In a cognitive radio network, non-reconstruction-based wideband compressive spectrum sensing poses challenges under the non-Gaussian noise environment. The maximum correntropy criterion (MCC) is robust to impulsive noise whereas, the Parzen window Renyi entropy is a good choice for spectrum sensing in the presence of Gaussian noise at a low signal-to-noise ratio (SNR). However, the detection performance of these algorithms depends on the accuracy of measured noise variance, which is sensitive to impulse noise. In this letter, we improve the sensing performance of both the aforementioned algorithms in a non-Gaussian noise environment by modifying the kernel and threshold using robust statistics. The robust technique minimizes the influence of impulsive noise in the received signal. Finally, we carry out the simulation results to illustrate the superior performance of robust sensing algorithms under both Bernoulli’s distribution and symmetric $\alpha $ stable ( $\text{S}\alpha \text{S}$ ) distribution channel noise. The performance is compared with the non-robust counterpart for sensing multi-carrier Universal-Filtered Multi-Carrier (UFMC) signal.

Cardiovascular diseases (CVD) are the dominant cause of deaths in the world, of which 90% are curable. The electrocardiogram (ECG) measures the electrical stimulus of the heart noninvasively. Convolutional neural networks (CNN) act as one of the powerful machine learning techniques to classify ECG arrhythmia classification and other CVDs. Nonetheless, they have some functional flaws like ignorance of spatial hierarchies between the features and are unable to acquire a rotational invariance. To overcome these problems of CNN, a novel neural network named capsule network (CapsNet) is proposed as an efficient algorithm to provide error‐free implementation of deep learning over the databases. The main focus of this work is to apply and implement CapsNet for ECG signal classification from the MIT‐BIH database and compare its efficiency with the pretrained CNN networks.

Aim: Today, Internet of Things (IoT) applications are extended from smart homes to ehealth, cyber security, data analytics, logistics and management of assets. There are many upcoming IOT solutions and platforms like ThingWorx, Xively, and Yaler. However, the existing eco-systems are not vibrant because of the high entry-level barrier and low potential for any stakeholder. Especially, the academic community requires a comprehensible way to create IoT services, develop their skillsets and build applications around them. In this regard, this work presents an affordable and scalable IoT eco-system with an easily programmable hardware platform, a private web server on the cloud and a user-friendly mobile application. Background: Home automation controls the devices and appliances in the home environment to increase the comfort and convenience. To design a typical immune home automation system, we need to incorporate different sensors, wireless networking and a central node that can collect data and act as a gateway for the internet connection. Objective: Delivering an IoT solution involves the use of multiple technologies that cut across traditional engineering stream boundaries - sensors, microcontrollers, wireless networking, network protocols, web programming, and mobile app development. The open challenge is to put the entire ecosystem together either through new development or through configuartion of existing components. Methods: In this work, we incorporated a suitable hardware platform that can be easily programmed. The platform is open so that new sensors and actuators can be added as per requirement. The existing web services are used to post and retrieve data from the cloud. In addition, mobile apps can be developed to make data available to the user. Results: A custom-built GISMO based IoT cloud system is developed with sensors and nodes to form an infrastructure. The framework will assure standard design that establishes a functional link between hardware, software and web applications. A private web service using HTTP server and MQTT broker is designed with access from anywhere with a public IP. The web services are coded in PHP and since it is an in-house development, the addition of new services and maintenance of existing services are easy. Conclusion: The IoT eco-system developed provides a platform for a quick out-of-the-box implementation of an IoT project. Other sensors such as a PIR sensor, an RTC module, an ultrasonic sensor, a soil moisture sensor have been interfaced to the GISMO module using the IOs brought out onto the expansion header. The GISMO thus can serve as a generic hardware platform for sensor/actuator interface in the IoT scheme. The eco-system can be replicated in other institutes and can serve as a base to implement applications like sensor-to-cloud interface, cloud-to-actuator interface, cloud-based alerts and notifications, HTTP and MQTT protocols usage.

Despite the growing prevalence of wireless body area networks (WBAN), there are still unresolved issues about the technology's efficacy and efficiency in transmitting electroencephalography (EEG) data. Utilising a voting-based system, an adaptive fuzzy predictor, and a tri-stage entropy encoder are the components that comprise this technique. There are three steps that make up the tri-stage entropy encoder. A CMOS technology with a resolution of 0.18 micrometres and 8405 gates was used in the manufacturing process of the design using this technology. The clock speed was 100 megahertz, and the power consumption that was simulated to occur during operation was 2.58 milliwatts. When applied to the CHB-MIT Scalp EEG Database, the method that was recommended was successful in achieving an average compression rate of 2.35 for the 23 channels that were examined. This research resulted in a 14.6% increase in compression rates while simultaneously resulting in a 77.3 percent decrease in hardware expenses. This was accomplished by comparing the results of this study to the presently existing hardware-oriented lossless EEG compression methodologies.

Providing safety, security, operational convenience and energy savings are some of the vital goals of a home automation system. A single unified point of interaction for the user is also an important consideration. Each feature requires a different sensor, powering scheme and mounting arrangement. To address each requirement, custom node development is to be prepared with all nodes being part of a personal area network for data transfer. A central node with Internet connectivity can push data to the cloud and the user can be provided access to data on a mobile app. This paper presents the practical implementation of such a distributed home automation scheme with custom-designed microcontroller-based ZigBee enabled hardware nodes for liquid petroleum gas leak detection, window open detection, lights switching control based on occupancy and intrusion detection. Messages are triggered by events and are not periodic to avoid traffic congestion. A central node built around a Raspberry Pi with a ZigBee dongle will serve as a data aggregator which sends messages to our own MQTT (Message Queuing Telemetry Transport) broker. An android mobile app will work as an MQTT client, helps the user to connect the MQTT broker, subscribe to the topics published and makes alerts to the user as soon as an event occurs. The mobile app will also enable the user to switch on lights and other appliances through command messages sent to the MQTT broker and subscribed to Raspberry Pi.