top of page
fiomilmillsadjo

Advanced Computing and Systems for Security: Volume Three - Computational Intelligence, Data Mining,



The series "Advances in Intelligent Systems andComputing" contains publications on theory, applications, and designmethods of Intelligent Systems and Intelligent Computing. Virtually alldisciplines such as engineering, natural sciences, computer and informationscience, ICT, economics, business, e-commerce, environment, healthcare, lifescience are covered. The list of topics spans all the areas of modernintelligent systems and computing such as: computational intelligence, softcomputing including neural networks, fuzzy systems, evolutionary computing andthe fusion of these paradigms, social intelligence, ambient intelligence,computational neuroscience, artificial life, virtual worlds and society,cognitive science and systems, Perception and Vision, DNA and immune basedsystems, self-organizing and adaptive systems, e-Learning and teaching,human-centered and human-centric computing, recommender systems, intelligentcontrol, robotics and mechatronics including human-machine teaming,knowledge-based paradigms, learning paradigms, machine ethics, intelligent dataanalysis, knowledge management, intelligent agents, intelligent decision makingand support, intelligent network security, trust management, interactiveentertainment, Web intelligence and multimedia.




Advanced Computing and Systems for Security: Volume Three (Advances in Intelligent Systems and 12



Since these cameras and sensors compile a huge amount of information and need to process it instantly to avoid the car in the next lane, autonomous vehicles require high-performance computing, advanced algorithms, and deep learning systems to adapt to new scenarios. This means that software is the key, not the physical car or truck itself.30 Advanced software enables cars to learn from the experiences of other vehicles on the road and adjust their guidance systems as weather, driving, or road conditions change.31


As an NVIDIA Tesla qualified GPU computing platform, PowerEdge T640 is one of the most compact systems. It is a unique industrial-grade edge AI platform supporting dual NVIDIA Tesla T640 GPU cards. The system allows innovators to run multiple models simultaneously such as engaging advanced applications with false-fail and redundant GPU configurations or assign the two T640s to separate tasks, set one for video transcoding while setting the other for AI inference tasks. It supports 2 Intel Xeon Silver 4214 2.2Ghz 12 core CPUs with expansion capabilities. It also features compact dimensions and low power consumption characteristics. With T640 boosted AI inference processing power, it is ideal for medical image and video analysis, deep learning machine vision, autonomous machines and more.


Arend Hintze, an assistant professor of integrative biology and computer science and engineering at Michigan State University, explained in a 2016 article that AI can be categorized into four types, beginning with the task-specific intelligent systems in wide use today and progressing to sentient systems, which do not yet exist. The categories are as follows:


Jeanna N. Matthews is a Professor of Computer Science at Clarkson University and an affiliate of the non-profit research organization Data & Society. Her current work focuses on securing societal decision-making processes and supporting the rights of individuals in a world of automation. She has published research in a broad range of systems topics from virtualization and cloud computing to social media security and distributed file systems. In her interview, Matthews discusses her interest in operating systems, civil discourse in computing, fair and transparent algorithms, and more.


The Internet of Things (IoT) is an emerging paradigm that enables the communication between electronic devices and sensors through the internet in order to facilitate our lives. IoT use smart devices and internet to provide innovative solutions to various challenges and issues related to various business, governmental and public/private industries across the world [1]. IoT is progressively becoming an important aspect of our life that can be sensed everywhere around us. In whole, IoT is an innovation that puts together extensive variety of smart systems, frameworks and intelligent devices and sensors (Fig. 1). Moreover, it takes advantage of quantum and nanotechnology in terms of storage, sensing and processing speed which were not conceivable beforehand [2]. Extensive research studies have been done and available in terms of scientific articles, press reports both on internet and in the form of printed materials to illustrate the potential effectiveness and applicability of IoT transformations. It could be utilized as a preparatory work before making novel innovative business plans while considering the security, assurance and interoperability.


Smart city is one of the trendy application areas of IoT that incorporates smart homes as well. Smart home consists of IoT enabled home appliances, air-conditioning/heating system, television, audio/video streaming devices, and security systems which are communicating with each other in order to provide best comfort, security and reduced energy consumption. All this communication takes place through IoT based central control unit using Internet. The concept of smart city gained popularity in the last decade and attracted a lot of research activities [9]. The smart home business economy is about to cross the 100 billion dollars by 2022 [10]. Smart home does not only provide the in-house comfort but also benefits the house owner in cost cutting in several aspects i.e. low energy consumption will results in comparatively lower electricity bill. Besides smart homes, another category that comes within smart city is smart vehicles. Modern cars are equipped with intelligent devices and sensors that control most of the components from the headlights of the car to the engine [11]. The IoT is committed towards developing a new smart car systems that incorporates wireless communication between car-to-car and car-to-driver to ensure predictive maintenance with comfortable and safe driving experience [12].


The IoT architecture consists of five important layers that defines all the functionalities of IoT systems. These layers are perception layer, network layer, middleware layer, application layer, business layer. At the bottom of IoT architecture, perception layer exists that consists of physical devices i.e. sensors, RFID chips, barcodes etc. and other physical objects connected in IoT network. These devices collects information in order to deliver it to the network layer. Network layer works as a transmission medium to deliver the information from perception layer to the information processing system. This transmission of information may use any wired/wireless medium along with 3G/4G, Wi-Fi, Bluetooth etc. Next level layer is known as middleware layer. The main task of this layer is to process the information received from the network layer and make decisions based on the results achieved from ubiquitous computing. Next, this processed information is used by application layer for global device management. On the top of the architecture, there is a business layer which control the overall IoT system, its applications and services. The business layer visualizes the information and statistics received from the application layer and further used this knowledge to plan future targets and strategies. Furthermore, the IoT architectures can be modified according to the need and application domain [19, 20, 37]. Besides layered framework, IoT system consists of several functional blocks that supports various IoT activities such as sensing mechanism, authentication and identification, control and management [38]. Figure 6 illustrates such functional blocks of IoT architecture.


Moreover, increasing amount of massive data being generated through the communication between IoT sensors and devices is a new challenge. Therefore, an efficient architecture is required to deal with massive amount of streaming data in IoT system. Two popular IoT system architectures are cloud and fog/edge computing that supports with the handling, monitoring and analysis of huge amount of data in IoT systems. Therefore, a modern IoT architecture can be defined as a 4 stage architecture as shown in Fig. 8.


In stage 1 of the architecture, sensors and actuators plays an important role. Real world is comprised of environment, humans, animals, electronic gadgets, smart vehicles, and buildings etc. Sensors detect the signals and data flow from these real world entities and transforms into data which could further be used for analysis. Moreover, actuators is able to intervene the reality i.e. to control the temperature of the room, to slow down the vehicle speed, to turn off the music and light etc. Therefore, stage 1 assist in collecting data from real world which could be useful for further analysis. Stage 2 is responsible to collaborate with sensors and actuators along with gateways and data acquisition systems. In this stage, massive amount of data generated in stage 1 is aggregated and optimized in a structured way suitable for processing. Once the massive amount of data is aggregated and structured then it is ready to be passed to stage 3 which is edge computing. Edge computing can be defined as an open architecture in distributed fashion which allows use of IoT technologies and massive computing power from different locations worldwide. It is very powerful approach for streaming data processing and thus suitable for IoT systems. In stage 3, edge computing technologies deals with massive amount of data and provides various functionalities such as visualization, integration of data from other sources, analysis using machine learning methods etc. The last stage comprises of several important activities such as in depth processing and analysis, sending feedback to improve the precision and accuracy of the entire system. Everything at this stage will be performed on cloud server or data centre. Big data framework such as Hadoop and Spark may be utilized to handle this large streaming data and machine learning approaches can be used to develop better prediction models which could help in a more accurate and reliable IoT system to meet the demand of present time. 2ff7e9595c


0 views0 comments

Recent Posts

See All

Comentarios


bottom of page