We have thoroughly investigated the effect of on-chip decoupling capacitors on the simultaneous switching noise (SSN) and the radiated emission. Furthermore, we have successfully demonstrated an efficient design method for on-chip decoupling capacitors on an 8-bit microcontroller without increasing the die size, which results in more than 10 dB of suppressed radiated emission.

In this letter, we propose a novel supervised pre-training technique for deep neural network (DNN)-hidden Markov model systems to achieve robust speech recognition in adverse environments. In the proposed approach, our aim is to initialize the DNN parameters such that they yield abstract features robust to acoustic environment variations. In order to achieve this, we first derive the abstract features from an early fine-tuned DNN model which is trained based on a clean speech database. By using the derived abstract features as the target values, the standard error back-propagation algorithm with the stochastic gradient descent method is performed to estimate the initial parameters of the DNN. The performance of the proposed algorithm was evaluated on Aurora-4 DB, and better results were observed compared to a number of conventional pre-training methods.

Prefix caching improves the performance of IP lookup by exploiting spatial and temporal locality of IP references. However, it cannot cache non-leaf prefixes, so several prefix expansion schemes have been proposed to handle those prefixes. Such schemes have some drawbacks to incur modification of routing table or severe miss penalty. We propose an efficient prefix expansion scheme which achieves good performance without additional burden to lookup scheme. In the proposed scheme a non-leaf prefix is expanded to the length of the longest immediate descendant prefix when it is cached. Evaluation result shows our scheme achieves very low miss ratio even though it does not increase the size of routing table and cache miss penalty.

In this letter, we propose a new approach to estimate the degree of noise masking based on a sophisticated model for clean speech distribution. This measure, named as noise masking probability (NMP), is incorporated into the feature compensation technique to achieve robust speech recognition in noisy environments. Experimental results show that the proposed approach improves the performance of the baseline recognition system in the presence of various background noises.

This letter presents a novel frame splitting scheme for an error-robust audio streaming over packet-switching networks. In our approach to perceptual audio coding, an audio frame is split into several subframes based on the network configuration such that each packet can be decoded independently at the receiver. Through a subjective comparison category rating (CCR) test, it is discovered that our approach enhances the quality of the decoded audio signal under the lossy packet-switching networks environment.

In this letter, we introduce new parameters for classifying digitally modulated unknown QAM and PSK signals. Our two parameters for the classification are the variance of magnitude ratios and the mean of mod 2π phase differences. The gain adjustments of amplitudes are not required for the classification. Five different types of QAM constellations and three different types of PSK constellations are tested and the characteristics of our classification parameters are investigated in various SNR environments. Simulation results demonstrate the effectiveness of our proposed technique.

This letter introduces a pre-rejection technique for wireless channel distorted speech with application to automatic speech recognition (ASR). Based on analysis of distorted speech signals over a wireless communication channel, we propose a method to reject the channel distorted speech with a small computational load. From a number of simulation results, we can discover that the pre-rejection algorithm enhances the robustness of speech recognition operation.

In time difference of arrival-based signal source location estimation, geometrical errors are caused by the location of multiple unmanned aerial vehicles (UAV). Herein, we propose a divide-and-conquer algorithm to determine the optimal location for each UAV. Simulations results confirm that multiple UAVs shifted to an optimal position and the location accuracy improved.

Broadband subscriber loop system and ATM switching system are the key equipment for construction of Broadband networks. In this paper, we describe the architecture of access network and the implementation of ATM switching system with multi-service interface for construction of broadband access network. We also represent the design of MAIN-AN (Multi-service Access Integrated Network--Access Node) system as integrated access network platform which enables to accommodate ATM/SDH-based and ATM/PON-based FTTx (Fiber-To-The-x) access architecture simultaneously. The system has a Cross-point ATM Switch Fabric with 10 Gbits/sec throughput and it has been implemented using 0.5 µm CMOS technology. For performance evaluation of it, we simulate it under burst traffic conditions. In addition, we show the implementation of prototype of ASIC (Application Specific Integrated Circuit), MAIN system and its core PBA (Printed circuit Board Assembly) and so on.

This paper presents a CMOS A/D converter based on the folding and interpolating technique. A current steering folder composed of differential pairs allows low power operation and an interpolation is used for high speed with low supply voltage. In a folding circuit, only twenty-three MOSFETs are required to have eight reference voltages of an 8-b A/D converter. The interpolation is implemented with a current division technique to generate 32 folding signals. This approach requires much less area and power consumption than other conventional flash A/D converter. The simulation in a 0.35 µm CMOS process achieves 8-b resolution at 250 Msample/s with power consumption 70 mW at 3.3 V power supply. The preliminary experiment indicates the current steering folder and coarse bits operate as expected.

In this paper, the planar subarray structure to be optimized by using random search method for large active array antenna is presented. Although MPSL of the optimized subarray structure is 1.09dB higher, G/T of the optimized subarray structure is 2.07dB higher than the reference subarray structure.

The literature presents a digitized pension system based on a consortium blockchain, with the aim of overcoming existing pension system challenges such as multiparty collaboration, manual intervention, high turnaround time, cost transparency, auditability, etc. In addition, the adoption of hyperledger fabric and the introduction of smart contracts aim to transform multi-organizational workflow into a synchronized, automated, modular, and error-free procedure.

This paper proposes a fault-resilient ROS platform supporting rapid fault detection and recovery. The platform employs heartbeat-based fault detection and node replication-based recovery. Our prototype implementation on top of the ROS Melodic shows a great performance in evaluations with a Nvidia development board and an inverted pendulum device.

Two simple and high performance multichannel distributed queue dual bus (MDQDB) protocols based on the wavelength division multiplexing (WDM) network technology are proposed, and the network architecture and operations are presented. To be suited for a high speed network, they inherit the advantages of the DQDB protocol such as node simplicity, network flexibility and distributed operations of individual nodes. The network capacity can also be greatly increased with marginal increase of node complexity. Simulation has been done to estimate the performances such as throughput and average access delays for individual nodes. The influence of the bandwidth balancing mechanism on the two protocols is considered at medium, high, and overload conditions. We also investigate the fairness characteristics in asymptotic conditions for various initial states.

MEMS storage devices are new non-volatile secondary storages that have outstanding advantages over magnetic disks. MEMS storage devices, however, are much different from magnetic disks in the structure and access characteristics in the following ways. They have thousands of heads called probe tips and provide the following two major access facilities: (1) flexibility : freely selecting a set of probe tips for accessing data, (2) parallelism: simultaneously reading and writing data with the set of probe tips selected. Due to these characteristics, it is nontrivial to find data placements that fully utilize the capability of MEMS storage devices. In this paper, we propose a simple logical model called the Region-Sector (RS) model that abstracts major characteristics affecting data retrieval performance, such as flexibility and parallelism, from the physical MEMS storage model. We also suggest heuristic data placement strategies based on the RS model. To show the usability of the RS model, we derive new data placements for relational data and two-dimensional spatial data by using these strategies. Experimental results show that the proposed data placements improve the data retrieval performance by up to 4.7 times for relational data and by up to 18.7 times for two-dimensional spatial data of approximately 320 Mbytes compared with those of existing data placements. Further, these improvements are expected to be more marked as the database size grows.

We propose an effective 2d image based end-to-end deep learning model for malware detection by introducing a black & white embedding to reserve bit information and adapting the convolution architecture. Experimental results show that our proposed scheme can achieve superior performance in both of training and testing data sets compared to well-known image recognition deep learning models (VGG and ResNet).

Internal user threats such as information leakage or system destruction can cause significant damage to the organization, however it is very difficult to prevent or detect this attack in advance. In this paper, we propose an anomaly-based insider threat detection method with local features and global statistics over the assumption that a user shows different patterns from regular behaviors during harmful actions. We experimentally show that our detection mechanism can achieve superior performance compared to the state of the art approaches for CMU CERT dataset.

To implement a fast and reliable question-answering system in Korean, we propose a two-pass answer indexer using co-occurrence information between answer candidates and adjacent content words. The two-pass indexer scans documents twice for obtaining local scores and global scores. Then, the two-pass indexer calculates the degrees of association between answer candidates and co-occurring content words. Using this technique, the proposed QA system shortens the response time and enhances the precision.

The objective is to generate a suboptimal weight vector for an adaptive array system operating in a multipath fading CDMA (Code Division Multiple Access) channel. The total computational load of the proposed procedure is about including the update of autocovariance matrix as well as the weight update itself, where N is the number of antenna elements. The performance of the proposed array system is shown in terms of BER (Bit Error Rate), allowable capacity, and convergence rate in practical CDMA signal environments such as IS95 and IS2000 1X.

A 12 Gb/s 10-level pulse amplitude modulation (PAM) serial-link transmitter was implemented using a 0.18 µm CMOS process. The proposed 10-PAM transmitter achieves a channel efficiency of 4 bit/symbol by dual-mode amplitude modulations using 10 differential-mode levels and 3 common-mode levels. The measured maximum data-rate was 12 Gb/s over 0.7-m cable and 2-cm printed circuit board (PCB) traces. The entire transmitter consumes 432 mW such that the figure of merit of the transmitter is 36 pJ/bit. The present work demonstrates the greater channel efficiency of 4 bit/symbol than the currently reported multi-level PAM transmitters.