Education Column

Karl F. Warnick Department of Electrical and Computer Engineering Brigham Young University Provo, UT 84602 USA Tel: +1 (801) 422-1732; Fax +1 (801) 422-0201 E-mail: [email protected]

2014 Design Challenge Final Report

Technical Article

n this issue, we present a report from the 2014 AP-S Student Design Challenge joint second prize winning team, from the Indian Institute of Technology Kanpur: Gaurangi Gupta, Bhanu Pratap Singh, Amrita Bal, Deepam Kedia, and A. R. Harish.

This issue’s Education Column technical article, by Prashant K. Mishra, Dhananjay R. Jahagirdar, and Girish Kumar, is a review of techniques to design dual-polarized microstrip antennas with high isolation and broad bandwidth.

I

Orientation Detection Using Passive UHF RFID Technology Gaurangi Gupta, Bhanu Pratap Singh, Amrita Bal, Deepam Kedia, and A. R. Harish Department of Electrical Engineering Indian Institute of Technology Kanpur Kanpur, Uttar Pradesh, India 208016 E-mail: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]

Abstract A simple, compact, rechargeable handheld radio-frequency identification (RFID) reader was designed to detect the orientation of tagged objects. The system uses a linearly polarized reader antenna and the orientation sensor of the mobile device to identify the orientation of the tag. The system works in the ultra-high-frequency (UHF) RFID band. Detailed testing was done in different surroundings with various materials and tags. The system was able to correctly identify the orientation in several cases with an average angular error of less than 10°. This system can also be used to demonstrate the concept of polarization. Keywords: Accelerometer; Android; artificial magnetic conductor (AMC); dipole antennas; high impedance surface (HIS); orientation detection; polarization; radiofrequency identification (RFID); RFID tags IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 221

221

12/27/2014 2:03:42 PM

1. Introduction

D

etection of the orientation of objects is very important in several applications. If the object is visible, e.g., during an assembly process, optical techniques could be used to detect the orientation [1-4]. There are several situations where it is not possible to view the object but it is required to know the orientation of the object, e.g., bottles stored in a carton box ready for transportation. With the proliferation of item-level tagging using RFID tags – which are usually linearly polarized [5-8] – it is possible to use an RFID system to detect the orientation of the tags, and hence that of the object to which the tag is attached. The power received by an antenna depends on the polarization of the incoming wave (which depends on the polarization of the transmitting antenna) and the polarization of the receiving antenna. The polarization loss factor is given by 2

2

ˆ w ρˆ a = τ p ρ= cosψ p ,

measure the received power as a function of orientation of the reader antenna. As the relative orientation of the two antennas changes, the RSSI also changes monotonically. Therefore, by measuring the RSSI value using the reader, it is possible to estimate the orientation of the tag. In this work we designed an RFID system consisting of a commercial off-the-shelf (COTS) UHF RFID reader, a power charging system with a battery backup, and a mobile phone running the Android operating system. We designed and fabricated in the laboratory a low-profile linearly polarized antenna for the reader, which used a high-impedance surface as a reflector. The orientation of the mobile device, and hence the orientation of the reader antenna, is determined using the orientation sensor (accelerometer) available in the mobile phone. An Android application was designed and implemented that apart from providing a user interface, also performed data processing and housekeeping actions. Again, we used commercial off-the-shelf UHF RFID tags in this study.

(1)

where ρˆ w and ρˆ a are respectively the polarization vectors of the wave and the receiving antenna, and ψ p is the angle between the two unit vectors. The polarization loss factor can be used to compute the reduction in the received power when there is a mismatch between the two polarizations [9]. Consider a transmitting antenna with changing orientation ( φ ) and a receiving antenna with fixed orientation. If the transmitting antenna is circularly polarized and the receiving antenna is linearly polarized, the polarization loss factor, and hence the received power, are independent of the orientation ( φ ) of the transmitting antenna. This can be seen from Figure 1, where the received power is equal for the transmitter-receiver combination kept at two different orientations of 0° and 90°, respectively. On the other hand, if the transmitting antenna is linearly polarized, the received power is a function of φ , as can be seen in Figure 2. The RFID tags are usually linearly polarized. The RFID reader antennas are thus preferred to be circularly polarized, in order to ensure that the tags are read irrespective of the relative orientation of the tag and the reader antennas [10, 11]. Instead of a circularly polarized antenna, if a linearly polarized antenna is used for the reader, the power received by the tag will be maximum when the polarization vectors of the reader and the tag are aligned. An RFID reader with the capability to record the RSSI (received signal strength indicator) can be used to

222

AP_Mag_Dec_2014_Final.indd 222

Figure 1. The received power with a circularly polarized transmitting antenna in two different orientations.

Figure 2. The received power with a linearly polarized transmitting antenna in two different orientations.

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

12/27/2014 2:03:42 PM

In Section 2, we present the system description, design methodology, and operation. The measured system performance is presented in Section 3, and finally in Section 4, we present the conclusions.

2. System Description The block schematic of the system is shown in Figure 3, and each of the sub-systems is explained in this section.

2.1 Antenna Description The proposed antenna (Figure 4) consisted of a bowtie dipole, placed on a double-layered via-less high-impedance surface (HIS). A bowtie dipole is linearly polarized, and has an omnidirectional pattern with nulls along the axis of the dipole [9]. Generally, a reflector made of a perfect electric conductor (PEC) is kept a quarter-wavelength away from the antenna to enhance the gain in one direction. As the distance between the dipole and the ground plane reduces, the antenna’s performance deteriorates [12]. A high-impedance surface – which works like a perfect magnetic conductor (PMC) over a frequency range – can be used as a reflector, and can be placed very close to the dipole without degrading the performance of the antenna [1216]. The high-impedance surface consisted of two layers of periodic metal patches printed on two sides of a dielectric substrate. This structure was backed by a grounded dielectric slab. We used FR4 as the dielectric material to construct the antenna. Metal patches in the top layer were coupled capacitively with the adjacent patches on the same layer. Patches in the middle layer overlapped with four adjacent patches in the top layer, and provided additional capacitance [17]. The bowtie dipole antenna (length of 93 mm, feed-gap of 0.5 mm, smaller width of 2.5 mm, larger width of 15 mm) had a resonance frequency of 1.2 GHz when placed in free space. It had a gain of 1.98 dBi, and a bandwidth (corresponding to S11 < −10 dB) of 13.1%. This dipole antenna was placed on a 4 × 4 array of high-impedance surface ( a = 30 mm, ab = 48 mm, g = 1 mm, height of lower substrate of 3.2 mm, height of upper substrate of 1.6 mm, ε r = 4.4 ) at a height of 1.6 mm from the top surface. The overall height of the antenna was 6.4 mm, and its size was 234 mm × 234 mm.

respectively. From these results, we could conclude that the antenna had a −10 dB S11 bandwidth of 11.55%, and a peak gain of 6.75 dBi. The antenna had an efficiency of 76.7%, and at the center frequency, the directivity was 7.9 dBi. The radiation patterns of the proposed antenna in two orthogonal planes at 0.92 GHz are shown in Figure 7. The antenna had a front-toback ratio of 20 dB. The cross-polarization level along the direction of maximum was also better than 20 dB. The low cross-polarization level helped in accurately distinguishing the two orthogonal components, and hence made this antenna a good choice for detecting the orientation of the linearly polarized RFID tags.

2.2 Hardware Description and Assembly The system hardware consisted of four main off-the-shelf components, viz., the RFID reader, Android phone, power supply and communication board, and housing (Figures 3 and 8).

2.2.1 RFID Reader A Winnix HYM730 UHF RFID Reader Module Version V1.0 [18] was used for the system. It provided a UART baud rate of 115200 bps. The received signal strength indication (RSSI) was extracted as a measurement of the power present in the radio signal. This indicator was used in the proposed system as a measure of the relative orientation between the tag and the reader. Another indicator could have been the read count, which is the number of times a tag is read per unit time. When the tag was completely polarization aligned with the reader antenna, the read count significantly increases, and it decreases as the two go out of alignment. It was observed that the RSSI was a better measure of alignment compared to the read count.

2.2.2 Android Phone In the proposed system, a Micromax A116 with Android 4.1.2 (API level 16), USB interface and accelerometer was used. Any Android phone with an API level 16 or higher

The performance of the antenna–high-impedance-surface system was optimized using an FEM-based commercial EM solver. Since the in-house fabrication technique could not completely remove the air gap between the dielectric layers, we assumed an air gap of 0.1 mm in the simulation. The dipole was directly fed by a coax probe on one arm, and the other arm was shorted to the ground plane. The measured and simulated S11 values and the gain of the designed antenna are compared in Figure 5 and Figure 6, IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 223

Figure 3. A block diagram of the system. 223

12/27/2014 2:03:43 PM

Figure 4. The antenna’s geometry.

224

AP_Mag_Dec_2014_Final.indd 224

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

12/27/2014 2:03:43 PM

Figure 5. The input reflection coefficient of the antenna.

Figure 6. The gain of the antenna as a function of frequency.

Figure 7. The normalized two-dimensional radiation patterns: (a) 0.92 GHz ( xz plane); (b) 0.92 GHz ( yz plane) (solid line is simulated, squares are measured).

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 225

225

12/27/2014 2:03:44 PM

and equipped with orientation sensors could also have been used for this application.

2.2.3 Power Supply and Communication Board Charging circuitry along with a rechargeable battery was used so that a continuous power supply was not required. In order to establish serial-to-UART communication between the Android device and the RFID reader, an FT232BL single-chip USB-to-serial-UART interface IC was used [19]. It provided a 384-byte receiving buffer and a 128-byte transmitting buffer for high data throughput. All of the above-mentioned components were assembled as shown in Figure 8 and Figure 9. Finally, the fabricated antenna (Figure 10a) was connected to the reader via an SMA cable, and fitted with the device using a plastic bracket. The final hardware is shown in Figure 10b.

2.3 Software Description

Figure 9a. The reader assembly and interfaces: the Android phone.

An Android application was developed in the opensource software Eclipse IDE installed with the Android Development Tools (ADT) Package v.22.0.5 [20]. Function libraries present in the Android Software Development Kit (SDK) running on the Java run-time environment v.1.7.0 [21] were used in the application, which consisted of the following functional modules.

2.3.1 Tag Data Import A PHP script ran on an Apache server hosted on a personal computer (PC). The tag data (item name and tag id) were Figure 9b. The reader assembly and interfaces: the USB interface and antenna connector.

Figure 8. The RFID reader components. 226

AP_Mag_Dec_2014_Final.indd 226

Figure 9c. The reader assembly and interfaces: the power charging port and indicator. IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

12/27/2014 2:03:46 PM

Depending on the number data points for a tag, we used two different algorithms to predict its orientation. If the RSSI values for 100 or more distinct angles were available, we used a “filter-based algorithm” to predict the orientation of the tag; otherwise, we used the “direct algorithm.”

2.3.3 Filter-Based Algorithm The averaged raw data consisted of significant noise (green trace in Figure 12). It was possible to remove the highfrequency noise by going into the Fourier domain, choosing only the lowest five spectral components, and transforming the data back into the angle domain. The filtered data was much smoother (blue trace in Figure 12). It was observed that five spectral components was a tradeoff between eliminating noise and maintaining sufficient variation so as to extract features.

Figure 10a. The fabricated antenna.

2.3.4 Feature Extraction The finite-difference technique was used on the filtered data to compute the derivative of the filtered RSSI values with respect to angle. All zero-crossing points in the derivative domain were selected and stored in a list. These points were the local maxima and local minima of the RSSI data. In this list, we merged those adjacent minima and maxima points into common points the difference of which in the RSSI levels was less than 3 dB. For example, merging of two local maxima into a common point is shown in Figure 12. This reduced the number points for the next stage. We then found the global maximum ( θ max ) and the global minimum ( θ min ) occurring in the angular region of ±120° , and reported the orientation of the tag ( θtag ) as

θ= tag θ max ± Figure 10b. The complete assembly of the RFID reader.

saved in the local memory of the PC in XML format. Using the import button on the first screen of the application (Figure 11a), the PHP script was called, which in turn passed the tag data into the phone’s memory.

2.3.2 Data Collection and Processing During the data collection process, the rotation angle of the device was obtained from the orientation detection application that was preinstalled in the Android phone. As we rotated the device, the orientation information and the corresponding RSSI values for each of the tags were collected. The raw data was averaged over 1° angular intervals, and saved in the phone’s memory. IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 227

( 90 − θmax − θmin ) , 2

(2)

where the positive sign was chosen if θ max > θ min and the negative sign was chosen, otherwise. The correction factor was introduced so that the difference between the global maximum and minimum was made 90°.

2.3.5 Direct Processing Algorithm In cases when items were at a large distance, when the number of items was large, or in the presence of other tags, the number of tag reads dropped significantly, i.e., not enough RSSI values at distinct angles was available. In order to predict the tag orientation with limited data, the global maximum of the complete data set was picked. We then selected all the data points having an RSSI within 3 dB from the global maximum, and angles that were within ±30° about the global maximum (as shown in Figure 13). The orientation of the tag was the mean of these angles. This algorithm was sometimes prone to noise, but could serve as a reasonable estimate when tag reads were low. 227

12/27/2014 2:03:47 PM

Figure 11a. The application GUI: the data import screen.

Figure 11b. The application GUI: the main activity screen.

Figure 12. The application of the filtering algorithm and the merging of two maxima. 228

AP_Mag_Dec_2014_Final.indd 228

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

12/27/2014 2:03:47 PM

Figure 13. The application of the direct-processing algorithm.

Figure 14b. the straight and folded-back tag used for system evaluation.

Figure 14a. The square tags used for system evaluation.

Figure 14c. The S-shaped tag used for system evaluation. IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 229

229

12/27/2014 2:03:48 PM

2.4 RFID Tag As per the specifications, passive unidirectional square paper tags of dimensions 50 mm × 50 mm were selected (Figure 14a). Foam spacers were used to ensure that the tags did not come in direct contact with the object on which they were attached. This was done to minimize the effect of the electrical properties of the object on the performance of the tag. With the maximum power level specified (4 W EIRP), these tags could be read at a distance of 5 m when attached to empty plastic bottles. Rectangular tags are more commonly available and used in RFID applications. During this project, we had access to commercial tags that were 100 mm × 10 mm (Figure 14b). However, these did not meet the project specification. We modified this tag by folding it in two different ways, as shown in Figures 14b and 14c. In the first version, the tag was folded onto itself, and a 4 mm thick foam spacer was introduced in the middle. This maintained a reasonably low cross-polarization level; however, the reading range decreased. By folding the ends of the tag to form an S shape (Figure 14c), we observed that the reading range did not significantly drop. However, due to the reorientation of the currents, the cross-polarization level went up, and hence affected the ability of the system to accurately detect angles. EM simulation results were used to confirm the performance of the straight, folded-back and S-shaped tags. Figure 15 shows the comparison of the gains for these tags. It was observed that the gain was reduced to 0.9 dBi at 0.9 GHz for the folded-back tag, as compared to 1.6 dBi for the straight tag. This justified the reduction in reading range for the folded-back tag. Figure 16 shows the radiation patterns for these tags. We observed that the cross-polarization level was only 15 dB below the co-polarization level for the S-shaped tag, which made it a poor choice for this application. The cross-polarization levels for the straight and the folded-back tags were better than 40 dB.

2.5 System Operation The inventory was started by pressing the start button on the main activity screen (Figure 11b). Whenever a tag was read by the reader, the corresponding RSSI value and the orientation of the device were fetched and stored in the local database. The read count of the corresponding tag was also increased by one. As the device was rotated about its axis, the antenna also rotated, and the polarization of the EM waves radiated by the antenna also changed. This resulted in a change in RSSI of the tag read by the reader. The system continuously recorded the RSSI values and the corresponding orientation of the reader for all the tag reads. While rotating the reader, care had to be taken to keep the rotation axis perpendicular to the antenna’s plane. If the axis of rotation changed, the antenna’s gain along the direction of the tag changed, thus affecting the obtained RSSI value levels, which could lead to false readings. 230

AP_Mag_Dec_2014_Final.indd 230

Figure 15. A comparison of the gains for different tag configurations.

3. Experimental Scenarios and System Evaluation The system has been tested in different environments, consisting of several tags and various types of objects. In order to compare the performance, we calculated the average error in predicted orientation of the tag.

3.1 Non-Metallic Environment We considered four empty plastic bottles, each of which had an RFID tag attached to it. These bottles were kept on a wooden table (Figure 17). Using the proposed system, we estimated the orientation of these bottles. For example, when the distance between the tags and the reader was 1 m, the average error in the orientation was only 5°. The average angular error (in degrees) for various combinations of orientations and as a function of distance between the reader and the tags is shown in Figure 18. As the distance between the tags and the reader increased, the error also increased. The average error was the highest when all the bottles were placed horizontally, but it was only 12°. In the next experiment, we considered empty plastic bottles, plastic bottles containing water, and metal cans (Figures 19a, 19b, and 19c). The distance between the reader and the tag was maintained at 1 m. We plotted the average error in the orientation as a function of the number of bottles tilted to 90° (Figure 20). We saw that the accuracy of prediction was higher for empty bottles; however, the average angular error even for the metal cans was better than 14°. IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

12/27/2014 2:03:48 PM

Figure 16a. The radiation patterns for the straight tag (H plane on the left, E plane on the right, red is co-polar, blue is crosspolar).

Figure 16b. The radiation patterns for the folded-back tag (H plane on the left, E plane on the right, red is co-polar, blue is cross-polar).

Figure 16c. The radiation patterns for the S-shaped tag (H plane on the left, E plane on the right, red is co-polar, blue is cross-polar). IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 231

231

12/27/2014 2:03:49 PM

Figure 17. Orientation detection of four bottles using the RFID reader.

Figure 18. The average angular error (in degrees) as a function of distance for different orientations.

Figure 19. Sample scenarios: (a) empty bottles, (b) filled bottles, (c) metal cans, (d) bottles on the top shelf of a metal rack, (e) empty bottles in the middle shelf of a metal rack, and (f) filled bottles in the middle shelf of a metal rack. 232

AP_Mag_Dec_2014_Final.indd 232

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

12/27/2014 2:03:50 PM

3.2 Metal Environment Some experiments were conducted by placing tagged plastic bottles on the top and middle shelves of a metal rack (Figures 19d, 19e, and 19f). A comparison of the average angular error is shown in Figure 21. The prediction accuracy was very good for the bottles placed on a wooden table (labeled as normal environment). However, the performance slightly degraded when the bottles were placed on the top shelf of the metal rack (error of 12°). With the bottles placed on the middle shelf, the error increased to 24°, especially when three or four bottles were placed horizontally. One of the reasons for this was the effect of the metallic environment on the polarization purity of the tags, especially when the bottles were placed horizontally.

Figure 20. A comparison of the angular error for different object materials and object orientations.

The performance was very similar for both empty bottles (denoted by hollow markers) and water filled bottles (denoted by filled markers).

3.3 Performance with Different Tags We then compared the performance of three different types of tags, shown in Figure 14. The tags were attached to empty bottles placed on a wooden table, and the reader was placed at a distance of 1 m. The square tag had the best accuracy (average angular error of 4°), but the folded-back tag was slightly inferior (average angular error of 7°). With the S-shaped tag, the accuracy significantly degraded (Figure 22), due to the increased cross-polarization level. The reading range of these tags was measured by placing them on different objects (Table 1). The reading range decreased slightly as the number of objects was increased. However, the reading range was significantly affected by the object type. From the data presented in the table, we also observed that the reading ranges of the square tag and the S-shaped tag were very similar, but that of the folded-back tag was significantly inferior. Figure 21. The angular error as a function of the number of horizontal bottles in a metallic environment: unfilled markers are empty bottles, filled markers are water­filled bottles.

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 233

From these measurements, we could conclude that in order to ensure good reading range as well as accurate orientation prediction, the tag should have high gain and low crosspolarization level.

233

12/27/2014 2:03:50 PM

Table 1. Reading ranges in different scenarios. Range Scenario

Empty bottles Filled bottles Metal Cans

No. of Items

Square Tags

FoldedBack Tags

S-Shaped Folded Tags

2

4.49 m

2.2 m

4.4 m

4

4m

1.7 m

3.8 m

2

1.5 m

1.3 m

2.3 m

4

1.03 m

1m

1.9 m

1

1.1 m

1m

1.3 m

2

0.8 m

0.7 m

1m

Figure 22. A comparison of the performance of different tags.

Figure 23. The power levels as a function of the rotation angle for (a) a linearly polarized antenna, (b) a circularly polarized antenna. 234

AP_Mag_Dec_2014_Final.indd 234

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

12/27/2014 2:03:51 PM

Table 2. Component list and budget. Part

Hardware

Source

Cost (US $)

HYM730 UHF RFID Reader

Winnix

1

550

Charging circuit and battery

Iaito Infotech

1

150

Android Phone

Micromax A116

1

225

1

100

In-house

1

350

Alien

10

–

Iaito

10

–

Packaging Antenna Fabrication Antenna

Qty.

Passive Tags Total

1375

3.4 Demonstration of Polarization

7. References

So far, we used a linearly polarized antenna for the reader. As we rotated the reader about its axis and recorded the RSSI as a function of orientation, we got the pattern shown in Figure 23a. If we replaced the reader’s antenna by a circularly polarized antenna and repeated the experiment, the RSSI versus angle plot will be almost a straight line (with some amount of noise), as shown in Figure 23b. For this experiment, we used a helical antenna that was fabricated in the laboratory. The system can also therefore be used to demonstrate the polarization of the reader’s antenna.

1. R. Morris and L. Rubin, Technique for Object Orientation Detection Using a Feed-Forward Neural Network, United States Patent No. 5060276 A, October 1991.

4. Conclusion In this work, we proposed the use of a passive UHF RFID reader to detect the orientation of the tagged object, using the concept of polarization and built-in Android orientation sensors. We carried out the performance testing of the system under different conditions, viz., objects with different electrical properties, a metallic environment, and different tag geometries. The results demonstrated that the average angular error was less than 15° in most of the cases. We also observed that in an extreme metallic environment, the prediction accuracy degraded.

5.Acknowledgement The authors would like to thank IAITO Infotech for hardware support. The authors would also like to thank Dr. Buon Kiong Lau for his support.

6. Appendix The component list and budget is shown in Table 2. IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 235

2. S. Sritulanont, W. Viriya, and A. Hongmala, Method and System for Detection of Integrated Circuit Package Orientation in a Tape and Reel System, United States Patent No. 6475826 B1, November 2002. 3. J. Ouellette, Reject Bottle Detection and Ejection Mechanisms, United States Patent 20030105550 A1, Jun 2003. 4. J. Stricker, Product Packaging Arrangement Using Invisible Marking for Product Orientation, United States Patent No. 6370844 B1, April 2002. 5. J. García, A. Arriola, F. Casado, X. Chen, J. I. Sancho and D. Valderas, “Coverage and Read Range Comparison of Linearly and Circularly Polarised Radio Frequency Identification UltraHigh Frequency Tag Antennas,” IET Microwaves, Antennas & Propagation, 6, 9, June 2012, pp. 1070-1078. 6. N. M. Faudzi, M. T. Ali, I. Ismail, H. Jumaat and N. H. M. Sukaimi, “A Compact Dipole UHF-RFID Tag Antenna,” 2013 IEEE International RF and Microwave Conference (RFM), December 2013, pp. 314-317. 7. A. E. Abdulhadi and R. Abhari, “Tunable Compact Printed Monopole Antenna for Passive UHF RFID Tags,” 2012 IEEE International Symposium on Antennas and Propagation, July 2012, pp. 1-2. 8. A. E. Abdulhadi and R. Abhari, “Design and Experimental Evaluation of Miniaturized Monopole UHF RFID Tag Antennas,” IEEE Antennas and Wireless Propagation Letters, 11, 2012, pp. 248-251. 9. C. A. Balanis, Antenna Theory: Analysis and Design, Third Edition, New York, Wiley, 2005. 235

12/27/2014 2:03:51 PM

10. Z. Chen, X. Qing and H. L. Chung, “A Universal UHF RFID Reader Antenna,” IEEE Transactions on Microwave Theory and Techniques, 57, 5, May 2009, pp. 1275-1282.

Introducing the Authors

11. Y. K. Jung and B. Lee, “Dual-Band Circularly Polarized Microstrip RFID Reader Antenna Using Metamaterial BranchLine Coupler,” IEEE Transactions on Antennas and Propagation, AP-60, 2, February 2012, pp. 786-791. 12. H. Mosallaei and K. Sarabandi, “Antenna Miniaturization and Bandwidth Enhancement Using a Reactive Impedance Substrate,” IEEE Transactions on Antennas and Propagation, AP-52, 9, September 2004. 13. L. Akhoondzadeh, D. J. Kern, P. S. Hall and D. H. Werner, “Wideband Dipoles on Electromagnetic Bandgap Ground Planes,” IEEE Transactions on Antennas and Propagation, AP55, 9, September 2007.

Gaurangi Gupta received the Masters in Electrical Engineering from Indian Institute of Technology Kanpur, Kanpur, India. She is currently pursuing her PhD in Electrical Engineering from Indian Institute of Technology Kanpur. Her research interests include antenna design, radio-frequency integrated circuits, and RFID technology.

14. P. Deo, A. Mehta, D. Mirshekar-Syahkal, P. J. Massey and H. Nakano, “Thickness Reduction and Performance Enhancement of Steerable Square Loop Antenna Using Hybrid High Impedance Surface,” IEEE Transactions on Antennas and Propagation, AP-58, 5, May 2010. 15. S. Raza, M. A. Antoniades and G. V. Eleftheriades, “A Compact Low-Profile High-Impedance Surface for Use as an Antenna Ground Plane,” 2011 IEEE International Symposium on Antennas and Propagation, July 2011, pp. 1832-1835. 16. S. R. Best and D. L. Hanna, “ Design of a Broadband Dipole in Close Proximity to an EBG Ground Plane,” IEEE Antennas and Propagation Magazine, 50, 6, December 2008. 17. G. Gupta and A. R. Harish, “A Broadband Dipole on a Double Layered Via-less High Impedance Surface,” 2014 IEEE International Symposium on Antennas and Propagation, Memphis, USA, July 2014.

Bhanu Pratap Singh received the Bachelors in Electrical Engineering from Indian Institute of Technology Kanpur, Kanpur, India. He is currently working on a seed stage startup idea. His research interests include Web and mobile applications, algorithms and data science.

18. http://www.winnix.net/uploadfile/20140225153417773.pdf 19. http://www.ftdichip.com/Support/Documents/DataSheets/ ICs/DS_FT232BL_BQ.pdf 20. http://developer.android.com/ 21. http://docs.oracle.com/javase/7/docs/api/

236

AP_Mag_Dec_2014_Final.indd 236

Amrita Bal received the Masters in Electrical Engineering from the Indian Institute of Technology Kanpur, Kanpur, India. She is currently working as an Engineer in Qualcomm Technologies, Inc. Her research interests include LTE 4G communication, antenna design and RFID technology.

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

12/27/2014 2:03:51 PM

Deepam Kedia completed the Bachelors in Electrical Engineering from Indian Institute of Technology Kanpur, Kanpur, India. He is currently working as a Senior Analyst at Goldman Sachs. His research interests include analog circuit designs, machine learning algorithms, and data mining.

IEEE Antennas and Propagation Magazine, Vol. 56, No. 6, December 2014

AP_Mag_Dec_2014_Final.indd 237

A. R. Harish received the PhD in Electrical Engineering from the Indian Institute of Technology Kanpur, Kanpur, India. He was with Com Dev Wireless, Dunstable, UK, and was a visiting faculty at the University of Kansas. He is currently a professor with the Department of Electrical Engineering, Indian Institute of Technology Kanpur. His current research interests include antenna analysis, microwave measurements, microwave circuits, radio-frequency identification, and computational electromagnetics.

237

12/27/2014 2:03:52 PM