Accurate traffic flow prediction is the precondition for many applications in Intelligent Transportation Systems, such as traffic control and route guidance. Traditional data driven traffic flow prediction models tend to ignore traffic self-features (e.g., periodicities), and commonly suffer from the shifts brought by various complex factors (e.g., weather and holidays). These would reduce the precision and robustness of the prediction models. To tackle this problem, in this paper, we propose a CNN-based multi-feature predictive model (MF-CNN) that collectively predicts network-scale traffic flow with multiple spatiotemporal features and external factors (weather and holidays). Specifically, we classify traffic self-features into temporal continuity as short-term feature, daily periodicity and weekly periodicity as long-term features, then map them to three two-dimensional spaces, which each one is composed of time and space, represented by two-dimensional matrices. The high-level spatiotemporal features learned by CNNs from the matrices with different time lags are further fused with external factors by a logistic regression layer to derive the final prediction. Experimental results indicate that the MF-CNN model considering multi-features improves the predictive performance compared to five baseline models, and achieves the trade-off between accuracy and efficiency.

The explosive dynamics present in on-line social networks, typically represented by flaming phenomena, can have a serious impact on not only the sustainable operation of information networks but also on activities in the real world. In order to counter the flaming phenomenon, it is necessary to understand the mechanism underlying the generation of the flaming phenomena within an engineering framework. This paper discusses a new model of the generating mechanism of the flaming phenomena. Our previous study has shown that the cause of flaming phenomena can, by reference to an oscillation model on networks, be understood complex eigenvalues of the matrix formed to describe oscillating phenomena. In this paper, we show that the flaming phenomena can occur due to coupling between degenerated oscillation modes even if all the eigenvalues are real numbers. In addition, we investigate the generation process of flaming phenomena with respect to the initial phases of the degenerated oscillation modes.

This study proposes a Peak-to-Average Power Ratio (PAPR) reduction method using an adaptive Finite Impulse Response (FIR) filter in Orthogonal Frequency Division Multiplexing systems. At the transmitter, an iterative algorithm that minimizes the p-norm of a transmitted signal vector is used to update the weight coefficients of the FIR filter to reduce PAPR. At the receiver, the FIR filter used at the transmitter is estimated using pilot symbols, and its effect can be compensated for by using an equalizer for proper demodulation. Simulation results show that the proposed method is superior to conventional methods in terms of the PAPR reduction and computational complexity. It also shows that the proposed method has a trade-off between PAPR reduction and bit error rate performance.

In recent years, capsule endoscopy has attracted attention as one of the medical devices that examine internal digestive tracts without burdening patients. Wireless power transmission of the capsule endoscope has been researched now, and the power transmission efficiency can be improved by knowing the capsule location. In this paper, we develop a localization method wireless power transmission. Therefore, a simple algorithm for using received signal strength (RSS) has been developed so that position estimation can be performed in real time, and the performance is evaluated by performing three-dimensional localization with eight receiving antennas.

LINE is currently the most popular messaging service in Japan. Communications using LINE are protected by the original encryption scheme, called LINE Encryption, and specifications of the client-to-server transport encryption protocol and the client-to-client message end-to-end encryption protocol are published by the Technical Whitepaper. Though a spoofing attack (i.e., a malicious client makes another client misunderstand the identity of the peer) and a reply attack (i.e., a message in a session is sent again in another session by a man-in-the-middle adversary, and the receiver accepts these messages) to the end-to-end protocol have been shown, no formal security analysis of these protocols is known. In this paper, we show a formal verification result of secrecy of application data and authenticity for protocols of LINE Encryption (Version 1.0) by using the automated security verification tool ProVerif. Especially, since it is claimed that the transport protocol satisfies forward secrecy (i.e., even if the static private key is leaked, security of application data is guaranteed), we verify forward secrecy for client's data and for server's data of the transport protocol, and we find an attack to break secrecy of client's application data. Moreover, we find the spoofing attack and the reply attack, which are reported in previous papers.

This paper presents recognition of anomalously deformed Kana sequences in Japanese historical documents, for which a contest was held by IEICE PRMU 2017. The contest was divided into three levels in accordance with the number of characters to be recognized: level 1: single characters, level 2: sequences of three vertically written Kana characters, and level 3: unrestricted sets of characters composed of three or more characters possibly in multiple lines. This paper focuses on the methods for levels 2 and 3 that won the contest. We basically follow the segmentation-free approach and employ the hierarchy of a Convolutional Neural Network (CNN) for feature extraction, Bidirectional Long Short-Term Memory (BLSTM) for frame prediction, and Connectionist Temporal Classification (CTC) for text recognition, which is named a Deep Convolutional Recurrent Network (DCRN). We compare the pretrained CNN approach and the end-to-end approach with more detailed variations for level 2. Then, we propose a method of vertical text line segmentation and multiple line concatenation before applying DCRN for level 3. We also examine a two-dimensional BLSTM (2DBLSTM) based method for level 3. We present the evaluation of the best methods by cross validation. We achieved an accuracy of 89.10% for the three-Kana-character sequence recognition and an accuracy of 87.70% for the unrestricted Kana recognition without employing linguistic context. These results prove the performances of the proposed models on the level 2 and 3 tasks.

We present an OpenACC-based parallelization implementation of stochastic algorithms for simulating biochemical reaction networks on modern GPUs (graphics processing units). To investigate the effectiveness of using OpenACC for leveraging the massive hardware parallelism of the GPU architecture, we carefully apply OpenACC's language constructs and mechanisms to implementing a parallel version of stochastic simulation algorithms on the GPU. Using our OpenACC implementation in comparison to both the NVidia CUDA and the CPU-based implementations, we report our initial experiences on OpenACC's performance and programming productivity in the context of GPU-accelerated scientific computing.

In this work, we investigate a joint transmit beamforming and artificial noise (AN) covariance matrix design in a multiple-input multiple-output (MIMO) cognitive radio (CR) downlink network with simultaneous wireless information and power transfer (SWIPT), where the malicious energy receivers (ERs) may decode the desired information and hence can be treated as potential eavesdroppers (Eves). In order to improve the secure performance of the transmission, AN is embedded to the information-bearing signal, which acts as interference to the Eves and provides energy to all receivers. Specifically, this joint design is studied under a practical non-linear energy harvesting (EH) model, our aim is to maximize the secrecy rate at the SR subject to the transmit power budget, EH constraints and quality of service (QoS) requirement. The original problem is not convex and challenging to be solved. To circumvent its intractability, an equivalent reformulation of this secrecy rate maximization (SRM) problem is introduced, wherein the resulting problem is primal decomposable and thus can be handled by alternately solving two convex subproblems. Finally, numerical results are presented to verify the effectiveness of our proposed scheme.

In-phase and quadrature-phase imbalance (IQI) at transceivers is one of the serious hardware impairments degrading system performance. In this paper, we study the overall performance of massive multi-user multi-input multi-output (MU-MIMO) systems with IQI at both the base station (BS) and user equipments (UEs), including the estimation of channel state information, required at the BS for the precoding design. We also adopt a widely-linear precoding based on the real-valued channel model to make better use of the image components of the received signal created by IQI. Of particular importance, we propose estimators of the real-valued channel and derive the closed-form expression of the achievable downlink rate. Both the analytical and simulation results show that IQI at the UEs limits the dowlink rate to finite ceilings even when an infinite number of BS antennas is available, and the results also prove that the widely-linear precoding based on the proposed channel estimation method can improve the overall performance of massive MU-MIMO systems with IQI.

This paper introduces an energy-efficient transmit design for multiple-input single-output (MISO) energy-harvesting cognitive radio (CR) networks in the presence of external eavesdroppers (Eves). Due to the inherent characteristics of CR network with simultaneous wireless information and power transfer (SWIPT), Eves may illegitimately access the primary user (PU) bands, and the confidential message is prone to be intercepted in wireless communications. Assuming the channel state information (CSI) of the Eves is not perfectly known at the transmitter, our approach to guaranteeing secrecy is to maximize the secrecy energy efficiency (SEE) by jointly designing the robust beamforming and the power splitting (PS) ratio, under the constraints of total transmit power, harvested energy at secondary receiver (SR) and quality of service (QoS) requirement. Specifically, a non-linear energy harvesting (EH) model is adopted for the SR, which can accurately characterize the property of practical RF-EH circuits. To solve the formulated non-convex problem, we first employ fractional programming theory and penalty function to recast it as an easy-to-handle parametric problem, and then deal with the non-convexity by applying S-Procedure and constrained concave convex procedure (CCCP), which enables us to exploit the difference of concave functions (DC) programming to seek the maximum worst-case SEE. Finally, numerical results are presented to verify the performance of the proposed scheme.

In recent years, a rise in healthy eating has led to various food management applications which have image recognition function to record everyday meals automatically. However, most of the image recognition functions in the existing applications are not directly useful for multiple-dish food photos and cannot automatically estimate food calories. Meanwhile, methodologies on image recognition have advanced greatly because of the advent of Convolutional Neural Network (CNN). CNN has improved accuracies of various kinds of image recognition tasks such as classification and object detection. Therefore, we propose CNN-based food calorie estimation for multiple-dish food photos. Our method estimates dish locations and food calories simultaneously by multi-task learning of food dish detection and food calorie estimation with a single CNN. It is expected to achieve high speed and small network size by simultaneous estimation in a single network. Because currently there is no dataset of multiple-dish food photos annotated with both bounding boxes and food calories, in this work we use two types of datasets alternately for training a single CNN. For the two types of datasets, we use multiple-dish food photos annotated with bounding boxes and single-dish food photos with food calories. Our results showed that our multi-task method achieved higher accuracy, higher speed and smaller network size than a sequential model of food detection and food calorie estimation.

An efficient resistive random access memory (ReRAM) structure is developed for accelerating convolutional neural network (CNN) powered by the in-memory computation. A novel ReRAM cell circuit is designed with two-directional (2-D) accessibility. The entire memory system is organized as a 2-D array, in which specific memory cells can be identically accessed by both of column- and row-locality. For the in-memory computations of CNNs, only relevant cells in an identical sub-array are accessed by 2-D read-out operations, which is hardly implemented by conventional ReRAM cells. In this manner, the redundant access (column or row) of the conventional ReRAM structures is prevented to eliminated the unnecessary data movement when CNNs are processed in-memory. From the simulation results, the energy and bandwidth efficiency of the proposed memory structure are 1.4x and 5x of a state-of-the-art ReRAM architecture, respectively.

This paper discusses the impact of stacking on-chip inductor on power/ground network. Stacking inductor on other circuit components can reduce the circuit area drastically, however, the impact on signal and power integrity is not clear. We investigate the impact by a field-solver, a circuit simulator and real chip measurement. We evaluate three types of power/ground network and various multi-layered inductors. Experimental results show that dense power/ground structures reduce noise although the coupling capacitance becomes larger than that of sparse structures. Measurement in a 65-nm CMOS shows a woven structure makes the noise voltage half compared to a sparse structure.

In December 2018, satellite broadcasting for 4K/8K ultra-high-definition television (UHDTV) will begin in Japan. It will be provided in the 12-GHz (11.7 to 12.75GHz) band with right- and left-hand circular polarizations. BSAT-4a, a satellite used for broadcasting UHDTV, was successfully launched in September 2017. This satellite has not only 12-GHz-band right- and left-hand circular polarization transponders but also a 21-GHz-band experimental transponder. The 21-GHz (21.4 to 22.0GHz) band has been allocated as the downlink for broadcasting satellite service in ITU-R Regions 1 (Europe, Africa) and 3 (Asia Pacific). To receive services provided over these two frequency bands and with dual-polarization, we implement and evaluated a dual-band and dual-circularly polarized parabolic reflector antenna fed by 12- and 21-GHz-band microstrip antenna arrays with a multilayer structure. The antenna is used to receive 12- and 21-GHz-band signals from in-orbit satellites. The measured and experimental results prove that the proposed antenna performs as a dual-polarized antenna in those two frequency bands and has sufficient performance to receive satellite broadcasts.

This paper presents a low-power small-area-overhead physical random number generator utilizing SAR ADC embedded in sensor SoCs. An unpredictable random bit sequence is produced by an existing comparator in typical SAR ADCs, which results in little area overhead. Unlike the other comparator-based physical random number generator, this proposed technique does not require an offset calibration scheme since SAR binary search algorithm automatically converges the two input voltages of the comparator to balance the differential circuit pair. Although the randomness slightly depends on an quantization error due to sharing AD conversion scheme, the input signal distribution enhances the quality of random number bit sequence which can use for various security countermeasures such as masking techniques. Fabricated in 180nm CMOS, 1Mb/s random bit generator achieves high efficiency of 0.72pJ/bit with only 400μm2 area overhead, which occupies less than 0.5% of SAR ADC, while remaining 10-bit AD conversion function.

In this paper, a high-frequency resolution Digital Frequency Discriminator (DFD) IC using a Time to Digital Converter (TDC) and an edge counter for Instantaneous Frequency Measurement (IFM) is proposed. In the proposed DFD, the TDC measures the time of the maximum periods of divided RF short pulse signals, and the edge counter counts the maximum number of periods of the signal. By measuring the multiple periods with the TDC and the edge counter, the proposed DFD improves the frequency resolution compared with that of the measuring one period because it is proportional to reciprocal of the measurement time of TDC. The DFD was fabricated using 0.18-um SiGe-BiCMOS. Frequency accuracy below 0.39MHz and frequency precision below 1.58 MHz-RMS were achieved during 50 ns detection time in 0.3 GHz to 5.5 GHz band with the temperature range from -40 to 85 degrees.

With advances in information technology and the development of big data, manual operation is unlikely to be a smart choice for stock market investing. Instead, the computer-based investment model is expected to bring investors more accurate strategic analysis and more effective investment decisions than human beings. This paper aims to improve investor profits by mining for critical information in the stock data, therefore helping big data analysis. We used the R language to find the technical indicators in the stock market, and then applied the technical indicators to the prediction. The proposed R package includes several analysis toolkits, such as trend line indicators, W type reversal patterns, V type reversal patterns, and the bull or bear market. The simulation results suggest that the developed R package can accurately present the tendency of the price and enhance the return on investment.

To minimize the annotation costs associated with training semantic segmentation models and object detection models, weakly supervised detection and weakly supervised segmentation approaches have been extensively studied. However most of these approaches assume that the domain between training and testing is the same, which at times results in considerable performance drops. For example, if we train an object detection network using only web images showing a large object at the center, it can be difficult for the network to detect multiple small objects. In this paper, we focus on training a CNN with only web images and achieve object detection in the wild. A proposal-based approach can address the problem associated with differences in domains because web images are similar to images of the proposal. In both domains, the target object is located at the center of the image and the ratio of the size of the target object to the size of the image is large. Several proposal methods have been proposed to detect regions with high “object-ness.” However, many of these proposals generate a large number of candidates to increase the recall rate. Considering the recent advent of deep CNNs, methods that generate a large number of proposals exhibit problems in terms of processing time for practical use. Therefore, we propose a CNN-based “food-ness” proposal method in this paper that requires neither pixel-wise annotation nor bounding box annotation. Our method generates proposals through backpropagation and most of these proposals focus only on food objects. In addition, we can easily control the number of proposals. Through experiments, we trained a network model using only web images and tested the model on the UEC FOOD 100 dataset. We demonstrate that the proposed method achieves high performance compared to traditional proposal methods in terms of the trade-off between accuracy and computational cost. Therefore, in this paper, we propose an intermediate approach between the traditional proposal approach and the fully convolutional approach. In particular, we propose a novel proposal method that generates high“food-ness” regions using fully convolutional networks based on the backward approach by training food images gathered from the web.

The real-time state-of-health (SOH) estimation of lithium-ion batteries for electric vehicles (EV) is essential to EV maintenance. According to situations in practical applications such as long EV battery capacity test time, unavailability of regular daily tests, and availability of full-life-cycle charge data of EV recorded on the charging facility big data platform, this paper studies an online in-use EV state-of-health estimation method using iterated extended Gaussian process regression-Kalman filter (GPR-EKF) to incorporate lithium-ion battery data at the macro time scale and the micro time scale based on daily charge data of electric vehicles. This method proposes a kernel function GPR (Gaussian process regression) integrating neutral network with cycles to conduct fitting for data at the macro time scale to determine colored measurement noise; in addition, fragment charge data at the micro time scale is adjusted with real-time iteration to be used as the state equation, which effectively addresses issues of real-time SOC calibration and nonlinearization. The pertinence, effectiveness and real-time performance of the model algorithm in online battery state-of-health estimation is verified by actual data.

The 3D holographic display has long been expected as a future human interface as it does not require users to wear special devices. However, its heavy computation requirement prevents the realization of such displays. A recent study says that objects and holograms with several giga-pixels should be processed in real time for the realization of high resolution and wide view angle. To this problem, first, we have adapted a conventional FFT algorithm to a GPU cluster environment in order to avoid heavy inter-node communications. Then, we have applied several single-node and multi-node optimization and parallelization techniques. The single-node optimizations include a change of the way of object decomposition, reduction of data transfer between the CPU and GPU, kernel integration, stream processing, and utilization of multiple GPUs within a node. The multi-node optimizations include distribution methods of object data from host node to the other nodes. Experimental results show that intra-node optimizations attain 11.52 times speed-up from the original single node code. Further, multi-node optimizations using 8 nodes, 2 GPUs per node, attain an execution time of 4.28 sec for generating a 1.6 giga-pixel hologram from a 3.2 giga-pixel object. It means a 237.92 times speed-up of the sequential processing by CPU and 41.78 times speed-up of multi-threaded execution on multicore-CPU, using a conventional FFT-based algorithm.