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.

We introduce a method to estimate the attractiveness of a food photo. It extracts image features focusing on the appearances of 1) the entire food, and 2) the main ingredients. To estimate the attractiveness of an arbitrary food photo, these features are integrated in a regression scheme. We also constructed and released a food image dataset composed of images of ten food categories taken from 36 angles and accompanied with attractiveness values. Evaluation results showed the effectiveness of integrating the two kinds of image features.

A novel adaptive cross-layer optimal power allocation (OPA) scheme over physical layer and data-link layer for two-way relaying system with amplify-and-forward policy (TWR-AF) is proposed in this paper. Our goal is to find the optimal power allocation factors under each channel state information (CSI) to maximize the sum throughput of two sources under total transmit power constraint in the physical layer while guaranteeing the statistical delay quality-of-service (QoS) requirement in the data-link layer. By integrating information theory with the concept of effective capacity, the OPA problem is formulated into an optimization problem to maximize the sum effective capacity. It is solved through Lagrange multiplier approach, and the optimal power allocation factors are presented. Simulations are developed and the results show that the proposed cross-layer OPA scheme can achieve the best sum effective capacity with relatively low complexity when compared with other schemes. In addition, the proposed cross-layer OPA scheme achieves the maximal sum effective capacity when the relay is located in (or near) the middle of the two source nodes, and the sum effective capacity becomes smaller when the difference between two QoS exponents becomes larger.

In this paper, we study the power allocation (PA) scheme design for energy efficiency (EE) maximization with discrete-rate adaptive modulation (AM) in the downlink distributed antenna system (DAS). By means of the Karush-Kuhn-Tucker (KKT) conditions, an optimal PA scheme with closed-form expression is derived for maximizing the EE subject to maximum transmit power and target bit error rate (BER) constraints, where the number of active transmit antennas is also derived for attaining PA coefficients. Considering that the optimal scheme needs to calculate the PA of all transmit antennas for each modulation mode, its complexity is extremely high. For this reason, a low-complexity suboptimal PA is also presented based on the antenna selection method. By choosing one or two remote antennas, the suboptimal scheme offers lower complexity than the optimal one, and has almost the same EE performance as the latter. Besides, the outage probability is derived in a performance evaluation. Computer simulation shows that the developed optimal scheme can achieve the same EE as the exhaustive search based approach, which has much higher complexity, and the suboptimal scheme almost matches the EE of the optimal one as well. The suboptimal scheme with two-antenna selection is particularly effective in terms of balancing performance and complexity. Moreover, the derived outage probability is in good agreement with the corresponding simulation.

Joint transmit and receive antenna selection (JTRAS) for transceive spatial modulation (TRSM) is investigated in this paper. A couple of low-complexity and efficient JTRAS algorithms are proposed to improve the reliability of TRSM systems by maximizing the minimum Euclidean distance (ED) among all received signals. Specifically, the QR decomposition based ED-JTRAS achieves near-optimal error performance with a moderate complexity reduction as compared to the optimal ED-JTRAS method. The singular value decomposition based ED-JTRAS achieves sub-optimal error performance with a significant complexity reduction. Simulation results show that the proposed methods remarkably improve the system reliability in both uncorrelated and spatially correlated Rayleigh fading channels, as compared to the conventional norm based JTRAS method.

Accurate segmentation of the region in the iris picture has a crucial influence on the reliability of the recognition system. In this letter, we present an end to end deep neural network based on U-Net. It uses dense connection blocks to replace the original convolutional layer, which can effectively improve the reuse rate of the feature layer. The proposed method takes U-net's skip connections to combine the same-scale feature maps from the upsampling phase and the downsampling phase in the upsampling process (merge layer). In the last layer of downsampling, it uses dilated convolution. The dilated convolution balances the iris region localization accuracy and the iris edge pixel prediction accuracy, further improving network performance. The experiments running on the Casia v4 Interval and IITD datasets, show that the proposed method improves segmentation performance.

In this letter, we investigate the physical layer security in multi-user multi-relay networks, where each relay is not merely a traditional helper, but at the same time, can become a potential eavesdropper. We first propose an efficient low-complexity user and relay selection scheme to significantly reduce the amount of channel estimation as well as the amount of potential links for comparison. For the proposed scheme, we derive the closed-form expression for the lower bound of ergodic secrecy rate (ESR) to evaluate the system secrecy performance. Simulation results are provided to verify the validity of our expressions and demonstrate how the ESR scales with the number of users and relays.

Temporal behavior is a primary aspect of business process executions. Herein, we propose a temporal outlier detection and analysis method for business processes. Particularly, the method performs correlation analysis between the execution times of traces and activities to determine the type of activities that significantly influences the anomalous temporal behavior of a trace. To this end, we describe the modeling of temporal behaviors considering different control-flow patterns of business processes. Further, an execution time matrix with execution times of activities in all traces is constructed by using the event logs. Based on this matrix, we perform temporal outlier detection and correlation-based analysis.

In this research, we propose a novel method to determine fingerprint liveness to improve the discriminative behavior and classification accuracy of the combined features. This approach detects if a fingerprint is from a live or fake source. In this approach, fingerprint images are analyzed in the differential excitation (DE) component and the centralized binary pattern (CBP) component, which yield the DE image and CBP image, respectively. The images obtained are used to generate a two-dimensional histogram that is subsequently used as a feature vector. To decide if a fingerprint image is from a live or fake source, the feature vector is processed using support vector machine (SVM) classifiers. To evaluate the performance of the proposed method and compare it to existing approaches, we conducted experiments using the datasets from the 2011 and 2015 Liveness Detection Competition (LivDet), collected from four sensors. The results show that the proposed method gave comparable or even better results and further prove that methods derived from combination of features provide a better performance than existing methods.

This paper introduces a method that identifies human activity from the height and Doppler Radar Cross Section (RCS) information detected by Multiple-Input Multiple-Output (MIMO) radar. This method estimates the three-dimensional target location by applying the MUltiple SIgnal Classification (MUSIC) method to the observed MIMO channel; the Doppler RCS is calculated from the signal reflected from the target. A gesture recognition algorithm is applied to the trajectory of the temporal transition of the estimated human height and the Doppler RCS. In experiments, the proposed method achieves over 90% recognition rate (average).

A novel configuration data compression technique for coarse-grained reconfigurable architectures (CGRAs) is proposed. Reducing the size of configuration data of CGRAs shortens the reconfiguration time especially when the communication bandwidth between a CGRA and a host CPU is limited. In addition, it saves energy consumption of configuration cache and controller. The proposed technique is based on a multicast configuration technique called RoMultiC, which reduces the configuration time by multicasting the same data to multiple PEs (Processing Elements) with two bit-maps. Scheduling algorithms for an optimizing the order of multicasting have been proposed. However, the multicasting is possible only if each PE has completely the same configuration. In general, configuration data for CGRAs can be divided into some fields like machine code formats of general perpose CPUs. The proposed scheme confines a part of fields for multicasting so that the possibility of multicasting more PEs can be increased. This paper analyzes algorithms to find a configuration pattern which maximizes the number of multicasted PEs. We implemented the proposed scheme to CMA (Cool Mega Array), a straight forward CGRA as a case study. Experimental results show that the proposed method achieves 40.0% smaller configuration than a previous method for an image processing application at maximum. The exploration of the multicasted grain size reveals the effective grain size for each algorithm. Furthermore, since both a dynamic power consumption of the configuration controller and a configuration time are improved, it achieves 50.1% reduction of the energy consumption for the configuration with a negligible area overhead.

Spectrum sensing is a fundamental requirement for cognitive radio, and it is a challenging problem in impulsive noise modeled by symmetric alpha-stable (SαS) distributions. The Gaussian kernelized energy detector (GKED) performs better than the conventional detectors in SαS distributed noise. However, it fails to detect the DC signal and has high computational complexity. To solve these problems, this paper proposes a more efficient and robust detector based on a Gaussian function (GF). The analytical expressions of the detection and false alarm probabilities are derived and the best parameter for the statistic is calculated. Theoretical analysis and simulation results show that the proposed GF detector has much lower computational complexity than the GKED method, and it can successfully detect the DC signal. In addition, the GF detector performs better than the conventional counterparts including the GKED detector in SαS distributed noise with different characteristic exponents. Finally, we discuss the reason why the GF detector outperforms the conventional counterparts.

In recent years, intelligent transportation system (ITS) techniques have been widely exploited to enhance the quality of public services. As one of the worldwide leaders in recycling, Taiwan adopts the waste collection and disposal policy named “trash doesn't touch the ground”, which requires the public to deliver garbage directly to the collection points for awaiting garbage collection. This study develops a travel time prediction system based on data clustering for providing real-time information on the arrival time of waste collection vehicle (WCV). The developed system consists of mobile devices (MDs), on-board units (OBUs), a fleet management server (FMS), and a data analysis server (DAS). A travel time prediction model utilizing the adaptive-based clustering technique coupled with a data feature selection procedure is devised and embedded in the DAS. While receiving inquiries from users' MDs and relevant data from WCVs' OBUs through the FMS, the DAS performs the devised model to yield the predicted arrival time of WCV. Our experiment result demonstrates that the proposed prediction model achieves an accuracy rate of 75.0% and outperforms the reference linear regression method and neural network technique, the accuracy rates of which are 14.7% and 27.6%, respectively. The developed system is effective as well as efficient and has gone online.

Machine learning algorithms are becoming more and more popular in current era. Data preprocessing especially feature selection is helpful for improving the performance of those algorithms. A new powerful feature selection algorithm is proposed. It combines the advantages of ant colony optimization and brain storm optimization which simulates the behavior of human beings. Six classical datasets and five state-of-art algorithms are used to make a comparison with our algorithm on binary classification problems. The results on accuracy, percent rate, recall rate, and F1 measures show that the developed algorithm is more excellent. Besides, it is no more complex than the compared approaches.

This paper proposed an iterative soft interference canceller (IC) referred to as turbo equalizer for the self-coherent detection, and extrinsic information transfer (EXIT) chart based irregular low density parity check (LDPC) code optimization for the turbo equalizer in optical fiber short-reach transmissions. The self-coherent detection system is capable of linear demodulation by a single photodiode receiver. However, the self-coherent detection suffers from the interference induced by signal-signal beat components, and the suppression of the interference is a vital goal of self-coherent detection. For improving the error-free signal detection performance of the self-coherent detection, we proposed an iterative soft IC with the aid of forward error correction (FEC) decoder. Furthermore, typical FEC code is no longer appropriate for the iterative detection of the turbo equalizer. Therefore, we designed an appropriate LDPC code by using EXIT chart aided code design. The validity of the proposed turbo equalizer with the appropriate LDPC is confirmed by computer simulations.

Pre-trained CNNs on ImageNet have been widely used in object tracking for feature extraction. However, due to the domain mismatch between image classification and object tracking, the submergence of the target-specific features by noise largely decreases the expression ability of the convolutional features, resulting in an inefficient tracking. In this paper, we propose a robust tracking algorithm with low-dimensional target-specific feature extraction. First, a novel cascaded PCA module is proposed to have an explicit extraction of the low-dimensional target-specific features, which makes the new appearance model more effective and efficient. Next, a fast particle filter process is raised to further accelerate the whole tracking pipeline by sharing convolutional computation with a ROI-Align layer. Moreover, a classification-score guided scheme is used to update the appearance model for adapting to target variations while at the same time avoiding the model drift that caused by the object occlusion. Experimental results on OTB100 and Temple Color128 show that, the proposed algorithm has achieved a superior performance among real-time trackers. Besides, our algorithm is competitive with the state-of-the-art trackers in precision while runs at a real-time speed.

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.

Due to progressing process technology, yield of chips is reduced by timing violation caused by delay variation of gates and wires in fabrication. Recently, post-silicon delay tuning, which inserts programmable delay elements (PDEs) into clock trees before the fabrication and adjusts the delays of the PDEs to recover the timing violation after the fabrication, is promising to improve the yield. Although post-silicon delay tuning improves the yield, it increases circuit area and power consumption since the PDEs are inserted. In this paper, a PDE structure is taken into consideration to reduce the circuit area and the power consumption. Moreover, a delay selection algorithm, and a clustering method, in which some PDEs are merged into a PDE and the PDE is inserted for multiple registers, are proposed to reduce the circuit area and the power consumption. In computational experiments, the proposed method reduced the circuit area and the power consumption in comparison with an existing method.

This work presents an FSK inductive-coupling transceiver using a load-modulated transmitter and LC-oscillator-based receiver for energy-budget-unbalanced applications. By introducing the time-domain load modulated transmitter for FSK instead of the conventional current-driven scheme, energy reduction of the transmitter side is possible. For verifying the proposed scheme, a test chip was fabricated in 65nm CMOS, and two chips were stacked for verifying the inter-chip communication. The measurement results show 0.64fJ/bit transmitter power consumption while its input voltage is 60mV, and the communication distance is 150μm. The footprint of the transmitter is 0.0016mm2.

Background subtraction algorithms generate a background model of the monitoring scene and compare the background model with the current video frame to detect foreground objects. In general, most of the background subtraction algorithms fail to detect foreground objects when the scene illumination changes. An entropy based background subtraction algorithm is proposed to address this problem. The proposed method adapts to illumination changes by updating the background model according to differences in entropy value between the current frame and the previous frame. This entropy based background modeling can efficiently handle both sudden and gradual illumination variations. The proposed algorithm is tested in six video sequences and compared with four algorithms to demonstrate its efficiency in terms of F-score, similarity and frame rate.