当前位置:首页 >> 信息与通信 >>

Implementation of an adaptive Leakage Cancellation Control for passive UHF RFID Readers


2011 IEEE International Conference on RFID

Implementation of an adaptive Leakage Cancellation Control for passive UHF RFID Readers
Iker Mayordomo and Josef Bernhard Fraunhofer

Institute for Integrated Circuits IIS Am Wolfsmantel 33, 91058 Erlangen, Germany iker.mayordomo@iis.fraunhofer.de, josef.bernhard@iis.fraunhofer.de

Abstract— In this paper the implementation of an automatic leakage cancellation for UHF (Ultra High Frequency) RFID (Radio Frequency Identi?cation) readers is presented. The system architecture and the control algorithms are described in detail. The proposed architecture has been completely implemented and measurements have been carried out for evaluation. The system is demonstrated to adaptively change the control signals to achieve an optimal leakage cancellation for every situation. It is shown that the Control Module keeps the leakage cancellation level between 41 dB and 46 dB, despite the leakage properties being changed by means of a phase shifter. It is also demonstrated that without such a module, the leakage power increases exponentially as the leakage properties change; e.g. from -41 dBm to -3 dBm with 30? phase shift. Finally, the system time response has been measured and the system operation has been evaluated in a re?ective environment.

[9] a closed-loop control algorithm has been proposed and simulated. Finally, a carrier suppression mechanism based on locked-loop architectures has been developed by [10]. In this paper, a fully automatic adaptive leakage cancellation control is presented. The control architecture is based on a microcontroller, which provides ?exibility for testing different control algorithms. The system has been also implemented and the concept is demonstrated by measurements. The paper is organized as follows: Section II presents the system architecture. Section III explains the developed control algorithms. Section IV describes the test scenario and presents the evaluation results. Finally, conclusions and future work are presented in section V. II. S YSTEM A RCHITECTURE A. The Leakage Cancellation Module (LCM) The leakage cancellation module that has been used for the presented implementation is basically made up of a vector modulator, an ampli?er and a directional coupler. As shown in Fig. 1, the module takes a sample of the transmission signal (leakage) and adapts it in phase an amplitude so that it has the same amplitude as the leakage in the received signal and 180? phase difference. Both signals, the received signal with the leakage and the leakage canceler, are combined by means of a directional coupler. The leakage-free output signal is afterwards sent to the RFID reader receiver. The theory behind leakage cancellation has already been explained for example by [2], [10]. The main component in the LCM is the vector modulator, which modi?es the leakage sample in phase and amplitude by means of I and Q control signals. In the present design, 40 dB amplitude range and 360? phase range is provided by this component. The ampli?er after the vector modulator is required to reach the power levels necessary to cancel the leakage. It must be noticed that this signal is attenuated 10 dB by the directional coupler. A directional coupler was used instead of a power combiner in order to reduce the insertion losses in the reception path at the cost of needing higher power levels of leakage canceler. Besides, the insertion of additional distortion and noise that would be directly added (after 10 dB attenuation) to the reception signal should be minimized. For that reason, low-noise components with high linearity and high-enough OP1dB (1 dB output compression point) should be chosen.

I. I NTRODUCTION NE of the main issues that must be overcome in the design of an UHF RFID reader is the transmission leakage into the receiver. Passive RFID readers transmit a continuous wave and receive the backscattered data from the RFID tag at the same time and at the same frequency. This leakage coming from the transmitter into the receiver results in problems such as very high dynamic range requirements, DC offsets at baseband in the receiver or degraded signalto-noise ratio because of the leakage phase noise [1], [2]. In order to avoid these issues and thus improve RFID readers performance, several leakage cancellation methods have been developed during the last years [2], [3], [4], [5]. These techniques have been demonstrated to remove the leakage in the receiver and performance improvements have been demonstrated [6]. However, these methods are rather static and the control is carried out manually. That means, the leakage cancellation works only for the actual conditions. In real applications, conditions can change at any moment, for example because of re?ections from objects surrounding the reader antenna, and thus an adaptive leakage cancellation is required. There are some publications in the ?eld of leakage cancellation that already deal with the implementation of an adaptive automatic control. In [7] a microcontroller has been included in the leakage cancellation architecture; however the automatic control has not been implemented yet. In [8] the proposed leakage cancellation control is based on storing references for different frequencies on a memory. In

O

978-1-4244-9606-8/11/$26.00 ?2011 IEEE

121

Tx/Rx Antenna Rx Signal without Leakage

Power Detector
2,5

Leakage Vector Modulator

Vout (V)

RFID Reader Transmitter

Tx Signal

Rx Signal

Leakage Sample

Leakage canceller

RFID Reader Receiver

2

1,5

Amp

I Ctrl

Q Ctrl

Leakage Cancellation Module (LCM)

1

0,5

0

Fig. 1. The leakage cancellation module (LCM) takes a sample of the transmission leakage and adapts it in phase and amplitude to cancel the transmission leakage. I and Q signals control the operation of the vector modulator.
RFID Reader Transmitter RFID Reader Receiver
Rx Signal without Leakage Tx Signal Leakage Sample Rx Signal

-90

-80

-70

-60

-50

-40 Pin (dBm)

-30

-20

-10

0

10

Fig. 3.

Measured power detector control voltage.

C. The Power Detector
Tx/Rx Antenna

LCM

Rx Signal Sample I Ctrl Q Ctrl

Power Detector Control Module
Ctrl Voltage

ADC

ADC

Microcontroller

Fig. 2. System architecture including the Control Module that provides the I and Q signals to the LCM.

Further details about the developed LCM architecture and its evaluation in a static environment (i.e. I and Q control signals are set manually) can be found in [6]. B. The Control Module As explained in the previous section, the LCM operation is controlled by the I and Q signals, which modify the amplitude and the phase of the leakage canceler. The Control Module, which is the core of the implementation presented in this paper, provides these two signals. Furthermore, it automatically and adaptively modi?es them always looking for the optimal cancellation point for the actual conditions. The architecture of the Control Module is shown in Fig. 2. In the implemented architecture, the sample of the transmitted signal (leakage) is provided to the LCM by means of a 10 dB directional coupler. Furthermore, this coupler also allows the simultaneous transmission and reception of RFID signals. This way, a circulator is no longer needed. The main advantage of this con?guration is the higher isolation provided by directional couplers compared to standard circulators. However, the received signal reaches the receiver after 10 dB attenuation, which increases the RFID reader receiver noise ?gure.

For the control algorithm to work properly, monitoring of the received signal after the leakage cancellation is necessary. As shown in Fig. 2, this is achieved by means of a directional coupler at the output of the LCM. This sample signal is converted to a voltage by a power detector. Afterwards this control voltage is digitalized by the microcontroller so that it can be used by the control algorithm. At this point, some assumptions regarding the power detection range were made. For the implemented architecture, and considering a typical output power level of 33 dBm e.r.p. (effective radiated power) and 27 dB isolation between transmitter and receiver (through the directional coupler), typical intrinsic leakage levels of 3 dBm were expected at the receiver input. Taking into account the 10 dB attenuation of the monitoring coupler, a leakage value of -7 dBm at the power detector input was expected. These values can vary in real applications mainly because of re?ections from surrounding objects or also because of antenna matching changes, which are typical in re?ective environments. Further information about the design criteria used for the LCM, what the main contributors to the leakage are and the estimations done above, can be found in [6]. For this application, the Analogue Devices AD8313 power detector was chosen. It features a 70 dB dynamic range with a slope of 20 mV/dB. The detection slope was measured and the results are shown in Fig. 3. As it can be seen, the detection linear range goes from -65 dBm to -5 dBm approximately. This means that the Control Module will be able to detect leakage levels between -55 dBm and 5 dBm, which in principle matched our power requirements estimations. In case other ranges are needed, for example because of higher leakage levels, the power detector slope can be modi?ed or a coupler with different coupling can be used. D. The Microcontroller The microcontroller chosen to implement the control algorithms is the ATMEGA-16 from ATMEL. An internal 10-bit ADC (Analogue to Digital Converter) is used to digitalize

122

90

Q
Leakage 180 Leakage canceler 0

Pmax

Variables Initialisation: I, Q, step

I

0

No

Input signal? Yes Find minimum for both I and Q

270

Fig. 4. By means of the I and Q control signals the leakage canceler must be modi?ed to have the same amplitude as the leakage and opposite phase.

Update values

the power detector control voltage. Besides, a SPI (Serial Peripheral Interface) interface is used to control the two DACs (Digital to Analogue Converter) that provide both the I and Q control voltages with 12 bits resolution. Internal DACs were not available in this microcontroller. Finally, for debug purposes the AVR Dragon debug platform from ATMEL was used, which was connected to a PC. III. T HE C ONTROL A LGORITHM Once the hardware architecture has been described, it is time to introduce the developed control algorithms. The main objective of these algorithms is to automatically ?nd the optimal cancellation point for the actual conditions. As already explained in Section II, both the I and Q control values that make the leakage canceler have the same amplitude and opposite phase as the leakage must be found. This is illustrated by Fig. 4. Furthermore, both the I and Q optimal values must be worked out continuously in real time based on the feedback provided by the power detector, which monitors the signal after the LCM. The implemented control algorithm is illustrated in Fig. 5. Firstly the values for I and Q are initialized. Before starting, the algorithm checks always if an input signal is present. This was achieved by de?ning an input power threshold. As a result, unstable states because of absence of input signal are avoided. Once the input signal is detected, the algorithm operation can begin. The main variable in the algorithm is the step. This variable de?nes the increment with which the I and Q will be modi?ed when looking for the optimal leakage cancellation point. This variable is initialized with a relatively big value and then it is decreased until it reaches its minimum value. The algorithm ?nds the I and the Q values that minimize the leakage for the given step size. It updates the I and Q values with the optimal ones and also updates the step size. If this is not the minimum step size, it reduces the step and looks again for the I and Q values that minimize the leakage. If now this is the minimum step size, the algorithm enters the optimal region. In this stage, the algorithm keeps the I and Q values constant and only monitors that the conditions remain unchanged. If the conditions do change, it updates the

Minimum step? Yes Optimal region: monitoring

No

Reduce step

No Estimate optimal step Yes Conditions changed?

Fig. 5.

The automatic control algorithm main part.

step value and begins again looking for the optimal I and Q values. This optimal region is explained in more detail below. One of the key points in the algorithm is the way it looks for the optimal I and Q values. An important assumption that is made is that only one minimum exists for both I and Q and thus only one optimal point, which was experimentally observed by operating the LCM manually. Fig. 7 shows the way the optimal values for I and Q are found. Firstly, the algorithm looks for the optimal point in the forward direction by incrementing the I or Q voltage with the current step size. It stops when the next value is worse than the actual value and it assumes that this is the optimal value. If this happens in the ?rst iteration, then the algorithm looks in the other direction by decreasing the I or Q voltage and stops when the results get worse, just as before. The algorithm carries out the same procedure for I and Q and when both optimal values have been found it reduces the step size and begins again. This method is repeated until the minimum step size is reached. When that happens, the algorithm assumes that the optimal point has been found and enters the optimal region. This methodology can be better understood by means of Fig. 7. The algorithm starts with step1 and measures the leakage power at points 1, 2, 3 and 4. Then it realizes that point 3

123

Look for minimum in forward direction No Minimum found? Yes
Yes

Minimum step? Yes Power Monitoring: I & Q unchanged

Minimum = initial point?

Variation within range? No Adjust step according to variation

Yes Look for minimum in backward direction No

Leave optimal point monitoring

Update values

Fig. 8. The optimal region algorithm where the I and Q values are kept constant. The algorithm only wakes up when a change in the conditions is detected.

Fig. 6.

The “?nd minimum” function.

1

2

3/9

8 10 5

7

4/6

Step1

Step4 Step3 Step2

Fig. 7. The algorithm ?nds the optimal value by progressively reducing the step size.

is an optimal point for the actual step size. The step is then decreased (to the half in this case), and it measures points 5 and 6. Then it decides point 5 is an optimal point for this step and the step is reduced again. This time, the algorithm measures point 7 and notices that in this direction there is no improvement, so it decides to go in the other direction and measures points 8 and 9. It stops at point 8 and reduces the step again. With this step size the algorithm ?nds the optimum at point 10. When the algorithm reaches the minimum step size, it enters the optimal region. Here the algorithm only monitors that the conditions have not changed and keep the I and Q values constant. The way the optimal region works is

illustrated by means of Fig. 8. The algorithm keeps I and Q unchanged until it notices that the conditions have changed. This is done by measuring the value of the actual leakage and comparing it with the optimal value. If the difference is bigger than some prede?ned range, then it is considered that the conditions have changed. In our application, we set this limit to 5 dB difference between the optimal value and the actual value. For the algorithm to react promptly to changes in the conditions, instead of starting the algorithm from the beginning with the biggest step size, the new step size is worked out taking into account how big the difference was. The reason for that is that the time the algorithm will need to ?nd the optimal value again will depend on the initial step size. If the difference was big, then we need a big step size to ?nd the optimal point quickly; if the difference was small, then we can use a small step size. This way, big leakage measurements that can interfere with the RFID communication are avoided. Finally, some limits must also be set to the control algorithm to make sure that it does not exceed the maximum/minimum values allowed by the vector modulator. In our case, the control voltage range is between 0.5 V and 2.5 V. Furthermore, it must be also somehow supervised that the control algorithm does not measure leakage levels that can saturate the RFID receiver. This can happen for example if too big step sizes are used when looking for the optimal I and Q values. IV. M EASUREMENTS AND RESULTS In order to test the performance of the proposed automatic leakage control, the test system illustrated by Fig. 9 was implemented. The main idea is to change the phase of the leakage with respect to the leakage sample to check if the

124

Signal generator

Tx Signal PA Rx Signal without Leakage Leakage Sample Phase shifter LCM Rx Signal Sample Control Module I Q

60

50
Leakage Cancellation (dB)

55 50 45 40 35 30 25 20 0 30 60 90 120 150 180 210 240 270 300 Added delay (° )

Spectrum analyzer

Leakage

Phase control

Fig. 9. Test environment for the evaluation of the implemented system. A phase shifter is placed in the signal reception path before the LCM.

Fig. 11. Leakage cancellation for different leakage conditions. The phase difference is changed in 30? steps for a 300? range.
10 0 -10 With automatic LCM -20 -30 -40 -50 -60 0 30 60 90 120 150 180 210 240 270 300 Added delay (° ) With static LCM Wihtout LCM

Fig. 10. Measurement set-up including the fabricated boards, the phase shifter and the AVR Dragon debug module.

Fig. 12. Leakage power for different leakage conditions. The phase difference is changed in 30? steps for a 300? range.

algorithm can ?nd the new optimal point as this phase changes. As explained before, the algorithm decisions are based on the feedback provided by the power detector, which monitors the received signal and thus the actual leakage level. The received signal is characterized by means of a spectrum analyzer, which also allows us to check the operation of the control algorithm. The signal generator transmits a continuous carrier at 868 MHz; 30 dBm output power are reached by means of a PA (Power Ampli?er) module specially designed for RFID applications [11]. This module provides a transmission signal sample as well as the possibility to be operated either in mono- or bi-static mode. The coupler isolation is about 27 dB for the frequency band and thus leakage values of 3 dBm are expected at the receiver input in mono-static mode. The measurement set-up implemented in the laboratory is shown by Fig. 10. A phase shifter was placed between the coupler and the LCM to modify the leakage phase. This phase shifter allows introducing a phase shift in the range of 0? - 312? at 868 MHz. The phase shift was changed in 30? steps between 0? and 300? and the leakage cancellation was measured when the algorithm reached the optimal region (minimum step size). Leakage cancellation is de?ned as the

difference between the initial leakage power in the system and the remaining leakage power after leakage cancellation, as shown in Eq. 1. Leakcanc [dB] = Leakinit [dBm] ? LeakLCM [dBm]. (1) As algorithm parameters, an initial step size of 0.5 V and a minimum step size of 2 mV were chosen. The algorithm started with 1 V voltage for both I and Q. The results are shown in Fig. 11. It can be noticed that the control algorithm achieves a leakage cancellation between 41 dB and 46 dB for the whole 300? range. It is also interesting to see that different cancellation levels are achieved for the different phase shift values. However, the maximum difference between the maximum and the minimum levels is only of 5 dB, which indicates that the algorithm is quite stable. It is worth pointing out that this was carried out without reseting the system; the algorithm reacted to the changes in the conditions and found again the optimal point. Another interesting measurement was carried out to check what happens if an automatic control mechanism is not implemented. In other words, see what happens if after reaching the optimal point for certain conditions, these conditions change and however the I and Q values are kept constant. The results are shown by Fig. 12. It can be seen that by implementing the automatic control

125

Leakage Power (dBm)

2500 2300 2100 I & Q Voltages (mV) 1900 1700 1500 1300 1100 900 700 500 0 30 60 90 120 150 Added delay (° ) 180 210 240 270 300 I Voltage Q Voltage

Fig. 13.
input

Evolution of the I and Q control signals for a 300? range.
leakage leakage sample

+ +
Leakage canceller

output output sample

LCM

Fig. 15. Time response measured from start point until the optimal region.

Feedback loop

Fig. 14.

Control system block diagram.

proposed by this paper the leakage power is always kept around -40 dBm. However, if I and Q voltages are kept with the out-of-date optimal values, the leakage power increases exponentially as the leakage phase changes. For example, for only 30? phase shift, the leakage increases from -41 dBm to -3 dBm, which most probably would degrade the RFID reader performance. Furthermore, from 60? phase shift the leakage level is even higher than the intrinsic system leakage (3 dBm). That means that the leakage canceler is being added to the intrinsic leakage, which makes the situation even worse than without LCM. In order to demonstrate how the control algorithm follows the changes in the leakage properties, Fig. 13 illustrates the evolution of both I and Q optimal control voltages as the phase changes. It is worth pointing out that 0.5 V and 2.5 V are the minimum and maximum control voltages allowed by the vector modulator, respectively. It was also checked that once the algorithm reaches the minimum step size, it enters the optimal region where the I and Q signals are kept constant. This maintains the leakage in the optimal value and avoids the leakage from springing from one value to another all the time. It was also checked that the algorithm can be woken up by changing the leakage characteristics (either phase or amplitude). The control reacts immediately and looks again for the optimal values. The system time response was also measured. The time response can be de?ned as the time the system needs to reach the optimal region either from the start point or after some alteration in the conditions. This measurement was done following the set-up illustrated by Fig. 14. After reseting the system, the leakage power along time was measured by means of a spectrum analyzer. The measurement was triggered as soon as the input signal was

applied to the system. That means, we measured the time it takes the system to reach the optimal region after reset. The results are shown in Fig. 15. As it can be seen, it takes approximately 330 ms to reach the optimal region after reset. It can also be seen that some minimum points are found before reaching the optimal region. However, they are not stored and the system kind of loose them when changing the step size. The reason for that can be that with bigger step size it is dif?cult to tune the measurements and when returning to the same I and Q values the values do not correspond to a minimum anymore. Anyway, after reducing the step size the algorithm ?nds again the optimal point. Once the optimal region has been reached, I and Q voltages are kept constant and the leakage power is kept at -41 dBm, which corresponds to 44 dB leakage cancellation. It was also considered interesting to check what the behavior of the system is when a typical UHF RFID reader antenna is connected to the system output, instead of the 50 ? termination shown in Fig. 9. This way, how re?ections can in?uence the algorithm performance could also be tested. For this purpose, a circularly polarized antenna with a 5 m long cable from Alien Technology was chosen. This antenna can operate in the whole 865 - 965 MHz band and has a maximum gain of 5.5 dBiL. The measurements were done in a laboratory, where any kind of re?ections could take place. Firstly, the intrinsic leakage level was measured. Results showed a little higher leakage level than with a 50 ? termination, but this increment was less than 1 dB. It was also veri?ed that changing the radiation direction of the antenna, or just moving us around it, increased and decreased the leakage level, because of antenna mismatching and re?ections. Furthermore, antenna mismatching can also alter the coupler isolation, which will directly modify the leakage amplitude and/or phase. Under these conditions, the developed control module was demonstrated to be able to reduce the leakage to the same levels presented above (i.e. around -40 dBm leakage level).

126

In contrast to the ?rst test scenario, where after reaching the optimal region the leakage level remained stable, in this second scenario the leakage power ranged approximately between -60 dBm and -33 dBm while changes around the antenna where taking place. The control algorithm was continuously woken up in order to adapt itself to the new leakage properties (i.e. different phase and/or different amplitude). However, when the situation got stable again, it returned to the typical cancellation level and remained constant. V. C ONCLUSIONS In real RFID applications, where conditions can change at any moment, the implementation of an automatic leakage cancellation control is essential. In this paper, a Control Module has been presented, which automatically works out the optimal control signals that minimize the transmission leakage in the RFID receiver. After introducing the hardware platform and the basics of leakage cancellation, the core of the system, which is the control algorithm, has been presented and described in detail. It is based on the assumption that only an optimal I and Q value exists, which was observed experimentally. The algorithm tunes the search by reducing the step variable until it ?nds the optimal values for the minimum size. The de?nition of an optimal region, where I and Q voltages are kept constant, has been also recommended to prevent the algorithm from springing continuously from one value to another. Any change in the conditions will wake up the control algorithm and the new optimal point will be found. The system has been evaluated by measurements. The control module achieves a leakage cancellation range between 41 dB and 46 dB for a 300? range by modifying the I and Q control signals accordingly. The need for such a control system was also demonstrated. If the values are not updated, small changes in the leakage phase greatly increase the leakage level. Furthermore, it can even increase the intrinsic leakage level of the system. The system time response was measured from reset until the optimal region was reached. It was shown to be 330 ms, which can be considered fast enough for typical applications. Measurements with an RFID reader antenna showed that the Control Module can also work in re?ective environments and adapt itself to changing situations. In comparison with other state-of-the-art pieces of research, this work focuses on the leakage cancellation control algorithms, which are described in detail. Furthermore, a fully-automatic 44 dB leakage cancellation, which is equivalent to a 71 dB Tx-to-Rx isolation, is achieved. In [7], the microcontroller included in the leakage cancellation architecture is used to control both the variable attenuator and the phase shifter. This way, a 55 dB Tx-to-Rx isolation is achieved. However, no intelligence or adaptive control algorithms have been implemented. In [8] similar results to [7] are obtained, but the microcontroller has a table where the optimal values for different frequencies are stored. However, this solution only eliminates the system intrinsic leakage but cannot adapt itself against changes around the reader

antenna. In [10], a carrier suppression mechanism based on locked-loop architectures has been developed. Results show a high Tx-to-Rx isolation of around 70 to 80 dB. However, no details are given about how the gain and phase detector output voltages are achieved. Our future work plan includes measurements and evaluation in real RFID applications, with objects moving around the reader antenna that continuously modify the leakage properties because of re?ections. Probably, for such situations the algorithms must still be improved in terms of time response. It can also be worth it to allow a higher difference range in the optimal region before activating the control algorithm in order to increase the system stability at the cost of allowing higher levels of leakage. VI. ACKNOWLEDGMENTS This work was partly supported by the Bavarian Government (Center for Intelligent Objects ZIO [12]). The authors are also grateful for the fruitful collaboration with TECNUN (San Sebastian, Spain). Finally, the authors would like to thank Xabier Moraza (TECNUN) and Rafael Psiuk (Fraunhofer IIS) for their valuable contribution. R EFERENCES
[1] K. Finkenzeller, RFID Handbook: Fundamentals and Applications in Contactless Smart Cards, Radio Frequency Identi?cation and NearField Communication, 3rd ed. John Wiley and Sons, 2010. [2] G. Lasser, R. Langwieser, and A. Scholtz, “Broadband suppression properties of active leaking carrier cancellers,” in IEEE International Conference on RFID, Orlando, USA, 2009. [3] J. Lee, J. Choi, K. H. Lee, B. Kim, M. Jeong, Y. Cho, H. Yoo, K. Yang, S. Kim, S.-M. Moon, J.-Y. Lee, S. Park, W. Kong, J. Kim, T.-J. Lee, B.-E. Kim, and B.-K. Ko, “A UHF mobile RFID reader IC with self-leakage canceller,” in Radio Frequency Integrated Circuits (RFIC) Symposium, 2007 IEEE, 2007, pp. 273 –276. [4] T. Brauner and X. Zhao, “A novel carrier suppression method for RFID,” IEEE Microwave and Wireless Components Letters, vol. 19, no. 3, pp. 128–130, 2009. [5] P. Pursula, M. Kiviranta, and H. Seppa, “UHF RFID reader with re?ected power canceller,” Microwave and Wireless Components Letters, IEEE, vol. 19, no. 1, pp. 48 –50, jan. 2009. [6] R. Psiuk, I. Mayordomo, and J. Bernhard, “Design criteria for a leakage cancellation module and evaluation on an RFID reader platform,” in The Third International EURASIP Workshop on RFID Technology RFID, Spain, 2010. [7] J.-W. Jung, H.-H. Roh, J.-C. Kim, H.-G. Kwak, M. S. Jeong, and J.-S. Park, “Tx leakage cancellation via a micro controller and high tx-to-rx isolations covering an UHF RFID frequency band of 908-914 MHz,” Microwave and Wireless Components Letters, IEEE, vol. 18, no. 10, pp. 710 –712, 2008. [8] J.-W. Jung, H.-H. Roh, H.-G. Kwak, M. S. Jeong, and J.-S. Park, “Adaptive TRX isolation scheme by using TX leakage canceller at variable frequency,” Microwave and optical technologys Letters, vol. 50, no. 08, pp. 2043 – 2045, 2008. [9] J. Y. Wang, B. Lv, W. Z. Cui, W. Ma, J. T. Huangfu, and L. X. Ran, “Isolation enhancement based on adaptive leakage cancellation,” in Progress In Electromagnetics Research Symposium Proceedings, China, 2010, pp. 1059–1063. [10] D. Villame and J. Marciano, “Carrier suppression locked loop mechanism for UHF RFID readers,” in RFID, 2010 IEEE International Conference on, april 2010, pp. 141 –145. [11] I. Mayordomo, A. Jaschke, H. Solar, and J. Bernhard, “A wideband power module for passive UHF RFID readers,” in The Third International EURASIP Workshop on RFID Technology, Spain, 2010. [12] Center for Intelligent Objects ZIO. Fraunhofer IIS. [Online]. Available: http://www.zio.fraunhofer.de/EN/index.jsp

127


相关文章:
更多相关标签: