Given a large enough time series signal from an ECG signal, it is possible to identify and classify heartbeats not only into normal and abnormal classes but into multiple classes including but not limited to Normal beat, Paced beat, Atrial Premature beat and Ventricular flutter as originally suggested by benchmark electrocardiogram (ECG) datasets like the MIT-BIH Arrhythmia Dataset. There are multiple approaches that target ECG classifications using Machine and Deep Learning like One Class SVM, ELM, Anogan etc. These approaches require either very high computational resources, fail to classify classes apart from normal/abnormal classes or fail to classify all classes with an equivalent or near-equivalent accuracy. With these limitations in mind, this paper proposes a deep learning approach using Convolutional Neural Networks (CNNs) to classify multiple classes of heartbeats in an efficient, effective, and generalized manner. By using the MIT-BIH Arrhythmia dataset to filter and segment individual correctly structured heartbeats, we have designed a network which can be trained on different classes of heartbeats and present robust, accurate and efficient results. The class imbalance prevalent in the MIT-BIH dataset has been dealt with using Synthetic Minority Over-sampling Technique (SMOTE). The robustness of the model is increased by adding techniques of loss minimization such as dropout and early stopping. The approach gives an accuracy of approximately 96% and an extremely short time span for class prediction(classification), i.e., less than 1 second. The results are also illustrated over multiple (10) classes to exemplify the generality of the model. We have illustrated these results over multiple (10) classes to exemplify generality of the model.

No abstract

With over 16 million horses worldwide and nearly 60,000 sport horses registered to the International Federation for Equestrian Sports database, tracking the activities and performance of these equines is becoming an important aspect in horse management. To perform this activity recognition, Inertial Measurement Units (IMUs) are often used in combination with machine learning algorithms. These often require large labeled datasets to be trained. To this end, a data-efficient algorithm is proposed that requires only 3 minutes of labeled calibration data. This is achieved by combining supervised feature selection, using the tsfresh time-series feature calculation library and the Kendall rank correlation coefficient, with a distance-based clustering algorithm. The generalizability performance of the algorithm is tested by evaluating on a dataset captured with leg-mounted IMUs and on a dataset captured using a neck-mounted IMU. On both datasets, the algorithm achieved accuracies of 95%, comparable to state-of-the-art deep learning approaches, when calibrating and evaluating using the same horse. When the algorithm was calibrated on data from multiple horses and evaluated on horses that were not in the calibration dataset, a 15% drop in classification accuracy is observed. The proposed algorithm is compared with fully supervised algorithms like convolutional neutral network, support vector machine, and random forest in terms of accuracy achieved with respect to the size of the labeled data using calibration. Our approach achieved accuracies that were similar to these classical algorithms whilst only using 10-15% the amount of labeled data.

With the ongoing push towards an automated Industry 4.0, data-driven intelligent algorithms are getting more attention. Warehouse operators have traditionally required human labor to identify and register their assets. Autonomous flying drones will help alleviate this task by flying through the warehouse and detecting assets. This can be done based on vision, requiring expensive and energy consuming hardware, limiting drone flight time. In contrast, we propose a solution using radio-frequency identification (RFID) tags and machine learned algorithms to localize assets, which does not require a well-lit environment and can be processed in an energy efficient way. Our machine learning model achieves a 92-93 % accuracy, even when the drone is flying at different heights than the assets. Additionally, the model is easily implementable on off-the-shelf and low-energy consuming embedded hardware. This data-driven solution can easily be retrained for different environments and allows cheap RFID-based horizontal localization of assets in warehouses of the future.

<p>Dynamic Spectrum Sharing (DSS) is an enabler for a seamless transition from 4G Long Term Evolution (LTE) to 5G New Radio (NR) by utilizing existing LTE bands without static spectrum re- farming. In this paper, we propose a cross-band DSS scheme that utilizes the Multimedia Broadcast Multicast Service over a Single Frequency Network (MBSFN) feature of an LTE network and the Multicast Broadcast Service (MBS) feature of an NR network. The proposed DSS scheme utilizes LTE and NR resource controllers to assign muted MBSFN subframes on the LTE band and muted MBS subframes on the NR band based on traffic needs. In contrast to the state-of-the-art, the proposed DSS scheme does not require a coordination signaling channel between the LTE and NR networks. Instead, a machine learning-based Technology Recognition and Traffic Characterization (TRTC) system is used to identify and characterize traffic patterns. The LTE and NR resource controllers use the TRTC to sense the muted subframes and offload traffic accordingly.</p>