Search

Energy efficient routing in wireless sensor networks has attracted attention from researchers in both academia and industry, most recently motivated by the opportunity to use SDN (software defined network)-inspired approaches. These problems are NP-hard, with algorithms needing computation time which scales faster than polynomial in the problem size. Consequently, heuristic algorithms are used in practice, which are unable to guarantee optimally. In this short paper, we show proof-of-principle for the use of a quantum annealing processor instead of a classical processor, to find optimal or near-optimal solutions very quickly. Our preliminary results for small networks show that this approach using quantum computing has great promise and may open the door for other significant improvements in the efficacy of network algorithms.

Information-as-a-Service (InfaaS) is a key concept for collaborative disaster management applications. For developing a collaborative disaster management application based on InfaaS concept, the control in infrastructure-layer is required. We have been developing InfaaS-adaptive Disaster Management Application Platform (InfaaS-AP) to support the construction and operation of collaborative disaster management applications that satisfy InfaaS concept. InfaaS-AP leverages multiple existing Software-Defined technologies, each of which has a dedicated controller. All controllers are centrally managed by Resource Management System (RMS) in InfaaS-AP. Though availability is critical for InfaaS-AP, RMS in current InfaaS-AP is a single point of failure. It is due to the connectivity between RMS and the controllers. RMS with redundant configuration requires maintaining connectivity after fail-over. In addition, since each controller has its own redundancy mechanism, the behavior of the redundancy mechanism makes it difficult to maintain the connectivity. From such a situation, we propose the connectivity-aware redundancy control module. The proposed module not only handles the fail-over of RMS but also monitors and interrupts the behavior of the redundancy mechanism of each controller. In the evaluation, we conducted experiments in which one of the controllers or RMS was stopped and checked the behavior of the proposed module.

This study addresses the problem of automating the sewing process in the garment industry. While automatic feeding during sewing has been achieved, overlapping and feeding two a pair of cloths still require a human operator. The difficulty lies in manipulating a soft material. Instead of using a generic robot finger, we use a horizontal dual-arm manipulator with rollers as end effectors, which always hold down the cloth during manipulation. We validated our system through experiments with a prototype that can feed and sew a single piece of cloth in a single operation.

High-performance computing (HPC) cluster systems sometimes adopt a two-layered file system composed of local and global file systems to achieve both capacity and performance in storage. In such a cluster system, the input data of an application needs to be staged from the global storage into the local storage, and the output data needs to be staged from the local storage out to the global storage. This staging operation must be efficiently and quickly performed to gain higher job throughput because an inefficient staging operation prevents waiting job requests from being executed. In particular, in the case of the cluster system with the oversubscribed interconnect shared by the storage and the computing nodes, the inter-node communication and this staging operation traffic collides, which may degrade the job throughput. In this research, we focus on the traffic collision of the inter-node communication and the staging traffic to improve job throughput, targeting the cluster system with the oversubscribed interconnect where these two types of traffic flow. In other words, whether the dynamic control of the traffic flow derived from the staging operation leads to the improvement in the job throughput or not is investigated. For the investigation, we present a traffic collision avoidance method to dynamically configure a set of data paths for each type of the traffic only while the staging operation is conducted. The evaluation in this article shows that the proposed method avoids a traffic collision and accelerates the staging operation by 22.0% on our cluster system. Also, this evaluation indicates the overhead of the application incurred by the proposed method is negligible. Furthermore, 8.7% of the job execution time is reduced by the proposed method.

Neuromorphic computers offer the opportunity for low-power, efficient computation. Though they have been primarily applied to neural network tasks, there is also the opportunity to leverage the inherent characteristics of neuromorphic computers (low power, massive parallelism, collocated processing and memory) to perform non-neural network tasks. Here, we demonstrate how an approach for performing sparse binary matrix-vector multiplication on neuromorphic computers. We describe the approach, which relies on the connection between binary matrix-vector multiplication and breadth first search, and we introduce the algorithm for performing this calculation in a neuromorphic way. We validate the approach in simulation. Finally, we provide a discussion of the runtime of this algorithm and discuss where neuromorphic computers in the future may have a computational advantage when performing this computation.

Blockchain is a technology that is essentially defined as a decentralized storage system for transactions, or ledger of all transactions or digital events that have been shared among the participating groups. A ridesharing system helps riders to reach a particular destination driven by the owner of the ride. There are many existing ridesharing services available in the market today. But most of these services are centralized and hosted by a third party which gives them the authority to monitor features like fare calculation, user data, etc. Privacy and trust are major concerns in such a system. The purpose of our paper is to make a decentralized application for ride-sharing where all transactions, fare calculation, matching and information is stored on a Distributed Ledger. The ledger will be accessible to all the peers in the network. This will match users with rides in a decentralized way without relying on third parties of any centralized network which makes the system transparent and reliable. The data on blockchain is immutable. Hence, this technology is used to store rides and user information to maintain security and ensure user privacy.

This work demonstrates a novel 1Tb 3D Flash memory chip that has an area efficiency of 10.4Gb/mm2 in a 3b/cell technology. Using a circuit under array (CUA) design technique and over 170 word-line (WL) layers, the chip achieves 33% higher bit density than prior 3b/cell work [1], and better density than in 4b/cell technology [3]. This paper discusses the challenges advanced 3D Flash memories face: using over 100 WL layers results in large parasitic loads and decreases read/program speed, and its complicated operation increases test costs. On the other hand, as high bandwidth is also required, this chip supports a 2.0Gbps IO transfer rate, while maintaining signal integrity. This work introduces four new key technologies to address these difficulties. 1) Asynchronous independent plane read (AIPR), with a 4-plane architecture to improve system-level performance. 2) Enhanced sensing that enables faster read time $(t_{R})$. 3) IO-DCC (duty cycle correction) training for high-speed DDR operation. 4) A scan chain to improve test coverage and cost effectiveness.

An unsuitable clearance value may lead to secondary crack formation in sheet metal forming, especially in sheet metal blanking. Cracks lead to rough edges, resulting in a loss of productivity in terms of surface finish quality and dimensional precision. In the present work, tests are conducted with the uni-punch tool as a cutting tool and IS277GI as material for processing, on the power press. Punch penetration in blanking is varied to find a depth at which crack begins in sheet metal. After punching, shear angle and angle of fracture and punch penetration from the available ranges are considered as input parameters. The optimum clearance value obtained by plotting angles versus per cent clearance is experimentally calculated. The input values are fed to the formulated Neural Network. For various unknown input parameters, this qualified neural network forecasts the clearance. The results of predictions are in line of experimental concepts.

Quality control is an important issue in semiconductor manufacturing. Statistical process control (SPC) is known as a powerful method for accomplishing process stability and reducing variability. In this paper, we adopt the quality-oriented statistical process control (QOSPC) method. In QOSPC, product quality test data, such as electrical performance and product reliability, are incorporated into the process control procedure. QOSPC has two major challenges: extracting process variables that affect product quality, and determining quality control limits (QCLs) for each variable. In this work, we fully exploit a Bayesian approach to resolve both of these challenges simultaneously. We introduced a linear bathtub model that contains parameters corresponding to QCLs as obvious change points and fit the model to the observed data by Bayesian inference (BI). In our experiments with artificial datasets, we demonstrated that the values of QCLs and their confidence, by which we can judge whether the measured process variable is related to product quality, are estimated successfully by BI. We verified the robustness of our method by testing it repeatedly. The proposed method reduced the human labor cost for extracting quality-related process variables and determining QCLs by 93%.

Interconnection is the need for a widespread power system for the pooling of power in every crisis, enabling maximum utilization of the grid. Consequently, disturbance on a stressed line can induce low-frequency electromechanical oscillations. Identifying the modes is crucial to take corrective measures and to avoid the potential threat of grid collapse. Matrix Pencil Method is an efficient modal identification method used to identify these oscillations. The basic Matrix Pencil Method identifies trivial modes alongside dominant modes thereby leading to erroneous control. Trivial modes can be distinguished from the dominant modes based on their energy content. Therefore, the paper aims at identifying dominant modes using the Matrix Pencil Method combined with an energy sorting technique. Further, the performance of this approach is comparatively studied with energy sorted Prony Analysis.

In this paper, we introduce a distribution system of synthesized data of Japanese population using Interdisciplinary Large-scale Information Infra-structures in Japan. Synthetic population is synthesized based on the statistics of the census that are conducted by the government and publicly released. Therefore, the synthesized data have no privacy data. However, it is easy to estimate the compositions of households, working status in a certain area from the synthetic population. Therefore, we currently distribute the synthesized data only for public or academic purposes. For academic purposes, it is important to encourage scholars or researchers to use a large-scale data of households, we define protection levels for the attributes in the synthetic populations. According to the protection levels, we distribute the data with proper attributes to those who try to use them. We encourage researchers to use the synthetic populations to be familiar to large-scale data processing.

Today's deep learning models are primarily trained on CPUs and GPUs. Although these models tend to have low error, they consume high power and utilize large amount of memory owing to double precision floating point learning parameters. Beyond the Moore's law, a significant portion of deep learning tasks would run on edge computing systems, which will form an indispensable part of the entire computation fabric. Subsequently, training deep learning models for such systems will have to be tailored and adopted to generate models that have the following desirable characteristics: low error, low memory, and low power. We believe that deep neural networks (DNNs), where learning parameters are constrained to have a set of finite discrete values, running on neuromorphic computing systems would be instrumental for intelligent edge computing systems having these desirable characteristics. To this extent, we propose the Combinatorial Neural Network Training Algorithm (CoNNTrA), that leverages a coordinate gradient descent-based approach for training deep learning models with finite discrete learning parameters. Next, we elaborate on the theoretical underpinnings and evaluate the computational complexity of CoNNTrA. As a proof of concept, we use CoNNTrA to train deep learning models with ternary learning parameters on the MNIST, Iris and ImageNet data sets and compare their performance to the same models trained using Backpropagation. We use following performance metrics for the comparison: (i) Training error; (ii) Validation error; (iii) Memory usage; and (iv) Training time. Our results indicate that CoNNTTA models use 32 × less memory and have errors at par with the Backpropagation models.

We are conducting research on P-VCASS (Pedestrian - Vehicle Collision Avoidance Support System) to avoid collision between pedestrians and vehicles using wireless communications. However, there are only two judgments as to give a warning or not. To match the human senses, we have proposed a new risk calculation algorithm using fuzzy reasoning to decide the traffic state between vehicles and pedestrians is dangerous or safe. Through several experiments, we show that the proposed system can give warning at appropriate timing and properly.