A novel CMOS level converter with balanced rise and fall delays for arbitrary voltage conversion is presented. The proposed level converter was designed using a 90 nm CMOS process technology. The comparison result indicates that the maximum difference between the rise and fall delays of the proposed level converter was reduced by up to 92% compared to the conventional CMOS level converters.

We propose an elastic pipeline that can apply dynamic voltage scaling (DVS) to hardwired logic circuits. In order to demonstrate its feasibility, a hardwired H.264/AVC HDTV decoder is designed as a real-time application. An entropy decoding process is divided into context-based adaptive binary arithmetic coding (CABAC) and syntax element decoding (SED), which has advantages of smoothing workload for CABAC and keeping efficiency of the elastic pipeline. An operating frequency and supply voltage are dynamically modulated every slot depending on workload of H.264 decoding to minimize power. We optimize the number of slots per frame to enhance power reduction. The proposed decoder achieves a power reduction of 50% in a 90-nm process technology, compared to the conventional clock-gating scheme.

We propose a sub-mW H.264 baseline-profile motion estimation processor for portable video applications. It features a VLSI-oriented block partitioning strategy and low-power SIMD/systolic-array datapath architecture, where the datapath can be switched between an SIMD and systolic array depending on processing flow. The processor supports all the seven kinds of block modes, and can handle three reference frames for a CIF (352288) 30-fps to QCIF (176144) 15-fps sequences with a quarter-pixel accuracy. It integrates 3.3 million transistors, and occupies 2.83.1 mm2 in a 130-nm CMOS technology. The proposed processor achieves a power of 800 µW in a QCIF 15-fps sequence with one reference picture.

We propose a voltage control scheme for 6T SRAM cells that makes a minimum operation voltage down to 0.3 V under DVS environment. A supply voltage to the memory cells and wordline drivers, bitline voltage, and body bias voltage of load pMOSFETs are controlled according to read and write operations, which secures operation margins even at a low operation voltage. A self-aligned timing control with a dummy wordline and its feedback is also introduced to guarantee stable operation in a wide range of the supply voltage. A measurement result of a 64-kb SRAM in a 90-nm process technology shows that a power reduction of 30% can be achieved at 100 MHz. In a 65-nm 64-Mb SRAM, a 74% power saving is expected at 1/6 of the maximum operating frequency. The performance penalty by the proposed scheme is less than 1%, and area overhead is 5.6%.

A novel cache-based network processor (NP) architecture that can catch up with next generation 100-Gbps packet-processing throughput by exploiting a nature of network traffic is proposed, and the prototype is evaluated with real network traffic traces. This architecture consists of several small processing units (PUs) and a bit-stream manipulation hardware called a burst-stream path (BSP) that has a special cache mechanism called a process-learning cache (PLC) and a cache-miss handler (CMH). The PLC memorizes a packet-processing method with all table-lookup results, and applies it to subsequent packets that have the same information in their header. To avoid packet-processing blocking, the CMH handles cache-miss packets while registration processing is performed at the PLC. The combination of the PLC and CMH enables most packets to skip the execution at the PUs, which dissipate huge power in conventional NPs. We evaluated an FPGA-based prototype with real core network traffic traces of a WIDE backbone router. From the experimental results, we observed a special case where the packet of minimum size appeared in large quantities, and the cache-based NP was able to achieve 100% throughput with only the 10%-throughput PUs due to the existence of very high temporal locality of network traffic. From the whole results, the cache-based NP would be able to achieve 100-Gbps throughput by using 10- to 40-Gbps throughput PUs. The power consumption of the cache-based NP, which consists of 40-Gbps throughput PUs, is estimated to be only 44.7% that of a conventional NP.

For super-parallel video processing, we proposed a power- and area-efficient SRAM core architecture with a segmentation-free access, which means accessibility to arbitrary consecutive pixels, and horizontal/vertical access. To achieve these flexible accesses, a spirally-connected local-wordline select signal and multi-selection scheme in wordlines are proposed, so that extra X-decoders in the conventional multi-division SRAM can be eliminated. Consequently, the proposed SRAM reduces a power and area by 57-60% and 60%, respectively, when it is applied to a 128 parallel architecture. The proposed 160-kbit SRAM with 16-read ports (2-read port SRAM with eight-parallel architecture) is implemented to a search window buffer for an H.264 motion estimation processor core which dissipates 800 µW for QCIF 15-fps in a 130-nm technology.

This paper proposes a 15-bit 10-MS/s pipelined ADC based on the incomplete settling principle. The traditional complete settling stage is improved to the incomplete settling structure through dividing the sampling clock of the traditional stage into two parts for discharging the sampling and feedback capacitors and completing the sampling, respectively. The proposed ADC verifies the correction and validity of optimizing ADCs' conversion speed without additional power consumption through the incomplete settling. This ADC employs scaling-down scheme to achieve low power dissipation and utilizes full-differential structure, bottom-plate-sampling, and capacitor-sharing techniques as well as bit-by-bit digital self-calibration to increase the ADC's linearity. It is processed in 0.18 µm 1P6M CMOS mixed-mode technology. Simulation results show that 82 dB SNDR and 87 dB SFDR are obtained at the sampling rate of 10 MHz with the input sine frequency of 100 kHz and the whole static power dissipation is 21.94 mW.

In modern CMOS digital design, the noise immunity has come to have an almost equal importance to the power consumption. In the last decade, many low power design schemes have been presented. However, no one can simply judge which one is the best from the noise immunity point of view. In this paper, we investigate the noise immunity of the static CMOS low power design schemes in terms of logic and delay errors caused by different kinds of noise existing in the static CMOS digital circuits. To fulfill the aims of the paper, first a model representing the different sources of noise in deep submicron design is presented. Then the model is applied to the most famous low power design schemes to find out the most robust one with regard to noise. Our results show the advantages of the dual threshold voltage scheme over other schemes from the noise immunity point of view. Moreover, it indicates that noise should be carefully taken into account when designing low power circuits; otherwise circuit performance would be unexpected. The study is carried out on three circuits; each is designed in five different schemes. The analysis is done using HSPICE, assuming 0.18 µm CMOS technology.

A VCO for multi-standard transceiver should operate in wide-tuning range, while providing low-phase noise quadrature outputs with low power consumption. In this paper, a multi-standard CMOS LC QVCO is designed utilizing reconfigurable LC tank and low power low phase noise quadrature generation method. Designed in 0.18 µm CMOS technology, the VCO achieved very wide tuning characteristics in two separate bands with low power consumption.

A new method to reduce power consumption of a linear transconductor is proposed in this paper. The minimum tail current for the operation of the transconductor is supplied by a new current source circuit. The proposed circuit is based on a dynamic biasing current technique. Results of SPICE simulation show that the proposed technique is very effective to reduce power consumption of the transconductor.

This paper describes the design of a small-offset 12-bit CMOS charge-redistribution DAC using a weighted-mean flip-around sample-and-hold circuit (S/H). Flip-around S/H topology can realize small-offset characteristics, and it is effective to reduce power dissipation and chip area because independent feedback capacitors are not necessary. In this DAC the small-offset characteristic remains not only in amplification phase but also in sampling phase with the circuit technique. The design of 1.8 V, 50 MS/s fully differential DAC with output swing of 2 Vp-p has very small offset of 100 µV for the reset switch mismatch of 2%. A technique to improve dynamic performance measured by SFDR using damping resistors and switches at the output stage is also presented. The designed 12-bit DAC with 0.25 µm CMOS technology has low-power dissipation of 35 mW at 50 MS/s.

This paper proposes a low power scan test scheme and formulates a problem based on this scheme. In this scheme the flip-flops are grouped into N scan chains. At any time, only one scan chain is active during scan test. Therefore, both average power and peak power are reduced compared with conventional full scan test methodology. This paper also proposes a tabu search-based approach to minimize test application time. In this approach we handle the information during deterministic test efficiently. Experimental results demonstrate that this approach drastically reduces both average power and peak power dissipation at a little longer test application time on various benchmark circuits.

Sleep transistors such as MTCMOS and SCCMOS drastically reduce leakage current, but their ON resistances cause significant performance degradation. Larger sleep transistors reduce their ON resistances, but increase leakage current in a sleep mode. Decoupling capacitors beside sleep transistors reduce leakage current. Experimental results show that PMOS SCCMOS with a 4 pF decoupling capacitor reduces leakage current by 1/673 on a 64 bit adder in a 90 nm process.

This work proposes a 10b 100 MS/s 1.4 mm2 CMOS ADC for low-power multimedia applications. The proposed two-step pipeline ADC minimizes chip area and power dissipation at the target resolution and sampling rate. The wide-band SHA employs a gate-bootstrapping circuit to handle both single-ended and differential inputs of 1.2 Vp-p at 10b accuracy while the second-stage flash ADC employs open-loop offset sampling techniques to achieve 6b resolution. A 3-D fully symmetrical layout reduces the capacitor and device mismatch of the first-stage MDAC. The low-noise references are integrated on chip with optional off-chip voltage references. The prototype 10b ADC implemented in a 0.18 µm CMOS shows the maximum measured DNL and INL of 0.59LSB and 0.77LSB, respectively. The ADC demonstrates an SNDR of 53.7 dB, an SFDR of 61.5 dB, and the power dissipation of 56 mW at 100 MS/s.

In this paper, we propose a multicast protocol, called BAM (Branch Aggregation Multicast), for wireless sensor networks. The main contribution of BAM is a reduction in the radio communication load, which is a key determinant of sensor energy consumption. BAM does not use any control packets such as join/leave messages and does not manage multicast groups. BAM is highly compatible with existing wireless sensor protocols, such as routing protocols, MAC protocols, and other kinds of energy efficient protocols. BAM implementation is quite simple and BAM works on various networks even if some sensors are not BAM-capable. BAM is composed of two aggregation techniques. One is single hop aggregation (S-BAM) and the other is multiple paths aggregation (M-BAM). S-BAM aggregates radio transmission within a single hop and enables single transmission to multiple intended receivers. M-BAM aggregates multiple paths into fewer ones and limits the range of radio transmission. S-BAM is designed to reduce redundant communication at every branch while M-BAM is designed to reduce the number of branches. SM-BAM, the combination of S-BAM and M-BAM, can reduce the radio communication load thus enabling energy efficient multicast communication. We evaluate BAM in three ways, qualitative evaluation by theoretical analysis, quantitative evaluation through computer simulations, and experiments using CrossBow's MICA2. Our results show that BAM is a very energy efficient multicast protocol that well supports wireless sensor networks.

This work describes an 8b 240 MS/s CMOS ADC as one of embedded core circuits for high-performance displays based on low-noise on-chip references and dual-mode inputs with the requirements of limited pins, low power, and small size at high speed. The proposed ADC uses externally connected pins only for analog inputs, digital outputs, and supplies. The ADC employs (1) a two-step pipeline architecture to optimize power and chip size at the target sampling frequency of 240 MHz, (2) advanced bootstrapping techniques to achieve high signal bandwidth in the input SHA, and (3) RC filter-based on-chip current and voltage references to improve noise performance with a power-off function for portable applications. The prototype ADC is implemented in a 0.18 µm CMOS and simultaneously integrated in a DVD system with dual-mode inputs. The prototype ADC shows the measured DNL and INL within 0.49LSB and 0.69LSB, and the SNDR and SFDR exceeding 38 dB and 50 dB for inputs up to the Nyquist frequency at 240 MS/s. The ADC consumes 104 mW at 240 MS/s and an active die area is 1.36 mm2 .

Energy efficiency of cache memories is crucial in designing embedded processors. Reducing energy consumption in the instruction cache is especially important, since the instruction cache consumes a significant portion of total processor energy. This paper proposes a new instruction cache architecture, named Partitioned Instruction Cache (PI-Cache), for reducing dynamic energy consumption in the instruction cache by partitioning it to smaller (less power-consuming) sub-caches. When the proposed PI-Cache is accessed, only one sub-cache is accessed by utilizing the temporal/spatial locality of applications. In the meantime, other sub-caches are not accessed, leading to dynamic energy reduction. The PI-Cache also reduces dynamic energy consumption by eliminating the energy consumed in tag lookup and comparison. Moreover, the performance gap between the conventional instruction cache and the proposed PI-Cache becomes little when the physical cache access time is considered. We evaluated the energy efficiency by running a cycle accurate simulator, SimpleScalar, with power parameters obtained from CACTI. Simulation results show that the PI-Cache improves the energy-delay product by 20%-54% compared to the conventional direct-mapped instruction cache.

In recent years, digital watermarking has become a popular technique for labeling digital images by hiding secret information which can protect the copyright. The goal of this paper is to develop a DCT-based watermarking algorithm for low power and high performance. Our energy-efficient technique focuses on reducing computation required on block-based permutation. Instead of using spacial coefficients proposed by Hsu and Wu's algorithm [1], we use DCT coefficients to pair blocks directly. The approach is implemented by C language and estimated power dissipation using Wattch toolset. The experimental results show that our approach not only reduces 99% energy consumption of pairing mechanism, but also increase the PSNR by 0.414 db for the best case. Moreover, the proposed approach is robust to a variety of signal distortions, such as JPEG, image cropping, sharpening, blurring, and intensity adjusting.

We developed a CMOS watchdog sensor that simulates the changes in quality of perishables such as farm and marine products. The sensor can imitate a chemical reaction that causes the changes in the quality of perishables, with a wide range of activation energy from 0.1 eV to 0.7 eV. Attached to perishable goods, the sensor simulates the deterioration of the goods caused by surrounding temperatures. By reading the output of the sensor, consumers can determine whether the goods are fresh or not. This sensor consists of subthreshold CMOS circuits with a low-power consumption of 5 µW or less.

Circuit techniques for realizing low-voltage and low-power SoCs for 90-nm CMOS technology and beyond are described. A proposed SAFBB (self-adjusted forward body bias techniques), ATC (Asymmetric Three transistor Cell) DRAM, and ADC using an offset canceling comparator deal with leakage and variability issues for these technologies. A 32-bit adder using SAFBB attained 353-µA at 400-MHz operation at 0.5-V supply voltage, and 1 Mb memory array using ATC DRAM cells achieved 1.5 mA at 50 MHz, 0.5 V. The 4-bit ADC attained 2 Gsample/s operation at a supply voltage of 0.9 V.