Fukashi MORISHITA Wataru SAITO Norihito KATO Yoichi IIZUKA Masao ITO
This paper proposes novel test techniques for high accuracy measurement of ADCs and a ramp generator on a CMOS image sensor (CIS) chip. The test circuit for the ADCs has a dual path and has an ability of multi-functional fine pattern generator that can define any input for each column to evaluate CIS specific characteristics electrically. The test circuit for the ramp generator can realize an on-chip current cell test and reject the current cell failure within 1LSB accuracy. We fabricated the test sensor using 55nm CIS process and measured the IP characteristics. Measured results show INL of 14.6LSB, crosstalk of 14.9LSB and column interference noise of 5.4LSB. These measured results agree with the designed values. By using this technique, we confirmed the accurate ADC measurement can be realized without being affected by the ambiguity of the optical input.
This paper formulates minimal word-line (WL) delay time with pre-emphasis pulses to design the pulse width as a function of the overdrive voltage for large memory arrays such as 3D NAND. Circuit theory for a single RC line only with capacitance to ground and that only with coupling capacitance as well as a general case where RC lines have both grounded and coupling capacitance is discussed to provide an optimum pre-emphasis pulse width to minimize the delay time. The theory is expanded to include the cases where the resistance of the RC line driver is not negligibly small. The minimum delay time formulas of a single RC delay line and capacitive coupling RC lines was in good agreement (i.e. within 5% error) with measurement. With this research, circuit designers can estimate an optimum pre-emphasis pulse width and the delay time for an RC line in the initial design phase.
Yan CHEN Jing ZHANG Yuebing XU Yingjie ZHANG Renyuan ZHANG Yasuhiko NAKASHIMA
An efficient resistive random access memory (ReRAM) structure is developed for accelerating convolutional neural network (CNN) powered by the in-memory computation. A novel ReRAM cell circuit is designed with two-directional (2-D) accessibility. The entire memory system is organized as a 2-D array, in which specific memory cells can be identically accessed by both of column- and row-locality. For the in-memory computations of CNNs, only relevant cells in an identical sub-array are accessed by 2-D read-out operations, which is hardly implemented by conventional ReRAM cells. In this manner, the redundant access (column or row) of the conventional ReRAM structures is prevented to eliminated the unnecessary data movement when CNNs are processed in-memory. From the simulation results, the energy and bandwidth efficiency of the proposed memory structure are 1.4x and 5x of a state-of-the-art ReRAM architecture, respectively.
This paper reviews architectures and topologies for column-parallel analog-to-digital converters (ADCs) used for CMOS image sensors (CISs) and discusses the performance of CISs using column-parallel ADCs based on figures-of-merit (FoM) with considering noise models which behave differently at low/middle and high pixel-rate regions. Various FoM considering different performance factors are defined. The defined FoM are applied to surveyed data on reported CISs using column-parallel ADCs which are categorized into 4 types; single slope, SAR, cyclic and delta-sigma ADCs. The FoM defined by (noise)2(power)/(pixel-rate) separately for low/middle and high pixel-rate regions well explains the frontline of the CIS' performance in all the pixel rates. Using the FoM defined by (noise)2(power)/(intrascene dynamic range)(pixel-rate), the effectiveness of recently-reported techniques for extended-dynamic-range CISs is clarified.
Hejiu ZHANG Ningmei YU Nan LYU Keren LI
This letter presents a 12-bit column-parallel hybrid two-step successive approximation register/single-slope analog-to-digital converter (SAR/SS ADC) for CMOS image sensor (CIS). For achieving a high conversion speed, a simple SAR ADC is used in upper 6-bit conversion and a conventional SS ADC is used in lower 6-bit conversion. To reduce the power consumption, a comparator is shared in each column, and a 6-bit ramp generator is shared by all columns. This ADC is designed in SMIC 0.18µm CMOS process. At a clock frequency of 22.7MHz, the conversion time is 3.2µs. The ADC has a DNL of -0.31/+0.38LSB and an INL of -0.86/+0.8LSB. The power consumption of each column ADC is 89µW and the ramp generator is 763µW.
Nan LYU Ning Mei YU He Jiu ZHANG
This letter presents a new time-digital single-slope ADC (TDSS) architecture for CMOS image sensors. In the proposed ADC, a conventional single-slope ADC is used in coarse phase and a time to digital convertor is employed in fine phase. Through second comparison of the two different slope voltages (discharge input voltage and ramp voltage), the proposed ADC achieves low bit precision compensation. Compared with multiple-ramp single-slope (MRSS) ADC, the proposed ADC not only has a simple digital judgment circuit, but also increases conversion speed without complicated structure of ramp generator. A 10-bit TDSS ADC consisting of 7-bit conventional single-slope ADC and 3-bit time to digital converter was realized in a 0.13µm CIS process. Simulations demonstrate that the conversion speed of a TDSS ADC is almost 3.5 times faster than that of a single-slope ADC.
In this correspondence, a generic method of constructing optimal p2-ary low correlation zone sequence sets is proposed. Firstly p2-ary column sequence sets are constructed, then p2-ary LCZ sequence sets with parameters (pn-1, pm-1, (pn-1)/(pm-1),1) are constructed by using column sequences and interleaving technique. The resultant p2-ary LCZ sequence sets are optimal with respect to the Tang-Fan-Matsufuji bound.
Daichi TAKEUCHI Katsunori MAKIHARA Mitsuhisa IKEDA Seiichi MIYAZAKI Hirokazu KAKI Tsukasa HAYASHI
We fabricated highly dense Si nano-columnar structures accompanied with Si nanocrystals on W-coated quartz and characterized their local electrical transport in the thickness direction in a non-contact mode by using a Rh-coated Si cantilever with pulse bias application, in which Vmax, Vmin, and the duty ratio were set at +3.0V, -14V, and 50%, respectively. By applying a pulse bias to the bottom W electrode with respect to a grounded top electrode made of ∼10-nm-thick Au on a sample surface, non-uniform current images in correlation with surface morphologies reflecting electron emission were obtained. The change in the surface potential of the highly dense Si nano-columnar structures accompanied with Si nanocrystals, which were measured at room temperature by using an AFM/Kelvin probe technique, indicated electron injection into and extraction from Si nanocrystals, depending on the tip bias polarity. This result is attributable to efficient electron emission under pulsed bias application due to electron charging from the top electrode to the Si nanocrystals in a positively biased duration at the bottom electrode and subsequent quasi-ballistic transport through Si nanocrystals in a negatively biased duration.
Lei GUO Yuhua TANG Yong DOU Yuanwu LEI Meng MA Jie ZHOU
The effective bandwidth of the dynamic random-access memory (DRAM) for the alternate row-wise/column-wise matrix access (AR/CMA) mode, which is a basic characteristic in scientific and engineering applications, is very low. Therefore, we propose the window memory layout scheme (WMLS), which is a matrix layout scheme that does not require transposition, for AR/CMA applications. This scheme maps one row of a logical matrix into a rectangular memory window of the DRAM to balance the bandwidth of the row- and column-wise matrix access and to increase the DRAM IO bandwidth. The optimal window configuration is theoretically analyzed to minimize the total number of no-data-visit operations of the DRAM. Different WMLS implementationsare presented according to the memory structure of field-programmable gata array (FPGA), CPU, and GPU platforms. Experimental results show that the proposed WMLS can significantly improve DRAM bandwidth for AR/CMA applications. achieved speedup factors of 1.6× and 2.0× are achieved for the general-purpose CPU and GPU platforms, respectively. For the FPGA platform, the WMLS DRAM controller is custom. The maximum bandwidth for the AR/CMA mode reaches 5.94 GB/s, which is a 73.6% improvement compared with that of the traditional row-wise access mode. Finally, we apply WMLS scheme for Chirp Scaling SAR application, comparing with the traditional access approach, the maximum speedup factors of 4.73X, 1.33X and 1.56X can be achieved for FPGA, CPU and GPU platform, respectively.
Daichi TAKEUCHI Katsunori MAKIHARA Mitsuhisa IKEDA Seiichi MIYAZAKI Hirokazu KAKI Tsukasa HAYASHI
We have fabricated highly-dense Si nano-columnar structures accompanied with Si nanocrystals on W-coated quartz, and characterized their local electrical transport in the thickness direction using atomic force microscopy (AFM) with a conductive cantilever. By applying DC negative bias to the bottom W electrode with respect to a grounded top electrode made of 10-nm-thick Au on the sample surface, current images reflecting highly-localized conduction were obtained in both contact and non-contact modes. This result is attributable to electron emission due to quasi-ballistic transport through Si nanocrystals via nanocolumnar structure.
Railway operators adjust timetables, and accordingly reschedule rolling stock circulation and crew duties, when the train operations are disrupted by accidents or adverse weather conditions. This paper discusses the problem of rescheduling driver assignment to freight trains after timetable adjustment has been completed. We construct a network from the disrupted situation, and model the problem as an integer programming problem with set-covering constraints combined with set-partitioning constraints. The integer program is solved by column generation in which we reduce the column generation subproblem to a shortest path problem and such paths by utilizing data parallelism. Numerical experiments using a real timetable, driver scheduling plan and major disruption data in the highest-frequency freight train operation area in Japan reveal that our method provides a quality driver rescheduling solution within 25 seconds.
We propose a character size optimization technique to reduce the number of EB shots of multi-column-cell (MCC) lithographic systems in which transistor patterns are projected with multiple column cells in parallel. Each and every column cell is capable of projecting patterns with character projection (CP) and variable shaped beam (VSB) methods. Seeking the optimal character size of characters contributes to minimizing the number of EB shots and reducing the fabrication cost for ICs. Experimental results show that the character size optimization achieved 70.6% less EB shots in the best case with an available electron beam (EB) size. Our technique also achieved 40.6% less EB shots in the best case than a conventional character sizing technique.
Makoto SUGIHARA Yusuke MATSUNAGA Kazuaki MURAKAMI
Character projection (CP) lithography is utilized for maskless lithography and is a potential for the future photomask manufacture because it can project ICs much faster than point beam projection or variable-shaped beam (VSB) projection. In this paper, we first present a projection mask set development methodology for multi-column-cell (MCC) systems, in which column-cells can project patterns in parallel with the CP and VSB lithographies. Next, we present an INLP (integer nonlinear programming) model as well as an ILP (integer linear programming) model for optimizing a CP mask set of an MCC projection system so that projection time is reduced. The experimental results show that our optimization has achieved 33.4% less projection time in the best case than a naive CP mask development approach. The experimental results indicate that our CP mask set optimization method has virtually increased cell pattern objects on CP masks and has decreased VSB projection so that it has achieved higher projection throughput than just parallelizing two column-cells with conventional CP masks.
Junichi AKITA Hiroaki TAKAGI Keisuke DOUMAE Akio KITAGAWA Masashi TODA Takeshi NAGASAKI Toshio KAWASHIMA
Although the line-of-sight (LoS) is expected to be useful as input methodology for computer systems, the application area of the conventional LoS detection system composed of video camera and image processor is restricted in the specialized area, such as academic research, due to its large size and high cost. There is a rapid eye motion, so called 'saccade' in our eye motion, which is expected to be useful for various applications. Because of the saccade's very high speed, it is impossible to track the saccade without using high speed camera. The authors have been proposing the high speed vision chip for LoS detection including saccade based on the pixel parallel processing architecture, however, its resolution is very low for the large size of its pixel. In this paper, we propose and discuss an architecture of the vision chip for LoS detection including saccade based on column-parallel processing manner for increasing the resolution with keeping high processing speed.
R.S. Raja DURAI Naoki SUEHIRO Chenggao HAN
The class of complete complementary sequences (of fixed length) have the ideal correlation properties and are good at increasing the channel usage efficiency but lacks in desirable sequence lengths. In spread spectrum communication systems, sequences having nice correlation properties are important in many ways such as in suppressing multi-user interference, for reliable initial synchronization and in separation of the multipath components. It would be even good if the sequences are easy to construct and have desirable lengths for the system under consideration. In this paper, M sets of sequences that constitute a complete complementary sequences with ith set containing N sequences of length Li each, i = 0, 1, ..., M - 1, is defined and a general method that constructs such a class of complete complementary sequences (of different lengths) is given. The proposed class of complete complementary sequences, constituted by sequence sets of different lengths, does not increase the data rates when short-length sequences are employed.
Shunsuke OKURA Tetsuro OKURA Bogoda A. INDIKA U.K. Kenji TANIGUCHI
This paper describes the design of a random access memory (RAM) bank with a 0.35-µm CMOS process for column-parallel analog/digital converters (ADC) utilized in CMOS imagers. A dynamic latch is utilized that expends neither input DC nor drain current during the monitoring phase. Accuracy analysis of analog/digital conversion error in the RAM bank is discussed to ensure low power consumption of a counter buffer circuit. Moreover, the counter buffer utilizes a combination of NMOS and CMOS buffers to reduce power consumption. Total power consumption of a 10-bit 800-column 40 MHz RAM bank is 2.9 mA for use in an imager.
Shingo TAKAHASHI Shuji TSUKIYAMA Masanori HASHIMOTO Isao SHIRAKAWA
In the design of an active matrix LCD (Liquid Crystal Display), the ratio of the pixel voltage to the video voltage (RPV) of a pixel is an important factor of the performance of the LCD, since the pixel voltage of each pixel determines its transmitted luminance. Thus, of practical importance is the issue of how to maintain the admissible allowance of RPV of each pixel within a prescribed narrow range. This constraint on RPV is analyzed in terms of circuit parameters associated with the sampling switch and sampling pulse of a column driver in the LCD. With the use of a minimal set of such circuit parameters, a design procedure is described dedicatedly for the sampling switch, which intends to seek an optimal sampling switch as well as an optimal sampling pulse waveform. A number of experimental results show that an optimal sampling switch attained by the proposed procedure yields a source driver with almost 18% less power consumption than the one by manual design. Moreover, the percentage of the RPVs within 1001% among 270 cases of fluctuations is 88.1% for the optimal sampling switch, but 46.7% for the manual design.
Seiko ICHIKAWA Katsumi SUEKUNI Masatoshi ISHIMARU Hiroyuki NAKATANI Takao UNATE Akira NAKASUGA
Large liquid crystal display (LCD) panels have several cell-gap problems. For example, gravity defects are observed as thicker cell-gap areas at the bottom of an LCD panel at a high temperature, and cold-bubble defects are observed as bubbles in an LCD panel at a low temperature. We have developed a gap simulation to investigate these problems. The calculation was carried out for both column and ball spacers. It was shown that gap defects can be substantially reduced using ball spacers.
Chen ZHENG Noriaki MIYAZAKI Toshinori SUZUKI
Effective and simply realizable rate compatible low-density parity-check (LDPC) codes are proposed. A parity check matrix is constructed with the progressively increased column weights (PICW) order and adopted to achieve a punctured LDPC coding scheme for a wide range of the code rates of the rate compatible systems. Using the proposed rate compatible punctured LDPC codes, low complex adaptive communication systems, such as wireless communication systems, can be achieved with the reliable transmissions.
Hidenori KUWAKADO Hatsukazu TANAKA
We propose a method for reducing the size of a share in visual secret sharing schemes. The proposed method does not cause the leakage and the loss of the original image. The quality of the recovered image is almost same as that of previous schemes.