Abstract In order to achieve perfection in security field, the biometric recognition has been used as a tool, because it is based on physiological and behavioral characteristics of a person which unlike the traditional prevailing methods of human identification, cannot be forgotten, stolen, eliminated or changed with time. In this paper, a theoretical and general presentation of an online signature based... In order to achieve perfection in security field, the biometric recognition has been used as a tool, because it is based on physiological and behavioral characteristics of a person which unlike the traditional prevailing methods of human identification, cannot be forgotten, stolen, eliminated or changed with time. In this paper, a theoretical and general presentation of an online signature based biometric system using a reduced set of features. The authenticity of a signer is determined by comparing an input signature to a stored reference set (template) based on a limited number of simple extracted features like the total length of the lines, pressure, the distance between the first and the last point, the average of horizontal displacement...etc. The proposed system was compared to QU-PRIP system which used more extracted features, this system which is the winner system in ICDAR2011 Signature Verification Competition. After several experiments, the obtained results show that even though the overall system performance are not very good enough but it is approximately similar to the one of the QU-PRIP system that uses more extracted features, the fact that affect directly on the total space of representation and the total time necessary for authentication. Read More Read Less

Abstract IOT devices suffer from the drawbacks of limited computational resources at local level. IOT devices are required to gather and process large quantity of data with limited resources. One solution is to upload data from IOT devices to cloud based systems, which can process data and send results back to devices. This technique suffers from delay in processing and results in... IOT devices suffer from the drawbacks of limited computational resources at local level. IOT devices are required to gather and process large quantity of data with limited resources. One solution is to upload data from IOT devices to cloud based systems, which can process data and send results back to devices. This technique suffers from delay in processing and results in delayed response from the cloud in real time scenarios. Another technique to solve the problem is to introduce a middleware processing node also called edge node. The processing of data from edge nodes is known as edge computing or fog computing. In this paper, the authors discuss the security aspects of fog computing. After careful study of related literature, the authors identified several challenges related to security of data processing at edge nodes. The authors identified challenges such as data security and programming related security. The authors present several solutions to the challenges. Read More Read Less