Electromagnetic Energy Harvesting and Wireless Power ... - IEEE Xplore

INVITED PAPER

Electromagnetic Energy Harvesting and Wireless Power Transmission: A Unified Approach This paper presents a rigorous procedure for the circuit-level analysis and design of entire systems, developed to provide power wirelessly in a very efficient way. By Alessandra Costanzo, Senior Member IEEE , Marco Dionigi, Member IEEE , Diego Masotti, Member IEEE , Mauro Mongiardo, Fellow IEEE , Giuseppina Monti, Luciano Tarricone, Senior Member IEEE , and Roberto Sorrentino, Life Fellow IEEE

ABSTRACT | In this paper, a rigorous procedure for the circuitlevel analysis and design of entire systems, developed to pro-

KEYWORDS | Energy harvesting; inductive coupling; rectenna; wireless power transfer

vide power wirelessly, is presented. A unified theoretical approach is first introduced, based on a two-port-equivalent circuit representation, to describe the wireless power transfer link when the transmitter and the receiver are either in the near-field or in the far-field region reciprocally. This approach allows one to compute in a straightforward manner the system figure of merit, namely the power transfer efficiency. Specific guidelines for the two configurations are then intensively discussed together with the adopted software tools based on the combination of full-wave analysis and nonlinear harmonic balance techniques. Several practical examples based on this design procedure are presented, demonstrating predicted and experimental behavior of unconventional devices for both near-field and far-field power transfer usage.

Manuscript received May 20, 2014; revised August 11, 2014; accepted August 29, 2014. Date of publication October 7, 2014; date of current version October 16, 2014. A. Costanzo and D. Masotti are with the Department of Electric Engineering and Information ‘‘Guglielmo Marconi,’’ University of Bologna, Bologna 40136, Italy (e-mail: [email protected]; [email protected]). M. Dionigi, M. Mongiardo, and R. Sorrentino are with the Department of Engineering, University of Perugia, Perugia 06125, Italy (e-mail: [email protected]; [email protected]; [email protected]). G. Monti and L. Tarricone are with the Department of Engineering for Innovation, University of Salento, Lecce 73100, Italy (e-mail: [email protected]; [email protected]). Digital Object Identifier: 10.1109/JPROC.2014.2355261

I . INTRODUCTION In many current application domains, such as medical and environmental monitoring, industrial automation, wireless sensor networks, intelligent transportation systems, etc., the need for battery-free ultralow-power devices, possibly wearable or implantable, is increasing dramatically. Ambient or structural monitoring based on the use of a large number of distributed battery-less microsystems with sensing capabilities is one of the main application areas toward the paradigm of zero power, battery-less systems. Such devices are normally off and need to be interrogated a few times to provide wirelessly the information about their monitoring activity [1], [2]. Depending on various environmental conditions, two basic approaches can be followed. According to the first one, which is typically ambient oriented, energyharvesting (EH) systems exploit energy sources already present in the environment, namely electromagnetic (EM), sunlight, mechanical, and thermal, or a combination of them. The corresponding scenario, in the case of EM sources, is illustrated in Fig. 1, where the energy to sustain device operations is provided by RF sources, today commonly present in any humanized environment [3]. Antennas thus

0018-9219 Ó 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

1692

Proceedings of the IEEE | Vol. 102, No. 11, November 2014

Costanzo et al.: Electromagnetic Energy Harvesting and Wireless Power Transmission: A Unified Approach

Fig. 1. Energy-harvesting scenario in presence of multiple, randomly distributed RF sources.

require featuring broadband or multiband behaviors to cover all the wireless standards and with circular polarization to ensure signal reception in any link conditions [4]–[8]. Since the ultimate target is to scavenge all RF sources at the same time, at any possible frequency, polarization, angle of arrival, and power intensity, it is apparent that the design of such harvesting systems is a very demanding task. In order to accurately design the whole harvesting system, EM theory, full-wave simulation, and nonlinear harmonic-balance (HB) analyses should be concurrently employed in order to make a circuit-level design of the entire microwave link possible. With this approach, the

radio channel between the source and the harvester can be modeled as a subsystem block [9], [10]. In the second approach, the radio-frequency (RF) source is known, so that the EH system design takes advantage from the knowledge of the RF frequency and (possibly) the direction of arrival of the field and its polarization, as is the case of wireless power transmission (WPT) techniques based on the far field (FF–WPT) [11]. For example, dedicated RF sources ready to provide the necessary low energy when requested [12] may be installed. This case is similar to powering passive RF identification (RFID) tags. This scenario is depicted in Fig. 2 for one out of many applications [13]. Ultrahigh-frequency (UHF) energy showers periodically provide the requested amount of energy to battery-less tags (possibly realized in eco-compatible materials), equipped with EH systems targeted at 868 MHz. As an alternative, WPT techniques based on near-field coupling (NF–WPT) can be employed to recharge devices with the energy needed for their operation. Resonant NF–WPT has indeed received renewed interest after the experiment reported in [14]. Several investigations have then been presented in the literature; see, e.g., [15]–[21] to cite just a few. Such investigations have been carried out either by experiments or by theoretical developments, based on, for example, network theory or full-wave EM simulators. In NF–WPT, several situations may occur [22] depending on the application, thus on power level, efficiency,

Fig. 2. UWB tags with onboard harvesting system deploying energy coming from known ‘‘energy showers’’ (GRETA project).

Vol. 102, No. 11, November 2014 | Proceedings of the IEEE

1693


vester for spurious emissions of compact fluorescent lamps are illustrated.

I I. THEORETICAL APPROACH TO MODE LI NG OF E NERGY HARVESTING AND RESONANT WIRELESS POWER TRANSFER

Fig. 3. Schematic of the moving field inductive power transfer system proposed in [23].

the distance between transmitter(s) (TXs) and receiver(s) (RXs), the applicable EMC normative, operating frequency, etc. A typical example is the automotive scenario, where TXs and RXs are usually in quite close proximity, but too distant for purely inductive coupling. Examples of NF–WPT charging of moving vehicles have been described, for example, in [23] and [24]. In particular, in [23], a moving field inductive power transfer system for electric vehicles has been proposed (see Fig. 3), while in [24], a capacitive coupling has been suggested. In this paper, a theoretical approach for the design of powering devices, located either in the near field or in the far field, using EM waves, is illustrated in Section II. To accurately quantify the energy/power levels involved, the power link transfer efficiency is introduced as the system figure of merit, and a method for its accurate evaluation is presented. Indeed, the knowledge of the actual transfer efficiency is mandatory when the radio channel between the energy source and the device is a complex environment, such as harsh civil or industrial sites or body tissue layers. In particular, the use of the conjugate image parameters is shown to be the most appropriate circuit representation in the case of complex impedances, for example, for near-field (reactive) links. Then, a general-purpose multidomain approach for the accurate computation and optimization of power efficiency of the system is presented, based on the combination of EM and circuit simulations with complex channel characterization and EM theory. In Section III, a selected variety of prototypes and the associated experimental results on both FF–WPT and NF– WPT systems are then discussed. More specifically, with regard to far-field systems, two rectifying antennas (rectennas) are presented: a wearable rectenna on textile materials and a tetra-band rectenna. Concerning NF–WPT systems, a wireless resonant energy link for energization of implanted medical devices (IMDs; such as pacemakers, gastrostimulators, drug-delivery pumps, etc.) and a har1694

It is possible to consider in the same unified manner a system for EH, for WPT, and a system of antennas. In all cases, we can consider M sources and N receivers, and we can describe the entire system by a linear network with K ¼ M þ N ports. Naturally, when the source(s) and the receivers(s) are in the reciprocal Fraunhofer (far-field) region, only active power transfer is involved, while when they are in the near-field region, the reactive components play an important role. As noted in [25], a system of closely spaced antennas can be also considered as a WPT structure. Noticeably, the results that we will illustrate match perfectly with those presented in [25]. In the following, for simplicity, we will refer to WPT systems with M ¼ N ¼ 1. However, in the case of EH, the case with M > 1 is also considered. It is important to note that in NF–WPT and FF–WPT systems the designer has control over the TX part, whereas in EH applications the TX part can be characterized but may not be under the designer’s control. In several instances, in the characterization of EH systems, it may be convenient to look directly into the output port of the twoport network so as to obtain an equivalent Thevenin– Norton characterization. In fact, when just one receiver is present (i.e., N ¼ 1), as for the case of a simple EH antenna, the goal of achieving maximum power transfer (MPT) is realized by conjugate matching. In the case of a WPT system, when we have one transmitter and one receiver, we can either maximize the power transfer to the load (at the expense of the efficiency) or we can maximize the efficiency by using the conjugate image impedances. The two different solutions, while sharing the same reactive part of the load impedance, require different resistive parts. In the following, we will focus on the solution that maximizes efficiency.

A. Evaluation of Power Transfer Efficiency in the Near-Field Region of the Source When the TX and the RX are in close proximity of each other, significant coupling takes place via the respective reactive fields. For measurement and simulation purposes, especially at high frequencies, it is often convenient to refer to the scattering formalism [26]–[28]. The definition of scattering parameters, in turn, requires the selection of reference impedances; typically, real reference impedances are selected and the relevant formalism is generally denoted by traveling waves [29]. In the case of complex reference impedances, the formalism of pseudowaves has been introduced in [30]. As noted in [31]–[36], however,



The conjugate image impedances Zci are then given by Zc1 ¼ r11 ðr jx Þ jx11 Zc2 ¼ r22 ðr þ jx Þ jx22 :

(4) (5)

Fig. 4. Definition of conjugate image impedances.

when dealing with power transmission problems, the conventional definition of scattering parameters may not be fully adequate. Power waves, a differently defined reflection coefficient, along with appropriate complex reference impedances may be a more convenient choice. The differences between the various formalisms, their relative merits, and drawbacks have recently been discussed in some detail in a couple of very interesting papers [37], [38]. In the Appendix, we briefly recall the terminology used in the various scattering formulations. Here, we discuss how to avoid the difficulties related to complex reference impedances, and we introduce the conjugate image impedances, which are the basis for obtaining the MPT in both EH and WPT networks. Naturally, the characterization of the two-port network can also be done using, for example, the impedance representation, but, as described next, the maximum efficiency is obtained when the conjugate image impedances are used as terminating impedances of the RF link. Moreover, their adoption provides a considerable insight into the design of both EH and WPT systems [31], [35]. Let us recall the definition of conjugate image impedance, not to be confused with the conventional image impedances. As depicted in Fig. 4, Zc1 (here and in the following, the asterisk denotes the complex conjugate) is the input impedance at port 1 when port 2 is terminated with Zc2 ; similarly, when port 1 is terminated with Zc1 , the impedance Zc2 is seen at port 2. Let us denote the impedance matrix of the two-port network as

The efficiency of the two-port network, defined as the ratio between the power delivered to the load ðPRX Þ and the power available from the generator ðPTX Þ

RFRF ¼

PRX PTX

(6)

can be easily expressed on the conjugate-image basis. In fact, when power waves are employed, and the corresponding generalized scattering matrix [29] is used, the efficiency can be directly expressed as S21 . By contrast, as observed in [37], the expression of efficiency in terms of the standard reflection coefficients requires a quite lengthy expression. MAX The ‘‘maximum efficiency’’ RFRF is obtained when generator of impedance Zc1 is connected to the input port, while load Zc2 is connected to the output port. Following [31], by introducing parameter c

c ¼

1 r þ jx 1 þ r þ jx

(7)

it is shown that the maximum efficiency is given by

MAX RFRF

z21 ¼ c : z

(8)

12

For reciprocal networks, the maximum efficiency is simply

Z11 Z21

Z12 Z22

(1)

with zik ¼ rik þ jxik ði; k ¼ 1; 2Þ. The conjugate image impedances can be calculated using the elegant procedure of [31]. For a reciprocal network, the following quantities are introduced: ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi s ffi r212 x212 r ¼ 1 1þ r11 r22 r11 r22 r12 x12 : x ¼ r11 r22

(2) (3)

MAX RFRF ¼ jc j:

(9)

Once the conjugate image impedances have been computed at the operating frequency, we can realize a matching network at the source and at the load that transform the source impedance into Zc1 and the load impedance into Zc2 . In a more direct approach, we can also add the imaginary parts of the conjugate image impedances directly to the two-port network. In this way, it is now sufficient to terminate the modified two-port network on the real part of the conjugate image impedances. Note that once we have added the reactive parts of the conjugate image impedances to the two-port network, we can use as reference impedances the real part of the conjugate image


1695


impedances, thus eliminating the problems related to the selection of traveling waves, pseudowaves, or power waves. An example of this procedure is given in the following part of this section; but first, let us point out the differences between conjugate image impedances and image impedances. The image parameter approach is a well-known filter design method based on the cascade of cells, all being matched to the common image impedances. The latter are defined as follows. When port 2 is terminated with image impedance Zi2 , image impedance Zi1 is seen at port 1; and vice versa, when port 1 is terminated with Zi1 , we see, at port 2, impedance Zi2 . Image impedances Zi1 and Zi2 can be expressed in terms of the impedance and admittance parameters of the two-port network as sffiffiffiffiffiffi Z11 Zi1 ¼ y11 sffiffiffiffiffiffiffi Z22 Zi2 ¼ : y22

(10)

(11)

It is worth noting that the conjugate image impedances are not the conjugate of the image impedances. This is clearly demonstrated by the following numerical example. Let us consider the WPT arrangement shown in Fig. 5, where the first and last coils (input/output ports) are inductively coupled with the second and third ones. The second and third coils are resonant at the operating frequency, while the first and fourth coils can be resonant or not. This structure is similar to the one used in [14] and provides a simple example of a WPT system.

Fig. 5. Experimental realization of a WPT system composed by a four-coil structure; the first and last coils (input/output) are inductively coupled with the second and third coils, which are resonant and which perform the sought wireless power transmission.

1696

Fig. 6. Measured scattering parameters of the WPT system of Fig. 5.

In order to characterize the WPT system of Fig. 5, we have measured the S-parameters using a vector network analyzer (VNA) operating with 50- reference impedances (Fig. 6). After converting the scattering parameters into the impedance parameters, the conjugate image parameters have computed using (2)–(4) and are shown in Fig. 7. For example, consider, at the frequency of 68 MHz, the measured scattering matrix ½S ¼

16:4641 þ j9:8042 12:3951 j79:1614

12:3951 j79:1614 : 13:2239 þ j32:3103

The resulting conjugate image impedances and image impedances are quoted in Table 1. From the above results, it is apparent that the conjugate image impedances are significantly different from the image impedances. The conjugate image matching can be obtained by putting two series capacitors at both ports, the real parts of the conjugate image parameters being the reference impedances. More specifically, by inserting two series capacitors of 27.6 and 25.3 pF at ports 1 and 2, respectively, and using 47.5 as the reference impedance at port 1, and 38.1 at port 2, the S-parameters shown in Fig. 8 are obtained. By comparison with Fig. 6, the importance of selecting the conjugate image impedances as the reference impedances is evident, since this choice provides the best power transfer efficiency.

Fig. 7. Conjugate image impedances of the WPT system of Fig. 5.



Table 1 Conjugate Image Impedances and Image Impedances of the WPT Arrangement of Fig. 5

In terms of EM field computation, it is worth pointing out that if the conjugate image impedances were to be used to terminate the network, a full-wave simulation with 50- terminating impedances would not provide the correct field amplitudes. A full-wave simulation with terminating conjugate image impedances should be employed instead.

B. Evaluation of Power Transfer Efficiency in the Far-Field Region of the Source When the TX and the RX are sufficiently apart, the farfield approximation can be used. If both antennas are located in the respective Fraunhofer regions, their interaction is due only to the radiated EM field, so that the reactive effects can be neglected in the equivalent twoport network of the TX/RX link. The field incident on the RX antenna does not present reactive components, and the receiving antenna does not perturb the transmitting one. The characterization of the link can be rigorously obtained by computing the Thevenin–Norton equivalent circuit of the current induced by the radiated EM field on the port of the RX antenna. This scenario is well suited for an EH system, where the designer can only operate on the RX end. A nonlinear design process is then carried out to match the Thevenin–Norton equivalent circuit to the rectifying system. The available RF power at the rectenna location ðPRX Þ represents the maximum power the antenna is able to deliver to the rectifying circuit. The accurate estimation of PRX is obviously needed to correctly define the rectenna

Fig. 8. Scattering parameters of the WPT system of Fig. 5 terminated with its conjugate image impedances. The component values at 68 MHz are C1 ¼ 27.6 pF, R1 ¼ 47.5 at port 1, and C2 ¼ 25.3 pF, R2 ¼ 38.1 at port 2.

efficiency [11]. Let us consider the case of multiple ðMÞ RF sources: the corresponding RF power collected by the receiving antenna to be used in HB analysis can be cast in the following way:

PRX

M Jeq ð!k Þ2 X ¼ 8Re½YH ð!k Þ k¼1

(12)

where YH ð!k Þ is the full-wave admittance matrix of the antenna, including its feeding network, defined in the frequency bands of interest, while Jeq ð!k Þ is the Norton current source, at the generic kth frequency, equivalent to the incident field at the same frequency ðEi ð!k ÞÞ. The direct application of EM theory allows one to rigorously evaluate such current contribution. Two cases can be distinguished depending on the reciprocal positions of the transmitting RF source and the receiving harvesting system. In a conventional link situation, with TX and RX antennas in the Fraunhofer region of each other and in the maximum link direction, the incident field Ei can be represented as a uniform plane wave. The corresponding equivalent current generator has the following expression [39], [40]:

Jeq ð!k Þ ¼ j

½1þR0 YH ð!k Þ U 2k rejr Ei ð!k Þ EH ðr; ; ; !k Þ: (13)

In (13), EH is the far field radiated by the harvester antenna computed in the direction of incidence ð; Þ at an arbitrary distance r from its phase center, when driven by a sinusoidal voltage source of frequency !k , electromotive force U, and internal impedance R0 , and is the freespace wave impedance. The hypotheses on which (13) is based typically hold in FF–WPT applications, where the knowledge of the RF source position allows one to establish a conventional link. On the other hand, many RF energy harvesting scenarios must be considered as ‘‘nonconventional’’ wireless links either because of limited TX/RX distances or because they are far from the maximum link direction.


1697


Fig. 9. Situation considered for the rigorous computation of the Norton equivalent current generator.

In such situations, the amplitude and phase errors with respect to a uniform incident plane wave Ei cannot be neglected. The EM theory permits again to achieve another rigorous expression of the Norton equivalent current generator by resorting to the situation depicted in Fig. 9. In this case, the equivalent generator turns out to be [39]

C. Optimization of the System Efficiency The theoretical approach to rectenna design is described in the following; it can be adopted for NF–WPT systems provided that suitable devices are employed. Once the actual rectenna excitations are established, the HB technique needs to be used in the nonlinear design of the rectifying system in order to account for the variable power range. The block representation of an entire FF–WPT system in the presence of a real radio channel and an RF source is shown in Fig. 10. This model is straightforwardly adoptable for the far-field scenarios described above. For the case of NF–WPT, it is still valid by substituting the radio channel and the receiving antenna with the twoport network discussed in Section II-B. In Fig. 10, the power quantities to be monitored at each subsystem section are outlined. These quantities are adopted to compute the overall efficiency [4] as a product of three contributions TOT ¼ RFRF RFdc dcdc ¼

Jeq ð!k Þ ¼

1 þ R0 YH ð!k Þ ^n ZU

½Ei ðP Þ HH ðP Þ Hi ðP Þd

(14)

where all the fields are evaluated at !k , on a plane, with normal versor ^n, placed between the TX and RX antennas. The nonconventional channel can be rigorously taken into account as well by including the neighboring objects affecting the propagation in the two full-wave simulations of the antennas. Once the antenna topology and the corresponding equivalent current generator (13) or (14) have been defined, the nonlinear behavior of the rectenna as a whole (see Fig. 10) can be obtained by resorting to the HB analysis method.

PRX Pdc PST : PTX PRX Pdc

These contributions are now discussed. As a preliminary step, the system design requires an accurate estimate of the radio-link efficiency at the RF band of interest (i.e., RFRF , as defined in Section II-A). Referring to (6), in the case of FF–WPT applications, PTX is the RF power available at the transmitting antenna input port, and PRX is the RF power received by the antenna. The quantity PTX is exactly known in case of intentional FF–WPT or it is approximately estimated in case of ambient energy harvesting [3]. It is noteworthy that the sole optimization variables at designer’s disposal are the receiving antenna radiating characteristics (i.e., efficiency, radiation surface, and polarization). In the next step, the linear subnetwork, representing the receiving antenna, and its feeding network must be optimized together with the rectifying circuit and the rectenna load to totally convert the RF power received by the antenna into direct current (dc) power ðPdc Þ. Accordingly, the maximization of RF–dc efficiency

RFdc ¼

Fig. 10. Building blocks of an RF energy harvesting system.

1698

(15)

Pdc PRX

(16)

must be carried out, focusing on those power levels typical of harvesting scenarios (e.g., 20 0 dBm). To reach this goal, an optimized matching network is needed between the antenna and the rectifier ports. Note that this is a cumbersome nonlinear design, if variable frequencies



mum link condition at each frequency. Each PTX value is associated to one excitation in the single-tone analysis, while it is equally distributed to the three tones in the multitone case. These PTX values are chosen in such a way that they roughly result in PRX values ranging from 20 to þ5 dBm in the bands of interest. It is clearly shown that the rectification of higher order intermodulation products actively contributes to the rectified power Pdc mainly at the lower input power levels. The last factor in (15) is optimized when power management units are included. In this case, the rectenna load consists of the input impedance of the dc–dc converter [11], [39], [44], which is optimized to dynamically track the rectenna optimum load. Thus, with reference to Fig. 10, the dc–dc efficiency is computed as

Fig. 11. Effects of diode package on RF–dc efficiency at 2.45 GHz.

and incident powers are involved [41]. For this reason, a space-mapping technique [42] can be useful in the nonlinear optimization loop in order to efficiently take into account the full-wave description of the linear subassembly. As regards the nonlinear rectifier, a full-wave peak-topeak diode-based topology should be adopted, which turns out to be the best choice for very low-power budget applications [43]: for example, the low-threshold Skyworks SMS7630 diode can be adopted. The inclusion of the diode package is recommended since the rectenna performance may sensibly vary, especially at high frequencies. As an example, Fig. 11 shows the simulated dependence of RF– dc efficiency of a wearable rectenna at 2.45 GHz on the SMS7630 diode package model. For multifrequency incident fields, the added value of the multiband optimization can be checked by a multitone HB analysis of the rectenna nonlinear regime in the presence of multiple ambient sources. This way the nonlinear effects of intermodulation products to the rectified output dc power can be further exploited. Table 2 shows the gain obtained in virtue of higher frequency harmonics for a triband wearable energy harvester (HB simulation spectrum given by the intermodulation up to the third order of the three input signals at 900, 1750, and 2450 MHz). In the table, PTX is the power transmitted by a resonant dipole located at 30 cm from the harvester and in the maxi-

dcdc ¼

PST : Pdc

(17)

A common specification for this subsystem operating at baseband is to keep the rectified voltage ðVdc Þ at about one half of the open-circuit voltage (VOPEN , when Idc ¼ 0), this condition being close enough to the maximum power point (MPP) region. Fig. 12 demonstrates this assumption for a wearable multiband energy harvester at two operating frequencies, for an RF available power of 14 dBm [45]. During this design process, the nonlinear simulation is carried out by time-domain transient analysis, with switching times and storage capacitor value as design parameters. The RF and baseband subsystem interactions are taken into account by including the time-domain model of the dispersive antenna [46]. A useful definition of the network function (17) during the energy transfer operation is obtained by using a suitable percentage p (e.g., p ¼ 90%) of the maximum energy storable by the output capacitor ðEST max Þ. Therefore, the design goal inside the timedomain simulator is given in terms of the following expression for (17):

dcdc ¼

PST pEST max ¼ Pdc Pdc TST

(18)

Table 2 Comparison of Single-Tone Versus Multiple-Tone Analyses


1699


Fig. 12. MPP zone and V OPEN values for a wearable harvester at 900 and 1750 MHz, for the same PRX level.

where TST is the storage time needed to charge the output capacitor at the desired percentage level.

III . EXPERIMENTAL RE SULT S In this section, some experimental results on devices for EH and WPT applications are presented. As previously discussed, two main strategies can be identified depending on the distance between the transmitter and the receiver. In both cases, the power link is generally implemented by using electromagnetically coupled resonant systems. In a far-field link, antennas are used to transmit and receive power, whereas in a near-field link, electrically or magnetically coupled systems are used. In the following part of this section, some examples of devices exploiting both near-field and far-field couplings with the source are presented.

A. Far-Field Devices This section focuses on rectenna devices. In more detail, the following devices are illustrated: / an ultrahigh-frequency (UHF) band wearable rectenna well suited for both energy harvesting and WPT; / a tetra-band genetic-based rectenna designed to harvest from the Global System for Mobile Communications (GSM), the Universal Mobile Telecommunications System (UMTS), and WiFi RF sources. 1) UHF Wearable Rectenna: Wearable devices play a key role in wireless body area network applications. In particular, efficient and compact antennas are crucial to implement the wireless communication among body sensor units (BSUs). In this regard, in order to achieve the energy autonomy of BSUs, wearable rectennas are of particular interest [47], [48]. The device proposed in [48] is a textile rectenna optimized for operation in the UHF band. Photographs of the 1700

Fig. 13. Photographs of the wearable rectenna presented in [48]. (a) Back view. (b) Front view: slotted square patch and (c) rectifier. (c) Example of placement of the textile rectenna.

front and back views are given in Fig. 13, while the stack layer adopted for realization is illustrated in Fig. 14. All the conductive parts of the antenna are fabricated by using a nonwoven low-cost adhesive fabric [49]. This fabric has the main advantage of avoiding fraying problems, thus allowing the simple realization of complicated geometries. In particular, according to the low-cost and time-saving method suggested in [50] for the realization of

Fig. 14. Stack layer adopted for the realization of the textile rectenna illustrated in Fig. 13.



Fig. 15. Schematic of the rectifier illustrated in Fig. 13(c).

RFID tags, the prototype illustrated in Fig. 13 was fabricated by using a common cutting plotter. The antenna is a square patch on a bilayer of pile ð"r ¼ 1:12Þ and jeans ð"r ¼ 1:67Þ; a compact geometry and an elliptical polarization are obtained by means of a rectangular slot on the diagonal and a modified T-slit on the edges. As for the rectifier, which is a full-wave bridge rectifier, the corresponding schematic is illustrated in Fig. 15. The optimization of the rectifier was performed by means of HB simulations taking into account experimental data of the input impedance of the wearable patch antenna. With reference to the use of the rectenna for powering high impedance sensors, a value of 1 k was assumed for the load. From Fig. 15, it can be seen that the matching network between the antenna and the rectifier consists of a microstrip line and a series varactor with a capacitance range of 0.65–2.5 pF. This variable capacitor allows adjusting the level of matching between the antenna and the rectifier during experimental tests, thus compensating the difference between numerical and experimental data related to the tolerance and the parasitic effects of the realization process. In order to preserve the load from unwanted RF signals, a dc pass 4.7 F capacitor in shunt configuration with the output port is also used. It is worth pointing out that all the microstrip lines were realized by using the same cutting plotter adopted for shaping the patch antenna. As for the soldering process of the lumped elements, it was realized by means of a standard soft soldering. Tests were performed by generating an RF signal by means of a software-defined radio (SDR) platform [51]. The PMM 8053A broadband field meter with the EP-183 isotropic probe was used to measure the power density incident on the rectenna (SRF ; see Fig. 10). Experimental data obtained in this way were used to calculate the RF–dc conversion efficiency (i.e., RFdc ) according to the definition introduced in (14) and (16). Figs. 16 and 17 show RFdc as a function of the RF power density incident on the textile antenna. To compute PRF in (16), the effective area of the antenna was used ðPRX ¼ SRF Aeff Þ.

For an incident power density of 14 W/cm2 the measured RF–dc conversion efficiency has a maximum of about 50% at 876 MHz. From Fig. 16(b), it can be also noted that for a load of 1 k the conversion efficiency assumes values higher than 45% for an incident power density in the range of W/cm2 [3], [16]. Fig. 17 shows the results obtained by varying the frequency of the signal generated by the SDR while keeping constant the power density and the load at 14 W/cm2 and 1 k, respectively. From experimental data, in the frequency range 860–918 MHz, RFdc is higher than 20%.

Fig. 16. RF–dc conversion efficiency of the rectenna illustrated in Fig. 13: (a) experimental data obtained at 876 MHz by varying the resistive load and (b) the power density incident on the antenna.


1701


Fig. 17. Measured RF–dc conversion efficiency of the rectenna of Fig. 13. Results of tests performed by varying the frequency of the RF input signal.

2) Tetra-Band Genetic-Based Rectenna: The simultaneous presence of different wireless RF sources in most of the humanized environments suggests the possibility to collect energy from multiple sources. The design of a single multiresonant highly efficient antenna, possibly with circular polarization (CP), becomes a demanding task. Here we describe the project presented in [5] of a multilayered, aperture-coupled printed rectenna, able to harvest from GSM 900, GSM 1800, UMTS, and WiFi standards. Due to the high number of involved degrees of freedom, we resort to a genetic algorithm (GA) optimization tool suitably combined with a full-wave solver. As a starting point of the optimization process, we adopt an annular ring planar antenna with inner and outer radii of 19.5 and 65 mm, respectively. A suitable choice for the ring resonant modes is the following: TM11, TM21, TM31, TM12, to resonate at 900, 1800, 2150, and 2450 MHz, respectively. The subdivision in pixels of the starting point topology is shown in Fig. 18(a): this digitalization process involves both the annular ring (trapezoidal pixels) and the two orthogonal apertures in the ground plane for CP purposes (rectangular pixels).

Fig. 18. Topology of the GA-based multilayered annular ring antenna: (a) starting; (b) final.

1702

Fig. 19. Measured and simulated scattering parameters of the two-port genetic-based antenna of Fig. 18(b).

For symmetry reasons, the GA algorithm manages only one eighth of the annular ring and one aperture; thus, the chromosome length results equal to 248, i.e., the total number of optimization variables (pixels). In Fig. 18(b), the exotic shapes of the patch metallization and of the apertures are shown. At the four frequency bands, design goals are specified for the port reflection and transmission coefficients, since both significantly affect the antenna radiation efficiency [5]. The results at the desired resonant frequencies in terms of port matching and decoupling are quite satisfactory, and this is also confirmed by measurements, as shown in the comparisons of Fig. 19. The poor decoupling behavior at 1800 and 2150 MHz (S21 > 6 dB) is due to the corresponding mode surface current patterns [5]. In spite of this, the simulated two-port antenna efficiencies are good: 70%, 53%, 51%, and 86% at the TM11, TM21, TM31, and TM12 resonant frequencies, respectively. The design of a broadband divider and 90 -phase shifter (for CP purposes) loaded by the dispersive EM description of the antenna is then carried out: the lengths and widths of the topology of Fig. 20(b) are used as design variables. The resulting layout is adopted as the starting point for the HB-based optimization of the entire rectenna: a multiband matching network is placed between the full-wave description of the antenna and the singlestage rectifier, and the dc load is included among the design variables, for each band of interest. Indeed, specifications on (16) are simultaneously given for the four fundamental frequencies of interest and for a set of possible incoming low-power levels (200 dBm). The final layout of the rectenna feeding and matching networks is given in Fig. 20, where the photo of the prototype is shown. Finally, the overall rectenna conversion efficiency is evaluated and measured in a realistic office scenario for two operating frequencies, in the presence of a single RF source. The results are shown in Fig. 21, for different values of the received available power: some differences between measurements and simulations are present, mainly



lamps (CFLs). The second example of application is a nearfield link optimized for powering IMDs. In both cases, a wireless power transfer is implemented by inductive coupling.

Fig. 20. Photo of the (a) front and (b) backside of the prototype with highlighted different sections.

in the lower band where the system has demonstrated higher sensitivity with respect to variations of the diode model parameters.

B. Near-Field Devices In this section, two examples of devices operating in the near-field region of the source are reported. The first one is a harvester optimized for power generation from spurious emissions of compact fluorescent

1) EH Devices (Resonant Energy Scavenger for Power Generation by Spurious Emissions From CFL): With regard to devices for EH, based on a far-field coupling, several rectennas optimized to harvest the EM energy associated to wireless communication systems have been proposed in the literature [47], [48], [51]–[54]. The main drawback of these devices is related to the characteristics of the source. The EM power density available for harvesting in common indoor environments in fact is typically very low and extremely variable in terms of both frequency and intensity. According to the previous observations, with reference to home monitoring applications, particularly attractive is the idea suggested in [55] and [56] of harvesting lowfrequency spurious emissions that can be found in common household environments. In particular, in [55] and [56], a device for power harvesting from spurious emissions by CFLs is proposed. CFLs represent an attractive source for EH applications being widely used for artificial lighting of indoor environments and emitting a relatively strong EM field in the frequency range from few tens to some hundreds of kilohertz [56]– [60]. According to the equivalent circuit of Fig. 22, the device proposed in [55] and [56] exploits a near-field magnetic coupling implemented by means of a resonant loop to harvest RF spurious emissions of CFLs. The goal here is to collect the energy that would otherwise be wasted associated to spurious emissions of CFLs. Such energy can be made available to power sensors for home monitoring or for battery recharging. Two realizations optimized to operate with a different relative position of the harvester with respect to the CFL are shown in Fig. 23(a)–(c). In the following, they will be referred to as resonators of type A and type B, respectively. Both harvesters consist of a resonator, a bridge rectifier for RF–dc conversion, a network matching the rectifier to the resonator, and a dc pass capacitor in shunt

Fig. 21. Measured and simulated conversion efficiencies for the

Fig. 22. Equivalent circuit of the resonator inductively coupled with

genetic-based rectenna at 900 and 2450 MHz.

spurious emissions of a CFL.


1703


detail, the photo shows the test performed with a 50- matched realization of the resonator of type B coupled with a 30-W warm white Beghelli CFL. The spectrum obtained this way is given in Fig. 24(b); the total RF power received in the frequency range 20–250 kHz was approximately equal to 2.5 mW. Similar results were obtained for the resonator of type A; in this case, the total RF received power was 2 mW. These results were taken into account in optimizing the harvesters by means of HB simulations. In more detail, according to the equivalent circuit illustrated in Fig. 22, each harvester was optimized by modeling the resonator coupled with a CFL with an alternating current (ac) voltage source. The spectrum of the input signal was the one obtained by means of measurements [see Fig. 24(b) for the harvester of type B]. As for the dc power delivered to a resistive load, measured data are illustrated in Fig. 25. It is evident that three type A harvesters in a lampshade configuration [see Fig. 23(b)] can deliver more than 2-mW dc power when the load is in the range of 0.5–2.5 M. The corresponding RF–dc conversion efficiency is given in Fig. 25(b); RFdc , as defined in (16), was calculated by using the experimental setup shown in Fig. 24(a). In more detail, the total RF received power obtained by integrating the RF spectrum received in the frequency range 20–250 kHz was used for PRX . As can be noted from Fig. 25(b), a maximum of 33% was obtained in the case of the harvester of type A.

Fig. 23. (a)–(b) Energy harvester proposed in [55] (harvester of type A): (a) front and back views (dimensions are in centimeters); and (b) lampshade configuration obtained by arranging three harvesters around the CFL. (c) Harvester proposed in [56] (harvester of type B).

configuration with the load. According to experimental results demonstrating a relative peak of spurious emissions of common CFLs in a frequency range around 50 kHz [61], the resonators were designed to have a resonance frequency at about 50 kHz. The geometry, dimensions, and relative position of the resonators with respect to the CFL were optimized to obtain a good compromise between overall dimensions and efficiency of the harvesters. As for the matching network, with reference to the use of the harvesters for powering high impedance sensors, it was optimized by considering a load ðRL Þ of 1 M for the rectifier. Fig. 24(a) illustrates the experimental setup adopted for measurements of the spectrum collected by the resonators when coupled with a CFL HP E4411B. In more 1704

2) WPT for Medical Applications (434-MHz Inductive Link for IMDs): Finally, we consider the application proposed in [62], where an inductive power link for powering IMDs is presented. IMDs for local stimulation (i.e., pacemakers, gastrostimulators, deep brain stimulators) are crucial for the treatment of important diseases. Today, these devices have built-in batteries and have to be replaced on a regular basis with evident discomforts and costs. Accordingly, in recent years, great efforts have been devoted to identifying alternative power sources. Among these, the use of WPT combined with rechargeable batteries is extremely appealing. The solution proposed in [62] consists of an inductive energy link operating in the industrial–scientific–medical (ISM) band centered at 434 MHz. The inductive coupling is implemented by means of two segmented planar resonators loaded by a lumped capacitor [see Fig. 26(a)]. With reference to the application of the link for energizing pacemakers, the resonators were optimized by means of full-wave simulations, using the configuration illustrated in Fig. 26(b). In more detail, the link consists of an external resonator (primary resonator) and an implanted resonator (secondary resonator) optimized to work on a layer of muscle and below a 2-mm layer of skin and a 3-mm layer of fat [see Fig. 26(b)]. For the specific



Fig. 24. Harvester of type B: (a) experimental setup used to estimate the RF spectrum received by the harvester when coupled with a 30-W CFL; and (b) spectrum measured with the HP E4411B spectrum analyzer.

experimental setup, the resonators were designed to be terminated by 50- impedances (i.e., Zc1 ¼ Zc2 ¼ 50 ). Experimental tests were performed by using a minced of pork leg; in fact, in the frequency range of interest, pork leg has EM parameters similar to the ones of human skin and muscle [62]. Experimental data were taken by using

the experimental setup illustrated in Fig. 27 with d1 ¼ d2 ¼ 0.5 cm; the corresponding results are given in Fig. 28(a). The measured S21 parameter has a maximum of 12.9 dB at 443 MHz. As for the power that the secondary resonator is able to deliver ðPRX Þ, with reference to a 50- load, it can be calculated by scattering parameters according to PRX ¼ PTX jS21 j2 1 jS11 j2 1 jS22 j2

Fig. 25. Measured dc output power generated by spurious RF emissions of a 30-W CFL. (a) Data obtained by using the harvester of Fig. 23(a) (the harvester of type A), the harvester of Fig. 23(c) (the harvester of type B), and three harvesters of type A in a lampshade configuration, as illustrated in Fig. 23(b). (b) Experimental data obtained for the RF–dc conversion efficiency.

(19)

Fig. 26. (a) Geometry of the segmented spiral loop resonator adopted in [62] to implement the inductive link. (b) Configuration adopted in full-wave simulations for the optimization of the inductive link: the secondary resonator is inside a three-layer medium consisting of a first layer of skin, a second layer of fat, and a third layer of musle. (c) EM parameters assumed in full-wave simulations for human tissues (data taken from [63]–[67]).


1705


Fig. 27. Setup adopted for the characterization of the inductive link proposed in [62]. As illustrated in the inset, the secondary resonator was placed inside a minced of pork leg at a depth d1 , while the primary resonator was at a distance ðd1 þ d2 Þ.

where PTX is the power delivered to the primary resonator, while Sij are the scattering parameters measured using a 50- normalization impedance. Results obtained for different values of d2 are given in Fig. 28(b). When the resonators are at a distance of 1 cm (d1 ¼ d2 ¼ 0.5 cm) and assuming 1 W of power supplied by the transmitter to the primary resonator, the secondary resonator is able to deliver a power of about 51 mW. The corresponding radio link efficiency ðRFRF Þ calculated according to (6) is given in Fig. 28(c); a maximum of about 5% has been obtained at a distance of 1 cm.

IV. CONCLUSION In this contribution, we have presented a complete procedure for the design of wireless power transfer system, based on rigorous circuit level description of each part. Starting from a general representation of a WPT link, we have discussed various design perspectives that should be pursued, based on the application scenarios. Indeed, if intentional wireless power transfer is to be realized, not only the source and the receiver are under the designer control but also their reciprocal position, and thus the wireless link can be accurately characterized and accounted for. On the contrary, when harvesting from ambient sources is exploited, the receiver only can be optimized and the wireless link conditions could be significantly variable. In FF–WPT, this means that different antenna characteristics 1706

Fig. 28. (a) S-parameters measured by using the experimental setup illustrated in Fig. 27 with d1 ¼ d2 ¼ 0.5 cm. (b) Power delivered to a 50- load by the secondary resonator; data calculated for different values of d2 by using (19) and the measured data of the scattering parameters. (c) Measured radio link efficiency in the band of interest.

are required: narrowband, directive antennas are more suitable in the first case, while multiband, omnidirectional, and circularly polarized ones are required in the second case. Similarly, when NF–WPT is adopted, design goals should take into account weather the transmitter and receiver distance/alignment is known a priori, in order to ensure the best power transfer efficiency in any possible configurations. For this purpose, the conjugate image impedances at the WPT link ports need to be ensured. By adopting the described general purpose design framework, combining full-wave simulation and nonlinear analysis, we have demonstrated that it is possible to take into account all these constraints. We have presented a number of significant design examples, and their associated prototypes, demonstrating the accuracy of the entire approach. These systems have been considered in realistic operating conditions, for NF–WPT and FF–WPT, to demonstrate their feasibility in actual application scenarios.



We believe that an accurate design of such systems, for the exploitation of EM power sources, allows zero-power (battery-less) systems to operate both in distributed environments and for implantable purposes.

APPENDIX The conventional approach in microwave circuit theory is to use the so-called traveling waves (TWs), which provide a description of forward and backward voltage and current waves referred to a transmission line with real characteristic impedance Zc [29], [68]. To deal with complex reference impedances, next we introduce the pseudowaves [30] from which the TW can be obtained as a special case when considering real reference impedance.

A. Pseudowaves By denoting with V0þ and V0 the amplitudes of the forward and reflected traveling waves, respectively, the voltage and the current along a transmission line with propagation constant and characteristic impedance Zc can be written as VðzÞ ¼ V0þ e z þ V0 eþ z IðzÞ ¼

V0þ z V0 þ z e e : Zc Zc

(20) (21)

Putting I 0 ¼ V0 =Zc , the total voltage and current at z ¼ 0 are

Fig. 29. Thevenin equivalent circuit of a generator VR , generator impedance ZR , and a load impedance ZL .

The time-average power can be computed as usual as (here and in the following, the asterisk denotes the complex conjugate) 1 P ¼ ReðV I Þ 2

1 Z2cr 2 1 2jImða bÞ 2 ¼ jaj jbj þ Zcr Re : (28) 2 Z2cr þ Z2ci 2 Zcr

The above expression shows that, for complex Zc , the total power is not equal to the difference between the powers carried by the incident and reflected waves: this property only holds for real impedances, i.e., for TW. As noted in [37] for real Zc , (28) also satisfies superposition of power: a principle that is generally not valid in electrical engineering. Let us now consider the case of a terminating impedance ZL , as shown in Fig. 29. By considering V ¼ ZL I, the reflection coefficient for TW (24) and (25) is given by

V ¼ V0þ þ V0

(22)

I ¼ Iþ 0 I0 :

(23)

b ZL ZR G ¼ ¼ : a ZL þ ZR

V þ Zc I a ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 ReðZc Þ

(24)

For complex ZL ; ZR , we know that the MPT is achieved when ZL ¼ ZR . But in this case, assuming ZR ¼ RR þ jXR , we have

V Zc I b ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi : 2 ReðZc Þ

(25)

(29)

The normalized wave amplitudes are defined as

The inverse relations provide voltages and currents in terms of traveling waves a; b pffiffiffiffiffiffiffiffiffiffiffiffiffiffi V ¼ ða þ bÞ ReðZc Þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi ReðZc Þ I ¼ ða bÞ : Zc

(26) (27)

Z ZR XR G MPT ¼ R ¼ j : ZR þ ZR RR

(30)

Thus, when the circuit is matched ðZL ¼ ZR Þ, we do not have MPT, while if we realize MPT, then G MPT 6¼ 0. In addition, from (30), the amplitude of the reflection coefficient can be greater than one. In conclusion, for complex loads, the conventional definition (29) of the reflection coefficient is unsuited to represent the power transfer from the source to the load.


1707


B. Power Waves As a possible remedy to the unsatisfying power management associated with the traveling wave definition of scattering parameters, the concept of power waves (PWs) was introduced by various authors [32]–[34], [36], [69]. PWs differ from TWs only by the complex conjugate of the reference impedance ZR ¼ RR þ jXR [see (24) and (25)]

We now have for the power 1 1 1 Ref2jImða bZR Þg P ¼ ReðV I Þ ¼ jaj2 jbj2 þ 2 2 2RR 1 ¼ jaj2 jbj2 : (35) 2 Using (21) and (22), the PW reflection coefficient is

V þ ZR I a ¼ pffiffiffiffiffi 2 RR

(31)

V Z I b ¼ pffiffiffiffiffiR : 2 RR

(32)

Conversely, voltages and currents are obtained from the PW as

V¼

ZR a þ ZR b pffiffiffiffiffi RR

(33)

ab I ¼ pffiffiffiffiffi : RR

(34)

REFERENCES [1] K. W. Del Lui, O. H. Murphy, and C. Toumazou, ‘‘32-W wirelessly-powered sensor platform with a 2-m range,’’ IEEE Sensors J., vol. 12, no. 6, pp. 1919–1924, Jun. 2012. [2] V. Leonov, T. Torfs, C. Van Hoof, and R. J. M. Vullers, ‘‘Smart wireless sensors integrated in clothing: An electrocardiography system in a shirt powered using human body heat,’’ Sensors Transducers J., vol. 107, no. 8, pp. 165–176, Aug. 2009. [3] M. Pinuela, P. D. Mitcheson, and S. Lucyszyn, ‘‘Ambient RF energy harvesting in urban and semi-urban environments,’’ IEEE Trans. Microw. Theory Tech., vol. 61, no. 7, pp. 2715–2726, Jul. 2013. [4] D. Masotti, A. Costanzo, P. Francia, M. Filippi, and A. Romani, ‘‘A load-modulated rectifier for RF micropower harvesting with start-up strategies,’’ IEEE Trans. Microw. Theory Tech., vol. 62, no. 4, pp. 994–1004, Apr. 2014. [5] D. Masotti, A. Costanzo, M. Del Prete, and V. Rizzoli, ‘‘A genetic-based design of a tetra-band high-efficiency RF energy harvesting system,’’ IET Microw. Antennas Propag., vol. 7, no. 15, pp. 1254–1263, 2013. [6] N. Borges Carvalho et al., ‘‘Wireless power transmission: R&D activities within europe,’’ IEEE Trans. Microw. Theory Tech., vol. 62, no. 4, pp. 1031–1045, Apr. 2014. [7] H. Sun, Y.-X. Guo, M. He, and Z. Zhong, ‘‘A dual-band rectenna using broadband yagi antenna array for ambient RF power harvesting,’’ IEEE Antennas Wireless Propag. Lett., vol. 12, pp. 918–921, 2013. [8] J. A. Hagerty, F. B. Helmbrecht, W. H. McCalpin, R. Zane, and Z. B. Popovic, ‘‘Recycling ambient microwave energy

1708

G¼

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

b ZL ZR ¼ : a ZL þ ZR

(36)

Equation (36) corresponds to a result obtained earlier by Roberts [31]. As noted in [38], however, also PWs are not fully satisfactory. As an example, the reflection coefficient of a short circuit ðZL ¼ 0Þ is usually different from 1. In addition, all measurements at microwave frequency are always done on real reference impedances. A virtue of the approach presented in this paper, i.e., the use of conjugate impedances, is to allow to use both approaches, keeping in mind their respective limits of validity. h

with broad-band rectenna arrays,’’ IEEE Trans. Microw. Theory Tech., vol. 52, no. 3, pp. 1014–1024, Mar. 2004. V. Rizzoli, A. Costanzo, D. Masotti, P. Spadoni, and A. Neri, ‘‘Prediction of the end-to-end performance of a microwave/RF link by means of nonlinear/electromagnetic co-simulation,’’ IEEE Trans. Microw. Theory Tech., vol. 54, no. 12, pp. 4149–4160, Dec. 2006. V. Rizzoli et al., ‘‘Integration of non-linear, radiation, and propagation CAD techniques for MIMO link design,’’ Int. J. Microw. Wireless Technol., vol. 4, no. 2, pp. 223–232, Apr. 2012. Z. Popovic, E. A. Falkenstein, D. Costinett, and R. Zane, ‘‘Low-power far-field wireless powering for wireless sensors,’’ Proc. IEEE, vol. 101, no. 6, pp. 1397–1409, Jun. 2013. H. J. Visser and R. J. M. Vullers, ‘‘Wireless sensors remotely powered by RF energy,’’ in Proc. 6th Eur. Conf. Antennas Propag., 2012, DOI: 10.1109/EuCAP.2012.6206234. National Italian Project, ‘‘Green tags and sensors with ultrawideband identification and localization (GRETA),’’ Feb. 2012–Feb. 2015. A. Kurs et al., ‘‘Wireless power transfer via strongly coupled magnetic resonances,’’ Science, vol. 317, no. 5834, pp. 83–86, 2007. Z. N. Low, R. A. Chinga, R. Tseng, and J. Lin, ‘‘Design and test of a high-power high-efficiency loosely coupled planar wireless power transfer system,’’ IEEE Trans. Ind. Electron., vol. 56, no. 5, pp. 1801–1812, Oct. 2009. B. L. Cannon, J. F. Hoburg, D. D. Stancil, and S. C. Goldstein, ‘‘Magnetic resonant coupling as a potential means for wireless power transfer to multiple small receivers,’’ IEEE


[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

Trans. Power Electron., vol. 24, no. 7, pp. 1819–1825, Oct. 2009. A. P. Sample, D. A. Meyer, and J. R. Smith, ‘‘Analysis, experimental results, and range adaptation of magnetically coupled resonators for wireless power transfer,’’ IEEE Trans. Ind. Electron., vol. 58, no. 2, pp. 544–554, Feb. 2011. M. Dionigi, A. Costanzo, F. Mastri, and M. Mongiardo, ‘‘Magnetic resonant wireless power transfer,’’ in Wireless Power Transfer. Aalborg, Denmark: River Publishers, 2012, p. 159. M. Dionigi and M. Mongiardo, ‘‘CAD of wireless resonant energy links (WREL) realized by coils,’’ in IEEE MTT-S Int. Microw. Symp. Dig., 2010, pp. 1760–1763. M. Dionigi and M. Mongiardo, ‘‘CAD of efficient wireless power transmission systems,’’ in IEEE Microw. Symp. Dig., 2011, DOI: 10.1109/MWSYM.2011.5972606. M. Dionigi, S. Koziel, and M. Mongiardo, ‘‘Full-wave computer-aided optimization of wireless power transfer systems,’’ in IEEE MTT-S Int. Microw. Symp. Dig., 2014, DOI: 10.1109/MWSYM.2014.6848314. H. Shoki, ‘‘Issues and initiatives for practical deployment of wireless power transfer technologies in Japan,’’ Proc. IEEE, vol. 101, no. 6, pp. 1312–1320, Jun. 2013. J. A. Russer, M. Dionigi, M. Mongiardo, and P. Russer, ‘‘A moving field inductive power transfer system for electric vehicles,’’ in Proc. Eur. Microw. Conf., Oct. 6–10, 2013, pp. 519–522. Y. Suzuki, T. Sugiura, N. Sakai, M. Hanazawa, and T. Ohira, ‘‘Dielectric coupling from electrified roadway to steel-belt tires characterized for miniature model car running demonstration,’’ in Proc. Microw. Workshop Ser. Innovative Wireless Power


[25]

[26]

[27]

[28]

[29]

[30]

[31] [32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

Transm., Tech. Syst. Appl., May 10–11, 2012, pp. 35–38. Q. Chen, K. Ozawa, Q. Yuan, and K. Sawaya, ‘‘Antenna characterization for wireless power-transmission system using near-field coupling,’’ IEEE Antennas Propag. Mag., vol. 54, no. 4, pp. 108–116, Aug. 2012. N. Marcuvitz, ‘‘On the representation and measurement of waveguide discontinuities,’’ Proc. IRE, vol. 36, no. 6, pp. 728–735, 1948. H. J. Carlin, ‘‘The scattering matrix in network theory,’’ IRE Trans. Circuit Theory, vol. 3, no. 2, pp. 88–97, 1956. E. W. Matthews, ‘‘The use of scattering matrices in microwave circuits,’’ IRE Trans. Microw. Theory Tech., vol. 3, no. 3, pp. 21–26, 1955. R. Collins, Foundation for Microwave Engineering, 2nd ed. New York, NY, USA: Wiley, 2001, pp. 268–276. R. B. Marks and D. F. Williams, ‘‘A general waveguide circuit theory,’’ J. Res. Nat. Inst. Standards Technol., vol. 97, pp. 533–533, 1992. S. Roberts, ‘‘Conjugate-image impedances,’’ Proc. IRE, vol. 34, no. 4, pp. 198–204, 1946. K. Kurokawa, ‘‘Power waves and the scattering matrix,’’ IEEE Trans. Microw. Theory Tech., vol. MTT-13, no. 2, pp. 194–202, Mar. 1965. P. Penfield, ‘‘Noise in negative-resistance amplifiers,’’ IRE Trans. Circuit Theory, vol. 7, no. 2, pp. 166–170, 1960. J. Rahola, ‘‘Power waves and conjugate matching,’’ IEEE Trans. Circuits Syst. II, Exp. Briefs, vol. 55, no. 1, pp. 92–96, Jan. 2008. N. Inagaki, ‘‘Theory of image impedance matching for inductively coupled power transfer systems,’’ IEEE Trans. Microw. Theory Tech., vol. 62, no. 4, pt. 2, pp. 901–908, Apr. 2014. D. A. Frickey, ‘‘Conversions between S, Z, Y, H, ABCD, T parameters which are valid for complex source and load impedances,’’ IEEE Trans. Microw. Theory Tech., vol. 42, no. 2, pp. 205–211, Feb. 1994. S. Llorente-Romano, A. Garca-Lamperez, T. K. Sarkar, and M. Salazar-Palma, ‘‘An exposition on the choice of the proper S parameters in characterizing devices including transmission lines with complex reference impedances and a general methodology for computing them,’’ IEEE Antennas Propag. Mag., vol. 55, no. 4, pp. 94–112, Aug. 2013. D. Williams, ‘‘Traveling waves and power waves: Building a solid foundation for microwave circuit theory,’’ IEEE Microwave, vol. 14, no. 7, pp. 38–45, Nov./Dec. 2013. A. Costanzo, A. Romani, D. Masotti, N. Arbizzani, and V. Rizzoli, ‘‘RF/baseband co-design of switching receivers for multiband microwave energy harvesting,’’ J. Sensors Actuators A, Phys., vol. 179, no. 1, pp. 158–168, Mar. 2012. A. Georgiadis, G. Andia, and A. Collado, ‘‘Rectenna design and optimization using reciprocity theory and harmonic balance analysis for electromagnetic (EM) energy

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

harvesting,’’ IEEE Antennas Wireless Propag. Lett., vol. 9, pp. 444–446, 2010. V. Rizzoli, A. Costanzo, D. Masotti, F. Donzelli, and G. Bichicchi, ‘‘CAD of multi-resonator rectenna for micro-power generation,’’ in Proc. Eur. Microw. Conf., Oct. 2009, pp. 1684–1687. S. Koziel, S. Ogurtsov, J. W. Bandler, and Q. S. Cheng, ‘‘Reliable space-mapping optimization integrated with EM-based adjoint sensitivities,’’ IEEE Trans. Microw. Theory Tech., vol. 61, no. 10, pp. 3493–3502, Oct. 2013. J. Essel, D. Brenk, J. Heidrich, and R. Weigel, ‘‘A highly efficient UHF RFID frontend approach,’’ in Proc. IEEE MTT-S Int. Microw. Workshop Wireless Sens. Local Positioning RFID, Croatia, 2009, DOI: 10.1109/IMWS2. 2009.5307866. A. Dolgov, R. Zane, and Z. Popovic, ‘‘Power management system for online low power RF energy harvesting optimization,’’ IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 7, no. 7, pp. 1802–1811, Jul. 2010. M. Dini et al., ‘‘A fully-autonomous integrated RF energy harvesting system for wearable applications,’’ in Proc. Eur. Microw. Conf., Oct. 6–11, 2013, pp. 987–990. V. Rizzoli, A. Costanzo, F. Mastri, and A. Neri, ‘‘A general SPICE model for arbitrary linear dispersive multiport components described by frequency-domain data,’’ in IEEE MTT-S Int. Microw. Symp. Dig., Jun. 2003, vol. 1, pp. 9–12. D. Masotti, A. Costanzo, and S. Adami, ‘‘Design and realization of a wearable multi-frequency RF energy harvesting system,’’ in Proc. 5th Eur. Conf. Antennas Propag., 2011, pp. 517–520. 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. ADFORS Saint-Gobain. [Online]. Available: http://www.sg-adfors.com/Technologies/ Fabrics/Glassmat L. Catarinucci, R. Colella, and L. Tarricone, ‘‘Smart prototyping techniques for UHF RFID tags: Electromagnetic characterization and comparison with traditional approaches,’’ Progr. Electromagn. Res., vol. 132, pp. 91–111, 2012. G. Monti and F. Congedo, ‘‘UHF rectenna using a bowtie antenna,’’ Progr. Electromagn. Res. C, vol. 26, pp. 181–192, 2012. G. Monti, F. Congedo, D. De Donno, and L. Tarricone, ‘‘Monopole-based rectenna for microwave energy harvesting of UHF RFID systems,’’ Progr. Electromagn. Res. C, vol. 31, pp. 109–121, 2012. U. Olgun, C.-C. Chen, and J. L. Volakis, ‘‘Investigation of rectenna array configurations for enhanced RF power harvesting,’’ IEEE Antennas Wireless Propag. Lett., vol. 10, pp. 262–265, 2011. K. Niotaki et al., ‘‘A compact dual-band rectenna using slot-loaded dual band folded dipole antenna,’’ IEEE Antennas Wireless Propag. Lett., vol. 12, pp. 1634–1637, 2013. G. Monti, F. Congedo, P. Arcuti, and L. Tarricone, ‘‘Resonant energy scavenger

[56]

[57]

[58]

[59]

[60]

[61]

[62]

[63]

[64]

[65]

[66]

[67]

[68] [69]

for sensor powering by spurious emissions from compact fluorescent lamps,’’ IEEE Sensors J., vol. 14, no. 7, pp. 2347–2354, Jul. 2014, DOI: 10.1109/JSEN.2014.2310235. G. Monti, P. Arcuti, F. Congedo, and L. Tarricone, ‘‘Power generation by spurious emissions from compact fluorescent lamps,’’ presented at the Eur. Microw. Week, Rome, Italy, Oct. 2014. T. Letertre, F. Gaudaire, A. Azoulay, A. Destrez, and C. Martinsons, ‘‘Characterization of compact fluorescent lights RF emissions in the perspective of human exposure,’’ in Proc. Int. Symp. Electromagn. Compat., Kyoto, Japan, 2009, pp. 473–476. G. Schmidt and I. Berta, ‘‘Radiated radiofrequency emission from the plasma of compact fluorescent lamps,’’ Int. J. Plasma Environ. Sci. Technol., vol. 5, no. 1, pp. 84–92, 2011. V. Sekar, T. G. Palanivelu, and B. Revathi, ‘‘Effective tests and measurements mechanisms for EMI level identification in fluorescent lamp operation,’’ Eur. J. Sci. Res., vol. 34, no. 4, pp. 495–505, Aug. 2009. W. G. Fano, ‘‘RF emissions of compact fluorescent lights,’’ Interference Technology, Dec. 2012. E. Coca, V. Popa, and G. Buta, ‘‘Compact fluorescent lamps electromagnetic compatibility measurements and performance evaluation,’’ in Proc. IEEE Int. Conf. Comput. Tool, Lisbon, Portugal, 2011, DOI: 10.1109/ EUROCON.2011.5929269. G. Monti, L. Tarricone, and C. Trane, ‘‘Experimental characterization of a 434 MHz wireless energy link for medical applications,’’ Progr. Electromagn. Res. C, vol. 30, pp. 53–64, May 2012. F. J. Huang et al., ‘‘Rectenna application of miniaturized implantable antenna design for triple-band biotelemetry communication,’’ IEEE Trans. Antennas Propag., vol. 59, no. 7, pp. 2646–2653, Jul. 2011. S. Gabriel, R. W. Lau, and C. Gabriel, ‘‘The dielectric properties of biological tissues: III. Parametric models for the dielectric spectrum of tissues,’’ Phys. Med. Biol., vol. 41, pp. 2271–2293, 1996. C. Gabriel, S. Gabriel, and E. Corthout, ‘‘The dielectric properties of biological tissues: I. Literature survey,’’ Phys. Med. Biol., vol. 41, pp. 2231–2249, 1996. S. Gabriel, R. W. Lau, and C. Gabriel, ‘‘The dielectric properties of biological tissues: II. Measurements in the frequency range 10 Hz to 20 GHz,’’ Phys. Med. Biol., vol. 41, pp. 2251–2269, 1996. S. Gabriel, R. W. Lau, and C. Gabriel, ‘‘The dielectric properties of biological tissues: III. Parametric models for the dielectric spectrum of tissues,’’ Phys. Med. Biol., vol. 41, pp. 2271–2293, 1996. D. M. Pozar, Microwave Engineering, 4th ed. New York, NY, USA: Wiley, 2012. D. C. Youla, ‘‘On scattering matrices normalized to complex port numbers,’’ Proc. IRE, vol. 49, no. 7, p. 1221, 1961.


1709


ABOUT THE AUTHORS Alessandra Costanzo (Senior Member, IEEE) received a Doctor degree in electrical engineering (magna cum laude) from the University of Bologna, Bologna, Italy. Since 2001, she has been an Associate Professor of EM fields with the University of Bologna. She has coauthored over 130 scientific publications in peer-reviewed international journals and conferences and three chapter books. She holds three international patents. Her research activities have focused on several topics such as electrothermal characterization and modeling of radio-frequency (RF)/microwave nonlinear devices; simulation and design of active microwave integrated circuits (ICs); broadband design of self-oscillating circuits and systems for electrical, stability, and noise performance. More recently, she has been involved with an innovative software platform for the nonlinear and EM cosimulation of RF systems excited by modulated sources. She has demonstrated circuitlevel analysis of entire multiple-input–multiple-output (MIMO) and ultrawideband (UWB) links, including realistic channel models. She is currently involved in the development of energy autonomous sensors based on wearable systems and energy harvesting technologies and in the development of WPT systems. Her research and industrial collaborations are plentiful, and have been conducted via significant funding sources from both institutional and private investors. Dr. Costanzo is a member of the Editorial Board of the International Journal of Microwave and Wireless Technologies. She is the Executive Editor of the Wireless Power Transfer Journal (Cambridge University Press). She is a member of the Management Committee of the recently approved European Union (EU) COST action ‘‘Wireless power transfer for sustainable electronics’’ (WiPE). She is a member of the Technical Program Committee (TPC) board of the IEEE Microwave Theory and Techniques Society (IEEE MTT-S) International Microwave Symposium (IMS) and European Microwave Week (EUMW), and the Steering Committee, 2014 EUMW, Rome, Italy. She is Vice Chair of the IEEE MTT-S TC-26 Wireless Energy Transfer and Conversion and a member of the IEEE MTT-S TC-24 RFID Technologies. She serves on the editorial board of many international journals, such as the IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, the IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF ICS AND SYSTEMS, and IEEE MICROWAVE AND GUIDED WAVE LETTERS.

Marco Dionigi (Member, IEEE) received the Ph.D. degree in electronic engineering from the University of Perugia, Perugia, Italy, in 1996. He is currently an Assistant Professor with the Department of Electronic and Information Engineering, University of Perugia. His current research interests are in the field of microwave and millimeter-wave waveguide component modeling and optimization, wireless power transfer systems and modeling, microwave sensor modeling and design, and microwave and ultrawideband system design.

Diego Masotti (Member, IEEE) received the Dr.Ing. degree in electronic engineering and the Ph.D. degree in electric engineering from the University of Bologna, Bologna, Italy, in 1990 and 1997, respectively. In 1998, he joined the University of Bologna as a Research Associate of Electromagnetic Fields. He has coauthored more than 120 scientific publications on peer-reviewed international journals and conferences. His research interests are in the areas of nonlinear microwave circuit simulation and design, and nonlinear/electromagnetic codesign of integrated subsystems/systems.

1710

Dr. Masotti has been a member of the Paper Review Board of the IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, IEEE COMMUNICATION LETTERS, IET Circuit Devices & Systems, IEEE MICROWAVE AND WIRELESS COMPONENTS LETTERS, since 2004, 2010, 2011, and 2013, respectively. He is also a member of the Technical Program Committee (TPC) board of the IEEE Microwave Theory and Techniques Society (IEEE MTT-S) European Microwave Week (EUMW). Mauro Mongiardo (Fellow, IEEE) received the Laurea degree (cum laude) in electronic engineering from the University of Rome ‘‘La Sapienza,’’ Rome, Italy, in 1983. In 1991, he became an Associate Professor of Electromagnetic Fields and, in 2001, a full Professor of Electromagnetic Fields at the University of Perugia. He is the author or coauthor of over 200 papers and articles in the fields of microwave components, microwave CAD, and antennas. He is the coauthor of the books Open Electromagnetic Waveguides (London, U.K.: IEE, 1997), and Electromagnetic Field Computation by Network Methods (New York, NY, USA: Springer-Verlag, 2009). Recently, he has coauthored a chapter in the book Wireless Power TransferVPrinciples and Engineering Explorations (Rijeka, Croatia: Intech, 2012), and a chapter in the book Wireless Power Transfer (Aalborg, Denmark: River Publishers, 2012). His scientific interests are concerned primarily with the numerical modeling of electromagnetic wave propagation both in closed and open structures. His research interests involve computer-aided design (CAD) and optimization of microwave components and antennas and, more recently, modeling of wireless power transfer systems. Prof. Mongiardo has been elected Fellow of the IEEE ‘‘for contributions to the modal analysis of complex electromagnetic structures’’ in 2011. He has been serving on the Technical Program Committee of the IEEE International Microwave Symposium since 1992. Since 1994, he has been a member of the Editorial Board of the IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES. During 2008–2010, he was an Associate Editor of the IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES. Giuseppina Monti received the Laurea degree in telecommunication engineering (with honors) from the University of Bologna, Bologna, Italy, in 2003 and the Ph.D. degree in information engineering from the University of Salento, Lecce, Italy, in 2007. She is currently with the Department of Innovation Engineering, University of Salento, as a Temporary Researcher and Lecturer in computeraided design (CAD) of microwave circuits and antennas. Her current research interest includes the analysis and applications of artificial media, the design and realization of microwave components, reconfigurable antennas and devices, wireless power transmission, and energy harvesting technologies. She has coauthored a book chapter and approximately 90 papers in international conferences and journals. Luciano Tarricone (Senior Member, IEEE) received the Laurea degree in electronic engineering (cum laude) and the Ph.D. degree from the Rome University ‘‘La Sapienza,’’ Rome, Italy, in 1989 and 1994, respectively. From 1990 to 1992, he was a Researcher with the IBM Rome Scientific Centers. From 1992 to 1994, he was with the IBM European Center for Scientific and Engineering Computing, Rome, Italy. Between 1994 and 1998, he was a Researcher with the University of Perugia, Perugia, Italy, and, between 1998 and 2001, he



was a ‘‘Professore Incaricato’’ of electromagnetic (EM) fields and EM compatibility. Since November 2001, he has been a Faculty Member with the Department of Innovation Engineering, University of Salento, where he is a Full Professor of EM fields and coordinates a research group of about 15 people. He has authored and coauthored approximately 300 scientific papers. His main contributions are in the modeling of microscopic interactions of EM fields and biosystems, and in numerical methods for efficient computer-aided design (CAD) of microwave circuits and antennas. He is currently involved in bioelectromagnetics, electromagnetic energy harvesting and wireless power transmission, novel CAD tools and procedures for microwave circuits, RFID, and EM highperformance computing.

Roberto Sorrentino (Life Fellow, IEEE) received the Laurea degree in electronic engineering from the La Sapienza University of Rome, Rome, Italy, in 1971. In 1974, he became an Assistant Professor of Microwaves at the La Sapienza University of Rome. He was an Adjunct Professor at the University of Catania, the University of Ancona, Ancona, Italy, and the La Sapienza University of Rome (1977–1982), where he then was an Associate Professor from 1982 to 1986. In 1983 and 1986, he was appointed a Research Fellow at the University of Texas at Austin, Austin, TX, USA. From 1986 to 1990, he was a Professor at the University of Rome ‘‘Tor Vergata,’’ Rome, Italy. Since November 1990, he has been a Professor at the University of Perugia, Perugia, Italy, where he was the Chairman of the Electronic Department, Director of the Computer Center (1990–1995), and Dean of the Faculty of Engineering (1995–2001). He is the author or coauthor of more than 150 technical papers in international journals and 200 refereed conference papers. In 2007, he founded RF Microtech, a spinoff company of the University of Perugia dealing with radiofrequency microelectromechanical systems (RF–MEMS), microwave systems, and antennas. He has edited a book by R. Sorrentino Numerical Methods for Passive Microwave Structures (New York, NY, USA: IEEE Press, 1989), and coauthored four books: Advanced Modal Analysis (New York, NY, USA: Wiley, 2000), RF and Microwave Engineering (New York, NY, USA: McGrawHill, 2006, in Italian), Electronic Filter Simulation and Design (New York, NY, USA: McGrawHill, 2007), and RF and Microwave Engineering (New York, NY, USA: Wiley, 2010). His research activities

have been concerned with various technical subjects, such as the electromagnetic wave propagation in anisotropic media, the interaction of electromagnetic fields with biological tissues, but mainly with numerical methods and computer-aided design (CAD) techniques for passive microwave structures and the analysis and design of microwave and millimeter-wave circuits, including filters and antennas. In recent years, he has been involved in modeling and design of radio-frequency microelectromechanical systems (RF–MEMS) and their applications on tunable and reconfigurable circuits and antennas. Prof. Sorrentino became a Fellow of the IEEE for ‘‘contribution to the modeling of planar and quasi-planar microwave and millimeter-wave circuits’’ in 1990. In 1993, he was the recipient of the IEEE Microwave Theory and Techniques Society (IEEE MTT-S) Meritorious Service Award and, in 2000, he was one of the recipients of the IEEE Third Millennium Medal. In 2004, he received the Distinguished Educator Award from IEEE MTT-S. In 2010, he received the Distinguished Service Award from the European Microwave Association. In 2012, he received, with S. Bastioli and C. Tomassoni, the Microwave prize for the paper ‘‘A new class of waveguide dual-mode filters using TM and nonresonating modes,’’ which appeared in the IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, vol. 58, no 12, pp. 3909–3917, Dec. 2010. He has been active within the IEEE MTT Society. From 1984 through 1987, he was the Chairman of the IEEE Section of Central and South Italy and was the founder of the local MTT/AP Chapter that he chaired from 1984 to 1987. From January 1995 through April 1998, he was the Editor-in-Chief of IEEE MICROWAVE AND GUIDED WAVE LETTERS. From 1998 to 2005, he served on the Administrative Committee of IEEE MTT-S. He was elected again in MTT AdCom for the term 2011–2013. He is also a member of Technical Committees MTT-15 on Field Theory and MTT-01 on Computer-Aided Design, which he chaired in 2003–2004. He served the International Union of Radio Science (URSI) as Vice Chair (1993–1996), and then Chair (1996–1999) of the Commission D (Electronics and Photonics). Since 2007, he has been the President of the Italian Commission of URSI. In 2002, he was among the founders and first President of the Italian Electromagnetic Society (SIEm) that he chaired until 2008. From 1998 to 2005, he was a member of the High Technical Council of the Italian Ministry of Communications. In 1996, during the restructuring of the management of the European Microwave Conference, he was elected Chairman of the Steering Committee. In 1998, an entirely new structure, the European Microwave Week, was launched, soon underpinned by a new not-for-profit organization called the European Microwave Association (EuMA), of which he was one of the founders and President from its constitution until 2009.


1711