3012 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 57, NO. 11, NOVEMBER 2010

Transcription

1 3012 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 57, NO. 11, NOVEMBER 2010 An Advanced External Compensation System for Active Matrix Organic Light-Emitting Diode Displays With Poly-Si Thin-Film Transistor Backplane Hai-Jung In, Member, IEEE, Kyong-Hwan Oh, Member, IEEE, Inhwan Lee, Member, IEEE, Do-Hyung Ryu, Sang-Moo Choi, Keum-Nam Kim, Hye-Dong Kim, and Oh-Kyong Kwon, Member, IEEE Abstract An advanced method for externally compensating the nonuniform electrical characteristics of polycrystalline silicon thin-film transistors (TFTs) and the degradation of organic light-emitting diode (OLED) devices is proposed, and the method is verified using a 14.1-in active matrix OLED (AMOLED) panel. The proposed method provides an effective solution for highimage-quality AMOLED displays by removing IR-drop and temperature effects during the sensing and displaying operations of the external compensation method. Experimental results show that the electrical characteristics of TFTs and OLEDs are successfully sensed, and that the stained image pattern due to the nonuniform luminance error and the differential aging of the OLED is removed. The luminance error range without compensation is from 6.1% to 9.0%, but it is from 1.1% to 1.2% using the external compensation at the luminance level of 120 cd/m 2 in a 14.1-inch AMOLED panel. Index Terms Compensation, flat-panel displays, light-emitting diode (LED) displays, thin-film transistors (TFTs). I. INTRODUCTION RECENTLY, efforts to commercialize active matrix organic light-emitting diode (AMOLED) displays have been made by many researchers [1] [9]. OLED is an excellent device for display applications due to fast response time, high color reproducibility, low power consumption, and thin form factor [1]. However, a suitable high-image-quality AMOLED display has not yet been produced because of the critical drawbacks of OLED, which are the luminance nonuniformity due to the electrical characteristic variations of polycrystalline silicon (poly-si) thin-film transistors (TFTs) [2] and the image sticking due to the differential aging of OLED devices [3]. Previous works attempted to solve these problems by Manuscript received November 18, 2009; revised July 26, 2010; accepted July 30, Date of publication September 20, 2010; date of current version November 5, This work was supported in part by the Korean Research Foundation Grant (KRF J04101). The review of this paper was arranged by Editor J. Kanicki. H.-J. In, K.-H. Oh, I. Lee, and O.-K. Kwon are with the Division of Electronics and Computer Engineering, Hanyang University, Seoul , Korea ( okwon@hanyang.ac.kr). D.-H. Ryu, S.-M. Choi, K.-N. Kim, and H.-D. Kim are with the Research and Development Center, Samsung Mobile Display Company, Ltd., Yongin , Korea. Digital Object Identifier /TED developing AMOLED pixel structures using various compensation techniques [4] [9]. To compensate the threshold voltage variation, storing and compensating the threshold voltage of the driving TFT in the pixel was used [4], [5]. The drawback of this technique is that it cannot compensate the mobility variations of TFTs. Current programming methods [6], [7] can compensate both the threshold voltage and the mobility variations of driving TFTs, but they suffer from a long programming time for a lowlevel current. A current-scaling method using mirroring TFTs [8] was used to overcome the long programming time problem in low gray levels, but the mismatch of the electrical characteristics of mirroring TFTs caused the deviation of emission current. A voltage-boosting method with current programming [9] was proposed to overcome the long programming time in low gray levels and also the mismatch of mirroring TFTs, but it could not properly compensate the mobility variation of TFTs using a simple voltage-boosting method in every gray level. All these methods [4] [9] were adding TFTs in a pixel to perform various functions for the compensation of the nonuniform electrical characteristics of TFTs, but they could not compensate the degradation of OLEDs. The pixel structure should become more complex if we want to compensate the electrical characteristic variation of TFTs, as well as the degradation of OLEDs. On the other hand, using more TFTs leads to a complex structure and thus the decrease in panel yield. A feasible solution is externally sensing and compensating the electrical characteristic deviations of TFTs and the degradation of OLEDs. In our previous work, we proposed an external compensation algorithm with a simple pixel structure and a fastsensing method and verified the algorithm using AMOLED test pixels [10]. However, the I V characteristic variation of the OLED with respect to temperature variation and IR-drop on the sensing path of matrix type display panel structure causes errors during the sensing operation. In this paper, we propose an advanced external compensation algorithm to remove IR-drop and temperature effects during the sensing operation and verify the algorithm using a 14.1-inch AMOLED panel. II. COMPENSATION ALGORITHM Fig. 1 shows the block diagram of the external compensation system and the circuit diagram of the unit pixel and dummy /$ IEEE

2 IN et al.: COMPENSATION SYSTEM FOR AMOLED DISPLAYS WITH POLY-Si TFT BACKPLANE 3013 Fig. 1. Block diagram of the pixel and data driving circuit of the external compensation method. operation, sensing the electrical characteristics of driving TFTs or sensing OLED degradation is never performed. Fig. 2. Measured emission current generated from T1 with respect to time after the initial aging is performed. A. Sensing the Electrical Characteristics of Driving TFTs During the step for sensing the electrical characteristics of driving TFTs, two voltage levels for each unit pixel are sensed and memorized. First, the scan, sense, and em signals are forced to low to turn on the T2, T3, and T4 TFTs, and the VSS is forced to high, so that the OLED is in the cutoff state. Then, we turn sw1 on, and turn sw2, sw3, sw4, and sw5 off, so that I MAX flows through T1, T4, T3, data line, and sw1. Here, I MAX is the required current for the OLED to generate the maximum gray level luminance. If T1 is in the saturation region, the drain current of T1, I D,T1, can be expressed as I D,T1 = 1 2 μ T 1C ox ( W L ) (VDD V G,T 1 V th,t 1 ) 2 (1) OLED. The external compensation can be divided into three steps, namely: the step for sensing the electrical characteristics of driving poly-si TFTs, the step for sensing OLED degradation, and the step for display operation. During the step for sensing the electrical characteristics of driving poly-si TFTs, two points on the I V curve of the driving TFT in each pixel are sensed. The sensed two-point data for each pixel are stored in a nonvolatile memory and subsequently used for extracting the threshold voltage and the mobility factor of the driving TFT in each pixel. This step is performed only once after the panel is fabricated, and the initial aging of TFTs is performed because the long-term aging effect of poly-si TFTs is negligible, as shown in Fig. 2. The percentage degradation of the OLED device in each pixel is determined and memorized during the step for sensing OLED degradation every time the display system is turned on or off. During the normal display operation, the video input data is modulated using stored data in the memory, so the threshold voltage and mobility variations of driving TFTs and the degradation of OLEDs can be compensated using the proposed external compensation algorithm. During the normal display where μ T 1, C ox, (W/L), V G,T 1, and V th,t 1 are the mobility, the gate capacitance per unit area, the channel-width-tochannel-length ratio, the gate voltage, and the threshold voltage of T1, respectively. The gate voltage of T1 with sinking I MAX can be expressed as V G,T 1(1) = VDD 2I MAX μ T 1 C ox ( ) L V th,t 1 (2) W and the sensed voltage V S1 at the input of the analog-to-digital converter (ADC) is represented by V S1 = V G,T 1(1) I MAX R data (3) where R data is the resistance of data line. There is no voltage drop due to T2 because no current flows through T2 when the V G,T 1(1) is settled. We can calculate R data and remove the IR-drop factor of the data line in (3) because we know the sheet resistance of the data line and the row number of the sensing pixel. As a result, we can obtain the value of V G,T 1(1).

3 3014 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 57, NO. 11, NOVEMBER 2010 Fig. 3. Measured anode voltage increment with respect to the normalized luminance of the OLED device under the condition of 1 ma/cm 2 current stress. Next, we turn sw1 off and sw2 on, so that (1/64)I MAX flows through T1, T4, T3, data line, and sw2. Then, similarly, the gate voltage V G,T 1(2) with sinking (1/64)I MAX can be expressed as V G,T 1(2) = VDD 1 8 2I MAX μ T 1 C ox ( ) L V th,t 1 (4) W Fig. 4. Measured anode voltage deviation of OLED samples from a reference OLED sample with respect to the normalized luminance of the OLED device under the condition of 1 ma/cm 2 current stress. Five OLED samples are measured, and one OLED sample is used as reference. and the sensed voltage at the input of the ADC is represented by V S2 = V G,T 1(2) 1 64 I MAX R data. (5) Again, we can obtain the value of V G,T 1(2) by removing the IR-drop of the data line in (5). At the end of this step, we store V G,T 1(1) and V G,T 1(2) of every pixel on the panel in the external nonvolatile memory. B. Sensing OLED Degradation The basic sensing idea of OLED degradation is that we can determine the degradation level by sensing the anode voltage of the OLED in each pixel with reference current sourcing. Fig. 3 shows the measured increase in the anode voltage with respect to the normalized luminance of the OLED device under dc stress condition. Because the anode voltage increases as the luminance of the OLED decreases, we can estimate the degradation level of the OLED by measuring the anode voltage of the OLED. Five OLED samples on the same glass are measured under same dc stress condition, and the deviation of the anode voltage of four OLED samples from one reference OLED sample is shown in Fig. 4. The anode voltage increment of OLEDs on the same glass are very similar due to the degradation on the constant current stress condition, as shown in Fig. 4. The anode voltage error ranges from 1.5 to 2.2 mv, which can only make the luminance error of the OLED from 0.96% to 1.33%. However, the voltage measured at the ADC is different from the anode voltage of the OLED due to the IR-drop of the data line and T3. Also, the I V characteristics of OLEDs depend on temperature, as shown in Fig. 5. Therefore, we use the double- Fig. 5. Measured current density due to the anode-to-cathode voltage of the OLED when temperature is 20 C, 40 C, and 80 C, respectively. sampling method to eliminate the IR-drop error and also use the dummy OLED to remove temperature dependency. To sense OLED degradation, the scan, em, and D sense signals are forced to high, and the sense signal is forced to low, so that T2, T4, and T5 are turned off, and T3 is turned on. Then, we turn sw1, sw2, sw4, and sw5 off and turn sw3 on, so that I ref flows through sw3, data line, T3, and OLED, where I ref is constant reference current. The sensed voltage at the input of the ADC can be expressed as V S3 = V OLED,Iref + V T 3,Iref + V data,iref (6) where V OLED,Iref, V T 3,Iref, and V data,iref are the anode-tocathode voltage of the OLED, the voltage drop between the source and the drain of T3, and the voltage drop at the data line from the ADC to T3, respectively, when I ref current flows. We can estimate the value of V data,iref because we know the sheet resistance of the data line and the level of I ref. However,

4 IN et al.: COMPENSATION SYSTEM FOR AMOLED DISPLAYS WITH POLY-Si TFT BACKPLANE 3015 voltage of the OLED of each pixel due to only the degradation level without temperature dependency. To measure the anode voltage of the dummy OLED, the scan, em, and sense signals are forced to high, and the D sense signal is forced to low, so that T2, T3, and T4 are turned off, and T5 is turned on. Then, we turn sw1, sw2, sw4, and sw5 off and turn sw3 on, so that I ref flows through sw3, data line, T5, and dummy OLED. The voltage sensed at the input of the ADC can be expressed as V S5 = V DOLED,Iref + V T 5,Iref + V Ddata,Iref (9) Fig. 6. Measured anode voltage of the 0%- and 50%-degraded OLEDs as the function of the temperature under the condition of 1 ma/cm 2 current sourcing. The measured anode voltage difference from the voltage at 20 C is plotted. we cannot estimate V T 3,Iref because the electrical characteristic of T3 varies randomly from pixel to pixel. So, we sense the I V characteristic of the OLED in the pixel one more time with 2I ref current sourcing to eliminate V T 3,Iref. That is, after V S3 is memorized, we turn sw3 off and turn sw4 on, so that 2I ref flows through sw4, data line, T3, and OLED. Then, the voltage sensed at the input of the ADC can be expressed as V S4 = V OLED,2Iref + V T 3,2Iref + V data,2iref (7) where V OLED,2Iref, V T 3,2Iref, and V data,2iref are the anode-tocathode voltage of the OLED, the voltage drop between the source and the drain of T3, and the voltage drop at the data line from the ADC to T3, respectively, when 2I ref current flows. Because T3 is operating in the deep triode region, it can be modeled as a resistor, which means that the V T 3,2Iref and V data,2iref are twice as much as V T 3,Iref and V data,iref, respectively. Therefore, we can obtain the value of V POLED,IRcomp eliminating the IR-drop error using (6) and (7) as follows: V POLED,Ircomp = 2 V S3 V S4 = 2 V OLED,Iref V OLED,2Iref. (8) The dummy OLED is located at the top or bottom row of the pixel array and only used during the step for sensing OLED degradation. Since it is never used during the normal display operation, the dummy OLED is rarely degraded and thus can be used as a reference of OLED degradation. We can make the temperature of every OLED in the pixel array including the dummy OLED similar by attaching the highly thermal conductive heat-spread material to the cathode plane. Then, anode voltages of the 0%-degraded dummy OLED and the OLED in each pixel are sensed at the same temperature. Fig. 6 shows the anode voltage decreasing from the reference anode voltage at 20 C of the 0%- and 50%-degraded OLEDs as temperature increases. It is shown that the amount of the decrement in the anode voltage of the 0%- and 50%-degraded OLED devices is almost same. By subtracting the anode voltage of the dummy OLED from the sensed anode voltage of the OLED in each pixel, we can achieve the incremented anode where V DOLED,Iref, V T 5,Iref, and V Ddata,Iref are the anode-tocathode voltage of the dummy OLED, the voltage drop across T5, and the voltage drop at the data line from the ADC to T5, respectively, when I ref current flows. After memorizing V S5, we turn sw3 off and turn sw4 on, so that 2I ref flows through sw4, data line, T5, and dummy OLED. Then, the voltage sensed at the input of the ADC can be expressed as V S6 = V DOLED,2Iref + V T 5,2Iref + V Ddata,2Iref (10) where V DOLED,2Iref, V T 5,2Iref, and V Ddata,2Iref are the anodeto-cathode voltage of the dummy OLED, the voltage drop across T5, and the voltage drop at the data line from the ADC to T5, respectively, when 2I ref current flows. Since V T 5,2Iref and V Ddata,2Iref are twice as high as V T 5,Iref and V Ddata,Iref, respectively, we can obtain the value of V DOLED,IRcomp eliminating the IR-drop error using (9) and (10) as follows: V DOLED,Ircomp = 2 V S5 V S6 = 2 V DOLED,Iref V DOLED,2Iref. (11) By using (8) and (11), we can calculate the value of V OLED,diff, which is independent of temperature variation, as well as the IR-drop of the data line and switching TFTs as follows: V OLED,diff = V POLED,Ircomp V DOLED,Ircomp. (12) Fig. 7 shows the measured V OLED,diff with respect to the normalized luminance of the OLED under the conditions of 1 and 2 ma/cm 2 dc stress. It is shown that the V OLED,diff increases as the percentage luminance of the OLED decreases, and that the increase in V OLED,diff with respect to the luminance of the OLED does not depend on the current density of the stressed condition. So, although the current level flowing through the OLED is different in each pixel during the normal display operation, we can determine the percentage luminance of the OLED device by sensing the V OLED,diff of each pixel. The determined percentage luminance of each pixel is memorized in the external memory. Because users might be waiting during the step for sensing OLED degradation, this sensing should be performed as quickly as possible. The time required for sensing OLED degradation for every pixel on the panel is only 576 ms in the case that the channel ADC is used for a 14.1-inch AMOLED panel with a highdefinition (HD) resolution format.

5 3016 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 57, NO. 11, NOVEMBER 2010 Fig. 7. Measured increment of V OLED,diff from the initial value with respect to the luminance of the OLED on 1 and 2 ma/cm 2 dc stress conditions. Two OLEDs are stressed differently with 1 and 2 ma/cm 2,buttheV OLED,diff and the luminance of OLEDs are measured with the same level of I ref. C. Display Operation During the normal display operation, data voltage is modulated using input video data, sensed V G,T 1(1) and V G,T 1(2), and the percentage luminance of the OLED in each pixel. The modulated data voltage during this step can be expressed as (100 )( ) γ i V data = VDD V M1 V M2 (13) α n where α, γ, i, and n are the memorized percentage luminance of the OLED device in each pixel during the step for sensing OLED degradation, the controllable parameter for gamma correction, the selected gray level from the input video data, and the total number of gray levels, respectively. V M1 and V M2 can be expressed as V M1 = 8 ( ) VG,T 1(2) V G,T 1(1) (14) 7 V M2 = VDD V G,T 1(1) V M1. (15) When V data is applied to the gate of T1, I D,T1 can be expressed as I D,T1 = 1 2 μ W T 1C ox L (VDD V data V th,t 1 ) 2 ( )( ) γ 100 i = I MAX. (16) α n It is shown that the drain current of T1 during display time is not the function of the threshold voltage and the mobility of T1. The degraded percentage luminance of the OLED device is compensated using the (100/α) factor. The gamma curve of the OLED can be controlled using the γ parameter, so gamma correction can easily be performed during the modulation of video data. III. EXPERIMENTAL SETUP For the verification of the proposed external compensation algorithm, a 14.1-inch HD AMOLED panel is driven by a test Fig. 8. (a) Block diagram of the test board for the external compensation method, (b) photographs of the organized test board, and (c) backside of an AMOLED panel. board. Excimer laser annealing is used for the crystallization of low-temperature poly-si TFTs on the panel. Fig. 8(a) (c) show the block diagram of the measurement setup, the photographs of the test board, and the backside of the panel, respectively. During the step to sense the TFT characteristics, the upper switch array disconnects the voltage programming data driver IC from the AMOLED panel. The decoder and the lower switch array connect ADCs to the AMOLED panel. The reference current block sinks I MAX or (1/64)I MAX. The scan and the

6 IN et al.: COMPENSATION SYSTEM FOR AMOLED DISPLAYS WITH POLY-Si TFT BACKPLANE 3017 Fig. 9. Thermal infrared image of the panel using infrared camera during the OLED-sensing operation. sense drivers sequentially turn the T2, T3, and T4 TFTs on in each pixel on the panel, one row after another, and the sensed voltage is converted to digital 8-bit data using the ADC. The measured values of V G,T 1(1) and V G,T 1(2) for each pixel are stored in an electrically erasable programmable read-only memory (EEPROM). During the step to sense OLED degradation, the upper switch array disconnects the voltage programming data driver IC from the AMOLED panel. The decoder and the lower switch array connect ADCs to the AMOLED panel. The reference current block sources I ref or 2I ref. The sense driver sequentially turns the T3 TFT on in each pixel on the panel, one row after another. The value of V OLED,diff is calculated, and the percentage luminance of the OLED is determined using field-programmable gate array (FPGA). The determined value of α for each pixel is stored in EEPROM as 7-bit data. During the display operation step, a static random-access memory (SRAM) is used as temporary registers. Stored V G,T 1(1), V G,T 1(2), and α in the EEPROM are transferred to the SRAM for high-speed calculation. In a practical way, a flash memory can be used instead of EEPROM. Once sensed data of every pixel is transferred from the EEPROM to the SRAM, the operation for the proposed external compensation is performed using FPGA. Using video input data with an 8-bit gray level, an 8-bit V G,T 1(1),an8-bitV G,T 1(2), and a 7-bit α, FPGA generates 10-bit modulated data. The eight-channel data per each color is processed in parallel using the clock frequency of 8 MHz. The modulated data is applied to the voltage programming data driver IC and the upper switch array is turned on. Then, the analog voltage generated using the 10-bit resolution voltage programming data driver IC is applied to the AMOLED panel. The scan driver sequentially turns the T2 TFT on in each pixel on the panel, one row after another. The data voltage is applied to the gate node of T1 in each pixel, and the panel shows the display image as a result. All of this operation is controlled by a personal computer. IV. EXPERIMENTAL RESULTS The electrical characteristics of TFTs on the panel differ from pixel to pixel because of the random formation of the crystallization on the poly-si TFT backplane. Fig. 9 shows the thermal infrared image of the panel during the OLED-sensing operation. It is shown that the temperature distribution of the panel is very uniform because every OLED only emits the light for 330 μs during the OLED-sensing operation. Fig. 10. Measured drain current of driving TFTs in three sample pixels as the function of gray level (a) without and (b) with the external compensation. Fig. 10(a) and (b) show the measured drain current of T1 in three sample pixels without and with the proposed external compensation. It is shown that the external compensation algorithm works well in various gamma values, so even though the TFT in each pixel has a different threshold voltage or mobility, the drain current of T1 in each pixel can be generated uniformly in every gray level. Fig. 11 shows the measured luminance for a portion of the panel with and without the proposed compensation. As shown in Fig. 11, the luminance error range without compensation is from 6.1% to 9.0%, but it is from 1.1% to 1.2% using the external compensation at the luminance level of 120 cd/m 2.It is shown that the uniformity of the luminance across the panel is improved dramatically after the external compensation of the threshold voltage and the mobility of TFTs. Fig. 12(a) and (b) show the display images on the 14.1-inch HD panel without and with the external compensation, respectively. To facilitate the comparison of luminance, the same gray level is applied to every pixel on the panel. It is shown that the intentionally made bright ring pattern at the right bottom of the panel in Fig. 12(a) disappeared in Fig. 12(b), and the nonuniformity of the luminance is also dramatically improved by compensating electrical characteristic

7 3018 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 57, NO. 11, NOVEMBER 2010 panel. It is expected that the proposed advanced external compensation system provides an effective solution for high-imagequality AMOLED displays. Fig. 11. Luminance of the portion of the panel with and without external compensation using sensed data. REFERENCES [1] G. Gu and S. R. Forrest, Design of flat-panel displays based on organic light-emitting devices, IEEE J. Sel. Topics Quantum Electron., vol. 4, no. 1, pp , Jan./Feb [2] X. Guo and S. R. P. Silva, Investigation on the current nonuniformity in current-mode TFT active-matrix display pixel circuitry, IEEE Trans. Electron Devices, vol. 52, no. 11, pp , Nov [3] S. C. Xia, R. C. Kwong, V. I. Adamovich, M. S. Weaver, and J. J. Brown, OLED device operational lifetime: Insights and challenges, in Proc. 41st Annu. IEEE Int. Symp. Reliab. Phys., Apr. 2007, pp [4] J.-C. Goh, H.-J. Chung, J. Jang, and C.-H. Han, A new pixel circuit for active matrix organic light emitting diodes, IEEE Electron Device Lett., vol. 23, no. 9, pp , Sep [5] Y.-H. Tai, Y.-H. Tai, B.-T. Chen, Y.-J. Kuo, C.-C. Tsai, K.-Y. Chiang, Y.-J. Wei, and H.-C. Cheng, A new pixel circuit for driving organic light-emitting diode with low temperature polycrystalline silicon thin-film transistors, J. Display Technol., vol. 1, no. 1, pp , Sep [6] A. Shin, B. Yoon, and M. Y. Sung, A novel current driving method using organic TFT pixel circuit for active-matrix OLED, in Proc. IEEE Int. Conf. Microelectron., Dec. 2007, pp [7] F. H. Wang, H. C. Lin, P. S. Shih, L. Y. Lin, and H. W. Liu, New current programmed pixel circuit for active-matrix organic light-emitting-diode displays, in Proc. IEEE Int. Conf. EDSSC, Dec. 2007, pp [8] J.-H. Lee, W.-J. Nam, B.-K. Kim, H.-S. Choi, Y.-M. Ha, and M.-K. Han, A new poly-si TFT current-mirror pixel for active matrix organic light emitting diode, IEEE Electron Device Lett., vol.27,no.10,pp , Oct [9] Y.-C. Lin, H.-P. D. Shieh, and J. Kanicki, A novel current-scaling a-si:h TFTs pixel electrode circuit for AM-OLEDs, IEEE Trans. Electron Devices, vol. 52, no. 6, pp , Jun [10] H.-J. In and O.-K. Kwon, External compensation of non-uniform electrical characteristics of thin film transistors and degradation of OLED devices in AMOLED displays, IEEE Electron Device Lett., vol.30,no.4, pp , Apr Fig. 12. Display image of an AMOLED panel (a) without and (b) with external compensation. Video data with the same gray level is applied to every pixel on the panel. Hai-Jung In (M 04) received the B.S. and M.S. degrees in electronics and computer engineering from Hanyang University, Seoul, Korea, in 2004 and 2006, respectively. He is currently working toward the Ph.D. degree in electronics and computer engineering at the same university. His research interests include low-power circuits, analog circuit design, system-on-panel and driving methods, and circuits for flat-panel displays. variations of driving TFTs and the degradation of the OLED using the proposed external compensation method. V. C ONCLUSION An advanced system for externally compensating the nonuniform electrical characteristic variations of TFTs and the degradation of OLEDs has been proposed and verified by driving a 14.1-inch AMOLED panel. By using the double-sampling method and the dummy OLED, the proposed method eliminates not only the IR-drop but also the temperature dependency problem during the sensing operation. Experimental results show that the proposed external compensation system provides a dramatic improvement in the luminance uniformity of the Kyong-Hwan Oh (M 09) received the B.S. degree in electronics and computer engineering from Hanyang University, Seoul, Korea, in He is currently working toward the M.S. degree in electrical and computer engineering at the same university. His research interests include mixed circuit design, high-speed circuit design, system-on-panel and driving methods, and circuits for flat-panel displays.

8 IN et al.: COMPENSATION SYSTEM FOR AMOLED DISPLAYS WITH POLY-Si TFT BACKPLANE 3019 Inhwan Lee (M 91) received the B.S. and M.S. degrees in electrical engineering from Seoul National University, Seoul, Korea, in 1979 and 1985, respectively, and the Ph.D. degree in electrical and computer engineering from the University of Illinois at Urbana Champaign, in He was a Researcher with the Korean Agency for Defense Development from 1979 to 1986 and was a Technical Staff Member with Tandem Computers from 1994 to Since 1997, he has been with the Division of Electronics and Computer Engineering, Hanyang University, Seoul, where he is currently a Professor. His research interests include display systems and reliable system design. Do-Hyung Ryu received the B.S. and M.S. degrees in electrical engineering from Donga and Busan National University, Busan, Korea, in 1999 and 2001, respectively. He joined Samsung SDI Company, Suwon, Korea, in 2001 in the cooperated research and development center and had engaged in the development of active matrix organic light-emitting diodes (AMOLEDs) for 8 years. In 2008, he moved to Samsung SMD Company, Ltd., Yongin, Korea, which is a joint company between Samsung SDI and Samsung SEC for the business of mobile liquid-crystal displays and AMOLEDs, and he is currently a Senior Engineer with the module technology group of the technology center for the development of advanced AMOLEDs. His research interests include AMOLED, timing controller, and driving methods and systems for flat-panel displays. Sang-Moo Choi received the B.S. and M.S. degrees in electrical and computer engineering from Hanyang University, Seoul, Korea, in 2002 and 2004, respectively. He is currently designing the active matrix organic light-emitting diode (AMOLED) panel as a Senior Engineer with Samsung Mobile Display Company, Ltd., Yongin, Korea. His research interests are pixel circuits, integrated driving circuits on panel, and optimization of panel layout design of AMOLED displays. Keum-Nam Kim received the B.S. and M.S. degrees in physics from Kyung Hee University, Seoul, Korea, in 1996 and 1999, respectively. He joined Samsung SDI Company, Suwon, Korea, in 1999 in the cooperated research and development center and had engaged in the development of polycrystalline silicon thin-film transistor liquid-crystal displays (TFT-LCDs) and active matrix organic light-emitting diodes (AMOLEDs) for 10 years. In 2008, he moved to Samsung SMD Company, Ltd., Yongin, Korea, which is a joint company between Samsung SDI and Samsung SEC for the business of mobile LCDs and AMOLEDs, and he is now partly a Leader of the design group in the technology center. His research interests include TFT-LCD, AMOLED, analog circuit design, and circuits for flat-panel displays. Hye-Dong Kim received the B.S. degree in metallurgical engineering from Inha University, Incheon, Korea, in 1988 and the M.S. and Ph.D. degrees in material science and engineering from the Korea Advanced Institute of Science and Technology, Daejeon, Korea, in 1991 and 1995, respectively. He joined Samsung SDI Company, Suwon, Korea, in 1994 in the cooperated research and development center and had engaged in the development of various next-generation displays like ferroelectric liquid-crystal displays (LCDs), polycrystalline silicon thin-film transistor (TFT)-LCDs, and active matrix organic light-emitting diodes (AMOLEDs) as a Project Leader and a Team Leader for 14 years. In 2008, he moved to Samsung SMD Company, Ltd., Yongin, Korea, which is a joint company between Samsung SDI and Samsung SEC for the business of mobile LCDs and AMOLEDs, and he is now a Group Leader of the technology center for the development of large-area AMOLEDs. His research interests include polycrystalline silicon TFTs, oxide TFTs, organic LED devices, and driving circuits for flat-panel displays. Dr. Kim is a member of SID Korea Chapter and Korea Information Display Society. He has authored and coauthored 48 international journals and conference papers and is the holder of over 20 U.S. patents. Oh-Kyong Kwon (M 83) received the B.S. degree in electronic engineering from Hanyang University, Seoul, Korea, in 1978 and the M.S. and Ph.D. degrees in electrical engineering from Stanford University, Stanford, CA, in 1986 and 1988, respectively. From 1987 to 1992, he was with the Semiconductor Process and Design Center, Texas Instruments Inc., Dallas, where he was engaged in the development of multichip module technologies and smart power integrated circuit technologies for automotive and flat-panel display applications. In 1992, he joined Hanyang University, Seoul, as an Assistant Professor with the Department of Electronic Engineering and is currently a Professor with the Division of Electronics and Computer Engineering at Hanyang University. He has authored and coauthored over 210 international journal and conference papers and is the holder of 65 U.S. patents. His research interests include interconnect and electrical noise modeling for high-speed system-level integration, wafer-scale chip-size packages, smart power integrated circuit technologies, circuit design for mixed mode signals, and driving methods and circuits for flat-panel displays. Dr. Kwon was an IEEE IEDM subcommittee member on solid-state devices from 1997 to 1998, a Technical Program Chair of the 1999 IEEE International Conference on VLSI and CAD, and a Workshop Cochair of the 2000 and 2001 Asia-Pacific Workshop on Fundamentals and Applications of Advanced Semiconductor Devices. He served as the Korean Chapter Chair of IEEE EDS from 2002 to He was the Program Manager of the Korean Polycrystalline Silicon Thin-Film Transistor Light-Emitting Diode Research and Development Program from 1993 to He was the Program Manager of the Korean Flat Panel Research and Development Program from 1998 to He was a Technical Program Chair of the International System-on-a-Chip Conference in 2004, a Technical Program Chair of the International Meeting on Information Display/International Display Manufacturing Conference in 2006, and an Executive Chair of the International Meeting on Information Display in 2007 and He is currently the Vice President of the Institute of Electronics Engineers of Korea and a Technical Program Committee Member of the Society for Information Display, the International Solid State Circuit Conference, and the International Display Workshops.