A Novel Speech Intelligibility Enhancement Model based on Canonical Correlation and Deep Learning.

Current deep learning (DL) based approaches to speech intelligibility enhancement in noisy environments are often trained to minimise the feature distance between noise-free speech and enhanced speech signals. Despite improving the speech quality, such approaches do not deliver required levels of speech intelligibility in everyday noisy environments. Intelligibility-oriented (I-O) loss functions have recently been developed to train DL approaches for robust speech enhancement. Here, we formulate, for the first time, a novel canonical correlation based I-O loss function to more effectively train DL algorithms. Specifically, we present a canonical-correlation based short-time objective intelligibility (CC-STOI) cost function to train a fully convolutional neural network (FCN) model. We carry out comparative simulation experiments to show that our CC-STOI based speech enhancement framework outperforms state-of-the-art DL models trained with conventional distance-based and STOI-based loss functions, using objective and subjective evaluation measures for case of both unseen speakers and noises. Ongoing future work is evaluating the proposed approach for design of robust hearing-assistive technology.

Ensemble learning-based IDS for sensors telemetry data in IoT networks.

The Internet of Things (IoT) is a paradigm that connects a range of physical smart devices to provide ubiquitous services to individuals and automate their daily tasks. IoT devices collect data from the surrounding environment and communicate with other devices using different communication protocols such as CoAP, MQTT, DDS, etc. Study shows that these protocols are vulnerable to attack and prove a significant threat to IoT telemetry data. Within a network, IoT devices are interdependent, and the behaviour of one device depends on the data coming from another device. An intruder exploits vulnerabilities of a device's interdependent feature and can alter the telemetry data to indirectly control the behaviour of other dependent devices in a network. Therefore, securing IoT devices have become a significant concern in IoT networks. The research community often proposes intrusion Detection Systems (IDS) using different techniques. One of the most adopted techniques is machine learning (ML) based intrusion detection. This study suggests a stacking-based ensemble model makes IoT devices more intelligent for detecting unusual behaviour in IoT networks. The TON-IoT (2020) dataset is used to assess the effectiveness of the proposed model. The proposed model achieves significant improvements in accuracy and other evaluation measures in binary and multi-class classification scenarios for most of the sensors compared to traditional ML algorithms and other ensemble techniques.

Scheduling Sensor Duty Cycling Based on Event Detection Using Bi-Directional Long Short-Term Memory and Reinforcement Learning.

A smart home provides a facilitated environment for the detection of human activity with appropriate Deep Learning algorithms to manipulate data collected from numerous sensors attached to various smart things in a smart home environment. Human activities comprise expected and unexpected behavior events; therefore, detecting these events consisting of mutual dependent activities poses a key challenge in the activities detection paradigm. Besides, the battery-powered sensor ubiquitously and extensively monitors activities, disputes, and sensor energy depletion. Therefore, to address these challenges, we propose an Energy and Event Aware-Sensor Duty Cycling scheme. The proposed model predicts the future expected event using the Bi-Directional Long-Short Term Memory model and allocates Predictive Sensors to the predicted event. To detect the unexpected events, the proposed model localizes a Monitor Sensor within a cluster of Hibernate Sensors using the Jaccard Similarity Index. Finally, we optimize the performance of our proposed scheme by employing the Q-Learning algorithm to track the missed or undetected events. The simulation is executed against the conventional Machine Learning algorithms for the sensor duty cycle, scheduling to reduce the sensor energy consumption and improve the activity detection accuracy. The experimental evaluation of our proposed scheme shows significant improvement in activity detection accuracy from 94.12% to 96.12%. Besides, the effective rotation of the Monitor Sensor significantly improves the energy consumption of each sensor with the entire network lifetime.

A Multi-Objective Approach for Optimal Energy Management in Smart Home Using the Reinforcement Learning.

Maintaining a fair use of energy consumption in smart homes with many household appliances requires sophisticated algorithms working together in real time. Similarly, choosing a proper schedule for appliances operation can be used to reduce inappropriate energy consumption. However, scheduling appliances always depend on the behavior of a smart home user. Thus, modeling human interaction with appliances is needed to design an efficient scheduling algorithm with real-time support. In this regard, we propose a scheduling algorithm based on human appliances interaction in smart homes using reinforcement learning (RL). The proposed scheduling algorithm divides the entire day into various states. In each state, the agents attached to household appliances perform various actions to obtain the highest reward. To adjust the discomfort which arises due to performing inappropriate action, the household appliances are categorized into three groups i.e., (1) adoptable, (2) un-adoptable, (3) manageable. Finally, the proposed system is tested for the energy consumption and discomfort level of the home user against our previous scheduling algorithm based on least slack time phenomenon. The proposed scheme outperforms the Least Slack Time (LST) based scheduling in context of energy consumption and discomfort level of the home user.