論文 - 大槻　知明

Learning and Analysis of Damping Factor in Massive MIMO Detection Using BP Algorithm with Node Selection

Tachibana J., Ohtsuki T.

IEEE Access （IEEE Access）  8   96859 - 96866 2020年

　概要を見る

© 2013 IEEE. In a massive multiple-input multiple-output (MIMO) system, belief propagation (BP) detection is known as a method to separate and detect received signals. In BP detection, a MIMO channel is represented by a factor graph and the transmitted symbols are estimated by message passing. However, the convergence property of BP deteriorates due to multiple loops included in the MIMO channel. As a method to improve the convergence property and the detection performance, the damped BP that averages two successive messages with a weighing factor (called damping factor) is known. To train the damping factors off-line for each antenna configuration, deep neural network-based damped BP (DNN-dBP) has been reported. The problem with DNN-dBP is that the detection performance deteriorates when there is a difference of the channel correlation between training and test. This is because the optimal damping factors vary with the channel correlation. In this paper, to solve this issue, we derive the damping factors of BP with the node selection (NS) method that selects nodes to be updated to lower spatial correlation using DNN-dBP. By applying the NS method, the channel correlation among the selected nodes in BP detection is lowered. Therefore, the proposed method can improve the detection performance deterioration due to the mismatches of the channel correlations between training and test in DNN-dBP. In addition, the convergence property of BP is improved by applying the NS method. Therefore, the proposed method has the same detection performance with low computational complexity as the conventional DNN-dBP. By computer simulation, it is shown that the proposed method significantly improves the bit error rate (BER) performance deterioration due to the mismatches of the channel correlations between training and test in DNN-dBP. The results also show that the proposed method can show the same BER performance with low computational complexity as the conventional DNN-dBP. We also investigate the distribution of the trained damping factors and evaluate the tendency of that.

• Access to Document (DOI)

Energy efficiency analysis in cache-enabled D2D-Aided heterogeneous cellular networks

Panahi F.H., Panahi F.H., Ohtsuki T.

IEEE Access （IEEE Access）  8   19540 - 19554 2020年

　概要を見る

© 2013 IEEE. Small cells (SCs) offer a promising approach to meeting the exponentially growing data rate demands. However, dense deployment of SCs can degrade the energy-efficiency (EE) of the network due to the additional deployed base stations (BSs). Under heterogeneous network (HetNet) deployments, SCs can be dynamically switched off for energy saving when traffic load decreases. We do this by defining a load-dependent transmission power coefficient (TPC) for SC BSs. In addition, with Device-to-Device (D2D) communication, mobile users in proximity can establish a direct link and bypass the BSs, thereby offloading the network infrastructure and providing further improvement of EE. In this paper, the objective is to rely on both D2D communications and sleeping SC BSs to offload traffic from the mains powered macrocell BSs leading to energy saving in the network. Furthermore, in order to consider D2D requirements under the practical application scenarios, we assume a framework for wireless video sharing, where users can store popular video files and share files via D2D communication. To begin with, we derive the EE expression of the cache-enabled D2D-aided HetNet. Then, to maximize the EE of the cache-enabled D2D-aided HetNet, the optimal load-dependent TPC for SC BSs is obtained under constraints on both network's coverage and rate. Using simulations, we will investigate the potential effects of the TPC defined for SC BSs as well as the introduced D2D layer on the network EE.

• Access to Document (DOI)

ECG Signal Reconstruction via Doppler Sensor by Hybrid Deep Learning Model with CNN and LSTM

Yamamoto K., Hiromatsu R., Ohtsuki T.

IEEE Access （IEEE Access）  8   130551 - 130560 2020年

　概要を見る

© 2013 IEEE. An Electrocardiogram (ECG) is a typical method used to detect heartbeat, and an ECG signal analysis enables the detection of some heart diseases. However, the ECG-based heartbeat detection requires device attachment, which is not preferred for daily use. A Doppler sensor could be a device used to enable the non-contact heartbeat detection. In this paper, we propose a Doppler sensor-based ECG signal reconstruction method by a hybrid deep learning model with Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). An ECG signal can be reconstructed by relating features of a heartbeat signal obtained by a Doppler sensor to those of the ECG signal. Thus, we construct the deep learning model that extracts the spatial and temporal features from the heartbeat signal by CNN and LSTM. Based on the extracted features, the ECG signal is reconstructed. We conducted experiments to observe heartbeat against 9 healthy subjects without heart disease. The experimental results showed that our method performed ECG signal reconstruction with a correlation coefficient of 0.86 between the reconstructed and actual ECG signals, even without attaching devices. The results indicate that it is possible to remotely reconstruct an ECG signal from a heartbeat signal via a Doppler sensor.

• Access to Document (DOI)

CNN-based respiration rate estimation in indoor environments via MIMO FMCW radar

Yamamoto K., Toyoda K., Ohtsuki T.

2019 IEEE Global Communications Conference, GLOBECOM 2019 - Proceedings （2019 IEEE Global Communications Conference, GLOBECOM 2019 - Proceedings）  2019年12月

ISSN  9781728109626

　概要を見る

© 2019 IEEE. Respiration is known to reflect our health condition, which motivates researchers to develop various radar-based respiration rate estimation methods. However, these conventional methods do not work, when a subject is not right in front of the radar. In this paper, we propose a novel CNN (Convolutional Neural Network)-based respiration rate estimation method in indoor environments via a MIMO (Multiple-Input Multiple-Output) FMCW (Frequency Modulated Continuous Wave) radar. A MIMO FMCW radar can estimate the DoA (Direction of Arrival) and the distance between a MIMO FMCW radar and an object. Thus, the respiration can be captured based on the phase variation at a subject's location. However, even when the advanced signal processing, e.g., MUSIC (MUltiple SIgnal Classification) algorithm, is used, it is difficult to estimate the DoA and the distance in indoor environments due to the large effect of multipath. To deal with this problem, in the proposed method, phase variations against various locations are calculated from the received signals of a MIMO FMCW radar, and then spectrograms are calculated from the phase variations. Each spectrogram is subsequently fed into the CNN that outputs the respiration rates, e.g., 0.1 Hz, 0.2 Hz, and non-respiration, i.e., a spectrogram without the effect of respiration, where is one of the deep learning techniques that have been successfully applied to the image recognition. Through the experiments we confirmed that except for when microwaves were not transmitted directly toward a subject's chest due to furniture, the proposed method accurately estimated the respiration rate, regardless of the situation.

• Access to Document (DOI)

FMCW radar-based anomaly detection in toilet by supervised machine learning classifier

Takabatake W., Yamamoto K., Toyoda K., Ohtsuki T., Shibata Y., Nagate A.

2019 IEEE Global Communications Conference, GLOBECOM 2019 - Proceedings （2019 IEEE Global Communications Conference, GLOBECOM 2019 - Proceedings）  2019年12月

ISSN  9781728109626

　概要を見る

© 2019 IEEE. With the increase in the population of elderly, a technique for monitoring them is receiving more and more attention all over the world. In particular, it is highly demanded to develop the system that detects a fainting elderly in a toilet, since such an accident is likely to get worse due to the late detection. In this paper, we propose an anomaly detection method in toilet using an FMCW(Frequency Modulated Continuous Wave) radar, which can measure the distance between an object and an FMCW radar. As anomalies, we determine as the situations where a subject faints while stooping on the toilet seat, leaning on the backrest, leaning on the side wall, and lying on the floor after the fall. With an FMCW radar, a fainting subject could be detected based on whether the estimated distance between a subject and an FMCW radar is almost constant. However, a subject might be sitting still, and also the distance to not a subject but a floor could be estimated depending on a subject's posture. To address this problem, in the proposed method, the time when the estimated distance is almost constant is detected, and then the detected time is classified into the time due to a faint or a non-faint by a supervised machine learning classifier. The used features are extracted based on the data, e.g., the estimated distance, just before the detected time. To evaluate the anomaly detection accuracy of the proposed method, we conducted the experiments to observe not only the aforementioned abnormal behavior but also the normal behavior, e.g., the button operation and sitting down. As a result, in terms of the classification of abnormal and normal behavior, our proposed method achieved the F- measure of 0.91, where F-measure is the overall score of recall and precision rates.

• Access to Document (DOI)

Dual-ascent inspired iterative transmit precoding approaches for multiple access spatial modulation

Cao Y., Ohtsuki T.

2019 IEEE Global Communications Conference, GLOBECOM 2019 - Proceedings （2019 IEEE Global Communications Conference, GLOBECOM 2019 - Proceedings）  2019年12月

ISSN  9781728109626

　概要を見る

© 2019 IEEE. In this paper, we investigate the multiple access spatial modulation (MASM) in multiple-input multiple-output (MIMO) systems with the aid of dual-ascent inspired iterative transmit precoding (TPC) methods. To find a global optimum against the maximum minimum Euclidean distance (MMD) problem, as well as to reach a fast convergence rate with low-complexity, we first study the peculiarities of the convex optimization methods that take the dual- ascent method into account. This beneficially provides us an insight into developing a new dual- ascent method which is capable of bringing robustness to the existing dual-ascent method. Then, we introduce an evolving MASM-MIMO system by imposing non-stationary time-varying TPC parameters, and thereby resulting in a guaranteed Euclidean distance (GED) aided TPC method. Numerical results reveal that our proposals are capable of promoting a significant bit error rate (BER) performance improvement comparing with the existing methods for the MASM-MIMO systems, as well as facilitating a fast convergence with low- complexity.

• Access to Document (DOI)

Orthogonality Structure Designs for Generalized Precoding Aided Spatial Modulation

Cao Y., Ohtsuki T.

IEEE Wireless Communications Letters （IEEE Wireless Communications Letters）  8 （ 5 ） 1406 - 1409 2019年10月

ISSN  21622337

　概要を見る

© 2012 IEEE. In this letter, we propose an orthogonality structure design (OSD) for the generalized precoding aided spatial modulation (GPSM) to overcome the performance degradation caused by correlated fading channels. To improve the bit error rate (BER) performance of GPSM system, we devise an efficient receive antenna subset selection (RAS) technique relied on the mechanisms of maximizing the equivalent channel gain. To facilitate a better trade-off between performance and computational complexity, we study the peculiarities of the Hermitian matrix which provides an important insight for conceiving orthogonality conditions to the channel matrix of GPSM system. We prove that the performance gain in terms of BER can be attained with the aids of the proposed efficient RAS, and OSD-aided schemes with low-complexity. In particular, our results demonstrate that the OSD-aided scheme is capable of providing significant BER performance improvement for GPSM systems, compared to that of current strategies.

• Access to Document (DOI)

Low complexity and high accuracy channel interpolation with dividing URA into Small URAs for 3D Massive MIMO

Kuriyama M., Ohtsuki T.

IEEE Vehicular Technology Conference （IEEE Vehicular Technology Conference）  2019-September 2019年09月

ISSN  9781728112206

　概要を見る

© 2019 IEEE. Massive multiple-input multiple-output (MIMO) is a technology that uses many antennas at base station (BS), and can realize advanced beamforming and spatial multiplexing. However, since Massive MIMO has many antennas, there exist problems such as the increase of overhead including pilot signals used for channel estimation and feedback of channel state information (CSI). To reduce the overhead, there is a technique of transmitting pilot signals only from some antennas and obtaining channels of the remaining antennas by the minimum mean square error (MMSE) interpolation based on spatial correlation. In addition, to perform the MMSE interpolation with low computational complexity, there is a method of interpolating channels by dividing the uniform rectangular antenna array (URA) into the vertical and horizontal linear directions. However, there is a problem that channel interpolation accuracy becomes low with dividing into the linear arrays. In this paper, we propose a method of channel interpolation with dividing the entire URA into a set of smaller URAs where interpolation is performed in each URA. Through computer simulation we show that the proposed method reduces the computational complexity compared with that without dividing, and improves the channel interpolation accuracy compared with that with dividing into linear arrays.

• Access to Document (DOI)

Multi-class sentiment analysis on twitter: Classification performance and challenges

Bouazizi M., Ohtsuki T.

Big Data Mining and Analytics （Big Data Mining and Analytics）  2 （ 3 ） 181 - 194 2019年09月

ISSN  20960654

　概要を見る

© 2020 The author(s). Sentiment analysis refers to the automatic collection, aggregation, and classification of data collected online into different emotion classes. While most of the work related to sentiment analysis of texts focuses on the binary and ternary classification of these data, the task of multi-class classification has received less attention. Multi-class classification has always been a challenging task given the complexity of natural languages and the difficulty of understanding and mathematically "quantifying"how humans express their feelings. In this paper, we study the task of multi-class classification of online posts of Twitter users, and show how far it is possible to go with the classification, and the limitations and difficulties of this task. The proposed approach of multi-class classification achieves an accuracy of 60.2% for 7 different sentiment classes which, compared to an accuracy of 81.3% for binary classification, emphasizes the effect of having multiple classes on the classification performance. Nonetheless, we propose a novel model to represent the different sentiments and show how this model helps to understand how sentiments are related. The model is then used to analyze the challenges that multi-class classification presents and to highlight possible future enhancements to multi-class classification accuracy.

• Access to Document (DOI)

MUSIC-based Non-contact Heart Rate Estimation with Adaptive Window Size Setting

Yamamoto K., Toyoda K., Ohtsuki T.

Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS （Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS）     6073 - 6076 2019年07月

ISSN  9781538613115

　概要を見る

© 2019 IEEE. Continuous HR (Heart Rate) monitoring enables the stress estimation in daily life. A Doppler sensor could be a key device to facilitate the non-contact HR estimation. As one of the Doppler sensor-based HR estimation methods, we have previously proposed a MUSIC (MUltiple SIgnal Classification)-based HR estimation method. MUSIC is the algorithm widely used as a tool to estimate DOA (Direction of Arrival). In our previous method, MUSIC spectrum is calculated in each sliding window, and then HR is estimated by the maximum peak detection over the MUSIC spectrum. However, when HR changes largely within the window, several peaks due to heartbeats appear over the MUSIC spectrum, which might cause the incorrect peak detection. Hence, an adaptive window is required so that only one peak appears. In this paper, we propose a MUSIC-based HR estimation method with an adaptive window size setting. When several peaks due to heartbeats appear over the MUSIC spectrum, our proposed method shortens the time window and re-calculates the MUSIC spectrum, which is repeated until only one peak appears. The experimental results showed that our method outperformed not only our previous one but also the other existing MUSIC-based HR estimation one in terms of the estimation accuracy of the HR, the stress indexes CVI (Cardiac Vagal Index) and CSI (Cardiac Sympathetic Index).

• Access to Document (DOI)

Overloaded Wireless MIMO Switching for Secure Wireless Communication Exchanging to Untrusted Relay

Takahashi A., Sasamori F., Handa S., Takyu O., Fujii T., Ohtsuki T.

International Conference on Ubiquitous and Future Networks, ICUFN （International Conference on Ubiquitous and Future Networks, ICUFN）  2019-July   121 - 125 2019年07月

ISSN  9781728113395

　概要を見る

© 2019 IEEE. Wireless MIMO Switch makes the information exchange among terminals by two time steps. As a result, the highly efficiency of information exchanging is achieved. If the relay is impersonated by eavesdropper, it can exploit the information during the relay process, where the relay exploiting the information is the untrusted relay. In the proposed wireless MIMO switch, the stations whose number is the number of antennas plus one access the relay. As a result, the relay station hardly decodes each information owing to the overloaded access. In each station, firstly, one information signal is decoded by the interference cancellation of its own signal and the zero forcing of relay filter. Since one information signal is shared to all the stations, each station can demodulate the information because the overloaded condition is mitigated by the shared information signal. We confirm the effect of proposed scheme by computer simulation.

• Access to Document (DOI)

Novel CA-CFAR Approach for Improvement of Doppler Sensor-based Heart Rate Variability Estimation

Hiromatsu R., Yamamoto K., Toyoda K., Ohtsuki T.

Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS （Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS）     796 - 799 2019年07月

ISSN  9781538613115

　概要を見る

© 2019 IEEE. Heart rate variability is one of major physiological parameters to reflect our stress, which has motivated researchers to investigate a Doppler sensor-based non-contact heartbeat interval estimation algorithm. As one of such methods, we have previously proposed a spectrogram-based method. In this method, the spectrum that might be due to heartbeats is integrated over a spectrogram, and then heartbeat interval is estimated by detecting peaks over the integrated spectrum. However, when a subject moves, undesired peaks with large amplitude appear, which causes the incorrect peak detection. As one of the technique to eliminate the undesired peaks with large amplitude, there is CA-CFAR (Cell Average-Constant False Alarm Rate). CA-CFAR is the technique to detect a signal, when the amplitude of a signal exceeds a threshold calculated with average amplitude of signals before and after the investigated one. However, depending on the duration of body movements, the influence of body movements might be included within the signals used for the threshold calculation, which might results in the detection failure of undesired peaks. This is because the length of GT (Guard Time) is fixed, where GT is the time to prevent the signal used for the threshold calculation from including the investigated signal components. To solve this problem, we propose a novel CA-CFAR in which the length of GT is set as the latest peak interval and only the signal before the investigated one is used so that the influence of body movements does not affect the threshold calculation. Through the experiments where a subject moves, i.e., typing, we confirmed that our spectrogram-based heart rate variability estimation method with the proposed CA-CFAR outperformed the one with CA-CFAR based on fixed GT by the RMSE (Root Mean Squared Error) between the estimated heartbeat interval and the ground truth value of the one.

• Access to Document (DOI)

Performance Evaluation of Information-Sharing Scheme among Multiple Destinations with Mirroring Null Steering in Single Antenna Networks

Noguchi T., Takyu O., Fujii T., Ohtsuki T.

International Conference on Information Networking （International Conference on Information Networking）  2019-January   417 - 420 2019年05月

ISSN  9781538683507

　概要を見る

© 2019 IEEE. This paper clarifies the secure capacity in a secure information sharing among multiple destinations by using artificial noise and mirroring null steering (MNS) constructed by cooperative relays in single antenna networks. The MNS can selectively pass the desired signal to all destinations and remove artificial noise. Because the destinations archive high-quality received signals and the other systems cannot, secure information sharing can be achieved from a physical layer security perspective.

• Access to Document (DOI)

Identification of Darknet Markets' Bitcoin Addresses by Voting Per-Address Classification Results

Kanemura K., Toyoda K., Ohtsuki T.

ICBC 2019 - IEEE International Conference on Blockchain and Cryptocurrency （ICBC 2019 - IEEE International Conference on Blockchain and Cryptocurrency）     154 - 158 2019年05月

ISSN  9781728113289

　概要を見る

© 2019 IEEE. Bitcoin is a decentralized digital currency whose transactions are recorded in a common ledger, so called blockchain. Due to the anonymity and lack of law enforcement, Bitcoin has been misused in darknet markets which deal with illegal products, such as drugs and weapons. Therefore from the security forensics aspect, it is demanded to establish an approach to identify newly emerged darknet markets' transactions and addresses. In this paper, we thoroughly analyze Bitcoin transactions and addresses related to darknet markets and propose a novel identification method of darknet markets' addresses. To improve the identification performance, we propose a voting based method which decides the labels of multiple addresses controlled by the same user based on the number of the majority label. Through the computer simulation with more than 200K Bitcoin addresses, it was shown that our voting based method outperforms the nonvoting based one in terms of precision, recal, and F1 score. We also found that DNM's addresses pay higher fees than others, which significantly improves the classification.

• Access to Document (DOI)

Underdetermined Blind Separation using Multi-Subspace Representation in Time-Frequency Domain

Wang L., Ohtsuki T.

IEEE International Conference on Communications （IEEE International Conference on Communications）  2019-May 2019年05月

ISSN  9781538680889

　概要を見る

© 2019 IEEE. Blind source separation (BSS) is a technique to recognize the multiple talkers from the multiple observations received by some sensors without any prior knowledge information. The problem is that the mixing is always complex, i.e., nonlinear, underdetermined mixture, such as the case where sources are mixed with some direction angles, or where the number of sensors is less than that of sources. In this paper, we propose a multi-subspace representation based BSS approach that allows the mixing process to be nonlinear and underdetermined. The approach relies on a multi-layer representation and sparse representation in time-frequency (TF) domain. By parameterizing such subspaces, we can map the observed signals in the feature space with the coefficient matrix from the parameter space. We then exploit the linear mixture in the feature space that corresponds to the nonlinear mixture in the input space. Once such subspaces are built, the coefficient matrix can be constructed by solving an l1-regularization on the coding coefficient vector. Relying on the TF representation, the target matrix can be constructed in a sparse mixture TF vectors with a fewer computational cost. The experiments are run on the observations that are generated from nonlinear functions, and that are collected with some direction angles in a virtual room environment. The proposed approach exhibits a higher separation accuracy than that of the conventional algorithms.

• Access to Document (DOI)

Spectrogram-Based Simultaneous Heartbeat and Blink Detection using Doppler Sensor

Yamamoto K., Toyoda K., Ohtsuki T.

IEEE International Conference on Communications （IEEE International Conference on Communications）  2019-May 2019年05月

ISSN  9781538680889

　概要を見る

© 2019 IEEE. Heartbeats and a blink are two major physiological signals that provide crucial information on our stress and drowsiness. Hence, it is highly demanded to detect simultaneously heartbeats and blinks in many applications. A Doppler sensor could be a key device to facilitate the non-contact heartbeat and blink detection in daily life. Although many Doppler sensor-based heartbeat and blink detection methods have been independently proposed, when heartbeats and blinks are detected simultaneously with one Doppler sensor, the detection accuracies of such heartbeat and blink detection methods get degraded because of at least two issues: (i) the low SNR (Signal-to-Noise Ratio) of each signal reflected from a subject's chest and face, and (ii) the similarity of the spectrum distribution of heartbeats and a blink. In this paper, we propose a spectrogram-based simultaneous heartbeat and blink detection using one Doppler sensor. In the proposed method, to extract the spectra that might be due to heartbeats and a blink, the spectra on a spectrogram are integrated. Blink detection is then performed by classifying the peaks of the integrated spectrum into a peak due to a blink or a non-blink based on a supervised machine learning classifier trained with a set of the time domain and the time-frequency domain features. Based on the non-blink peaks, heartbeats are detected considering the RRI (R-R Interval) estimated before the investigated peak to prevent the incorrect heartbeat detection. Through the experiments in the case where microwaves are transmitted from one Doppler sensor toward a body including a chest and a face, our proposal has been shown to be able to simultaneously detect heartbeats and blinks with high accuracy.

• Access to Document (DOI)

Secrecy Performance Analysis of Wireless Sensor Networks

Ai Y., Cheffena M., Ohtsuki T., Zhuang H.

IEEE Sensors Letters （IEEE Sensors Letters）  3 （ 5 ）  2019年05月

　概要を見る

© 2017 IEEE. The wireless sensor network (WSN) plays a fundamentally important role in the realization of Internet of Things and Industry 4.0. In this article, we study the physical layer security performance of a large-scale ad hoc and homogeneous WSN. To accurately evaluate the secrecy performance of the large-scale WSN, we take into account the randomness of the interfering nodes (both in terms of the number of nodes and their respective position) based on Poisson point process in our analysis. We derive novel expressions for the average secrecy capacity and secrecy outage probability, including the effects of fading, path loss, and the network's spatial randomness. The results demonstrate the impacts of several factors such as node intensity and transmission power in large-scale WSN on secrecy performance, which should be seriously taken into consideration while designing WSNs.

• Access to Document (DOI)

Precoding Aided Generalized Spatial Modulation with K Transmit Antenna Groups

Cao Y., Ohtsuki T.

IEEE International Conference on Communications （IEEE International Conference on Communications）  2019-May 2019年05月

ISSN  9781538680889

　概要を見る

© 2019 IEEE. In this paper, we divide antennas at the transmitter into K transmit antenna groups (TAGs) and extend the generalized spatial modulation (GSM) idea to a new precoding aided multiple-input multiple-output (MIMO) system, which is referred to as precoding aided GSM (PGSM) system. To select the receive antenna subset with low complexity, a class of efficient receive antenna selection techniques with low computational complexity are devised. To reduce the complexity of maximum likelihood (ML) detector, a low complexity iterative greedy (IG) algorithm relied ML (IG-ML) detector is also derived. Simulation results show that the bit error rate (BER) performance gain up to 3.5 dB is obtained by the proposed PGSM with IG-ML detector over the PGSM with ordered block minimum mean-squared error (OBMMSE) detector at a given BER of 10-4. It is also shown that the proposed PGSM with IG-ML detector achieves a similar BER performance in high signal-to-noise power ratio (SNR) region compared to that of an ideal precoding for the GSM with optimal ML detector.

• Access to Document (DOI)

Polynomial Networks Representation of Nonlinear Mixtures with Application in Underdetermined Blind Source Separation

Wang L., Ohtsuki T.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings （ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings）  2019-May   3687 - 3691 2019年05月

ISSN  9781479981311

　概要を見る

© 2019 IEEE. Similar to the deep architectures, a novel multi-layer architecture is used to extend the linear blind source separation (BSS) method to the nonlinear case in this paper. The approach approximates the nonlinearities based on a polynomial network, where the layer of our network begins with the polynomial of degree 1, up to build an output layer that can represent data with a small bias by a good approximate basis. Relying on several transformations of the input data, with higher-level representation from lower-level ones, the networks are to fulfill a mapping implicitly to the high-dimensional space. Once the polynomial networks are built, the coefficient matrix can be estimated by solving an l1-regularization on the coding coefficient vector. The experiment shows that the proposed approach exhibits a higher separation accuracy than the comparison algorithms.

• Access to Document (DOI)

Non-Negative Matrix Factorization-Based Blind Source Separation for Non-Contact Heartbeat Detection

Ye C., Toyoda K., Ohtsuki T.

IEEE International Conference on Communications （IEEE International Conference on Communications）  2019-May 2019年05月

ISSN  9781538680889

　概要を見る

© 2019 IEEE. Recently, through exploiting the spectral sparsity of heartbeat component, a heartbeat detection method using a stochastic gradient approach has enabled a high-resolution of heartbeat spectrum reconstruction by Doppler radar signal, which also suppresses the residual noises after signal decomposition. However, the interference from respiration and/or body motion often corrupts the decomposition of signal by singular spectrum analysis (SSA), resulting in an inaccurate extraction of heartbeat component. In this paper, a non-negative matrix factorization (NMF)-based blind source separation (BSS) is first applied to non-contact heartbeat detection for better heartbeat extraction, incorporating the stochastic gradient approach. Specifically, motion noise is taken into account as one of sources, achieving relatively stable separation in various scenarios. In our proposed BSS approach, the spectrogram originated from radar signal is decomposed twice by NMF, which is used to learn the basis spectra (BS) relying on spectral correlation. Experimental results showed the improved accuracy and robustness of our method over conventional methods, on the heart rate (HR) measurement against subjects' sitting still or typewriting.

• Access to Document (DOI)