In this paper, we propose a real-time BSS (Blind Source Separation) system with two microphones that extracts only desired sound sources. Under the assumption that the desired sound sources are close to the microphones, the proposed BSS system suppresses distant sound sources as undesired sound sources. We previously developed a BSS system that can estimate the distance from a microphone to a sound source and suppress distant sound sources, but it was not a real-time processing system. The proposed BSS system is a real-time version of our previous BSS system. To develop the proposed BSS system, we simplify some BSS procedures of the previous system. Simulation results showed that the proposed system can effectively suppress the distant source signals in real-time and has almost the same capability as the previous system.

In this paper, the application of minimum mean square error (MMSE) pre-cancellation prior to belief propagation (BP) is proposed as a detection scheme for overloaded multiple-input multiple-output (MIMO) systems. In overloaded MIMO systems, the loops in the factor graph degrade the demodulation performance of BP. Therefore, the proposed scheme applies MMSE pre-cancellation prior to BP and reduces the number of loops. Furthermore, it is applied to the selected transmit and receive nodes so that the condition number of an inverse matrix in the MMSE weight matrix is minimized to suppress the residual interference and the noise after MMSE pre-cancellation. Numerical results obtained through computer simulation show that the proposed scheme achieves better bit error rate (BER) performance than BP without MMSE pre-cancellation. The proposed scheme improves the BER performance by 2.9-5.6dB at a BER of 5.0×10-3 compared with conventional BP. Numerical results also show that MMSE pre-cancellation reduces the complexity of BP by a factor of 896 in terms of the number of multiplication operations.

The need of telecommunications operators to reduce Capital and Operational Expenditures in networks which traffic is continuously growing has made them search for new alternatives to simplify and automate their procedures. Because of the different transport network segments and multiple layers, the deployment of end-to-end services is a complex task. Also, because of the multiple vendor existence, the control plane has not been fully homogenized, making end-to-end connectivity services a manual and slow process, and the allocation of computing resources across the entire network a difficult task. The new massive capacity requested by Data Centers and the new 5G connectivity services will urge for a better solution to orchestrate the transport network and the distributed computing resources. This article presents and demonstrates a Network Slicing solution together with an end-to-end service orchestration for transport networks. The Network Slicing solution permits the co-existence of virtual networks (one per service) over the same physical network to ensure the specific service requirements. The network orchestrator allows automated end-to-end services across multi-layer multi-domain network segments making use of the standard Transport API (TAPI) data model for both l0 and l2 layers. Both solutions will allow to keep up with beyond 5G services and the higher and faster demand of network and computing resources.

This paper deals with the problem of enumerating 3-edge-connected spanning subgraphs of an input plane graph. In 2018, Yamanaka et al. proposed two enumeration algorithms for such a problem. Their algorithm generates each 2-edge-connected spanning subgraph of a given plane graph with n vertices in O(n) time, and another one generates each k-edge-connected spanning subgraph of a general graph with m edges in O(mT) time, where T is the running time to check the k-edge connectivity of a graph. This paper focuses on the case of the 3-edge-connectivity in a plane graph. We give an algorithm which generates each 3-edge-connected spanning subgraph of the input plane graph in O(n2) time. This time complexity is the same as the algorithm by Yamanaka et al., but our algorithm is simpler than theirs.

Sound event detection (SED) and acoustic scene classification (ASC) are important research topics in environmental sound analysis. Many research groups have addressed SED and ASC using neural-network-based methods, such as the convolutional neural network (CNN), recurrent neural network (RNN), and convolutional recurrent neural network (CRNN). The conventional methods address SED and ASC separately even though sound events and acoustic scenes are closely related to each other. For example, in the acoustic scene “office,” the sound events “mouse clicking” and “keyboard typing” are likely to occur. Therefore, it is expected that information on sound events and acoustic scenes will be of mutual aid for SED and ASC. In this paper, we propose multitask learning for joint analysis of sound events and acoustic scenes, in which the parts of the networks holding information on sound events and acoustic scenes in common are shared. Experimental results obtained using the TUT Sound Events 2016/2017 and TUT Acoustic Scenes 2016 datasets indicate that the proposed method improves the performance of SED and ASC by 1.31 and 1.80 percentage points in terms of the F-score, respectively, compared with the conventional CRNN-based method.

C-C successive approximation register analog-to-digital converter (C-C SAR-ADC) is space-saving architecture compared to SAR-ADC with binary weighted capacitive digital-to-analog converter (CDAC). However, the accuracy of C-C SAR-ADC is degraded due to parasitic capacitance of floating nodes. This paper proposes an algorithm calibrating the non-linearity by γ-estimation to accurately estimate radix greater than 2 required to realize C-C SAR-ADC. Behavioral analyses show that the radix γ-estimation error become within 1.5, 0.4 and 0.1% in case of 8-, 10- and 12-bit resolution ADC, respectively. SPICE simulations show that the γ-estimation satisfies the requirement of 10-bit resolution C-C SAR-ADC. The C-C SAR-ADC using γ-estimation achieves 9.72bit of ENOB, 0.8/-0.5LSB and 0.5/-0.4LSB of DNL/INL.

The performance of packet processing applications is dependent on the memory access speed of network systems. Table lookup requires fast memory access and is one of the most common processes in various packet processing applications, which can be a dominant performance bottleneck. Therefore, in Network Function Virtualization (NFV)-aware environments, on-chip fast cache memories of a CPU of general-purpose hardware become critical to achieve high performance packet processing speeds of over tens of Gbps. Also, multiple types of applications and complex applications are executed in the same system simultaneously in carrier network systems, which require adequate cache memory capacities as well. In this paper, we propose a packet processing architecture that utilizes interleaved 3 Dimensional (3D)-stacked Dynamic Random Access Memory (DRAM) devices as off-chip Last Level Cache (LLC) in addition to several levels of dedicated cache memories of each CPU core. Entries of a lookup table are distributed in every bank and vault to utilize both bank interleaving and vault-level memory parallelism. Frequently accessed entries in 3D-stacked DRAM are also cached in on-chip dedicated cache memories of each CPU core. The evaluation results show that the proposed architecture reduces the memory access latency by 57%, and increases the throughput by 100% while reducing the blocking probability but about 10% compared to the architecture with shared on-chip LLC. These results indicate that 3D-stacked DRAM can be practical as off-chip LLC in parallel packet processing systems.

The traditional RANSAC samples uniformly in the dataset which is not efficient in the task with rich prior information. This letter proposes GUISAC (Guided Sample Consensus), which samples with the guidance of various prior information. In image matching, GUISAC extracts seed points sets evenly on images based on various prior factors at first, then it incorporates seed points sets into the sampling subset with a growth function, and a new termination criterion is used to decide whether the current best hypothesis is good enough. Finally, experimental results show that the new method GUISAC has a great advantage in time-consuming than other similar RANSAC methods, and without loss of accuracy.

This paper studies a novel approach to analysis of switched dynamical systems in perspective of bifurcation and multiobjective optimization. As a first step, we analyze a simple switched dynamical system based on a boost converter with photovoltaic input. First, in a bifurcation phenomenon perspective, we consider period doubling bifurcation sets in the parameter space. Second, in a multiobjective optimization perspective, we consider a trade-off between maximum input power and stability. The trade-off is represented by a Pareto front in the objective space. Performing numerical experiments, relationship between the bifurcation sets and the Pareto front is investigated.

Input devices based on direct touch have replaced traditional ones and become the mainstream interactive technology for handheld devices. Although direct touch interaction proves to be easy to use, its problems, e.g. the occlusion problem and the fat finger problem, lower user experience. Camera-based mobile interaction is one of the solutions to overcome the problems. There are two typical interaction styles to generate camera-based pointing interaction for handheld devices: move the device or move an object before the camera. In the first interaction style, there are two approaches to move a cursor's position across the handheld display: move it towards the same direction or the opposite direction which the device moves to. In this paper, the results of a comparison research, which compared the pointing performances of three camera-based pointing techniques, are presented. All pointing techniques utilized input from the rear-facing camera. The results indicate that the interaction style of moving a finger before the camera outperforms the other one in efficiency, accuracy, and throughput. The results also indicate that within the interaction style of moving the device, the cursor positioning style of moving the cursor to the opposite direction is slightly better than the other one in efficiency and throughput. Based on the findings, we suggest giving priority to the interaction style of moving a finger when deploying camera-based pointing techniques on handheld devices. Given that the interaction style of moving the device supports one-handed manipulation, it also worth deploying when one-handed interaction is needed. According to the results, the cursor positioning style of moving the cursor towards the opposite direction which the device moves to may be a better choice.

In this paper, we propose new generation methods of two-dimensional (2D) optical zero-correlation zone (ZCZ) sequences with the high peak autocorrelation amplitude. The 2D optical ZCZ sequence consists of a pair of a binary sequence which takes 1 or 0 and a bi-phase sequence which takes 1 or -1, and has a zero-correlation zone in the two-dimensional correlation function. Because of these properties, the 2D optical ZCZ sequence is suitable for optical code-division multiple access (OCDMA) system using an LED array having a plurality of light-emitting elements arranged in a lattice pattern. The OCDMA system using the 2D optical ZCZ sequence can be increased the data rate and can be suppressed interference by the light of adjacent LEDs. By using the proposed generation methods, we can improve the peak autocorrelation amplitude of the sequence. This means that the BER performance of the OCDMA system using the sequence can be improved.

Various software fault prediction models have been proposed in the past twenty years. Many studies have compared and evaluated existing prediction approaches in order to identify the most effective ones. However, in most cases, such models and techniques provide varying results, and their outcomes do not result in best possible performance across different datasets. This is mainly due to the diverse nature of software development projects, and therefore, there is a risk that the selected models lead to inconsistent results across multiple datasets. In this work, we propose the use of bandit algorithms in cases where the accuracy of the models are inconsistent across multiple datasets. In the experiment discussed in this work, we used four conventional prediction models, tested on three different dataset, and then selected the best possible model dynamically by applying bandit algorithms. We then compared our results with those obtained using majority voting. As a result, Epsilon-greedy with ϵ=0.3 showed the best or second-best prediction performance compared with using only one prediction model and majority voting. Our results showed that bandit algorithms can provide promising outcomes when used in fault prediction.

In this paper, we present a novel method for vehicle detection based on the Faster R-CNN frame. We integrate MobileNet into Faster R-CNN structure. First, the MobileNet is used as the base network to generate the feature map. In order to retain the more information of vehicle objects, a fusion strategy is applied to multi-layer features to generate a fused feature map. The fused feature map is then shared by region proposal network (RPN) and Fast R-CNN. In the RPN system, we employ a novel dimension cluster method to predict the anchor sizes, instead of choosing the properties of anchors manually. Our detection method improves the detection accuracy and saves computation resources. The results show that our proposed method respectively achieves 85.21% and 91.16% on the mean average precision (mAP) for DIOR dataset and UA-DETRAC dataset, which are respectively 1.32% and 1.49% improvement than Faster R-CNN (ResNet152). Also, since less operations and parameters are required in the base network, our method costs the storage size of 42.52MB, which is far less than 214.89MB of Faster R-CNN(ResNet50).

We developed an 8.4-inch electrostatic-tactile touch display using a segmented-electrode array (30×20) as both tactile pixels and touch sensors. Each pixel can be excited independently so that the electrostatic-tactile touch display allows presenting real localized tactile textures in any shape. A driving scheme in which the tactile strength is independent of the grounding state of the human body by employing two-phased actuation was also proposed and demonstrated. Furthermore, tactile crosstalk was investigated to find it was due to the voltage fluctuation in the human body and it was diminished by applying the aforementioned driving scheme.

Identifying the infection sources in a network, including the sponsor of a network rumor, the servers that inject computer virus into a computer network, or the zero-patient in an infectious disease network, plays a critical role in limiting the damage caused by the infection. A two-source estimator is firstly constructed on basis of partitions of infection regions in this paper. Meanwhile, the two-source estimation problem is transformed into calculating the expectation of permitted permutations count which can be simplified to a single-source estimation problem under determined infection region. A heuristic algorithm is also proposed to promote the estimator to general graphs in a Breadth-First-Search (BFS) fashion. Experimental results are provided to verify the performance of our method and illustrate variations of error detection in different networks.

When applying estimation methods, the issue of outliers is inevitable. The extent of their influence has not been clarified, though several studies have evaluated outlier elimination methods. It is unclear whether we should always be sensitive to outliers, whether outliers should always be removed before estimation, and what amount of precaution is required for collecting project data. Therefore, the goal of this study is to illustrate a guideline that suggests how sensitively we should handle outliers. In the analysis, we experimentally add outliers to three datasets, to analyze their influence. We modified the percentage of outliers, their extent (e.g., we varied the actual effort from 100 to 200 person-hours when the extent was 100%), the variables including outliers (e.g., adding outliers to function points or effort), and the locations of outliers in a dataset. Next, the effort was estimated using these datasets. We used multiple linear regression analysis and analogy based estimation to estimate the development effort. The experimental results indicate that the influence of outliers on the estimation accuracy is non-trivial when the extent or percentage of outliers is considerable (i.e., 100% and 20%, respectively). In contrast, their influence is negligible when the extent and percentage are small (i.e., 50% and 10%, respectively). Moreover, in some cases, the linear regression analysis was less affected by outliers than analogy based estimation.

Bluetooth is a common wireless technology that is widely used as a connection medium between various consumer electronic devices. The receivers mostly adopt the Viterbi algorithm to improve a bit error rate performance but are hampered by heavy hardware complexity and computational load due to a coherent detection and searching for the unknown modulation index. To address these challenges, a non-coherent maximum likelihood estimation detector with an eight-state Viterbi is proposed for Gaussian frequency-shift keying symbol detection against an irrational modulation index, without any knowledge of prior information or assumptions. The simulation results showed an improvement in the performance compared to other ideal approaches.

This paper investigates the security of KCipher-2 against differential attacks. We utilize an MILP-based method to evaluate the minimum number of active S-boxes in each round. We try to construct an accurate model to describe the 8-bit truncated difference propagation through the modular addition operation and the linear transformation of KCipher-2, respectively, which were omitted or simplified in the previous evaluation by Preneel et al. In our constructed model, the difference characteristics neglected in Preneel et al.'s evaluation can be taken into account and all valid differential characteristics can be covered. As a result, we reveal that the minimal number of active S-boxes is 25 over 15 rounds in the related IV setting and it is 17 over 24 rounds in the related IV-key setting. Therefore, this paper shows for the first time that KCipher-2 is secure against the related IV differential attack.

Incident ticket classification plays an important role in the complex system maintenance. However, low classification accuracy will result in high maintenance costs. To solve this issue, this paper proposes a fuzzy output support vector machine (FOSVM) based incident ticket classification approach, which can be implemented in the context of both two-class SVMs and multi-class SVMs such as one-versus-one and one-versus-rest. Our purpose is to solve the unclassifiable regions of multi-class SVMs to output reliable and robust results by more fine-grained analysis. Experiments on both benchmark data sets and real-world ticket data demonstrate that our method has better performance than commonly used multi-class SVM and fuzzy SVM methods.

The residue number system (RNS) is a method for representing an integer x as an n-tuple of its residues with respect to a given set of moduli. In RNS, addition, subtraction, and multiplication can be carried out by independent operations with respect to each modulus. Therefore, an n-fold speedup can be achieved by parallel processing. The main disadvantage of RNS is that we cannot efficiently compare the magnitude of two integers or determine the sign of an integer. Two general methods of comparison are to transform a number in RNS to a mixed-radix system or to a radix representation using the Chinese remainder theorem (CRT). We used the CRT to derive an equation approximating a value of x relative to M, the product of moduli. Then, we propose two algorithms that efficiently evaluate the equation and output a sign bit. The expected number of steps of these algorithms is of order n. The algorithms use a lookup table that is (n+3) times as large as M, which is reasonably small for most applications including cryptography.