Real-Time Detection of Intruders Using an Acoustic Sensor and Internet-of-Things Computing.

Modern home automation systems include features that enhance security, such as cameras and radars. This paper proposes an innovative home security system that can detect burglars by analyzing acoustic signals and instantly notifying the authorized person(s). The system architecture incorporates the concept of the Internet of Things (IoT), resulting in a network and a user-friendly system. The proposed system uses an adaptive detection algorithm, namely the "short-time-average through long-time-average" algorithm. The proposed algorithm is implemented by an IoT device (Arduino Duo) to detect people's acoustical activities for the purpose of home/office security. The performance of the proposed system is evaluated using 10 acoustic signals representing actual events and background noise. The acoustic signals were generated by the sounds of keys shaking, the falling of a small object, the shrinking of a plastic bag, speaking, footsteps, etc. The effects of different algorithms' parameters on the performance of the proposed system have been thoroughly investigated.

Cooperative Multiband Spectrum Sensing Using Radio Environment Maps and Neural Networks.

Cogitive radio networks (CRNs) require high capacity and accuracy to detect the presence of licensed or primary users (PUs) in the sensed spectrum. In addition, they must correctly locate the spectral opportunities (holes) in order to be available to nonlicensed or secondary users (SUs). In this research, a centralized network of cognitive radios for monitoring a multiband spectrum in real time is proposed and implemented in a real wireless communication environment through generic communication devices such as software-defined radios (SDRs). Locally, each SU uses a monitoring technique based on sample entropy to determine spectrum occupancy. The determined features (power, bandwidth, and central frequency) of detected PUs are uploaded to a database. The uploaded data are then processed by a central entity. The objective of this work was to determine the number of PUs, their carrier frequency, bandwidth, and the spectral gaps in the sensed spectrum in a specific area through the construction of radioelectric environment maps (REMs). To this end, we compared the results of classical digital signal processing methods and neural networks performed by the central entity. Results show that both proposed cognitive networks (one working with a central entity using typical signal processing and one performing with neural networks) accurately locate PUs and give information to SUs to transmit, avoiding the hidden terminal problem. However, the best-performing cognitive radio network was the one working with neural networks to accurately detect PUs on both carrier frequency and bandwidth.

On the Analysis of PM/FM Noise Radar Waveforms Considering Modulating Signals with Varied Stochastic Properties.

Noise Radar technology is the general term used to describe radar systems that employ realizations of a given stochastic process as transmit waveforms. Originally, carriers modulated in amplitude by a Gaussian random signal, derived from a hardware noise source, were taken into consideration, justifying the adopted nomenclature. With the advances made in hardware as well as the rise of the software defined noise radar concept, waveform design emerges as an important research area related to such systems. The possibility of generating signals with varied stochastic properties increased the potential in achieving systems with enhanced performances. The characterization of random phase and frequency modulated waveforms (more suitable for several applications) has then gained considerable notoriety within the radar community as well. Several optimization algorithms have been proposed in order to conveniently shape both the autocorrelation function of the random samples that comprise the transmit signal, as well as their power spectrum density. Nevertheless, little attention has been driven to properly characterize the stochastic properties of those signals through closed form expressions, jeopardizing the effectiveness of the aforementioned algorithms as well as their reproducibility. Within this context, this paper investigates the performance of several random phase and frequency modulated waveforms, varying the stochastic properties of their modulating signals.

Real-time identification of noise type contaminated in surface electromyogram signals using efficient statistical features.

Different types of noise contaminating the surface electromyogram (EMG) signal may degrade the recognition performance. For noise removal, the type of noise has to first be identified. In this paper, we propose a real-time efficient system for identifying a clean EMG signal and noisy EMG signals contaminated with any one of the following three types of noise: electrocardiogram interference, spike noise, and power line interference. Two statistical descriptors, kurtosis and skewness, are used as input features for the cascading quadratic discriminant analysis classifier. An efficient simplification of kurtosis and skewness calculations that can reduce computation time and memory storage is proposed. The experimental results from the real-time system based on an ATmega 2560 microcontroller demonstrate that the kurtosis and skewness values show root mean square errors between the traditional and proposed efficient techniques of 0.08 and 0.09, respectively. The identification accuracy with five-fold cross-validation resulting from the quadratic discriminant analysis classifier is 96.00%.

The Development and Implementation of Innovative Blind Source Separation Techniques for Real-Time Extraction and Analysis of Fetal and Maternal Electrocardiogram Signals.

This article presents an innovative approach to analyzing and extracting electrocardiogram (ECG) signals from the abdomen and thorax of pregnant women, with the primary goal of isolating fetal ECG (fECG) and maternal ECG (mECG) signals. To resolve the difficulties related to the low amplitude of the fECG, various noise sources during signal acquisition, and the overlapping of R waves, we developed a new method for extracting ECG signals using blind source separation techniques. This method is based on independent component analysis algorithms to detect and accurately extract fECG and mECG signals from abdomen and thorax data. To validate our approach, we carried out experiments using a real and reliable database for the evaluation of fECG extraction algorithms. Moreover, to demonstrate real-time applicability, we implemented our method in an embedded card linked to electronic modules that measure blood oxygen saturation (SpO2) and body temperature, as well as the transmission of data to a web server. This enables us to present all information related to the fetus and its mother in a mobile application to assist doctors in diagnosing the fetus's condition. Our results demonstrate the effectiveness of our approach in isolating fECG and mECG signals under difficult conditions and also calculating different heart rates (fBPM and mBPM), which offers promising prospects for improving fetal monitoring and maternal healthcare during pregnancy.

Improved power flow management with proposed fuzzy integrated hybrid optimized fractional order cascaded proportional derivative filter (1+ proportional integral) controller in hybrid microgrid systems.

The primary goal of power management in a hybrid microgrid is to maintain the active power balance among renewable energy sources, energy storage system, and loads. In a HMG system, an ESS is required to reduce power fluctuations caused by the intermittent behavior of RESs. A bi-directional dc-dc converter has been used to connect a battery to the dc bus. This paper proposes an improved power management scheme where a novel controller named as hybrid fuzzy integrated fractional order cascaded proportional derivative filter (1+proportional integral) is designed and implemented in a voltage-controlled loop of the BDC. This controller is introduced in this work to achieve optimum power flow management improves voltage stabilization, and enhances the system dynamic response. A hybrid modified sine cosine algorithm-pattern search algorithm is implemented here to tune the parameters of the proposed controller. The extensive time-domain analysis and robustness studies of the proposed controller have been performed and compared with the other conventional control methods Variable solar irradiance, solar temperature, wind speed, and load are used to test the effectiveness of proposed power management strategy. The proposed power management scheme with several case studies has been performed in MATLAB/Simulink and also validated through a Real-time digital platform (Opal-RT).

Robust Three-Microphone Speech Source Localization Using Randomized Singular Value Decomposition.

Direction-of-arrival (DOA) estimation is a fundamental technique in array signal processing due to its wide applications in beamforming, speech enhancement and many other assistive speech processing technologies. In this paper, we devise a novel DOA technique based on randomized singular value decomposition (RSVD) to improve the performance of non-uniform non-linear microphone arrays (NUNLA). The accurate and efficient singular value decomposition of large data matrices is computationally challenging, and randomization provides an effective tool for performing matrix approximation, therefore, the developed DOA estimation utilizes a modified dictionary-based RSVD method for localizing single speech sources under low signal-to-noise ratios (SNR). Unlike previous methods developed for uniform linear microphone arrays, the proposed approach with L-shaped three microphone setup has no 'left-right' ambiguity. We present the performance of our proposed method in comparison to other techniques. The demonstrated experiments shows at-least 20% performance improvement using simulated data and 25% performance improvement using real data when compared with similar DoA estimation techniques for NUNLA. The proposed method exploits frame-based online time delay of arrival (TDOA) measurements which facilitates the proposed algorithm to run on real-time devices. We also show an efficient real-time implementation of the proposed method on a Pixel 3 Android smartphone using its built-in three microphones for hearing aid applications.

Spectral Flux-Based Convolutional Neural Network Architecture for Speech Source Localization and Its Real-Time Implementation.

In this article, we present a real-time convolutional neural network (CNN)-based Speech source localization (SSL) algorithm that is robust to realistic background acoustic conditions (noise and reverberation). We have implemented and tested the proposed method on a prototype (Raspberry Pi) for real-time operation. We have used the combination of the imaginary-real coefficients of the short-time Fourier transform (STFT) and Spectral Flux (SF) with delay-and-sum (DAS) beamforming as the input feature. We have trained the CNN model using noisy speech recordings collected from different rooms and inference on an unseen room. We provide quantitative comparison with five other previously published SSL algorithms under several realistic noisy conditions, and show significant improvements by incorporating the Spectral Flux (SF) with beamforming as an additional feature to learn temporal variation in speech spectra. We perform real-time inferencing of our CNN model on the prototyped platform with low latency (21 milliseconds (ms) per frame with a frame length of 30 ms) and high accuracy (i.e. 89.68% under Babble noise condition at 5dB SNR). Lastly, we provide a detailed explanation of real-time implementation and on-device performance (including peak power consumption metrics) that sets this work apart from previously published works. This work has several notable implications for improving the audio-processing algorithms for portable battery-operated Smart loudspeakers and hearing improvement (HI) devices.

A Real-Time Convolutional Neural Network Based Speech Enhancement for Hearing Impaired Listeners Using Smartphone.

This paper presents a Speech Enhancement (SE) technique based on multi-objective learning convolutional neural network to improve the overall quality of speech perceived by Hearing Aid (HA) users. The proposed method is implemented on a smartphone as an application that performs real-time SE. This arrangement works as an assistive tool to HA. A multi-objective learning architecture including primary and secondary features uses a mapping-based convolutional neural network (CNN) model to remove noise from a noisy speech spectrum. The algorithm is computationally fast and has a low processing delay which enables it to operate seamlessly on a smartphone. The steps and the detailed analysis of real-time implementation are discussed. The proposed method is compared with existing conventional and neural network-based SE techniques through speech quality and intelligibility metrics in various noisy speech conditions. The key contribution of this paper includes the realization of CNN SE model on a smartphone processor that works seamlessly with HA. The experimental results demonstrate significant improvements over the state-of-the-art techniques and reflect the usability of the developed SE application in noisy environments.

Near Real-Time Implementation of An Adaptive Seismocardiography - ECG Multimodal Framework for Cardiac Gating.

Objective: Accurate gating for data acquisition of computed tomography (CT) is crucial to obtaining high quality images for diagnosing cardiovascular diseases. To illustrate the feasibility of an optimized cardiac gating strategy, we present a near real-time implementation based on fusing seismocardiography (SCG) and ECG. Methods: The implementation was achieved via integrating commercial hardware and software platforms. Testing was performed on five healthy subjects (age: 24-27; m/f: 4/1) and three cardiac patients (age: 41-71; m/f: 2/1), and compared with baseline quiescence derived from echocardiography. Results: The average latency introduced by computerized processing was 5.1 ms, well within a 100 ms tolerance bounded by data accumulation time for quiescence prediction. The average prediction error associated with conventional ECG-only versus SCG-ECG-based method over all subjects were 59.58 ms and 27.24 ms, respectively. Discussion: The results demonstrate that the multimodal framework can achieve improved quiescence prediction accuracy over the ECG-only-based method in near real-time.

A Convolutional Neural Network Smartphone App for Real-Time Voice Activity Detection.

This paper presents a smartphone app that performs real-time voice activity detection based on convolutional neural network. Real-time implementation issues are discussed showing how the slow inference time associated with convolutional neural networks is addressed. The developed smartphone app is meant to act as a switch for noise reduction in the signal processing pipelines of hearing devices, enabling noise estimation or classification to be conducted in noise-only parts of noisy speech signals. The developed smartphone app is compared with a previously developed voice activity detection app as well as with two highly cited voice activity detection algorithms. The experimental results indicate that the developed app using convolutional neural network outperforms the previously developed smartphone app.