networks for the implementation of communicating materials. Application to civil engineering industry,” Doctoral dissertation, INSA, Toulouse, France, 2021.
[86] A. Sidibe, A. Takacs, G. Loubet, and D. Dragomirescu, “Compact antenna in 3D configuration for rectenna wireless power transmission applications,” Sensors, vol. 21, no. 9, p. 3193, May 2021, doi: 10.3390/s21093193.
Digital Object Identifier 10.1109/MMM.2022.3210145