Optimizing Shoulder to Shoulder: A Coordinated Sub-Band Fusion Model for Real-Time Full-Band Speech Enhancement

Due to the high computational complexity to model more frequency bands, it is still intractable to conduct real-time full-band speech enhancement based on deep neural networks. Recent studies typically utilize the compressed perceptually motivated features with relatively low frequency resolution to filter the full-band spectrum by one-stage networks, leading to limited speech quality improvements. In this paper, we propose a coordinated sub-band fusion network for full-band speech enhancement, which aims to recover the low- (0-8 kHz), middle- (8-16 kHz), and high-band (16-24 kHz) in a step-wise manner. Specifically, a dual-stream network is first pretrained to recover the low-band complex spectrum, and another two sub-networks are designed as the middle- and high-band noise suppressors in the magnitude-only domain. To fully capitalize on the information intercommunication, we employ a sub-band interaction module to provide external knowledge guidance across different frequency bands. Extensive experiments show that the proposed method yields consistent performance advantages over state-of-the-art full-band baselines.

Review of data types and model dimensionality for cardiac DTI SMS-related artefact removal

As diffusion tensor imaging (DTI) gains popularity in cardiac imaging due to its unique ability to non-invasively assess the cardiac microstructure, deep learning-based Artificial Intelligence is becoming a crucial tool in mitigating some of its drawbacks, such as the long scan times. As it often happens in fast-paced research environments, a lot of emphasis has been put on showing the capability of deep learning while often not enough time has been spent investigating what input and architectural properties would benefit cardiac DTI acceleration the most. In this work, we compare the effect of several input types (magnitude images vs complex images), multiple dimensionalities (2D vs 3D operations), and multiple input types (single slice vs multi-slice) on the performance of a model trained to remove artefacts caused by a simultaneous multi-slice (SMS) acquisition. Despite our initial intuition, our experiments show that, for a fixed number of parameters, simpler 2D real-valued models outperform their more advanced 3D or complex counterparts. The best performance is although obtained by a real-valued model trained using both the magnitude and phase components of the acquired data. We believe this behaviour to be due to real-valued models making better use of the lower number of parameters, and to 3D models not being able to exploit the spatial information because of the low SMS acceleration factor used in our experiments.

Vehicle Trajectory Tracking Through Magnetic Sensors

Traffic surveillance is an important issue in Intelligent Transportation Systems(ITS). In this paper, we propose a novel surveillance system to detect and track vehicles using ubiquitously deployed magnetic sensors. That is, multiple magnetic sensors, mounted roadside and along lane boundary lines, are used to track various vehicles. Real-time vehicle detection data are reported from magnetic sensors, collected into data center via base stations, and processed to depict vehicle trajectories including vehicle position, timestamp, speed and type. We first define a vehicle trajectory tracking problem. We then propose a graph-based data association algorithm to track each detected vehicle, and design a related online algorithm framework respectively. We finally validate the performance via both experimental simulation and real-world road test. The experimental results demonstrate that the proposed solution provides a cost-effective solution to capture the driving status of vehicles and on that basis form various traffic safety and efficiency applications.

A Multi-Task Learning Model for Super Resolution of Wireless Channel Characteristics

Channel modeling has always been the core part in communication system design and development, especially in 5G and 6G era. Traditional approaches like stochastic channel modeling and ray-tracing (RT) based channel modeling depend heavily on measurement data or simulation, which are usually expensive and time consuming. In this paper, we propose a novel super resolution (SR) model for generating channel characteristics data. The model is based on multi-task learning (MTL) convolutional neural networks (CNN) with residual connection. Experiments demonstrate that the proposed SR model could achieve excellent performances in mean absolute error and standard deviation of error. Advantages of the proposed model are demonstrated in comparisons with other state-of-the-art deep learning models. Ablation study also proved the necessity of multi-task learning and techniques in model design. The contribution in this paper could be helpful in channel modeling, network optimization, positioning and other wireless channel characteristics related work by largely reducing workload of simulation or measurement.

A Combined Model for Noise Reduction of Lung Sound Signals Based on Empirical Mode Decomposition and Artificial Neural Network

Computer analysis of Lung Sound (LS) signals has been proposed in recent years as a tool to analyze the lungs' status but there have always been main challenges, including the contamination of LS with environmental noises, which come from different sources of unlike intensities. One of the common methods in noise reduction of LS signals is based on thresholding on Discrete Wavelet Transform (DWT) coefficients or Empirical Mode Decomposition (EMD) of the signal, however, in these methods, it is necessary to calculate the SNR value to determine the appropriate threshold for noise removal. To solve this problem, a combined model based on EMD and Artificial Neural Network (ANN) trained with different SNRs (0, 5, 10, 15, and 20dB) is proposed in this research. The model can denoise white and pink noises in the range of -2 to 20dB without thresholding or even estimating SNR, and at the same time, keep the main content of the LS signal well. The proposed method is also compared with the EMD-custom method, and the results obtained from the SNR, and fit criteria indicate the absolute superiority of the proposed method. For example, at SNR = 0dB, the combined method can improve the SNR by 9.41 and 8.23dB for white and pink noises, respectively, while the corresponding values are respectively 5.89 and 4.31dB for the EMD-Custom method.

Benchmarking real-time algorithms for in-phase auditory stimulation of low amplitude slow waves with wearable EEG devices during sleep

Auditory stimulation of EEG slow waves (SW) during non-rapid eye movement (NREM) sleep has shown to improve cognitive function when it is delivered at the up-phase of SW. SW enhancement is particularly desirable in subjects with low-amplitude SW such as older adults or patients suffering from neurodegeneration such as Parkinson disease (PD). However, existing algorithms to estimate the up-phase suffer from a poor phase accuracy at low EEG amplitudes and when SW frequencies are not constant. We introduce two novel algorithms for real-time EEG phase estimation on autonomous wearable devices. The algorithms were based on a phase-locked loop (PLL) and, for the first time, a phase vocoder (PV). We compared these phase tracking algorithms with a simple amplitude threshold approach. The optimized algorithms were benchmarked for phase accuracy, the capacity to estimate phase at SW amplitudes between 20 and 60 microV, and SW frequencies above 1 Hz on 324 recordings from healthy older adults and PD patients. Furthermore, the algorithms were implemented on a wearable device and the computational efficiency and the performance was evaluated on simulated sleep EEG, as well as prospectively during a recording with a PD patient. All three algorithms delivered more than 70% of the stimulation triggers during the SW up-phase. The PV showed the highest capacity on targeting low-amplitude SW and SW with frequencies above 1 Hz. The testing on real-time hardware revealed that both PV and PLL have marginal impact on microcontroller load, while the efficiency of the PV was 4% lower than the PLL. Active auditory stimulation did not influence the phase tracking. This work demonstrated that phase-accurate auditory stimulation can be delivered during home-based sleep interventions with a wearable device also in populations with low-amplitude SW.

uChecker: Masked Pretrained Language Models as Unsupervised Chinese Spelling Checkers

The task of Chinese Spelling Check (CSC) is aiming to detect and correct spelling errors that can be found in the text. While manually annotating a high-quality dataset is expensive and time-consuming, thus the scale of the training dataset is usually very small (e.g., SIGHAN15 only contains 2339 samples for training), therefore supervised-learning based models usually suffer the data sparsity limitation and over-fitting issue, especially in the era of big language models. In this paper, we are dedicated to investigating the \textbf{unsupervised} paradigm to address the CSC problem and we propose a framework named \textbf{uChecker} to conduct unsupervised spelling error detection and correction. Masked pretrained language models such as BERT are introduced as the backbone model considering their powerful language diagnosis capability. Benefiting from the various and flexible MASKing operations, we propose a Confusionset-guided masking strategy to fine-train the masked language model to further improve the performance of unsupervised detection and correction. Experimental results on standard datasets demonstrate the effectiveness of our proposed model uChecker in terms of character-level and sentence-level Accuracy, Precision, Recall, and F1-Measure on tasks of spelling error detection and correction respectively.

Toward Intention Discovery for Early Malice Detection in Bitcoin

Bitcoin has been subject to illicit activities more often than probably any other financial assets, due to the pseudo-anonymous nature of its transacting entities. An ideal detection model is expected to achieve all the three properties of (I) early detection, (II) good interpretability, and (III) versatility for various illicit activities. However, existing solutions cannot meet all these requirements, as most of them heavily rely on deep learning without satisfying interpretability and are only available for retrospective analysis of a specific illicit type. First, we present asset transfer paths, which aim to describe addresses' early characteristics. Next, with a decision tree based strategy for feature selection and segmentation, we split the entire observation period into different segments and encode each as a segment vector. After clustering all these segment vectors, we get the global status vectors, essentially the basic unit to describe the whole intention. Finally, a hierarchical self-attention predictor predicts the label for the given address in real time. A survival module tells the predictor when to stop and proposes the status sequence, namely intention. % With the type-dependent selection strategy and global status vectors, our model can be applied to detect various illicit activities with strong interpretability. The well-designed predictor and particular loss functions strengthen the model's prediction speed and interpretability one step further. Extensive experiments on three real-world datasets show that our proposed algorithm outperforms state-of-the-art methods. Besides, additional case studies justify our model can not only explain existing illicit patterns but can also find new suspicious characters.

Learning sparse auto-encoders for green AI image coding

Recently, convolutional auto-encoders (CAE) were introduced for image coding. They achieved performance improvements over the state-of-the-art JPEG2000 method. However, these performances were obtained using massive CAEs featuring a large number of parameters and whose training required heavy computational power.\\ In this paper, we address the problem of lossy image compression using a CAE with a small memory footprint and low computational power usage. In order to overcome the computational cost issue, the majority of the literature uses Lagrangian proximal regularization methods, which are time consuming themselves.\\ In this work, we propose a constrained approach and a new structured sparse learning method. We design an algorithm and test it on three constraints: the classical $\ell_1$ constraint, the $\ell_{1,\infty}$ and the new $\ell_{1,1}$ constraint. Experimental results show that the $\ell_{1,1}$ constraint provides the best structured sparsity, resulting in a high reduction of memory and computational cost, with similar rate-distortion performance as with dense networks.

Multi-level Adaptation for Automatic Landing with Engine Failure under Turbulent Weather

This paper addresses efficient feasibility evaluation of possible emergency landing sites, online navigation, and path following for automatic landing under engine-out failure subject to turbulent weather. The proposed Multi-level Adaptive Safety Control framework enables unmanned aerial vehicles (UAVs) under large uncertainties to perform safety maneuvers traditionally reserved for human pilots with sufficient experience. In this framework, a simplified flight model is first used for time-efficient feasibility evaluation of a set of landing sites and trajectory generation. Then, an online path following controller is employed to track the selected landing trajectory. We used a high-fidelity simulation environment for a fixed-wing aircraft to test and validate the proposed approach under various weather uncertainties. For the case of emergency landing due to engine failure under severe weather conditions, the simulation results show that the proposed automatic landing framework is robust to uncertainties and adaptable at different landing stages while being computationally inexpensive for planning and tracking tasks.