Integration of Slot Antennas in Commercial Photovoltaic Panels for ...

62

IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 61, NO. 1, JANUARY 2013

Integration of Slot Antennas in Commercial Photovoltaic Panels for Stand-Alone Communication Systems Roberto Caso, Member, IEEE, Andrea D’Alessandro, Andrea Michel, and Paolo Nepa, Member, IEEE

Abstract—The integration of slot antennas in a class of commercial photovoltaic (PV) panels is addressed. The basic idea is to exploit the room available between adjacent solar cells, also taking advantage of the presence of the cover glass layer that gives a valuable miniaturization effect. As test cases, two antenna designs are presented for stand-alone communication systems operating in the GSM/UMTS (1710–2170 MHz) and WiMAX (3300–3800 MHz) frequency bands, respectively. At early design stage, antenna design has been performed by resorting to a simplified numerical model; afterwards, the effects on the antenna performance of nearby solar cells and DC bus wires have been numerically evaluated. The impedance tuning of the final antenna design has been carried out through a measurement campaign with slot antenna prototypes attached to real PV panels. Index Terms—Integrated antennas, photovoltaic panels, slot antennas, solar cells, stand-alone communication systems.

I. INTRODUCTION

I

N autonomous communication systems, photovoltaic (PV) panels are usually physically separated from the antenna, and this demands for a compromise in the utilization of the available space. Moreover, in several applications, as monitoring, vehicular communications and satellite systems, using two distinct elements, namely the PV panel and the antenna, may be anti-aesthetic and expensive, and may cause a number of mechanical issues. For above reasons, antenna integration in PV panels is attractive. For example, designing antennas that can be easily integrated into a commercial PV panel can result in a low-cost implementation of a low data rate wireless link for the remote control and monitoring of each single PV panel in large solar fields. Transparent materials have been proposed to implement innovative transparent antennas that can be easily mounted directly on the solar cells. Transparent antennas have been optimized for different applications, at 3.4–3.8 GHz [1] and 2.5 GHz [2] frequency bands. Nevertheless, above transparent materials Manuscript received February 10, 2012; revised July 24, 2012; accepted August 20, 2012. Date of publication September 21, 2012; date of current version December 28, 2012. R. Caso is with the Department of Information Engineering, University of Pisa, Pisa, Italy (e-mail: [email protected]). A. D’Alessandro, A. Michel, and P. Nepa are with the Department of Information Engineering, University of Pisa, Pisa, Italy (e-mail: [email protected]; [email protected]; [email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TAP.2012.2220111

are still relatively expensive. Besides, meshed patch antennas printed on the top of PV cells are considered as a cost-friendly solution. An example of such antennas, optimized for a 2.52 GHz small-satellite application, is described in [3]. However, a 90% sunlight transparency needed for a proper functioning of the solar cell cannot be easily achieved [3]. Solar cells have been employed themselves as either a radiating patch [4] or a parasitic patch [5]–[7], for GPS vehicular applications, GSM frequency bands or 3.76 GHz satellite communications. Since the required operating frequency settles on the patch size, such arrangements cannot be applied to PV panels made of standardized commercial cells. Solar cells may also be used as a ground plane for an upper patch element [8]–[10], but with a reduction of the cell solar efficiency. Slot antennas placed between solar cells have been optimized for ultra-wideband (UWB, 3.1–10.6 GHz) applications [11], and at 2.4 GHz by requiring modifications of the DC bus wires path [12]. Also, slot antennas have been realized by etching them into the solar cell, but a reduction of the cell solar efficiency must be accepted [13], [14]. In this paper, two configurations of low-cost linearly-polarized slot antennas suitable for integration into a class of commercial large PV panels are presented. As test cases, two antennas operating at the GSM/UMTS and WiMAX frequency bands, respectively, have been designed to show the achievable performance in terms of compactness and percentage impedance bandwidth (Section II). A considerable reduction of the antenna physical size has been obtained by exploiting the presence of the cover glass layer. As a first step, antenna design has been performed by using a simplified model of the PV panel. Later on the effects on antenna performance of both the cells close to the slot and the DC bus wires have been numerically quantified. Finally, measurements on antenna prototypes attached to real PV panels have been used for a fine tuning of the antenna input impedance (Section III). The followed design procedure avoids using detailed numerical models that would result in complex models but not yet able to account for the several propagation phenomena involved in the periodic/multilayer structure of real PV panels. Concluding remarks are drawn in Section IV.

II. SLOT ANTENNA DESIGN The proposed antenna integration concept is based on the exploitation of the space available between adjacent solar cells of

0018-926X/$31.00 © 2012 IEEE

CASO et al.: INTEGRATION OF SLOT ANTENNAS IN COMMERCIAL PHOTOVOLTAIC PANELS FOR STAND-ALONE COMMUNICATION SYSTEMS

Fig. 1. Two typical arrangements for PV cells in large panels: (a) square cells separated by a distance D; (b) octagonal cells close to each other, with an uncovered square-shape space whose side length is denoted by J. Possible locations for slot antennas are also shown (with dark color), which are such that the slot aperture is not crossed by the DC bus wires.

some large PV panels, so without decreasing the panel solar efficiency. In particular, two cell arrangements often used in large PV panels are considered, and they are shown in Fig. 1. For such configurations, a linear or square slot antenna etched on a low-cost substrate can be just attached to the back side of the panel and positioned in such a way that the slot aperture is not obstructed by the cells (which basically behave like shielding conductive surfaces). Also, the DC bus wires must not cross the slot aperture, as it would result in an antenna shorting effect. For example, in the cell arrangement shown in Fig. 1(a), gaps parallel to the -axis are not functional because of the presence of the DC bus wires; on the other hand, more room is available along the -axis (it being only limited by the panel frame size), so allowing the allocation of either a few long linear slots (resonating at low frequencies) or high-directivity linear arrays made of several short linear slots (higher resonance frequencies). Electronic beam scanning can also be implemented through a proper feeding network. To better understand how to exploit the linear or square gaps shown in Fig. 1, the stack-up of two typical PV panels are shown in Fig. 2. Silicon or GaAs (Gallium Arsenide) solar cells are usually incorporated between two ethylene vinyl acetate (EVA) layers. Also, two cover glass layers are placed at the top and bottom sides of the PV panel. In some commercial panels, the bottom cover glass layer is replaced by a plastic back sheet, as a Tedlar film (Fig. 2(b)). Therefore, a slot antenna positioned according to the above mentioned criteria, would radiate in presence of cover glass and EVA layers. Both layers are low-loss dielectrics and their effect can be accounted for during the design process. EVA layer is relatively thin (usually less than 0.1 mm) and its effect can be neglected up to some Gigahertz. The glass layer is thicker, from a few millimeters up to 10 mm (when both top and bottom cover glass layers are present, see Fig. 2(a)). Notably, the glass layer of the PV panel that is going to cover the slot helps in reducing slot resonance frequency [15], so giving rise to an extremely advantageous antenna miniaturization effect (without requiring any antenna meandering or distributed reactive loading, or any other complex antenna layout). As test cases for the proposed design approach, a linear slot antenna and a square slot antenna have been designed and prototyped, to fit in commercial PV panel topologies like those in

63

Fig. 2. Stack-up of two typical commercial photovoltaic panels: (a) a glasscells-glass PV panel, with square solar cells and (b) a glass-cells-Tedlar® PV panel with octagonal solar cells. In both PV panel configurations, the solar cells matrices are encapsulated between two EVA layers.

Fig. 3. Layout of the proposed slot antennas suitable to be integrated in a PV panel: (a) three-stepped slot antenna fed by a CPW line, TSSA; (b) square slot antenna fed by a microstrip line, SSA.

Fig. 2(a) and (b), respectively. Fig. 3 shows the top view of the two proposed antennas: a three-stepped slot antenna (TSSA) [16] and a square slot antenna (SSA) [17]. Both slots are realized on a 1.6 mm thick FR4 substrate ; a CPW and a microstrip are used to feed the TSSA and SSA, respectively (both exhibit a 50- characteristic impedance). The three-stepped slot configuration has been chosen to check the best performance that can be achieved in terms of percentage impedance bandwidth, in the framework of slot antennas whose maximum width is limited by the presence of nearby PV cells separated by a given distance . Specifically, it is shown that the antenna can operate in frequency bands that allocate both GSM (1710–1910 MHz) and UMTS (1920–2170 MHz) applications (around 24% percentage bandwidth). The three-stepped slot is etched on the same side of the CPW feeding line. The main parameters related to the resonant frequency and the impedance bandwidth of the TSSA are the slot length and width ,

64


Fig. 4. Stack-up of the simplified numerical model: (a) TSSA and (b) SSA. The cover glass layer on the top side is the only layer of the PV cell that has been considered in the numerical model used for the slot antenna design.

TABLE I ANTENNA DIMENSIONS OPTIMIZED WITH COVER GLASS LAYER (mm)

Fig. 6. Stack-up of the (a) TSSA and (b) SSA, with a 250 mm-side square reflector plane.

it represents the lower frequency band that can be covered when the side of the square space available at the corner of octagonal cells is less than 30 mm (and a 5 mm thick cover glass layer is considered). The numerical simulations were performed using the commercial software CST MWS. Simulation results on the miniaturization effect due to the PV panel cover glass layer are discussed in Section II-A. The influence of a reflector behind the whole structure (needed if an unidirectional beam is required) is analyzed in Section II-B. Finally, the effects of the solar cells close to the antenna have been numerically checked, by adding square/octagonal metallic patches and wires, when all of them are placed in the middle of the glass layer (Section II-C). A. Miniaturization Effect Due to the Cover Glass Layer A simplified model of the PV cell multilayer structure has been adopted during the numerical design process (Fig. 4). Indeed, only a glass layer ( , , thickness equal to either 5 mm or 8 mm) has been considered. TSSA and SSA fit an overall volume of 200 70 1.6 and 60 60 1.6 , respectively. The TSSA slot is 92.4 mm long and 13 mm wide . The square slot exhibits an area of about 19.5 19.5 , with a stub long. The values of the geometrical parameters are shown in Table I. Fig. 5 shows the simulated reflection coefficient of the two antennas. To give evidence to the cover glass miniaturization effect, the reflection coefficient for an identical slot antenna without the glass layer is shown in the same figure. It has been estimated that the presence of the cover glass layer leads to a fairly important miniaturization level: 36% for the TSSA and 25% for the SSA. TSSA and SSA exhibit an antenna gain of around 6 dBi and 5 dBi, respectively. B. Adding a Metallic Reflector to Get an Unidirectional Radiation Pattern

Fig. 5. Simulated reflection coefficient for the slot antennas: (a) TSSA and (b) SSA. A reference curve of the reflection coefficient when the cover glass layer is removed is also shown.

respectively. An effective impedance fine tuning on a wide frequency bandwidth can be achieved by adjusting the lengths , and . In the SSA, the microstrip feeding line is in the opposite side with respect to the square slot. As in other square slot antennas, the length of the open-circuit stub behind the slot is optimized for input impedance tuning. The SSA has been designed to operate in the WiMAX 3300–3800 MHz frequency band, as

An aluminum reflector placed at a distance from the bottom side of the FR4 substrate (Fig. 6) can be used to achieve unidirectional radiation and increase the antenna gain. The effect on the reflection coefficient of a 250 mm-side aluminum square reflector is shown in Fig. 7 for different distances, . The presence of the reflector results in a slight impedance mismatching in the SSA configuration, which can be compensated by slightly varying the main geometrical parameters of the slot antenna. For , antenna gain increases up to about 8 dBi and 7 dBi, for the TSSA and SSA respectively. A back lobe amplitude less than is achieved for both antennas when the reflector is added.


65

Fig. 8. Stack-up of the proposed slot antennas for (a) the TSSA and (b) the SSA. The cover glass layer on the top is the only layer of the PV cell that has been considered in the numerical model used for the slot design, while simple metallic patches have been used to model the PV cells.

Fig. 7. Simulated reflection coefficient of the slot antennas versus the distance between the FR4 bottom and a metallic reflector: (a) TSSA and (b) SSA. The reference curve shows the antenna reflection coefficient when the reflector is absent.

C. Analysis of the Effect of the PV Cells and DC Bus Wires Located Nearby the Slot Antenna The effect of the closest solar cells to the slot (two for the TSSA and four for the SSA) has been numerically checked by adding square or octagonal metallic patches in the middle of the glass layer (Fig. 8). Fig. 9(a) shows the TSSA reflection coefficient, when varying the distance between the two PV cells from 10 mm up to 50 mm (two nearby PV cells modeled by 156 156 square metallic patches have been considered, with the centers of the TSSA and the PV cells aligned along the -direction). A significant impedance mismatching is noted when the distance between the two solar cells is smaller than the TSSA slot width , as expected since the slot is partially covered by the PV cell. Numerical results for the SSA are shown in Fig. 9(b) (four nearby PV cells modeled by 125 125 octagonal metallic patches have been considered). It results that for a slot side of 19.5 mm, the distance between neighboring octagonal PV cells (Fig. 1(b)) should be enough large to allow that the side of the square space in Fig. 2(b), , is at least greater than 26 mm. Indeed, the presence of the solar cells determines only a minor impedance mismatching when . Moreover in Fig. 9(a) curve that represents the combination of the solar cells and reflector (250 mm-side) effects has been added for both configurations. As previously analyzed (Section II-B), the presence of the reflector results in a slight impedance mismatching in the SSA configuration. Finally, the effect of additional cells nearby the slot antenna has been numerically checked. A total of cells is con-

Fig. 9. Simulated reflection coefficient for the slot antennas, as a function of and in the size of the room available between adjacent cells (parameters Fig. 1). PV cells have been modeled through simple metallic patches: (a) TSSA and (b) SSA. As a reference, a curve has been added for the simpler model where only the glass layer is present (with no neighboring PV cells).

sidered, where and represent the number of cells along the -axis and -axis, respectively. In the TSSA configuration, the centers of the linear slot and the two nearest PV cells (156 156 size, at a distance ) are aligned along the -direction (see geometry in Fig. 10). In the SSA configuration, the distance between octagonal solar cells (125 125 size) has been set to 3 mm and . In the numerical simulations, the glass layer size is such that it covers the entire area occupied by the solar cell array. Fig. 10 shows the simulated reflection coefficient when increasing the number of PV cells nearby the antenna. For both antennas, the reflection coefficient is below in the frequency bands of interest, whatever the number of cells that are added to the numerical model. It results that return loss performance is not deteriorated by the presence of additional cells other than the nearest ones. In Fig. 11, the radiation patterns ( and principal planes, at the center frequency of GSM, UMTS and WiMAX bands) of

66


Fig. 10. Simulated reflection coefficient for the (a) TSSA and the (b) SSA, as a , with or without the DC bus wires. PV function of the cells number cells have been modeled trough simple metallic patches; and represent the number of cells along the -axis and -axis, respectively.

the TSSA and SSA are shown, when increasing the number of solar cells nearby the slot antenna. In the -plane ( plane), a quite broad beam is observed. To reduce the antenna beamwidth in the -plane, an array can be realized along the -axis, but it should noted that the distance between the elements is induced by the solar cell size. As far as the influence of the solar cells close to the slot is considered, it is apparent that the radiation patterns do not change significantly when the cell number is more than 3 2 and 2 2 for the TSSA and SSA, respectively. In Figs. 10 and 11, a curve has also been added to show the effect of DC bus wires connecting the PV cells aligned along the -axis direction. From the numerical results it appears that such a wire connection does not perturb the slot antenna performance. On the other hand, it is worth to underlying that above simulations just allow to study what happens when adjacent cells are electrically connected (through DC bus wires), so that they behave as longer conducting plates close to slot. Actually, in PV panels DC bus wires are electrically long wires, series or shunt connected, so that more reliable simulation results should be relevant to a model of an entire panel. The two antennas (both they are linearly polarized antennas) exhibit a cross-polar component levels below 18 dB. Above numerical analysis has been obtained with an Intel Core i7-3930 K processor (3.2 GHz, 6 Cores, 12 Threads, 12 MB cache); for results relevant to the maximum number of cells, also in presence of the DC bus wires, the simulation time was 5.5 hours (7.5 million of mesh cells) and 1 hour (2.1 million of mesh cells) for the TSSA and SSA, respectively. III. EXPERIMENTAL RESULTS Fig. 12 shows TSSA and SSA prototypes. Two different commercial PV panels corresponding to the cell arrangements in

Fig. 11. Normalized radiation patterns in the principal planes, when varying the number of cells close to the slot (see geometry in Fig. 10): (a) TSSA, at GSM and UMTS center frequencies; (b) SSA, at WiMAX center frequency.

Fig. 12. TSSA (a) and SSA (b) prototypes realized on a 1.6 mm-thick FR4 substrate.

Fig. 2 have been used to estimate antenna performance, when the slot antennas are attached to a real PV panel. A. Measurement Results for the TSSA Prototype A 140-W-BRP6336064-140 PV panel [18] was employed to check the TSSA reflection coefficient for different locations of the slot with respect to the adjacent solar cells (Fig. 13). The panel consists of 36 high-quality polycrystalline silicon square solar cells (156 156 ); the size of the aluminum panel frame is 1655 991 , and its thickness is 40 mm. The total thickness of the top and bottom cover glass layers is 8 mm, which guarantees a good light permeability and protection from atmospheric agents. In the standard BRP6336064-140 panel, the distance between the PV cells is set to 25 mm. A panel with four columns at different distances ( , 30 and 50 mm) was specifically manufactured, to measure the effect of the cell


67

TABLE II PV PANEL INTEGRATED PROTOTYPE

Fig. 13. The commercial PV panel (BRP6336064-140) used for testing the , 30 TSSA. The panel is made of four columns at different distance: and 50 mm. The distance between cells along the vertical direction is 25 mm. The first and the last columns are at 133 mm from the panel frame border.

Fig. 15. Measured reflection coefficient of the TSSA for different widths of the gap between adjacent cell columns.

Fig. 14. Simulated and measured reflection coefficients of the TSSA, with and without the cover glass layer.

proximity on the antenna reflection coefficient. The distance between cells along the vertical direction is 25 mm. Simulated and measured values of the TSSA reflection coefficient are compared in Fig. 14, for the isolated antenna (without the glass layer) and when the antenna prototype is attached to the PV panel. The antenna has been located at a position close to the panel border, so to include the effect of the cover glass layer while minimizing the effect of the nearby solar cells. The miniaturization effect of the 8 mm-thick glass layer is apparent. The cover glass layer determines a frequency shift of about 350 MHz. A shift of about 150 MHz between simulated and measured curves is present, which could be due to an approximation of the glass dielectric permittivity. On the other hand, a relatively easy tuning of the slot antenna prototype can be done to achieve a return loss greater than 10 dB in the whole band of interest. The geometrical parameters of the final prototype are shown in Table II. Reflection coefficient measurements are shown in Fig. 15, when the TSSA is attached to the back side of the PV panel and placed between two adjacent solar cell columns (the distance between columns can be chosen between 20, 30 and 50 mm, as shown in Fig. 13). The centers of the TSSA and the PV cells are aligned along the horizontal direction, as shown in the photo included in Fig. 15. A slight impedance mismatching can be

Fig. 16. Measured reflection coefficient of the TSSA for different widths of the gap between adjacent cell columns, when a metallic reflector is placed at a from the panel bottom. distance

observed when , which can be easily compensated by varying some of the antenna geometrical parameters, as for example the length (see Fig. 3). In Fig. 16, the measured reflection coefficient is shown when a 400 300 metallic reflector is placed at a distance from the PV panel bottom layer (needed if an unidirectional radiation pattern is required). In Fig. 17, the TSSA reflection coefficient is shown when the antenna is placed in three different positions ( , , and ) along the vertical gap between two adjacent columns. The slot is located in the gap wide, which corresponds to the worst case in terms of impedance detuning (Fig. 15). In Position A, the centers of the slot antenna and the solar cells are aligned along the -direction (Fig. 17(a)). Instead, in Position B the slot center corresponds to the symmetry point of a group made of four cells (Fig. 17(b)). The Position C is an intermediate position between A and B. Measured results in Fig. 17(b) show that the antenna input impedance changes when the slot is shifted along -axis. This effect must be considered when an

68


also been plotted, when the stub length is set to 12.3 mm (without or with a metallic reflector at a distance of 30 mm from the panel bottom). The presence of a reflector resulted in a slight impedance mismatching. IV. CONCLUSIONS

Fig. 17. Measured reflection coefficient of the TSSA by varying the relative position of the slot antenna along the -axis with respect to the nearby solar cells. Three different positions have been considered. Position C is a intermediate position between Position A and Position B.

Fig. 18. (a) The RSP180S-50M PV panel used for testing the SSA; (b) simulated and measured reflection coefficients of the proposed SSA attached to the back side of a PV panel, with or without the reflector (the reflector was at a disfrom the FR4 bottom). The squared space between cells is tance wide. The SSA has been mounted in the center of the PV panel. 30 30

arrangement of slots is used to implement a linear array along the -axis, since the input impedance of each array element will be a function of its relative location with respect to nearby cells. In case of a single antenna, moving the slot with respect to the nearby cells could be a possible approach to improve impedance tuning without changing slot geometrical parameters. B. Measurement Results for the SSA Prototype A 180-W-RSP180S-50M PV panel [19] was employed to check the SSA input impedance performance (Fig. 18(a)). The 1580 808 panel consists of 72 monocrystalline silicon octagonal solar cells (125 125 ); a 30 mm-thick aluminum panel frame is present. The top cover glass layer is 4 mm-thick, while a 1 mm-thick Tedlar film is present at the bottom side. The available space between each group made of four octagonal solar cells is a 30 mm-side square. Fig. 18(b) shows the measured reflection coefficient of the SSA when it is attached to the PV panel backside (just behind the square empty space in the center of the PV panel), with or without a metallic reflector. The aluminum reflector can be easily fixed to the metallic frame, at a distance of 30 mm from the FR4 bottom side. To improve impedance matching of the antenna prototype when it is attached to the panel, it has been necessary to increase the stub length up to 12.3 mm. As reference, the simulated reflection coefficient of the SSA in presence of the glass layer, DC bus wires and four solar cells has

Two slot antenna configurations have been optimized to fit the room available between adjacent solar cells of a class of commercial PV panels. It has been shown that a linear slot (three-stepped slot antenna) 13 mm wide and 92.4 mm long can achieve a return loss greater than 10 dB in a 24% percentage frequency bandwidth, like that required for GSM/UMTS applications. Moreover, a square slot with a side smaller than 20 mm can be used to implement an antenna operating in the 3300–3800 MHz WiMAX band. The parametric analysis required for the slot impedance matching has been performed by using a simplified model that only accounts for the presence of the cover glass layer, which is always present in commercial PV panels. It has been verified that the glass layer determines a significant resonant frequency shift toward lower frequencies, so allowing an effective and valuable antenna miniaturization. On the other hand, it has been verified that the presence of solar cells and DC bus wires close to the radiating slot only cause a minor antenna impedance detuning, which can be easily compensated with a fine variation of some antenna geometrical parameters. The effectiveness of the followed design procedure has been verified through measurements on antenna prototypes attached to commercial PV panels. As far as the radiation patterns are concerned, it resulted that only the closest solar cells to the radiating slot should be included in a numerical model. The proposed slot antennas represent low-cost and compact solutions which can be integrated into commercial PV panels with a relatively easy mechanical process. Future work will be focused on the performance analysis of 2.4 GHz slot antennas integrated in PV panels of a large solar field, mainly to analyze the effectiveness of the polarization and radiation patterns of the proposed slot antennas in such propagation scenarios. REFERENCES [1] M. J. R. Ons, S. V. Shynu, M. J. Ammann, S. J. McCormack, and B. Norton, “Transparent patch antenna on a-Si thin-film glass solar module,” Electron. Lett., vol. 47, no. 2, pp. 85–86, Jan. 2011. [2] T. Yasin and R. Baktur, “Inkjet printed patch antennas on transparent substrates,” presented at the IEEE Antennas and Propagation Society Int. Symp., Toronto, ON, Canada, Jul. 11–17, 2010. [3] T. W. Turpin and R. Baktur, “Meshed patch antennas integrated on solar cells,” IEEE Antennas Wireless Propag. Lett., vol. 8, pp. 693–696, 2009. [4] N. Henze, A. Giere, H. Friichting, and P. Hofmann, “GPS patch antenna with photovoltaic solar cells for vehicular applications,” presented at the 58th IEEE Vehicular Technology Conf., Orlando, Oct. 6–9, 2003. [5] C. Bendel, J. Kirchhof, and N. Henze, “Application of photovoltaic solar cells in planar antenna structures,” in Proc. 3rd World Conf. on Photovoltaic Energy Conversion, May 11–18, 2003, vol. 1, pp. 220–223. [6] N. Henze, M. Weitz, P. Hofmann, C. Bendel, J. Kirchhof, and H. Fruchting, “Investigation of planar antennas with photovoltaic solar cells for mobile communications,” in Proc. 15th IEEE Int. Symp. on Personal, Indoor and Mobile Radio Communications, Sep. 5–8, 2004, vol. 1, pp. 622–626. [7] S. Vaccaro et al., “Integrated solar panel antennas,” Electron. Lett., vol. 36, no. 5, pp. 390–391, 2000.


[8] M. J. R. Ons, S.-V. Shynu, M. J. Ammann, S. McCormack, and B. Norton, “Investigation on proximity-coupled microstrip integrated PV antenna,” presented at the 2nd Eur. Conf. on Antennas and Propagation, Nov. 11–16, 2007. [9] S. V. Shynu, M. J. Ammann, and B. Norton, “Quarter-wave metal plate solar antenna,” Electron. Lett., vol. 44, no. 9, pp. 570–571, Apr. 2008. [10] S.-V. Shynu, M. J. R. Ons, P. McEvoy, M. J. Ammann, S. J. McCormack, and B. Norton, “Integration of microstrip patch antenna with polycrystalline silicon solar cell,” IEEE Trans. Antennas Propag., vol. 57, no. 12, pp. 3969–3972, Dec. 2009. [11] M. Danesh and J. R. Long, “Compact Solar cell Ultra-Wideband dipole antenna,” presented at the IEEE Antennas and Propagation Society Int. Symp., Toronto, ON, Canada, Jul. 11–17, 2010. [12] T. Wu, R. L. Li, and M. M. Tentzeris, “A mechanically stable, low profile, omni-directional solar cell integrated antenna for outdoor wireless sensor nodes,” presented at the IEEE Antennas and Propagation Society Int. Symp., Charleston, SC, Jun. 1–5, 2009. [13] S. V. Shynu, M. J. R. Ons, M. J. Ammann, B. Norton, and S. McCormack, “Dual band a-Si:H solar-slot antenna for 2.4/5.2 GHz WLAN applications,” in Proc. 3rd Eur. Conf. on Antennas and Propagation, Mar. 23–27, 2009, pp. 408–410. [14] N. Henze, C. Bendel, and J. Kirchhof, “Photovoltaic power supply and antennas in one device for wireless telecommunication equipment,” presented at the INTELEC 2005, Berlin, Germany, Sep. 18–22, 2005. [15] K. L. Chung and A. S. Mohan, “Effect of superstrate thickness on the performance of broadband circularly polarised stacked patch antenna,” in Proc. IEEE Antennas and Propagation Society Int. Symp., Jun. 20–25, 2004, vol. 1, pp. 687–690. [16] W. Kueathawikun, P. Thumwarin, N. Anantrasirichai, and T. Wakabayashi, “Wide-band slot antenna for IEEE 802.11b/g,” presented at the SICE-ICASE Int. Joint Conf., Bexco, Busan, Korea, Oct. 18–21, 2006. [17] J. Y. Jan and J. W. Su, “Bandwidth enhancement of a printed wide-slot antenna with a rotated slot,” IEEE Trans. Antennas Propag., vol. 53, no. 6, pp. 2111–2114, Jun. 2005. [18] [Online]. Available: http://www.brandonisolare.com/ [19] [Online]. Available: http://www.risenenergy.com Roberto Caso (M’12) was born in Vipiteno, Bolzano, Italy, in 1980. He received the Master Degree in telecommunications engineering and the Ph.D. degree in applied electromagnetism in electrical and biomedical engineering, electronics, smart sensors, nano-technologies from the University of Pisa, Italy, in 2007 and 2012, respectively. In 2008, he collaborated with the Department of Information Engineering of the University of Pisa on the design and characterization of microstrip antenna for base stations and subscriber units of mobile communications systems. His post-doc research interests focus on analysis, design and optimization of wideband and multiband integrated antennas for mobile communication systems.

69

Andrea Michel was born in Pisa, Italy, on January 20th, 1987. He received the B.E. and M.E. (summa cum laude) degrees in telecommunications engineering from the University of Pisa, Pisa, Italy, in 2009 and 2011, respectively, where he is working toward the Ph.D. degree. His current research topics focus on design of integrated antenna for communication systems and planar antennas for near field UHF-RFID readers.

Andrea D’Alessandro was born in Pietrasanta, Lucca, Italy, in 1982. He received the Master degree in telecommunications engineering from the University of Pisa, in 2010, where he is working toward the Ph.D. degree. In 2010 he collaborated with the Dept. of Information Engineering of the University of Pisa on design and characterization of compact DVB-T antennas. His current research interests include the design of integrated antennas for wireless communications and RFID systems.

Paolo Nepa (M’90) received the Laurea (Doctor) degree in electronics engineering (summa cum laude) from the University of Pisa, Italy, in 1990. Since 1990, he has been with the Department of Information Engineering, University of Pisa, where he is currently an Associate Professor. In 1998, he was at the ElectroScience Laboratory (ESL), The Ohio State University (OSU), Columbus, OH, as a Visiting Scholar supported by a grant of the Italian National Research Council. At the ESL, he was involved in research on efficient hybrid techniques for the analysis of large antenna arrays. His research interests include the extension of high-frequency techniques to electromagnetic scattering from material structures and its application to the development of radio propagation models for indoor and outdoor scenarios of wireless communication systems. He is also involved in the design of wideband and multiband antennas, mainly for base stations and mobile terminals of communication systems, as well as in the design of antennas optimized for near-field coupling and focusing. More recently he is working on channel characterization, wearable antenna design and diversity scheme implementation, for body-centric communication systems. In the context of RFID systems, he is working on techniques and algorithms for UHF-tag localization and smart shelves. Dr. Nepa received the Young Scientist Award from the International Union of Radio Science, Commission B, in 1998.