Introduction
Recently, owing to the progress of communication services and transport systems, major technical innovations have marked, by their importance, the growth of telecommunication. However, a new decentralized architecture based on vehicular communications has aroused real interest of telecommunication industries, research communities, and the automotive industry [Reference Liang, Li, Zhang, Wang and Bie1].
To reduce traffic congestion, improve road safety, and increase traffic efficiency, modern vehicles can contain several antennas for different wireless applications such as AM/FM radio, remote keyless entry, satellite navigation, digital satellite radio, and vehicle-to-everything (V2X) communications [Reference Pell, Sulic, Rowe, Ghorbani, John and Chiaberge2]. In addition to existing applications, vehicular communications will be covered by technologies such as DSRC/IEEE 802.11p and cellular technology/LTE [Reference Neira, Carlsson, Karlsson, Ström and Neira3–Reference Constantinescu, Borcoci, Rasheed and Hayes6]. The reliability and the efficiency of these technologies depend mostly on the quality of the communication link, where antennas are one among the critical modules [Reference Pell, Sulic, Rowe, Ghorbani, John and Chiaberge2,Reference Klemp7]. Indeed, finding optimal antenna performance is incredibly crucial. Besides, due to the high number of antennas and the limited space on the vehicle, the development and use of multi-band antennas is needed [Reference Pell, Sulic, Rowe, Ghorbani, John and Chiaberge2, Reference Constantinescu, Borcoci, Rasheed and Hayes6, Reference Klemp7].
Microstrip patch antennas (MPA) have many advantages in communication systems; their thickness, low cost, compact size, and easy integration into peripherals makes them appropriate for this kind of communication [Reference Pell, Sulic, Rowe, Ghorbani, John and Chiaberge2]. But the associated MPAs have disadvantages such as low gain, narrow bandwidth, and lower efficiency [Reference Bhatt, Mankodi, Desai, Patel, Howon and Dong-Chan8, Reference Palandöken9]. To surmount these drawbacks and with the increasing demand of multiband antennas, fractal geometry is a good option. Fractal antennas present two benefits [Reference Rusu, Baican and Minin10, Reference Gupta and Mathur11]. First, these antennas have a geometric specificity that allows them to resonate over several frequencies while possibly keeping the same electromagnetic characteristics, which is known by the self-similarity or similarity of geometric shape at different scales. Second, the space-filling efficiency of some fractal shapes gives hope for smaller antennas compared to the conventional ones. These benefits make them simpler and easier to manufacture multiband and wideband antennas. The application of this type of vehicular antennahas been studied by several researchers [Reference Pell, Sulic, Rowe, Ghorbani, John and Chiaberge2, Reference Neira, Carlsson, Karlsson, Ström and Neira3, Reference Bustamante, Inca, Chuchon, Adriano and Samaniego5, Reference Constantinescu, Borcoci, Rasheed and Hayes6, Reference Mondal, Samanta, Ghatak and Bhadra Chaudhuri12]. Therefore, much research has been conducted on designing antennas capable of covering different bands of frequencies [Reference Klemp7, Reference Mondal, Samanta, Ghatak and Bhadra Chaudhuri12–Reference Tiwari and Kumar14]. The authors in [Reference Neira, Carlsson, Karlsson, Ström and Neira3–Reference Constantinescu, Borcoci, Rasheed and Hayes6, Reference Neira, Carlberg, Carlsson, Karlsson and Ström15] have focused on the design and performance of LTE and DSRC antennas. The studies conferred in [Reference Satish13–Reference Usha Devi, Rukmini and Madhav22] projected modules of IEEE 802.11p/DSRC antennas. A compact antenna for V2X communication based on LTE and IEEE 802.11p technologies has presented in [Reference Neira, Carlsson, Karlsson, Ström and Neira3]. The study in [Reference Bustamante, Inca, Chuchon, Adriano and Samaniego5] presents a new antenna model appropriate to V2V communications based on LTE and the IEEE 802.11 standard. Considering the disadvantages of existing V2V communication antennas, the authors in [Reference Abishek, Raja, Kumar, Stephen and Raaza18] offer a regular broadband antenna with a relative bandwidth of 35.55%.
This study elaborates on the design of a hexagonal microstrip patch antenna (HMPA) of a compact size of $31 \times 28\times 1.6\, {\rm mm}^{3}$. The projected antenna illustrates a wider bandwidth (from 3.2 to 6.5 GHz) by the combination of Cantor fractal slots introduced on the patch and partial ground plane below. Mathematical modeling of the investigated HMPA and the antenna geometry design followed by a parametric study of critical geometrical variables are described in Section “Antenna design and parametric study.” While Section “Results and discussion” presents the discussion of the simulated and measurement results of the proposed antenna in terms of $S_{11}$
parameter, VSWR, bandwidth, gain, current distribution, and radiation patterns. Section “Antenna positioning on the vehicle” depicts the various antenna positioning on the vehicle. Finally, conclusion of the study is given.
Antenna design and parametric study
This section presents the mathematical modeling of an HMPA and simluated results given by a simple parametric study.
Antenna design
Using the cavity model, the resonant frequencies of the TM mode of a circular patch are given in equation (1) [Reference Balanis23]:
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn1.png?pub-status=live)
where $c$ is the light velocity in the free space, $\epsilon _r$
is the relative permittivity of the substrate, and $X_{np}$
are the zeros of the derivative of the Bessel function ${\rm Jn}\lpar x\rpar$
of order $n$
, as is true of TE-mode circular waveguides. The lowest-order mode, ${\rm TM}_{11}$
, uses $X_{11} = 1.84118$
and produces a linearly polarized field similar to a square patch. $a_{eff}$
is an effective radius of the patch taking fringing into account. As is known, the fringing gives a wider electrical aspect to the patch, and it has been taken into account by introducing a correction factor. For the circular patch, a correction is introduced using an effective radius $a_{eff}$
, to replace the real radius $a$
, given in [Reference Balanis23]:
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn2.png?pub-status=live)
where $h$ is the height of the substrate. Since the effective and physical radii of the patch are nearly the same, equation (2) may be iterated to compute the real radius $a$
given by equation (3):
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn3.png?pub-status=live)
where $h$ is in cm, and
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn4.png?pub-status=live)
By relating the areas of the circular and hexagonal patches, equation (1) can be applied for designing an HMPA. Let there be a circular patch of radius $a$; the area of this patch is:
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn5.png?pub-status=live)
A regular hexagon is constructed by drawing six equilateral triangles. As is known the area of an equilateral triangle of side $s$ is given by equation (6):
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn6.png?pub-status=live)
Then the area of a regular hexagon is given by equation (7):
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn7.png?pub-status=live)
The hexagon side is determined by comparing the areas of hexagonal and circular patches as shown in equation (8):
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn8.png?pub-status=live)
By taking fringing into account, the hexagonal side is:
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn9.png?pub-status=live)
Cantor set fractal theory
Cantor dust described by mathematician Georg Cantor in 1872 is probably the oldest known fractal and the easiest to create [Reference Ali, Abdulkareem, Hammoodi, Salim, Yassen, Rashed Hussan and Al-Rizzo24–Reference Reha, El Amri, Benhmammouch, Said, El Ouadih and Bouchouirbat26]. The construction of this structure is based on a straight line segment from which the central third is removed. The same operation is performed on the two remaining segments, then by successive iteration on the resulting smaller and smaller segments. This shape is characterized by a number of segments tending to become infinite with an almost zero length. The number of copies of the original shape obtained from one iteration to another is equal to 2 ($N = 2$) and the size of each new copy is equal to 1/3 of the original size ($r = 1/3$
). This geometry is described by using the iterated function system (IFS) which is represented by the following affine transformation [Reference Rusu, Baican and Minin10]:
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn10.png?pub-status=live)
The IFS coefficients for the Cantor set are given in Table 1. In general, the fractal geometry has a very particular feature that its dimension exceeds its topological dimension, indeed there are various types of representations used for the fractal dimensions such as Hausdorff dimension defined as:
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn11.png?pub-status=live)
where $r$ is the scaling factor and $N$
is the number of self-similar structure copies. The Cantor set fractal consists of $N = 2$
congruent subsets, where each gives the original set when it is enlarged by a factor $r = 3$
. Therefore, the fractal dimension of the Cantor set is defined as in equation (12):
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_eqn12.png?pub-status=live)
Table 1. IFS transformation coefficient for Cantor set fractal
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_tab1.png?pub-status=live)
The construction of the investigated antenna is started with a simple hexagonal patch. The antenna is manufactured on an FR4 epoxy substrate with an $\epsilon _r$ of 4.4 and a loss tangent of 0.0028. The volume of the antenna is $Ls \times Ws \times h = 31 \times 28 \times 1.6\, {\rm mm}^{3}$
, with a partial ground plane of $Lg \times Ws$
at the bottom of the substrate (Fig. 1). The $50\, \Omega$
impedance is achieved by adjusting the feed line width and length (respectively $wf$
and $Lf$
), the gap ($g$
) notch width, and the inset distance ($d$
) from the radiating edge. The antenna was designed and simulated using CST Microwave Studio, which uses the finite integration technique for computation. To show the effect of inserting fractal geometry into the patch, the structure has been designed with three iterations of the Cantor set fractal geometry that has been introduced on the hexagonal patch. The side $s$
of the hexagon shape is computed using equation (9). At a resonant frequency, the evaluated hexagonal side $s$
is of 21.92 mm.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig1.png?pub-status=live)
Fig. 1. The proposed antenna configuration: (a) front view and (b) back view.
Parametric study
The antenna performance is affected by several parameters such as the distance ($d$) between the radiating patch and feed line, the length ($Lg$
) of the ground plane, the length ($Lf$
) and width ($wf$
) of the feed line, the length ($Lc$
) and width ($Wc$
) of the Cantor set slots.
As shown in Fig. 2, changing the length of the ground plane $Lg$ not only changes the lateral dimensions of the antenna but also has a remarkable effect on the resonant characteristics of the proposed antenna. At the high $Lg$
values, the frequency performance degrades. The optimum ground plane length is $Lg = 11\, {\rm mm}$
, for which the desired band frequency is obtained.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig2.png?pub-status=live)
Fig. 2. Effect of various lengths of the ground plane ($Lg$) on $S_{11}$
.
To improve the attained characteristics, the proposed antenna is simulated by keeping the length $Lf = 12\, {\rm mm}$ and varying the width (wf) of the feed line. The return loss graphs of MPA along with various values of wf is shown in Fig. 3. It can be premeditated that the desired wideband characteristics have been exhibited with the feed line dimensions of $Lf = 12\, {\rm mm}$
and $wf = 3.5\, {\rm mm}$
. The distance ($d$
) between the radiation patch and the feed line has an important effect on the impedance matching of the antenna. From Figs 4 and 5, it can be elaborated that the desired features have been attained at the inset feed line distance of $d = 2\, {\rm mm}$
and a gap of $g = 1\, {\rm mm}$
. The Cantor fractal slots are introduced in the upper part of the hexagonal patch. A simple parametric study of the width ($Wc$
), the length ($Lc$
) of these slots, and the space between them ($Espc$
) have brought us back to results depicted in Fig. 6.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig3.png?pub-status=live)
Fig. 3. Effect of various widths of the feed line ($wf$) on $S_{11}$
.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig4.png?pub-status=live)
Fig. 4. $S_{11}$ for various inset feed line distances ($d$
).
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig5.png?pub-status=live)
Fig. 5. $S_{11}$ for various gaps ($g$
).
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig6.png?pub-status=live)
Fig. 6. $S_{11}$ for various Cantor set dimensions values.
It can be claimed that the Cantor slots width ($Wc$) of 12 mm, length ($Lc$
) of 1 mm, and ($Espc$
) of 0.5 mm are found to be optimum, for which the desired DSRC frequency band of 5.9 GHz is achieved.
It may be noted from Fig. 7 that the structure with Cantor fractal slots presents an improved matching and good return losses compared with the one without the fractal slots. The hexagonal-shaped structure with a Cantor fractal geometry covers the desired band exhibiting good performances.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig7.png?pub-status=live)
Fig. 7. Return loss comparison of both configurations.
Results and discussion
The optimized parameter dimensions of the HMPA are indexed in Table 2.
Table 2. The optimal dimensions of the proposed antenna
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_tab2.png?pub-status=live)
The antenna parameters such as return loss $S_{11}$ shown in Fig. 7, VSWR in Fig. 8, the impedance in Fig. 9, and the simulated bandwidth and gain are listed in Table 3.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig8.png?pub-status=live)
Fig. 8. VSWR of the proposed antenna.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig9.png?pub-status=live)
Fig. 9. Smith chart of the proposed antenna.
Table 3. $S_{11}$, VSWR, bandwidth, and gain of both configurations
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_tab3.png?pub-status=live)
The minimal values of $S_{11}$ are $-27.9\DIFdel {\, {\rm dB}}$
and $-41.53\, {\rm dB}$
at 3.71 and 5.9 GHz, respectively, which indicates a good adaptation. The efficiency of the antenna is evaluated by simulating its VSWR for the desired frequency band. It is clear from Fig. 10 that VSWR is 1.08 at 3.71 GHz, 1.01 at 5.9 GHz, and $\lt$
2 across the integral bandwidth which represents a suitable efficiency of the simulated antenna.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig10.png?pub-status=live)
Fig. 10. Radiation pattern ($E$-plane and $H$
-plane) at 3.71 and 5.9 GHz.
The value of the real part of the input impedance is $50\, \Omega$ as shown in Fig. 9, which represents a null imaginary part whence the good adaptation of the antenna.
Figure 10 shows the polar radiation patterns at both resonant frequencies of 3.71 and 5.9 GHz, it is observed that the antenna has an omnidirectional behavior in both $H$-plane (${\rm Theta} = 90$
) and $E$
-plane (${\rm Phi} = 90$
) at both resonant frequencies.
For the DSRC application, the antenna is commonly needed to hold an omnidirectional radiation pattern in the azimuth plane. From the simulated radiation patterns of the proposed HMPA at two frequencies namely 3.71 and 5.9 GHz it can be seen that the projected antenna is a good candidate for the vehicular applications. The simulated gain of the proposed antenna is illustrated in Fig. 11, which varies from 2.97 dB at the lower frequency to 5.11 dB at the upper frequency. The current distribution at the centered frequencies respectively of 3.71 and 5.9 GHz is depicted in Fig. 12. It may be claimed that the antenna principally behaves as a radiating slot formed between the hexagonal patch and the inset feed line. As displayed the current distribution at 5.9 GHz shows more concentration of the current near Cantor set fractal slots and inset feed line.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig11.png?pub-status=live)
Fig. 11. Simulated gain of the proposed antenna.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig12.png?pub-status=live)
Fig. 12. Current distribution at (a) 3.71 GHz and (b) 5.9 GHz.
Figure 13 illustrates the fabricated model of the proposed antenna. The $S_{11}$ of the fabricated antenna is measured using the ANRITSU MS2026C Network Analyzer. Figure 14 shows the measured and simulated $S$
-parameters of the proposed HMPA. It may be noted that the investigated design presents a bandwidth ($-10\, {\rm dB}$
) of about 3.28 GHz in simulation using the CST Studio Suite, about 3.15 GHz using HFSS software, and over than 2.92 GHz in measurement, respectively. Then it is clear that the proposed antenna displays suitable agreement in both simulation and measurements with good characteristics. A brief comparison between simulated and measured resonant frequencies and bandwidths of the simulated and manufacturing antenna is depicted in Table 4. The analysis of the results reveals that this antenna is qualified to cover many operating bands such as WLAN, WiMAX, V2X-LTE, and DSRC for vehicular communications.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig13.png?pub-status=live)
Fig. 13. Fabricated prototype of the proposed antenna: (a) front view and (b) back view.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig14.png?pub-status=live)
Fig. 14. Measured and simulated $S$-parameters of the proposed antenna.
Table 4. Comparison of measurement and simulation results
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_tab4.png?pub-status=live)
Table 5 summarizes a comparison of the performance of the projected antenna with some measured antennas in previously published studies, in terms of size, resonant frequencies, $S_{11}$ values, gain, and ($-10\, {\rm dB}$
) bandwidths.
Table 5. Comparison with previously published studies
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_tab5.png?pub-status=live)
Antenna positioning on the vehicle
The proposed antenna regards to operate among two bands for vehicular communication. It is indispensable to study the impact of the antenna placement when inserted on the vehicle platform on radiation performance and efficiency. Indeed, a CAD model of a car is taken, and three antenna positions are chosen to be tested on the roof top, side mirror, and front window of the vehicle. The prediction of far-field radiation characteristics of the antenna is performed employing an environment of the car body tool. Once the investigated antenna is installed at the intended locations on the car body, it will induce surface currents on the vehicle platform. These currents are then approximated by tracing high-density rays and then radiated to determine the influence of the body on the antenna radiation [Reference Madhav, Anilkumar and Kotamraju20]. The simulated radiation patterns are presented in the $XY$ (${\rm Phi} = 0$
), $YZ$
(${\rm Phi} = 90$
), and $XZ$
(${\rm Theta} = 90$
) planes and are shown in Figs 15, 16, and 17, respectively. The obtained radiation patterns for the antenna at 3.7 and 5.9 GHz are slightly changed. Figure 15 depicts the radiation pattern that simulated in the windshield mounting. It can be mentioned that at 3.7 and 5.9 GHz, the obtained radiation patterns of the antenna are toward omnidirectional. The antenna in this position exhibits a good radiation pattern and less distortion which indicates a fairly good reception area. Figure 16 shows plots of the radiation patterns of the antenna installed on the side-view mirror of the vehicle at 3.7and 5.9 GHz respectively. This antenna placement offered a more omnidirectional radiation pattern. From Fig. 17, when the antenna is positioned on the roof top of the vehicle, the direction of maximum radiation is observed in the region 30–90. The plots clearly show pattern distortion and several deep nulls, which is due to the reflections offered by the large perfectly electric conducting surface of the roof. The best placement for this antenna on the vehicle is the windshield. Even in this location, there still appears very little distortion due to the radiation pattern. It is worth to note that the back radiation exists in all the considered positions, and it is slightly suppressed due to the penetration of the radiating field through the car body. In the $XZ$
plane, the patterns are omnidirectional at 3.7 and 5.9 GHz DSRC band. Hence, the obtained radiation patterns are expected to use the proposed antenna in vehicular communications.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig15.png?pub-status=live)
Fig. 15. Far-field patterns of the proposed antenna mounting on front window.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig16.png?pub-status=live)
Fig. 16. Far-field patterns of the proposed antenna mounting on side mirror.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20210305142035049-0526:S1759078720000719:S1759078720000719_fig17.png?pub-status=live)
Fig. 17. Far-field patterns of the proposed antenna mounting on roof top.
Conclusion
An HMPA with Cantor set fractal slots on the radiating element has been designed, simulated, and measured. The antenna is designed for vehicular communication, including cellular V2X communication, DSRC communication based on the IEEE 802.11p protocol, and the antenna covers other wireless applications. To attain the desired characteristics, the effects of some geometrical parameters are analyzed. It has been noted that the projected antenna is suitable for vehicular communications including both LTE/WiMAX and V2X/DSRC. The printed antenna displays an omnidirectional radiation pattern in the horizontal plane for V2X. Reasonable results in terms of $S_{11}$, VSWR, and simulated gain are obtained within an effective bandwidth of 3.3 GHz from 3.22 to 6.5 GHz. The experimental and simulated results are in good agreement with a measurement bandwidth of 2.92 GHz and an omnidirectional radiation pattern in the $H$
-plane. These results implies that the suggested HMPA with the Cantor fractal slots is appropriate for many bands of wireless communication systems: 3.7 GHz for blind spot detection for vehicles, 5.8 GHz for WLAN, 3.5 GHz for WiMAX, and 5.9 GHz for DSRC.
Acknowledgements
This study was supported by the Laboratory of Electrical Systems and Telecommunications, Faculty of Science and Technology, Cadi Ayyad University – Marrakech (LSET-UCAM) in coordination with the Laboratory of Computer Science, Networks, Telecommunications and Multimedia of the National High School of Technology, Hassan II University of Casablanca where the measurements have been carried out, and the National Center of Scientific and Technical Research (CNRST) Morocco.
Fatima Ez-Zaki received her Bachelor of Sciences in industrial computer, electronic, electrotechnical and automatic from Cadi Ayyad University, Marrakesh, Morocco, in 2015. She received her Master of Science and Technology (M.Sc. Tech. (Eng)) in electrical engineering from Cadi Ayyad University, Marrakesh, Morocco, in 2017. She is currently working toward her Ph.D. degree at the Department of Applied Physics, Electrical Systems and Telecommunications Laboratory, Cadi Ayyad University of Marrakesh, Morocco. Her research interest includes telecommunications, antennas, and vehicular communications.
Hassan Belahrach received his B.S. degree in electrical and electronics from Mohamed V University, Rabat, Morocco, in 1986, his M.S. degree in electronics and microelectronics from Bordeaux I University, France, in 1987, his Ph.D. degree in microelectronics and technology from Bordeaux I University, France, in 1990, and his Ph.D. degree in microelectronics and telecommunication from Cady Ayyad University, Marrakech, Morocco, in 2001. Currently, he is a full University Professor of Analog and Digital Electronics at Royal Air Academy (Marrakech). His research areas include analysis and design packaging of very high-speed CMOS suitable for telecommunication applications.
Abdelilah Ghammaz received his Doctor of Electronic degree from the National Polytechnic Institute (ENSEEIHT) of Toulouse, France, in 1993. In 1994, he went back to the University of Cadi Ayyad of Marrakech, Morocco. Since 2003, he has been a Professor at the Faculty of Sciences and technology, Marrakech, Morocco. His research interests include the field of electromagnetic compatibility and antennas.