Data Transmission Using Transmitter Side Channel Estimation in Wireless Power Transfer System

Transcription

1 IEICE TRANS. FUNDAMENTALS, VOL.E98 A, NO.2 FEBRUARY PAPER Special Section on Wideband Systems Data Transmission Using Transmitter Side Channel Estimation in Wireless Power Transfer System Kazuki SUGENO a), Student Member, Yukitoshi SANADA b), Senior Member, and Mamiko INAMORI c), Member SUMMARY In recent years, wireless power transfer has been attracting a great deal of attention. In order to realize efficient power transfer, it is necessary to communicate data such as a frequency, required power, or error tolerance. In the proposed system, because of the use of the same antennas for power transmission and data transmission, the frequency response of a channel for the data transmission changes owing to load fluctuation and the distance between antennas. This paper investigates data transmission performance in the wireless power transfer system with frequency response estimation at the transmitter side. The numerical results obtained through computer simulation show that the proposed scheme can estimate the frequency response of the channel and the difference between the expected bit error rate (BER) and the BER with the estimation error is less than 0.5 db at the BER of key words: wireless power transfer, load fluctuation, OFDM, communication 1. Introduction Wireless power transfer enables devices to receive power supply without inconvenience rather than wired power connection [1] [3]. It is possible to use the wireless power transfer system for household electric appliances and electric vehicles [4]. Products with the power consumption of less than 50 W have been commercialized and the systems with the power consumption of over 50 W and electric vehicles will be commercialized in the future. Wireless power transfer has been implemented with three typical techniques. Among these techniques, a magnetic resonant coupling system has especially been attracting a great deal of attention [5] [7]. This paper focuses on a magnetic resonant coupling system. To transfer power efficiently and safely in a wireless power transfer system, data such as a frequency, required power, and error tolerance need to be transmitted reliably [8], [9]. However, if loop coils are used as antennas for data transmission, the frequency response of a channel is affected by the fluctuation of a load in a receiver and the distance between the antennas [10], [11]. The frequency response of the channel for exchanging data has to be detected at the transmitter side without feedback from the receiver at least at the beginning of the communication [12], [13]. Therefore, an estimation scheme for the frequency response of the channel without feedback from the receiver and a channel assignment scheme with the estimated frequency response is proposed in this paper. The frequency response is estimated with a directional coupler and a power meter in the transmitter side [12], [13]. In [12], the reflection coefficient is also measured in the wireless power transfer system. This system does not assume data transmission and the frequency response of the channel is not estimated. In the proposed system, the reflected signal from the transmit antenna is detected and the frequency response is estimated. The bandwidth to transmit the data is then decided. To satisfy the conditions for high speed and reliable communications, orthogonal frequency division multiplexing (OFDM) is applied as a modulation scheme. The bit error rate (BER) performance is evaluated through computer simulation. This paper is organized as follows. Section 2 introduces the system model. In Sect. 3, numerical results obtained through computer simulation are presented. Section 4 gives conclusions. 2. System Model 2.1 Data Transmission System The system model for data transmission is shown in Fig. 1. In this system, the data transmission block and the power transmission block in the transmitter are selected by the switch at the transmitter side. Loop coils are used as transmit and receive antennas. The reflected signal from the loop antenna is measured with the directional coupler and Manuscript received April 8, Manuscript revised August 29, The authors are with the Dept. of Electronics and Electrical Engineering, Keio University, Yokohama-shi, Japan. The author is with the Dept. of Electrical and Electronic Engineering, Tokai University, Hiratsuka-shi, Japan. a) sugeno@snd.elec.keio.ac.jp b) sanada@elec.keio.ac.jp c) inamori@tokai-u.jp DOI: /transfun.E98.A.589 Fig. 1 System model for wireless power transfer. Copyright c 2015 The Institute of Electronics, Information and Communication Engineers

2 590 IEICE TRANS. FUNDAMENTALS, VOL.E98 A, NO.2 FEBRUARY 2015 Fig. 4 Equivalent circuit of the wireless power transfer system. Fig. 2 Wireless power transfer system. Fig. 3 Equivalent circuit of single coil. Fig. 5 Top view of the loop coils. Table 1 Equivalent circuit parameters of single coil. Rs[Ω] L[μH] Cp[pF] Z[Ω] the power meter at the transmitter side. The bandwidth of the channel is determined on the basis of the measurement. The analog-to-digital converter (ADC) is inserted in parallel with the load and picks up the voltage at the load to receive the data signal during a data transmission period (that is separated from the power transfer period). Since the input impedance of the ADC is large enough, it does not affect the total amount of the output load. 2.2 Equivalent Circuit Model The assumed model of the loop coils for wireless power transfer in this paper is shown in Fig. 2. The loop coils with the diameter of 300 mm are used as the transmit and receive antennas and the resonance frequency is set to be MHz [14]. The equivalent circuit of the single coil is shown in Fig. 3 and the circuit parameters of the coil is shown in Table 1 [14]. The equivalent circuit of the wireless power transfer system is shown in Fig. 4. Z 0 and Z L represent the characteristic impedance and the load impedance. Rs 1, Cp 1, Rs 2,andCp 2 represent the resistance of the coil at the primary side, the capacitance of the coil at the primary side, the resistance of the coil at the secondary side, and the capacitance of the coil at the secondary side, respectively. Since the same coils are used as the transmit and receive antennas, it is assumed that Rs 1 = Rs 2 = Rs, Cp 1 = Cp 2 = Cp,andL 1 = L 2 = L. The mutual inductance, M, is derived with k and L as follows. M = k L 1 L 2 = kl (1) where k is the coupling coefficient. The top view of the loop coils which are used in this paper is shown in Fig. 5. The axes of the transmit coil, C 1,and Fig. 6 Side view and development view of loop coils. the receive coil, C 2, are located at the distance of d on the x axis. The distance between the elements of the coils, r 1 and r 2, is calculated on the x and y axes from the following equations. dx = d + R 2 cos θ 2 R 1 cos θ 1, (2) dy = R 2 sin θ 2 R 1 sin θ 1, (3) where θ 1 and θ 2 are the central angles, and R 1 and R 2 are the radii of the transmit coil and the receive coil, respectively. The side view and the development view of the loop coils are presented in Fig. 6. dc is the distance between the transmit coil and the receive coil. The distance between the elements of the coils, r 1 and r 2 in the z-axis, is calculated as follows. dz = dc + N 1 P 1 1 2π (P 1θ 1 P 2 θ 2 ) (4) where N 1 and N 2 represent the numbers of the turns and P 1 and P 2 represent the pitches in the transmit coil and the receive coil, respectively. From the above, the mutual inductance can be given as follows [15]. M = μ 0 4π 2π N1 2π N2 0 0 R 1 R 2 cos(θ 1 θ 2 ) dx2 + dy 2 + dz 2 dθ 2dθ 1. (5) The relationship of the distance between the coils and the

3 SUGENO et al.: DATA TRANSMISSION USING TRANSMITTER SIDE CHANNEL ESTIMATION IN WIRELESS POWER TRANSFER SYSTEM 591 Fig. 7 Distance between coils vs. coupling coefficient. Fig. 9 U-type distribution. Table 2 Coupling coefficient k. k Distance [cm] coefficient in the primary side, the distance between the antennas and the amount of the load in the secondary side are required. However, it is not possible to obtain those parameters in the primary side if there is no feedback from the secondary side. The estimation scheme employed in this paper does not require the feedback from the secondary side. The following relationship holds, S S 21 2 < 1 (6) Fig. 8 System model at transmitter side. coupling coefficient obtained from Eq. (5) is shown in Fig. 7. This figure shows the coupling coefficient for the casewhen the centers of the coils are at the same location from the top view, i.e. d = 0 in Fig. 5, and the distance means the length of the space between the coils, d c. In this paper, in order to evaluate the BER performance with different distances between the antennas, two coupling coefficients in Table 2 are employed. 2.3 Frequency Response Estimation The block diagram of a frequency response estimation scheme in the primary side is shown in Fig. 8. The directional coupler is used for the separation of the reflected wave and the traveling wave of the transmit signal that is generated by the signal generator. The power meter is used for the measurement of the reflected wave and the traveling wave as shown in Fig. 8. By detecting the reflected wave from the antenna and the traveling wave to the antenna at the transmitter the forward reflection coefficient, S 11, is measured. The forward transmission coefficient, S 21, of the channel is affected by the distance between the antennas and the fluctuation of the load. Communication performance is dependence on the value of S 21. In order to accurately detect the where S 11 is the reflection coefficient that can be measured at the primary side and the inequality is because of the loss of the system including radiation loss and conductive resistance. Since the bandwidth of the channel is limited to about 5 percent of the center frequency as shown in Sect. 4, the system loss is assumed to be constant over the measured channel [16]. The exact value of S 21 is not required for the estimation of the center frequency and the bandwidth of the communication channel, and the frequency dependence of S 21 can then be estimated in accordance with S 11. For the measurement of S 11 the power meter and the directional coupler are used. Uncertainty in the measurement using the directional coupler may occur [17], [18]. This paper assumes the mismatch between the antennas and the directional coupler and it causes a measurement error. The error due to the mismatch is derived as the probability density function of the U-type distribution with the limit of G as shown in Fig. 9 in which f (u) is the occurrence probability of the uncertainty, u [17], [18]. The limit, G, can be given by the following equation. G = 20 log 10(1 Γ D Γ A ) (7) where Γ D is the reflection coefficient of the directional coupler and Γ A is the reflection coefficient of the antenna. Γ A is given by the following equation. Γ A = 10 S (8) The uncertainty, u, effects on the forward transmission coefficient as S 11 = S 11 + u (9)

4 592 IEICE TRANS. FUNDAMENTALS, VOL.E98 A, NO.2 FEBRUARY 2015 where S 11 is the coefficient at the output of the power meter. The probability density function of the uncertainty has U- type distribution as shown in Fig. 9 and is then given as α(u β) 2, u G/2 f (u) = (10) 0, otherwise where 12 α = (g + g ), 3 (11) β = g + g +, 2 (12) g = G 2 is the lower limit of u, andg+ = G 2 is the upper limit of u [17]. 2.4 Communication System According to the distance between the antennas and the load fluctuation, resonance splitting may occur. When the resonance splitting occurs, the local minimum of the response at the vicinity of the resonance frequency appears. The center of the transmission channel is set to the frequency that has the maximum response. The transmission bandwidth is decided as the frequency range with the response above the half of its maximum as shown in Fig. 10. In the communication model for data transmission, the equivalent circuit used in the transmit and receive antennas is modeled as a band pass filter (BPF). OFDM is employed for data transmission due to frequency selectivity of the channel response. Suppose the data symbol on the mth subcarrier is S [m] (m = 0,..., N 1), the OFDM symbol is given as u[n] = 1 N 1 ( S [m]exp j2π nm ) N N m=0 (13) where n (n = 0,..., N 1) is the time index and N is the number of the subcarriers. The cyclic prefix is appended as a guard interval before data transmission. The baseband signal at the output of the filter is given by x(t) = P 1 n=0 u[n]h t(t nt s )whereh t (t) is the impulse response of the transmit filter, P is the length of the OFDM symbol including the cyclic prefix, and T s is the Nyquist sampling interval. In this system, the antennas are fixed and fading is not assumed. The received signal is given as P 1 y(t) = u[n]h(t nt s ) + v(t) (14) n=0 where v(t) is the additive white Gaussian noise (AWGN), h(t) is the impulse response of the channel including the transmit and receive antennas. The frequency response on the mth subcarrier, H[m], is equal to the forward transmission coefficient, S 21. Zero-forcing is used for the demodulation of the received signal as follows. Ŝ [m] = Y[m] H[m] where Y[m] = 1 N 1 ( y[n]exp j2π nm ) N N n=0 (15) (16) and y[n] is the sampled version of the received signal that is given as y[n] = y(nt s ). (17) 3. Numerical Results 3.1 Simulation Model BER performance is evaluated through computer simulation. The simulation conditions are shown in Table 3. Data bits are modulated with quadrature phase shift keying (QPSK) on each subcarrier. The number of data subcarriers is set to 64. The bandwidth of the OFDM signal is adjusted to the channel bandwidth that is estimated by using the power meter. The guard interval is set to 1/4ofthesymbol duration. The response of the communication channel estimated at the receiver side is assumed to be ideally obtained with pilot symbols appended at the beginning of data transmission. On the other hand, Ideal in the performance figures implies that those curves are obtained through computer simulation with the assumptions of no system loss in Eq. (6) and no uncertainty in Eq. (9) by means of the equivalent circuit presented in Fig Limit of Uncertainty Distribution The limit of the uncertainty, G, is given from Eq. (7), in which Γ D is assumed to be 0.5 and Γ A is derived from Eq. (8) Fig. 10 Channel assignment. Table 3 Simulation conditions. Modulation scheme 1st : QPSK 2nd : OFDM Number of data subcarriers 64 Channel estimation in receiver Ideal Number of trials 1,000,000

5 SUGENO et al.: DATA TRANSMISSION USING TRANSMITTER SIDE CHANNEL ESTIMATION IN WIRELESS POWER TRANSFER SYSTEM 593 Fig. 11 Frequency response of channel S 21 (k = 0.037). Fig. 13 Frequency response of channel S 21 (k = 0.037). Fig. 12 Frequency response of channel S 21 (k = 0.037). Fig. 14 Frequency response of channel S 21 (k = 0.150). Fig. 15 Frequency response of channel S 21 (k = 0.150). [17]. From Eq. (8), for the calculation of ΓA, the forward reflection coefficient, S 11, is then simulated by a circuit simulator with the equivalent circuit presented in Fig Estimation of Frequency Response To evaluate the effect on the bandwidth of the channel with load fluctuation, the load impedance, ZL, shown in Fig. 4 is varied from 5 to 200 Ω with each coupling coefficient, k. The frequency responses of the channel are shown in Figs The estimated responses with the uncertainty in the directional coupler are also shown in Figs Tables 4 and 5 show the 3 db channel bandwidth, W, the center frequency, fc, and their estimations for k = and with various load impedance values. As shown in these figures the frequency response of the channel are accurately estimated in the transmitter side. The accuracy within the channel bandwidth is better with the larger load impedance values such as ZL > 30 Ω for any coupling coefficients though the noise floor appears at the normalized forward reflection coefficient of around 8 db. If the load impedance is less than 30 Ω, 1 2 db differences are observed within the channel bandwidth. However, the center frequency and the 3 db bandwidth of the channel can be obtained accurately

6 594 IEICE TRANS. FUNDAMENTALS, VOL.E98 A, NO.2 FEBRUARY 2015 Fig. 16 Frequency response of channel S 21 (k = 0.150). Fig. 17 BER vs. SE b /N 0 (k = 0.037). Table 4 Channel bandwidth and center frequency at k = Ideal Estimation Z L [Ω] W [MHz] f c [MHz] W [MHz] f c [MHz] Table 5 Channel bandwidth and center frequency at k = Ideal Estimation Z L [Ω] W [MHz] f c [MHz] W [MHz] f c [MHz] Fig. 18 BER vs. SE b /N 0 (k = 0.037). with the estimated frequency response as observed in Tables 4 and BER Performance Figures show the BER performance with the coupling coefficients of k = and k = SE b /N 0 is defined as the bit energy at the transmitter side versus the noise spectrum density at the receiver side. This parameter includes the transmission loss between the loop antennas. In Figs. 17 and 19, the BER curves for k = with the frequency response estimation show db degradation as compared to those with ideal channel estimation at the BER of From Table 4, this is because the estimation bandwidth is wider than the ideal one. The same tendency can be Fig. 19 BER vs. SE b /N 0 (k = 0.037). observed with the coupling coefficient of k = However, the difference is smaller since the errors on the estimation of the 3 db channel bandwidth are smaller for the

7 SUGENO et al.: DATA TRANSMISSION USING TRANSMITTER SIDE CHANNEL ESTIMATION IN WIRELESS POWER TRANSFER SYSTEM Conclusions Fig. 20 BER vs. SE b /N 0 (k = 0.150). In this paper, data transmission for wireless power transfer systems has been investigated. In order to use the loop coils of wireless power transfer as the antennas for data transmission, the frequency response of the channel has to be estimated at the transmitter side. The frequency response changes according to the load fluctuation and the distance between the coils. The estimation scheme for the frequency response of the channel at the transmitter side has been proposed in this paper. The proposed scheme uses the directional coupler and the power meter in the transmitter side. The estimation of the frequency response is carried out by measuring the reflected wave and the traveling wave of the data signal in the transmitter. The estimation is not perfect because of the mismatch between the antennas and the directional coupler. However, the estimation in terms of the 3 db channel bandwidths and the center frequencies are accurate enough especially if the load impedance is more than 30 Ω. The BER performance of the OFDM system with the estimated frequency response has also been evaluated. The numerical results obtained through computer simulation have shown that the difference between the expected BER with ideal estimation and the BER with the estimation error is less than 0.5 db at the BER of Acknowledgement Fig. 21 BER vs. SE b /N 0 (k = 0.150). This work is supported in part by a Grant-in-Aid for Young Scientists (B) under Grant No from the Ministry of Education, Culture, Sport, Science, and Technology and Keio Gijuku Academic Development Funds in Japan. References Fig. 22 BER vs. SE b /N 0 (k = 0.150). coupling coefficient of k = [1] H. Kim and H. Lee, Design of an integrated wireless power transfer system with high power transfer efficiency and compact structure, th European Conference on Antennas and Propagation, pp , March [2] K. O Brien, R. Teichmann, and H. Gueldner, Magnetic field generation in an inductively coupled radio-frequency power transmission system, 2006 IEEE Power Electronics Specialists Conference, June [3] W. Zheng, X. Pi, X. Zheng, H. Liu, X. Xiao, and C. Peng, A wireless energy transmission system based on electromagnetism induction for remote controlled capsule, 2008 Automation Congress, Sept [4] T. Imura and Y. Hori, Maximizing air gap and efficiency of magnetic resonant coupling for wireless power transfer using equivalent circuit and Neumann formula, IEEE Trans. Ind. Electron., vol.58, no.10, pp , Oct [5] H. Hirayama, H. Yamada, N. Kikuma, and K. Sakakibara, Coupledresonant wireless power transfer technology from the viewpoint of electro-magnetic field, IEICE Technical Report, WPT , July 2013 (in Japanese). [6] A. Karalis, Efficient wireless non-radiative midrange energy transfer, Annals of Physics, vol.323, pp.34 48, April [7] A. Kurs, Wireless power transfer via strongly coupled magnetic resonances, Science Express, vol.317, no.5834, pp.83 86, 2007.

8 596 IEICE TRANS. FUNDAMENTALS, VOL.E98 A, NO.2 FEBRUARY 2015 [8] W.X. Zhong, X. Liu, and S.Y.R. Hui, A novel single-layer winding array and receiver coil structure for contactless battery charging systems with free-positioning and localized charging features, IEEE Trans. Ind. Electron., vol.58, no.9, pp , Sept [9] D. Wageningen and T. Staring, The Qi wireless power standard, 14th International Power Electronics and Motion Control Conference, pp.s15-25 S15-32, [10] K. Sugeno, S. Noguchi, M. Inamori, and Y. Sanada, Effect of load fluctuation in data transmission for wireless power transfer, IEICE Trans. Fundamentals, vol.e96-a, no.5, pp , May [11] T. Sugiura, Y. Suzuki, N. Sakai, T. Wuren, and T. Ohira, High efficiency RF rectifier for via-wheel power transfer to mobility scooter on electrified roadway, IEICE Technical Report, WPT , Dec (in Japanese). [12] K. Onizuka, N. Oodachi, H. Shoki, and O. Watanabe, Coupling factor estimation for efficiency optimization on wireless power transfer, IEICE General Conf., B-1-29, March 2010 (in Japanese). [13] M. Tsuboka, J. Vissuta, T. Imura, H. Fujimoto, and Y. Hori, Secondary parameter estimation for wireless power transfer system using magnetic resonance coupling, IEEJ IIC, vol.12, no.63, pp.77 80, March 2012 (in Japanese). [14] Y. Moriwaki, T. Imura, and Y. Hori, A study on reduction of reflected power using DC/DC converter in wireless power transfer system via magnetic resonant coupling, 2011 Annual Conference of IEE of Japan, Industry Application Society, pp , Sept (in Japanese) [15] B.S. Gurn and H.R. Hiziroglu, Electromagnetic Field Theory Fundamentals, second edition, Cambridge University Press, [16] H. Hirayama, Y. Okuyama, N. Kikuma, and K. Sakakibara, A consideration of equivalent circuit of magnetic-resonant wireless power transfer, the 5th European Conference on Antennas and Propagation, pp , [17] S. Ishigami and Y. Yamanaka, Uncertainty of EM-probe calibration by using a TEM Cell (2), IEICE Technical Report, EMCJ2004-3, April 2004 (in Japanese). [18] K. Fujii and Y. Yamanaka, Study on the measurement uncertainty of input power by using a directional coupler, IEICE General Conf., BS-13-4, March Yukitoshi Sanada was born in Tokyo in He received his B.E. degree in electrical engineering from Keio University, Yokohama Japan, his M.A.Sc. degree in electrical engineering from the University of Victoria, B.C., Canada, and his Ph.D. degree in electrical engineering from Keio University, Yokohama Japan, in 1992, 1995, and 1997, respectively. In 1997 he joined the Faculty of Engineering, Tokyo Institute of Technology as a Research Associate. In 2000 he joined Advanced Telecommunication Laboratory, Sony Computer Science Laboratories, Inc, as an associate researcher. In 2001 he joined Faculty of Science and Engineering, Keio University, where he is now a professor. He received the Young Engineer Award from IEICE Japan in His current research interests are in software defined radio, cognitive radio, and OFDM systems. Mamiko Inamori was born in Kagoshima, Japan in She received her B.E., M.E., and Ph.D. degrees in electronics engineering from Keio University, Japan in 2005, 2007, and 2009, respectively. Since April 2013, she has been a lecturer in Tokai University. She received the Young Scientist Award from Ericsson Japan in Her research interests are mainly concentrated on wireless communication and power electronics. Kazuki Sugeno was born in Kanagawa, Japan in He received his B.E. degrees in electronics engineering from Keio University, Japan in Since April 2012, he has been a graduate student in School of Integrated Design Engineering, Graduate School of Science and Technology, Keio University. His research interests are mainly concentrated on control system for wireless power transfer.