Wireless reprogramming is a critical issue in the Internet of Things (IoT). Current approaches designed for wireless sensor networks (WSN) are inadequate for IoT scenarios due to serious security vulnerabilities. To address this problem, we present Software Defined Function (SDF), a secure and wireless reprogramming architecture for IoT. The key is to implement a secure communication interface between the control layer and infrastructure layer. Four security protocols ensuring authentication and confidentiality are designed. In addition, the design takes into consideration the IoT devices' constrained capabilities in computation. We test the performance of SDF through a set of experiments and present a use case on a photovoltaic energy system. The evaluation results show that the proposed protocols can be implemented in real-world IoT applications.
Personal health records (PHRs) are valuable assets to individuals because they enable them to integrate and manage their medical data. Giving patients control over their medical data offers an advantageous realignment of the doctor-patient dynamic. However, today's PHR management systems fall short of giving reliable, traceable, trustful, and secure patients control over their medical data. In this paper, we propose blockchain-based smart contracts to give patients control over their data in a manner that is decentralized, immutable, transparent, traceable, trustful, and secure. The proposed system employs decentralized storage of interplanetary file systems (IPFS) and trusted reputation-based re-encryption oracles to securely fetch, store, and share patients' medical data.
This paper presents an improved version of RKEDA, one of the evolutionary algorithms for a spares part allocation problem in the telecom industry. The aim is to improve on previous approach and have the right spare parts at the right time at the right place. This will make a big difference to the quality of the service offered by a telecom organization by maximizing its utilization and revenues while minimizing the total cost. The improvement presented in this paper relates to ordering the way the initial sites are represented in the permutation in RKEDA algorithm. We describe proposed improvement, perform detail experimental analysis, and compare the performance of the new RKEDA to that of the original RKEDA and GA.
In this paper, we propose a deep-learning approach for human gender classification on RGB-D images. Unlike most of the existing methods, which use hand-crafted features from the human face, we exploit local information from the head and global information from the whole body to classify people's gender. A head detector is fine-tuned on YOLO to detect the head regions on the images automatically. Two gender classifiers are trained using head images and whole-body images separately. The final prediction is made by fusing the two classifiers' results. The presented method outperforms the state-of-art with an improvement in the accuracy of 2.6%, 7.6%, and 8.4% on three different test data of a challenging gender dataset which includes human standing, walking, and interacting scenarios.
A data-driven angle of arrival (AoA) estimation framework is proposed in this paper. A convolution neural network (CNN), which is a part of Deep Learning (DL), is employed to learn a mapping between the eigenvectors of the spatial covariance matrix of the received signals and the angles of arrival. This paper discusses the CNN architecture and provides a detailed description of the hyper-parameters. Simulation results show that the proposed approach outperforms traditional methods in a multipath environment with less execution time.
The growing adoption of microservice architecture in academia and industry has led to an increase in the amount of literature addressing microservices challenges, approaches, and practices in the last few years. Most of the published works did not consider the efficient allocation of resources of the proposed solutions. In addition, several aspects are still vague and scattered in the literature with regards to Microservices Architecture (MSA) and efficiency models. In this paper, we survey recent microservice frameworks for efficient resource utilization, and classify them based on their efficiency type. We further highlight current MSA trends and identify resource utilization and sharing amongst multiple microservices as important areas for future research.
Visual object tracking is one of the hottest and challenging research topics in computer vision. Due to its great potential for a wide range of real-world applications, many researchers endeavor to solve the problems in visual object tracking including illumination change, background clutter, crowd scenes, low resolution, and occlusion. This paper aims to investigate the state-of-the-art trackers to have a deep insight into the trackers for future research. First, three current research trends of visual object tracking are presented. Second, the fundamental concepts, intuitions, development, and contributions of the representative trackers in each trend are summarized.
Aligned with the UAE Space Strategy 2117, which aims to establish the first inhabitable human on the Martian Surface by 2117, with the enthuse toward space tourism, we propose a novel framework to assimilate the process of requirement specification for space Manned Mission. Deep Space manned missions are unique and characterized with a set of specific requirements that should be elicited from different sources and stakeholders to ensure the missions' success. In addition, these missions are highly dependent on the software components, which is used to control the spacecraft and interact with the astronauts. Our contribution consists of: (i) surveying current trends in space system requirements engineering from requirements elicitation to requirements specification; and (ii) introducing a new set of requirements for CDHS in space missions that are related to astronauts, particularly emotional requirements for deep space manned missions, which to the best of our knowledge have not been considered.
Cardiovascular disease (CVD) is the leading cause of premature death and disability worldwide. Effective Cardiac Rehabilitation (CR) could significantly improve mortality and morbidity rates in CVD patients; their CR uptake and adherence, however, are very low. Hence, a personalized home-based CR program, namely Care4MyHeart (C4MH), is presented here, to address such issue. C4MH incorporates a smart gamified platform that uses Personalized Serious Game Suite (C4MHPGS) to assist the CVD patients in their CR program. C4MHPGS employs gender and age specific CVD serious games related to exercise ('Exergames'), diet ('DietaryGames'), emotions ('EmoGames'), stress/smoking management ('Breathing/Stress- SmokeFreeGames') and personalized feedback (both to the user and to the related physician) associated with behavioral change support. Through employing machine learning and modeling techniques, the C4MH-PGS can provide a dynamic platform that normalizes technology to the CVD patients' needs, helping them to better self-manage their CR and, hence, their quality-of-life.
In this paper, a new multiport DC-DC converter is proposed for DC Microgrid applications. The bidirectional buck-boost structure of the proposed topology allows enhanced flexibility to connect sources and loads with different voltage and power levels. The control strategy is developed to achieve power control for renewable sources such as PV, in addition to a certain degree of resilience for DC sources availability maintaining boosted DC link voltage. A detailed steady state analysis is conducted to derive voltage relations between all ports. This proposed configuration has several benefits for the design and operation of DC microgrids such as reducing multiple power conversions, reduced number of elements, voltage boosting capability, and higher efficiency. A MATLAB/Simulink based simulation study is conducted to demonstrate the performance of proposed multiport converter topology under different operating conditions.
