Search

Background: Wearable and unobtrusive sensing devices are rapidly evolving for long-term cardiovascular monitoring. However, most of the cardiovascular device requires multi-channel physiological signals acquisition, especially in continuous blood pressure (BP) measurement using pulse transition time (PTT) based methods. The multi-devices implementation could impede wearable applications. Objective: This study developed a wearable neck patch device using single-channel impedance plethysmography (IPG) sensing for cardiovascular monitoring, including continuous BP and heart rate (HR) measurement. Methods: IPG-based BP model was derived based on the Bramwell-Hill equation. A patch IPG device was designed and installed above the carotid artery of the subject neck. To validate the BP and HR functions of our device, the Bland-Altman plots were performed to evaluate the estimation error between the reference and the proposed devices within 20 healthy subjects. Results: The BP performance indicates that systolic BP (SBP) estimation error was -0.16 ± 2.97 mmHg and 2.43 ± 1.71 mmHg in terms of mean error (ME) and mean absolute error (MAE), and 0.09 ± 3.30 mmHg and 2.83 ± 1.68 mmHg for diastolic BP (DBP) estimation. Moreover, the HR accuracy has the ME and MAE of 0.02 ± 0.17 bpm and 0.14 ± 0.08 bpm; mean percentage error (MPE) and mean absolute percentage error (MAPE) obtained 0.04 ± 0.23 % and 0.19 ± 0.12 %. Based on statistical results, the BP and HR function of our device satisfied with AAMI/ANSI criteria below 5 ± 8 mmHg and ± 5 bpm or ± 10%. Conclusion: This study implemented a wearable neck patch device with single-channel IPG acquisition that provided two significant cardiovascular parameters of continuous BP and HR, and its performance agreed with standard criteria based on validation with reference sensors. Significance: The proposed proof-of-concept IPG neck patch device has a high potential for wearable applications and low-cost manufacturing in cardiovascular monitoring.

Glycosylation is the most complex post-modification effect of proteins. It participates in many biological processes in the human body and is closely related to many disease states. Among them, N-linked glycosylation is the most contained glycosylation data. However, the current N-linked glycosylation prediction tool does not take into account the serious imbalance between positive and negative data. In this study, we used protein sequence and amino acid characteristics to construct an N-linked glycosylation prediction model called N-GlycoGo. Based on sequence, structure, and function, 11 heterogeneous features were encoded. Further, XGBoost was selected for modeling. Finally, independent testing of human and mouse prediction models showed that N-GlycoGo is superior to other tools with Matthews correlation coefficient (MCC) values of 0.397 and 0.719, respectively, which is higher than other glycosylation site prediction tools. We have developed a fast and accurate prediction tool, N-GlycoGo, for N-linked glycosylation. N-GlycoGo is available at http://ncblab.nchu.edu.tw/n-glycogo/.

Recently, with the emergence of retrieval requirements for certain individual in the same superclass, e.g., birds, persons, cars, fine-grained recognition task has attracted a significant amount of attention from academia and industry. In fine-grained recognition scenario, the inter-class differences are quite diverse and subtle, which makes it challenging to extract all the discriminative cues. Traditional training mechanism optimizes the overall discriminativeness of the whole feature. It may stop early when some feature elements has been trained to distinguish training samples well, leaving other elements insufficiently trained for a feature. This would result in a less generalizable feature extractor that only captures major discriminative cues and ignores subtle ones. Therefore, there is a need for a training mechanism that enforces the discriminativeness of all the elements in the feature to capture more the subtle visual cues. In this paper, we propose a Discrimination-Aware Mechanism (DAM) that iteratively identifies insufficiently trained elements and improves them. DAM is able to increase the number of well learned elements, which captures more visual cues by the feature extractor. In this way, a more informative representation is learned, which brings better generalization performance. We show that DAM can be easily applied to both proxy-based and pair-based loss functions, and thus can be used in most existing fine-grained recognition paradigms. Comprehensive experiments on CUB200-2011, Cars196, Market-1501, and MSMT17 datasets demonstrate the advantages of our DAM based loss over the related state-of-the-art approaches.

Panoptic segmentation is a challenging task aiming to simultaneously segment objects (things) at instance level and background contents (stuff) at semantic level. Existing methods mostly utilize a two-stage detection network to attain instance segmentation results, and a fully convolutional network to produce a semantic segmentation prediction. Post-processing or additional modules are required to handle the conflicts between the outputs from these two nets, which makes such methods suffer from low efficiency, heavy memory consumption and complicated implementation. To simplify the pipeline and decrease computation/memory cost, we propose an one-stage approach called Lightweight Panoptic Segmentation Network (LPSNet), which does not involve a proposal, anchor or mask head. Instead, we predict a bounding box and semantic category at each pixel upon the feature map produced by an augmented feature pyramid, and design a parameter-free head to merge the per-pixel bounding box and semantic prediction into panoptic segmentation output. Our LPSNet is not only efficient in computation and memory, but also accurate in panoptic segmentation. Comprehensive experiments on COCO, Cityscapes and Mapillary Vistas datasets demonstrate the promising effectiveness and efficiency of the proposed LPSNet.

A multiple-time programmable (MTP) twin-bit memory cell with metal floating gate coupled by n-well program gate is proposed and implemented by standard CMOS logic processes. Equipped with high programming gate coupling ratio controlled well by overlap area design, a twin bit cell with a pair of symmetric floating gates is successfully demonstrated. In addition, this memory also shows good endurance and low disturbance during the P/E cycling tests.

The voltage controlled oscillator (VCO) of the frequency modulation continuous wave (FMCW)radar is usually non-linear, which causes the radar range resolution to decrease. The non-linearity of frequency sweep is one of the main problems that limit the range resolution of chirp continuous wave radar. Although the traditional method based on polynomial model can correct most of the nonlinear distortion in the system, it is difficult to suppress the paired echoes caused by periodic nonlinearity, and there is still a large space for improving the quality of range compression. We propose a novel method to measure the systematic phase errors which are required as input to the compensation algorithm. Simulation results show that the proposed algorithm can effectively eliminate the periodic phase error and significantly improve the quality of range compression.

Aiming at the problem that obstacle avoidance algorithms do not distinguish the types of obstacles in the process of obstacle recognition and avoidance, this paper designs a robot obstacle avoidance system including knowledge base by combining prior knowledge with obstacle avoidance algorithm. The knowledge in the knowledge base can be defined by human as prior knowledge to classify obstacles. This makes it possible for the robot to identify obstacles according to the vision sensor in the obstacle avoidance process, and can choose different obstacle avoidance strategies for different obstacles according to the existing knowledge in the knowledge base. The knowledge base takes knowrob as the platform, and the experimental results show that the obstacle avoidance system can effectively avoid obstacles.

Integrity of rails is the foundation of safe rail transportation. It is critical to detect internal rail flaws in time, and one popular solution to this issue is ultrasonic techniques. On the other hand, long short-term memory (LSTM) has been proven in text classification to which we think the ultrasonic rail flaw detection can be quite similar. In this context, this paper proposes a novel text-style sequential modeling method for ultrasonic rail flaw data and a LSTM-based deep learning model for rail flaw detection. Comparative experiments proved the feasibility and remarkable computational efficiency of the proposed modeling method and model.

This paper presents a cell-based 50% Duty-Cycle Correction (DCC) design supporting a super-wide range of clock frequency from 10MHz to 1.2GHz, using a 90nm CMOS process. It can be integrated with a Delay-Locked Loop (DLL) as a convenient post-processing unit while achieving “zero phase shift” in a way that the phase locking result achieved by its precedent DLL is not affected at all. The unique features in this design include: (1) A wide-range and high-resolution Half-Period Tunable Delay Line (HP-TLD), (2) A fast-locking unit to enable our DCC to lock in to a new incoming clock frequency during frequency scaling, and (3) A wide-range and high-resolution Duty-Cycle Judge (DCJ) circuit as a feedback to guide the overall duty-cycle correction process. Post-layout simulation in a 90nm CMOS process is conducted to validate its effectiveness.

The driver's physiological status has enormous value to public traffic safety and cannot be ignored nowadays. In addition, heart rate (HR) is one of the most important indicators of human's health status. When detecting the driver's HR, using traditional contact-type devices might bring about the driver's distraction or discomfort. On the contrary, the remote photoplethysmography (rPPG) technique is a better way to monitor a driver's HR in vehicle applications simply by using a web-camera without interfering the driver. Most of the rPPG studies intended to reduce the interference caused by facial motion or luminance changes in the indoor or controlled scenario, but there are relatively fewer discussions on outdoor scenarios. Consequently, the purpose of this paper is to enhance the rPPG technique to make it suitable for the outdoor driving scenarios and for monitoring the driver's HR in different weather conditions, including daytime and nighttime. We first utilize artificial neural network (ANN) and train multiple personalized ANN models for each driver. For predicting the drivers' HR beat more precisely, we propose the approach, adaptive neural network model selection (ANNMS), which adaptively selects a personalized ANN model based on different noise conditions. Our algorithm eliminates the effect of noises caused by the variations of facial luminance in eight outdoor driving scenarios. The proposed driver's HR beat monitoring system has been evaluated against the state-of-the-art rPPG techniques that are Chrominance signal-based (CHRO) and k-nearest neighbors-based (kNN) algorithms. Compared with the CHRO and kNN algorithms, the ANNMS reduces the mean absolute error from 14.71 bpm (CHRO) and 9.91 (kNN) to 4.51 bpm (ANNMS) and enhances the success-rate-10, the probability in which the absolute error is below 10 bpm, from 44.1% (CHRO) and 56.3% (kNN) to 91.5% (ANNMS).

This paper designs and implements a smart proactive iBeacon system, called GuidingAll, using smartphones to detect abnormal events, guide mobile users, and find moving targets through Internet of Things (IoT) technologies. The GuidingAll system consists of the smartphone APP, guiding server, fixed iBeacon, and our designed smart iBeacon. The smartphone APP is used for real-time guiding and crowdsourced sensing through fixed iBeacons and nearby smartphones. The guiding server is employed for fastest path planning and moving target localization based on the sensing data reported by participating smartphones. The fixed iBeacons are deployed to provide indoor positioning information to smartphones with Bluetooth Low Energy (BLE) scanning, whereas the smart iBeacon is designed to enable proactive event detection and target BLE broadcasting through integrated sensors and associated iBeacon, respectively. In particular, mobile users with smartphones can form crowdsourced sensing networks in places with fixed iBeacons, and GuidingAll can continuously monitor and cooperatively find a specific moving target equipped with the smart iBeacon. An Android-based prototype with fixed and smart iBeacons is implemented to verify the feasibility and correctness of our GuidingAll system.

This paper discusses how to perform online integrity checking for a clock system spanning all over a heterogeneously integrated IC. A clock system in this work is assumed to consist of two major components - (1) the Delay-Locked Loop (DLL), commonly used for module-to-module clock synchronization, and (2) the clock distribution network (with clock buffers and interconnects). For the DLL part, we incorporate a “phase error monitoring scheme”, which is able to detect abnormal safety hazard, e.g., an instantaneous power glitch. For the clock distribution network, we incorporate a periodic self-test scheme featuring a “special short-pulse driven flush test procedure” to detect any worsening Clock Delay Fault (CDF). The proposed method can help identify a failure threat before it strikes havoc. Post-layout simulation results are presented to demonstrate the effectiveness of the proposed schemes.

TKIS-I helmet-mounted colorful imaging sonar is a mechanical scanning single-beam sonar which has high transmission frequency up to 675 KHz and remote detection ability. Currently, there are more than two dozen of it serving the navy of China, which plays an important role in underwater training and rescuing. However, there are some inherent shortcomings such as long scanning time, low imaging efficiency and low data transmission rate. The work of this paper aims to shorten the scanning time of sonar equipment without reducing the resolution of close-range targets. According to the working environment and performance requirements of sonar equipment. This paper researches the Uniform Linear Array (ULA), and designs a reasonable array structure that can form controllable directional beam, enhance target signal and suppress interference. ULA improves the resolutions of close range targets and greatly shorten the scanning time.

A low-capacitance silicon-controlled rectifier for high speed I/O pad protection is implemented with TSMC 7nm bulk FinFET technology. It can achieve much higher ESD robustness per capacitance with better dynamic on-resistance and faster turn-on speed for CDM protection as compared to the prior art.

Traditionally, it has been a difficult task to characterize the quality of a Clock Delay Fault (CDF). Here, a CDF is referred to a delay fault occurring in the clock network that causes an abnormal delay when a clock signal travels through it and thereby causing large-than-expected clock skews at the clock ports of some flip-flops. In a recent work [12], a modified flush test procedure taking short pulses as the test stimuli at selected clock cycles has been proven effective in characterizing a CDF. In this work, we extend this technique to support online Built-In Self-Test (BIST). We investigate two fault detection strategies, namely, valid-range criterion, and valid-span criterion, and we compare their fault detection abilities in terms of the minimum detectable CDF. With the proposed BIST scheme, the health condition of the clock network in a device operating in the field can be inspected on a regular basis so as to take precautions before the clock network breaks down due to deteriorating fault effects.

A novel ESD dual diode with stronger latch-up robustness has been proposed in FinFET technology. By inserting PW/NW straps, the proposed ESD diode can be built without changing DC-IV and ESD characteristics of conventional ESD diode. The experimental and simulation results demonstrate that the proposed structure has been successfully verified in FinFET CMOS platform.

Electric shared car systems have been introduced to a number of cities as a means of reducing traffic congestion and air pollution. There are two types of service: one-way and two-way service. In this paper, the objective is to optimize user demands and carsharing company profits, we investigate the ratio of one-way and two-way cars to satisfy the customers demands between two high-demand areas. First we propose a linear programming model. The revenue consists of customer charges and government subsidies. The cost consists of purchase and operating cost Second, we apply the model to a carsharing company called Xiaoming Travel in Hefei City, China. We consider round trips between two high demand areas, Xinqiao Airport and Mingzhu Square. We find maximal profit can be obtained when a proper ratio of one-way and two-way cars is provided. We also find the relocation rate and relocation coefficient influences the profit. Finally, we provide useful operational suggestions to decision makers of carsharing companies based on our results.

I/O activities of mobile storage are highly synchronous. Flash garbage collection activities in mobile storage introduce extra delay to write requests and negatively impact on user perceived-latency. Runtime write demand is subject to correlation between multiple parameters, such as network connectivity, GPS coordinates, and current time. We propose predicting write demand with a learning algorithm, XGBoost, and conducting background, rate-based garbage collection to optimize write latency without premature, excessive flash erasure. Our method reduced the 99-th percentile write latency by 56% compared to on-demand garbage collection and decreased flash erase count by 51% compared to unconditional background garbage collection.

In this paper, we present a cell-based wide-frequency-range DLL supporting clock frequencies from 10MHz to 1GHz in 90nm. To support fast locking (during frequency scaling), we present novel architecture with a socalled “fast-locking mode” - in which the control code for the Long-Range Delay Block (LRDB) can be quickly determined in a way that the overall delay across the underlying DLL can be roughly one or multiple clock cycle times of the input clock signal. One can use this LRDB as a common building block and integrate it with any arbitrary narrow-range DLL for DVFS (Dynamic Voltage and Frequency Scaling) applications while achieving fast frequency scaling. The functionality is validated by post-layout simulation using a 90nm CMOS process. With the proposed scheme, the long-range locking time can be reduced significantly from 24% to 98%, respectively, for different input clock frequencies.

Rapid detection of the chemical compositions is of great significance to stabilize the quality of cast iron production, which is usually achieved by thermal analysis in front of the furnace. However, the traditional thermal analyzer usually adopts a fixed parametric mathematical model. Under complex and changeable working conditions (such as fluctuation of furnace burden, changes in process conditions and sample cup batches, etc.), the fixed model often causes large deviations in chemical composition detection, and its parameters are not easy to adjust, which greatly limits the popularization and application of thermal analyzers in front of the furnace. To this end, this study is devoted to exploring a more adaptive thermal analysis model. The temperature eigenvalues of cooling curves are taken as input parameters, local weighted linear regression (LWLR) algorithm is used for local fitting, and particle swarm optimization (PSO) algorithm is used to optimize the bandwidth parameter of LWLR. Finally, a PSO-LWLR hybrid model is formed. In a large foundry, the thermal analysis temperature data and corresponding chemical composition data of more than 200 furnaces of hot metal were collected from actual production. By using the above PSO-LWLR model, and comparing with the multiple linear regression (MLR) model and BP artificial neural network (BP-ANN) model, it is proved that the proposed model can achieve higher prediction accuracy. Unlike the global model with fixed parameters, PSO-LWLR is a locally optimized Just-In-Time (JIT) learning model. It can build different optimized local models for different query samples, so it has stronger adaptability and robustness under time-varying working conditions.

In this paper, we present an end-to-end automatic speech recognition system, which successfully employs subword units in a hybrid CTC-Attention based system. The subword units are obtained by the byte-pair encoding (BPE) compression algorithm. Compared to using words as modeling units, using characters or subword units does not suffer from the out-of-vocabulary (OOV) problem. Furthermore, using subword units further offers a capability in modeling longer context than using characters. We evaluate different systems over the LibriSpeech 1000h dataset. The subword-based hybrid CTC-Attention system obtains 6.8% word error rate (WER) on the test_clean subset without any dictionary or external language model. This represents a significant improvement (a 12.8% WER relative reduction) over the character-based hybrid CTC-Attention system., Comment: accepted by ISCSLP 2018

With the rise of the Internet industry and the technique of artificial intelligence, personalized services are increasingly important in recent years for improving user experience and increasing corporates' competitiveness and profits. Precise prediction of customers' behaviors has shown great effects in modern business marketing, especially when making personalized decisions. In this paper, we develop a deep learning network to make personalized predictions of their behaviors among a list of potential choices. The architecture of this model combines each user's features and his historical event lists by sequence-to-sequence (Seq2Seq) structure and make predictions based on his recent event lists. We also modify the long-short- term memory (LSTM) cell forget gate's structure to enhance the attention ability. Such design, called the attetioned LSTM, converges quicker and better while still maintain the similar performance in open dataset IMDB. In addition, in dealing with personalized prediction problems in real-world datasets provided by our cooperative company, our attentioned LSTM achieves a 10% higher precision in average than the standard LSTM model. The advantage is confirmed in evaluation of this generic method on a real dataset of users' behaviors sequences and individuals' attribute profiles from Ant Financial. It also achieves a great result working on the real-world business scene. This model can also achieve a great performance working on the real-world business scene.

Automatic facial animation is a research topic of broad and current interest with widespread impact on various applications. In this paper, we present a novel joint audio-video driven facial animation system. Unlike traditional methods, we incorporate a large vocabulary continuous speech recognition (LVCSR) system to obtain phoneme alignments. The use of LVCSR reduces the high error rate associated with the traditional phoneme recognizer. We also introduce a knowledge guided 3D blendshapes modeling for each phoneme to avoid collecting training data and introducing bias from computer vision generated targets. To further improve the quality, we adopt video tracking and jointly optimize the facial animation by combining both sources. In the evaluations, we present both objective study and several subjective studies on three settings: audio-driven, video-driven, and joint audio-video driven. We find that the quality of our proposed system's facial animation generation surpasses that of the recent state-of-the-art systems.

How to efficiently use the limited wireless bandwidth is of paramount important in the radio wireless network area. This paper proposes a fairness aware and resource reuse algorithm for multicast in LTE network. The proposed algorithm consists two parts: the fairness aware scheme and the resource reuse scheme. Both of these schemes use the multicast transmission index to ensure that all users in multicast groups are able to receive the base layer. The fairness aware scheme always allocates resources to multicast groups according to the users with the poorest channel conditions in order to maximize fairness in resource allocation. The resource reuse scheme allocates resources to users in a given multicast group who are covered by macro eNodeB and pico eNodeBs. Pico eNodeBs are able to reuse resources allocated to other pico eNodeBs to a multicast group in which the users are not covered by pico eNodeBs. This helps to ensure that users covered by pico eNodeBs are able to receive higher video quality. Experimental results verify that the resource reuse scheme provides superior video quality when the number of users covered by the pico eNodeBs increases. In addition, our results demonstrate that the fairness aware scheme enhances fairness in distribution in LTE networks.

Precipitation prediction, such as short-term rainfall prediction, is a very important problem in the field of meteorological service. In practice, most of recent studies focus on leveraging radar data or satellite images to make predictions. However, there is another scenario where a set of weather features are collected by various sensors at multiple observation sites. The observations of a site are sometimes incomplete but provide important clues for weather prediction at nearby sites, which are not fully exploited in existing work yet. To solve this problem, we propose a multi-task convolutional neural network model to automatically extract features from the time series measured at observation sites and leverage the correlation between the multiple sites for weather prediction via multi-tasking. To the best of our knowledge, this is the first attempt to use multi-task learning and deep learning techniques to predict short-term rainfall amount based on multi-site features. Specifically, we formulate the learning task as an end-to-end multi-site neural network model which allows to leverage the learned knowledge from one site to other correlated sites, and model the correlations between different sites. Extensive experiments show that the learned site correlations are insightful and the proposed model significantly outperforms a broad set of baseline models including the European Centre for Medium-range Weather Forecasts system (ECMWF).

A FinFET SCR embedded ESD clamp for power supply protection with low leakage current is demonstrated. The proposed clamp is suitable for low power applications since it reduces ∼87% of leakage current per bigFET width and improves ESD robustness to ∼2X per footprint compared to the conventional RC-triggered clamp.

We report an interfacial modification of sol-gel ZnO/A1-doped ZnO (AZO) bilayer thin film as the electron transport layer (ETL) in the perovskite solar cells. We found that the power conversion efficiency (PCE) of perovskite solar cells was significantly increased from 11.28% to 14.24% with short-circuit current density (J sc ) of 19.53 rnA cm−2,open-circuit voltage (V oc ) of 1.12 V, and fill factor (FF) of 65.1% by employing the AZO thin film as the buffer layer. Moreover, the enhancement in performance was obtained through energy-level offset adjustment of AZO thin film, which allows the better charge separation of perovskite solar cells.

All electronic packages involve with a multi-material system, in which many of the materials or films are susceptible to moisture absorption. Despite dissimilar material properties, moisture transports in a multi-material system from a high "water activity" region to a low one, where water activity is a measure of water energy in a specific substance. This, however, has not been well recognized in electronic packaging industry. Furthermore, moisture concentration gradient is often deemed as the driving force for moisture transport, which inevitably poses a challenging discontinuity issue at interface for moisture diffusion in multi-material systems. Even though several normalization schemes have been developed in the literature, much confusion has existed on the fundamental principle of moisture diffusion. This paper derived an activity-based diffusion model using the concepts of chemical potential and water activity. We showed that the continuity of water activity at interface in dissimilar materials is warranted, and furthermore, many nonlinear water sorption isotherms can be applied in the activity-based model by using a new property called "generalized solubility". The activity-based model thus is capable to study complex moisture diffusion in multi-material system. Moreover, in this paper, the activity-based model was used to unify the different normalization theories, such as solubility-based and the so-called wetness normalization approaches. We also discussed how water sorption isotherm would affect the conventional moisture diffusivity, finding that only for some limiting cases (e.g., Henry sorption isotherm), the "effective moisture diffusivity" becomes independent of moisture concentration. We pointed out that the generalized solubility that are needed to solve the diffusion can be obtained using conventional terms such as saturated moisture concentration and solubility. As demonstration, a numerical example was performed in commercial finite element software to study the moisture diffusion through a bi-material interface under dynamic temperature and humidity conditions. The results from different nonlinear sorption isotherms were compared to demonstrate the capability and versatility of the model. We concluded that the activity-based moisture diffusion model is a unified and versatile approach to study and understand the moisture diffusion mechanism in IC packages.

This paper is to report our study on the moving object detection from surveillance images. For motion detection, some existed methods are used to find specific features between images and then to define the moving speeds of objects. However, human-created features may be difficult to define and to acquire especially when the objects are unknown. In this paper, the scale-invariant feature transform (SIFT) method is adopted to define features for motion detection. In the image, SIFT can be used to catch the properties of scale and rotation invariant. Even if the foreground target is partially obscured and the image is taken in a different angle and distance, SIFT can still have nice matching performances. However, when applied in detecting moving objects, SIFT does not work well due to finding incorrect features in the match. In this study, an enhanced edge detection method is proposed for Laplace of Gaussian (LoG) based SIFT. From the simulation results, it is evident that our proposed method can find more feature matching in the video image.

ITΌ-free inverted SMPV1:PC 71 BM solar cells with an Al doped ZnO (AZO) transparent electrodes are fabricated. The AZO thin film prepared by pulsed laser deposition (PLD) technique exhibits high transmission (>85%) and low sheet resistance (∼30 Ω/sq). The absorption spectrums of devices are also simulated by RCWA method to optimize the performance of photocurrent response. The power conversion efficiency of ITO-free inverted small molecule solar cell can reach 4.03% which is close to that of ITO based small molecule solar cell.

A mode-configurable filtering ΔΣ ADC is utilized as the analog baseband in a Wi-Fi 11ac direct-conversion receiver (RX). The filtering ΔΣ ADC providing both 2nd-order out-of-band filtering and 3rd-order in-band noise shaping is realized with only two opamps. A SAR-quantizer with built-in discrete-time (DT) excess-loop delay (ELD) compensation technique is also adopted. The filtering ΔΣ ADC is clocked at 480 MHz or 960 MHz and achieves 77-to-58 dB dynamic range (DR) in 10 MHz – 80 MHz bandwidth. With the aid of the filtering ability, the interferer DR at 4× bandwidth is at least 71.3 dB over modes. This work is fabricated in 28-nm low-power (LP) technology with 0.06 mm2 of active area. It consumes 3.97 mW or 6.39 mW over different clock rates, resulting a highest Schreier's FoM of 171 dB (BW20) and a best 4×-bandwidth Walden's FoM [2] of 13.3 fJ/c (BW160) among all modes.

An optimized SCR structure was proposed for high turn-on speed and low parasitic capacitance in FinFET CMOS process. Experimental results indicate that the proposed SCR structure delivers the best known results among the literatures (140mA/fF). By adopting the structure, ESD protection design for multi Gb/s transceiver can be simply realized.

We demonstrated novel non-bleaching contrast for super-resolution imaging based on saturation and on/off switching of scattering from plasmonic particles, for the first time. Our study opens up new paradigms for both plasmonics and super-resolution microscopy.

A new DC/DC converter with parallel circuits is presented for medium voltage and power applications. There are five pulse-width modulation (PWM) circuits in the proposed converter to reduce current stress at low voltage side for high output current. These five circuits have the same power switches so that the switch counts are reduced compared with the conventional parallel converters. In order to reduce the converter size, conduction loss and voltage stress of power semiconductors, the series connections of power MOSFETs with high switching frequency instead of IGBTs with low switching frequency are adopted. Thus, the voltage stress of MOSFETs is clamped at half of input voltage. The switched capacitor circuit is adopted to balance input split capacitor voltages. Asymmetric PWM scheme is adopted to generate the necessary switching signals of MOSFETs and regulate output voltage. Based on the resonant behavior at the transition interval of power switches, all MOSFETs are turned on under zero voltage switching (ZVS) from 50% load to 100% load. The circuit configuration, operation principle, converter performance and design example are discussed in detail. Finally, experimental verifications with a 1.92kW prototype are provided to verify the performance of the proposed converter.

The implementation of differential Pirani gauge for accurately measuring wafer-level package pressures was demonstrated. We proposed a multiple-sensor-solution, where two Pirani gauges were constructed under different pressures; one in sealed micro-cavity for measuring pressures and the other one in opened micro-cavity as a reference. Ambient pressure, structural dimension variations, and resistivity differences among wafers/lots, were captured through the differential scheme for error compensations, allowing accurate pressure determinations. Presented Pirani utilized small gaps (∼2µm) between heater and dual heat sinks to obtain wide operation range (0.05∼100 Torr) and high sensitivity (∼10000 ppm/Torr). With 5X error reductions and high stabilities, the proposed device was successfully used in examining reliabilities and monitoring processes of wafer-level packages.