Showing papers in "CSI Transactions on ICT in 2020"
TL;DR: The current state of the art, challenges to be solved, and the methodology that the authors have adapted for developing solutions in all these research areas are detailed.
Abstract: Smart city applications is not just limited to the use of technology in building better cities and improving the citizens’ quality of lives. In reality, it is capable of building comprehensive solutions for the development of sustainable communities, irrespective of being urban or rural. Different communities will have their own dynamics and complexity, some communities are widespread and distributed, while some are more dense and concentrated. Each of these characteristics can either become an advantage or a barrier while developing solutions for these communities, which can be addressed with the proper use of technology. Internet of things (IoT) and wireless sensor networks (WSN), are the technologies in particular that can help in building comprehensive solutions for creating sustainable communities as they have a wide variety of applications such as disaster management, healthcare, energy, water, surveillance, structural health monitoring and environmental monitoring. This paper describes about four major research works in smart city application areas such as (a) design and deployment of an IoT system for early warning of landslides, (b) heterogeneous wireless communication design and optimization for enabling real-time management of interoperation of heterogeneous devices and data services in a hybrid electrical network, (c) development of (low-cost) IoT sensors for remote patient monitoring and (d) design and deployment of systems and solution for providing mobile offshore connectivity for fishing vessels—internet over sea. This paper details the current state of the art, challenges to be solved, and the methodology that we have adapted for developing solutions in all these research areas. The extensive research in IoT, WSN and its applications has resulted in receiving the award, titled “World Center of Excellence in Landslide Risk Reduction” and five patents, namely (a) network-based system for predicting landslides and providing early warnings (b) mobile infrastructure for coastal region offshore communications and networks (c) wearable wireless tongue controlled devices (d) Systems, methods, and devices for remote health monitoring and management (e) Systems, methods, and devices for remote health monitoring and management using internet of things sensors
21 citations
TL;DR: This paper presents the detailed discussion in form of literature review of each aspect of WSN design starting from mote design and ending to associated application i.e. crop production or agriculture domain.
Abstract: From the last few decades, a lot of research work has been conducted to develop wireless sensor networks (WSN) for wide variety of applications such as agriculture and environment monitoring, healthcare, military applications etc. Nowadays, the academicians and researchers are working on resolving various research issues in this technology domain. A few of the research issues faced by researchers’ are designing low cost and low power motes, network topologies, deployment strategies, and energy efficient data routing protocols. A lot of research has already been done in each of the mentioned areas individually but only a few research articles or literature discusses research work about two or more areas collectively. The existing literature explores the simulation work only and a few has given the description of complete design, development and deployment of a WSN. This paper presents the detailed discussion in form of literature review of each aspect of WSN design starting from mote design and ending to associated application. Since a lot of applications are using WSNs for monitoring purpose, this paper focuses on one of the most crucial application for livelihood i.e. crop production or agriculture domain.
21 citations
TL;DR: In this work a lead free perovskite solar cell structure is proposed with NiO as the hole transport material (HTM), CH 3 NH 3 SnI 3 as the perovSKite absorber material and PCBM (phenyl C 61 butyric acid methyl ester) as the electron transport material.
Abstract: In this work a lead free perovskite solar cell structure is proposed with NiO as the hole transport material (HTM), CH3NH3SnI3 as the perovskite absorber material and PCBM (phenyl C61 butyric acid methyl ester) as the electron transport material (ETM). Numerical analysis of the designed solar cell is performed using Solar Cell Capacitance Simulator (SCAPS-1D) program. The power conversion efficiency (PCE) of the optimized device stack is found to be above 29% with Voc = 0.98 V, Jsc = 34.86 mA/cm2, FF = 85.64%. The lead free perovskite solar cell with different HTM and ETM may be investigated for high PCE.
10 citations
TL;DR: The proposed aspect-based sentiment analysis model uses polarity classification and sentiment extraction on reviews, and extracts the most interesting polarity aspects preferred by the customers automatically using both machine learning and deep learning algorithms.
Abstract: Extraction of positive or negative opinions from any online content has received more consideration from researchers during the past decade, since the number of internet users that actively use online review sites, social networks and personal blogs to express their opinions has been growing. Sentiment analysis is the study of automated techniques for extracting sentiments from written languages making use of natural language processing tasks to thoroughly pre-process the data and extract polarity from the data. Customers who want to purchase products or services as well as business organizations, often rely on online reviews for knowing the overall user sentiment. Based on the sentiment, customers can choose whether or not to purchase a product while the organizations get an overall picture of their product. Aspect-based sentiment analysis helps in extraction of important features called aspects because knowing the polarity only is not sufficient. The proposed aspect-based sentiment analysis model uses polarity classification and sentiment extraction on reviews, and extracts the most interesting polarity aspects preferred by the customers automatically using both machine learning and deep learning algorithms. A search engine to pull out tweets and reviews relevant to user specified keyword is developed and corresponding interesting aspects are displayed.
10 citations
TL;DR: The WAN network has been around for the past 20–25 years, but today, with the outburst of applications, the applications are no longer confined to the private data centers and have been moved to the public cloud.
Abstract: The WAN network has been around for the past 20–25 years. Earlier, in the campus networks, the people working in the campuses were accessing the applications hosted on a private cloud, which is mainly within their campuses (Fig. 1). The data center was nothing but the installation of the servers hosting those applications, and those were predominantly residing within the campus. It was easier for people to access those applications from those branches/campuses. They break out to the internet only on a need basis. But today, with the outburst of applications, the applications are no longer confined to the private data centers. There are a lot of applications that are hosted as a SaaS (Software as a Service) model. Microsoft Office 365 and Google Cloud platforms are examples of the SaaS model. IaaS (Infrastructure as a Service) model is offered by vendors like Amazon and Google. As a result, the applications in the private cloud have been moved to the public cloud, either in the SaaS form or in the IaaS form (Fig. 2). In the users’ side, the users earlier were predominantly accessing the application from the campus or the branches. Also, the users were using their compute devices like laptops or desktops. Nowadays, with the increase in the complexity, the users are mobile. The users want to access the applications using any device like smartphone and tablets. Therefore, it calls for an increase in the security to access those applications, so that the right people get access to the right applications. For example, a person could be accessing his corporate application from a public network. If the person is using his corporate provided device, he could be provided access. But, if the person uses his personal device, the network should be intelligent enough to understand that it is not a corporate provided device. Even though the person is a corporate user, if the person uses his personal device, the network should restrict the access. This kind of intelligence is expected from the network. In addition, there is a lot of IoT called as cloud computing. These devices also want to access the applications which are hosted on the data center. Thus, the WAN is becoming slightly complex than what it is used to be earlier. This calls for an intelligent connectivity as access to these applications become critical (Fig. 3). In order to find the reason for using SD-WAN, some of the business values which is being predicted by some of the industry analysts like IDC (Fig. 4): The cost of operations of SD-WAN is lower than the traditional WAN. The main reason is as follows. In the traditional WAN, people had to configure the devices, box by box manually. But, in SDWAN, overlay is done in which a common management platform defines the policy once and that gets pushed across to all the devices. Since it is easy to implement those policies, the network uptime is much higher than what it used to be. & Raghavan Kasturi Rangan rkasturi@cisco.com
8 citations
TL;DR: A short survey of programmable networks starting from pre-Software Defined Networking efforts to the more recent programmable data-plane, highlighting the benefits and challenges and presenting some of the challenges and future research directions.
Abstract: In this paper, we present a short survey of programmable networks starting from pre-Software Defined Networking efforts to the more recent programmable data-plane. In particular, we highlight the benefits of data-plane programmability and present some of the challenges and future research directions.
7 citations
TL;DR: A distributed HMIS architecture is presented which allows for integration of multiple stand-alone HMIS on a common platform that enables interoperability via EHR sharing and aggregation of standardized data for enabling analytics and visualization.
Abstract: The healthcare domain in India is rapidly evolving with multiple healthcare and Hospital Management Information Systems (HMIS) being deployed at various medical facilities. However, aggregation and analytics on standardized healthcare datasets is possible only when these isolated systems are integrated on a common platform for exchange of patient Electronic Health Record (EHR) and data interoperability is facilitated. This presents a challenge due to the heterogeneous nature of systems and operational or policy-based restrictions in integrating them. In this paper, we present a distributed HMIS architecture which allows for integration of multiple stand-alone HMIS on a common platform that enables interoperability via EHR sharing and aggregation of standardized data for enabling analytics and visualization. Additionally, the proposed architecture also facilitates local, standalone deployments at primary medical facilities which are configured for local requirements and also synchronize essential datasets at the central node for enabling analytics. The deployment of this distributed architecture has also enabled uniformity and transparency in healthcare service delivery, while also facilitating better management and monitoring of available resources for efficient administration and governance.
6 citations
TL;DR: The research work is mainly focused on the design and analysis of wireless communication networks and explores the spectral and energy efficiency aspects of future generation networks like smart grid communications.
Abstract: In this article, we summarize the research conducted for YFRFs of Visvesvaraya PhD Programme The research work is mainly focused on the design and analysis of wireless communication networks We also explore the spectral and energy efficiency aspects of future generation networks like smart grid communications This research is useful for improving the performance of current and future generation of communication systems
5 citations
TL;DR: This paper proposes an enhanced CRO framework which uses latch instead of inverter to build an RO and efficiently utilizes the resources found in a configurable logic block to increase the number of RO configurations while using the same area.
Abstract: Physically unclonable function (PUF) is one of the most advocated security primitives which extracts the uncontrollable intrinsic physical property of the fabrication process to generate secret bits for authentication, random number generation and key generation. Ring oscillator (RO) PUF is the widely adopted PUF design to implement in FPGA platform, but it is highly error prone to environmental noise (i.e. temperature and voltage). The configurable RO (CRO) PUF is advocated to resolve this issue without increasing the area overhead. This paper proposes an enhanced CRO framework which uses latch instead of inverter to build an RO. The use of dedicated latch (i.e. available in an FPGA) in place of inverter eliminates the restriction to use odd number of delay units (inverters) in an RO configuration. The proposed design efficiently utilizes the resources found in a configurable logic block to increase the number of RO configurations while using the same area. Also, it provides the flexibility to include a latch in an RO configuration which in turns improve the reliability and the security as well. Experimental results on Xilinx Spartan 3E FPGA establish that the proposed design exhibits high stability despite varying environmental conditions without using any error correcting code or post-processing technique.
5 citations
TL;DR: How key enablers of 5G like software defined networking, network function virtualization, quality of service handling, and service based interface model will handle faster connectivity, huge data demand with less backhaul congestion, large number of devices and real-time applications as compared to MEC in 4G is emphasized.
Abstract: Multi-access edge computing (MEC) is an extension of cloud computing technology to provide an application deployment platform close to the end users.
This platform helps to deploy real-time, compute and resource-intensive applications like video caching, augmented reality, virtual reality, and Internet of Things at the edge. MEC enhances the performance of 4G and 5G networks since it achieves reduced application latency and optimizes the backhaul bandwidth utilization. Due to these benefits, MEC has become one of the key enablers of 5G technology. This paper covers evolution, architecture, current standardization activities, and use cases of MEC in cellular networks. The paper also emphasizes on how key enablers of 5G like software defined networking, network function virtualization, quality of service handling, and service based interface model will handle faster connectivity, huge data demand with less backhaul congestion, large number of devices and real-time applications as compared to MEC in 4G. Further, MEC deployment approaches in 4G and 5G are discussed, which play a vital role in achieving the requirement of 5G services.
5 citations
TL;DR: This article synthesizes and optimizes MAC layer of IoT based protocols using a novel delta diagram synthesizer and accurately captures both the high heterogeneity of the IoT and the impact of the Internet as part of the MAC layer architecture.
Abstract: Recently, cross-layer optimization has been emphasized by researchers for Internet of Things (IoT) where billions of devices communicate intelligently making every day communication highly informative. Medium Access Control (MAC) sub-layer is part of data link layer in IoT protocol stack. The energy consumption of things is greatly affected by MAC protocol which controls the node radio functionalities. In this article, we propose the design of an efficient MAC for IoT by minimizing its energy consumption at various levels. The minimization model is subject to reduction in distortion levels while simultaneously extending network lifetime. This article synthesizes and optimizes MAC layer of IoT based protocols using a novel delta diagram synthesizer and accurately captures both the high heterogeneity of the IoT and the impact of the Internet as part of the MAC layer architecture. Furthermore a novel hybrid Whale-Artificial Bee Colony optimizer framework is proposed to obtain optimal nodes based on energy consumption and the communication parameters among things, by exploiting the interrelations among different layer functionalities in the IoT. This novel system saves considerable computation resources of the resource constrained IoT devices and adapts to various data sources with different distortion levels rapidly.
TL;DR: A novel OpenNetVM based 5G framework (ONVM-5G) which uses Data Plane Development Kit for the communication between the core NFs and thus, is a candidate solution for implementing 5G core (5GC) in a box.
Abstract: Software Defined Networking (SDN) and Network Functions Virtualization (NFV) are the two fundamental paradigms that introduce concealed flexibility in communication systems. 5G has adopted SDN/NFV to make Network Functions (NFs) as virtualized software modules which run on top of commodity hardware. 5G aims to expand its reach to various business verticals and serve diverse users using network slicing which creates multiple logical networks on top of a shared infrastructure. This paper proposes a novel OpenNetVM based 5G framework (ONVM-5G) which uses Data Plane Development Kit for the communication between the core NFs and thus, is a candidate solution for implementing 5G core (5GC) in a box. The proposed framework based deployment of network slices enhances the user experience by reducing the latency involved in performing various User Equipment activities and increasing data plane throughput. Moreover, ONVM-5G also helps in achieving the 5G goals by allowing the network operators to efficiently serve the users in small dense cells and make service level isolation among the diverse users through network slicing by providing a network in a box like deployment of 5GC.
TL;DR: In this paper, handwritten Kannada characters are captivated in document fashion and are subjected to Pre-processing and attribute extraction processes, which results in an impressive classification process deep convolution neural network classifier is preferred.
Abstract: In pattern recognition, identifying Kannada handwritten numerals are a complex knot. This paper portrays an avenue that pikes us to attain a most potent Kannada Numerals recognition process. In this, handwritten Kannada characters are captivated in document fashion and are subjected to Pre-processing and attribute extraction processes. Pre-processing entails steps like noise removal, binarization, normalization, skew amendment, and thinning. Features are extricated by exploiting strategies like Drift Length Count, Direction related progression code, DWT and Curvelet Transfiguration Wrapping. For an impressive classification process deep convolution neural network classifier is preferred. Isolation accuracy of Kannada numeral aimed here will outsource 96% of accuracy.
TL;DR: The various charecteristics of a slice, its vertical use cases, and how it can used for a specific V2X industry are discussed and how existing operators can leverage this technology to compete against the Over-The-Top (OTT) operators are discussed.
Abstract: Network requirements are evolving as more and more verticals have started using Internet. This will require operators to build a dynamic and virtualized network consisting of routers, switches and compute servers. This dynamic network can then offer service beyond just bandwidth services called "slices" which can be customized for each industry vertical. This paper discussed the various charecteristics of a slice, its vertical use cases, and how it can used for a specific V2X industry. This paper then discusses about the communication aspect of slice using SDN and how existing operators can leverage this technology to compete against the Over-The-Top (OTT) operators.
TL;DR: In this work, an optimal bidding strategy for a supplier has been developed, considering rival’s bidding behavior, in an hourly day-ahead pool market, as a bi-level multi objective optimization problem (BLMOOP).
Abstract: Global warming is one of the most alarming phenomena facing our planet today. There is a consensus among scientists that human-induced greenhouse gases (GHGs) should be regulated to slow down the heating of the Earth’s oceans and atmosphere. Energy consumption and CO2 emissions are continuously increasing in different countries, especially developing ones such as India, whose emerging economy and rapid economic development has caught the attention of the world. CO2 emission trading is executed by the various developed countries to alleviate the affect of GHG emissions. In this work, an optimal bidding strategy for a supplier has been developed, considering rival’s bidding behavior, in an hourly day-ahead pool market. Bidding problem has been formulated as a bi-level multi objective optimization problem (BLMOOP), where in the first level generator submit bid strategically to the ISO and in the next level a particle swarm optimization (PSO) approach has been employed for the maximization of social welfare with risk management. It is assumed that each generator should submit bid in sealed auction based on pay-as-bid MCP (market clearing price) mechanism with knowing the rival’s bidding behavior. The practicability of proposed optimization method is examined by an IEEE-30 bus system which consists of six suppliers.
TL;DR: This paper presents an enhanced and efficient CL Digital Multi-Receiver scheme with anonymity without bilinear pairing which is secure against foregeability, key replacement and key escrow attack, and compares the proposed scheme with some existing schemes in terms of computation and communication cost with higher level of security.
Abstract: Certificate-less (CL) signature schemes were presented to deal with key escrow issue in ID based signature. In recent years a number of digital signatures schemes with bilinear pairing have been presented to maintain information security and integrity. Acutely, the greater part of them endure intent security invasion. They don’t enforce in terms of computing and dissemination complications. Many researchers have proposed different schemes on anonymity for multi receiver identity based cryptography. In this paper an attempt has been made to propose an efficient CL Digital Multi-Receiver (CL-DMS) scheme with anonymity without bilinear pairing which is secure against foregeability, key replacement and key escrow attack. Our motivation is to design a CL DMS that simulcast a message with high level of computation efficiency and employs security that only proficient recipients can fetch the message. This will also include the non-repudiation as well as the time replay attack. We presented an enhanced and efficient scheme, which is secure against Type 1, and Type II attacks. We also compare the proposed scheme with some existing schemes in terms of computation and communication cost with higher level of security.
TL;DR: A new adaptive beamforming algorithm based onVariable step size is proposed in this paper called sign variable step size LMS algorithm and results show that it gives a better null placement in the direction of interference and the signal to noise ratio is improved upon the existing algorithms with less computational complexity.
Abstract: Beamforming algorithms are designed for generating main beams in the direction of interest and place nulls in the direction of interferences. Adaptive beamforming can be achieved by various optimization criteria. The criteria for an efficient design of algorithm is the speed of convergence, deeper null placement, better signal to noise ratio, and robustness. Motivated by the reduction in the computational complexity a new adaptive beamforming algorithm based on variable step size is proposed in this paper called sign variable step size LMS algorithm. The proposed algorithm is applied to the uniform linear array antenna. Simulation results show that it gives a better null placement in the direction of interference and the signal to noise ratio is improved upon the existing algorithms with less computational complexity.
TL;DR: The motivation, background and impact of these new frontiers of Artificial Intelligence (AI), Machine Learning (ML) and Natural Language Processing (NLP) have been described and two of the very recent research have been presented, where the first one attempts at incorporating courteousness in a dialogue agent, and the second one addresses natural language generation in a multi-modal setup involving text and images both.
Abstract: In this paper, some of the recent developments, aiming at building human-like Conversational Artificial Intelligence (AI) agents, have been presented very briefly. These robust dialogue systems are capable in dealing with the various affect attributes, such as sentiment, emotion, and courteousness. Firstly, the motivation, background and impact of these new frontiers of Artificial Intelligence (AI), Machine Learning (ML) and Natural Language Processing (NLP) have been described. Thereafter, two of our very recent research have been presented, where the first one attempts at incorporating courteousness in a dialogue agent, and the second one addresses natural language generation in a multi-modal setup involving text and images both.
TL;DR: Cognitive radio, non-orthogonal multiple access in heterogeneous networks, wireless caching and massive MIMO are explored for 5G and beyond systems, and simulation results corroborate the superiority of proposed techniques for next generation wireless communication systems.
Abstract: A vital challenge for 5G and beyond wireless communication system is to meet the ever-growing demand for high-speed data with massive connectivity with low latency. In this report, the research outcomes of four key technologies, namely cognitive radio, non-orthogonal multiple access in heterogeneous networks, wireless caching and massive MIMO are explored for 5G and beyond systems. The key ideas for each technology are described, followed by the proposed techniques. Simulation results corroborate the superiority of proposed techniques for next generation wireless communication systems.
TL;DR: The quantum transport in AlN/β-Ga2O3 based high electron mobility transistor (HEMT) along with the compact model development of GaN-based metal oxide semiconductor (MOS)-HEMt is presented for high frequency and high power electronics.
Abstract: III-nitride wide bandgap semiconductors such as alloys of (Al, Ga, In) N play a vital role in high frequency, power electronics, and opto-nanoelectronics applications. Apart from this recently, wide bandgap semiconductor, β-Ga2O3 has attracted the attention of worldwide researchers as an alternative of GaN substrates due to its low-cost processing and wide availability of large scale wafers. Due to its suitable material properties, β-Ga2O3 can also be used for high frequency and high power electronics applications. In this paper the quantum transport in AlN/β-Ga2O3 based high electron mobility transistor (HEMT) along with the compact model development of GaN-based metal oxide semiconductor (MOS)-HEMT is presented for high frequency and high power electronics. Finally, the modeling and simulations of InGaN based nanowire is developed for solar photovoltaics and opto-nanoelectronics applications.
TL;DR: A comparative analysis of the sensing performances for the above-mentioned three ternary hybrid gas sensors was illustrated and room temperature (27 °C) alcohol sensing with a very high detection range was achieved with appreciably fast response time and recovery time without compromising the response magnitude.
Abstract: Present paper aims to study the comparative analysis of the sensing performance between three fabricated metal oxide based ternary hybrid device structures [Pd/RGO/TiO2-nanotubes (NTs), Pd/RGO/MnO2 nanoflowers (NFs) and Pd/RGO/WO3-nanoflowers (NFs)]. In each of the ternary structure, the oxide nanostructure play the role of basic sensing matrix where due to extremely high surface to volume ratio of the corresponding nanostructure, the availability of adsorption sites increases, which leads to increased sensitivity or response magnitude (%RM) of the sensor. On the contrary, RGO acts on a high mobility (~ 1600 cm2/V-s) distributed connector among the neighbouring nanostructure leading towards improved response or recovery kinetics. Finally, Pd nanoparticles due to their catalytic activity (which reduces activation energy requirement for target species dissociation), bring forth as the substantial reduction in operating temperature. By employing those ternary hybrid structure sensing layers, room temperature (27 °C) alcohol sensing with a very high detection range (1–700 ppm) was achieved with appreciably fast response time (~ 12–20 s) and recovery time (~ 23–30 s) without compromising the response magnitude (80–98% at 700 ppm). Finally, a comparative analysis of the sensing performances for the above-mentioned three ternary hybrid gas sensors was illustrated.
TL;DR: A Costas array based key pre-distribution scheme for location dependent as well as location independent wireless sensor networks is proposed and comparative performance analysis of the scheme is shown.
Abstract: Security of wireless sensor networks is a major issue to protect sensor nodes from the attacker. Key management or key pre-distribution (KPS) is the preliminary step in the security of a sensor network, where secret keys are loaded into the sensor node’s memory before placement of the node in target positions. Various schemes for KPS have been proposed for location dependent as well as location independent wireless sensor networks. Costas arrays are
$$n \times n$$
matrices grid represented where dots are placed for the l’s and leave blanks for the 0’s of the matrix for a positive integer n with the property that the vectors connecting two dots of the grid are all distinct as vectors. Costas arrays have a wide range of application including the construction of sonar signal pattern, cryptography, etc. Due to the Costas arrays property of uniqueness among their elements, it can be used in key pre-distribution of wireless sensor networks, especially in a location-dependent grid-based network. In this paper, we have proposed a Costas array based key pre-distribution scheme. Simulations are done on different network scenarios and the results were analyzed. Comparative performance analysis of the scheme is shown for different grid sized network considering various order of Costas array.
TL;DR: Different modalities of Satvavajaya chikitsa which are similar to the steps of tele-counselling helps in providing medical management also.
Abstract: Corona Virus Disease (Covid 19) is a highly infectious disease declared as pandemic by WHO. Due to its fast
spreading nature this is creating a threatening situation throughout the world. Breaking the chain of transmission by keeping social distancing along with maintaining strict personal hygiene is the only way to control the spread of the disease. The panic situation created by the disease along with the unfamiliar way of conduct imposed several psychological issues on human beings. Chittodvega (Anxiety disorder) is one among the commonly found psychological problem during this pandemic. An important modality called psychotherapy, mentioned in Ayurveda i.e. Satvavajaya chikitsa can be adopted in such situation. Tele-counselling which is coming under indirect counselling helps to handle the needy people with psychological issues, without directly approaching the medical practitioner during lockdown situation. Different modalities of Satvavajaya chikitsa which are similar to the steps of tele-counselling helps in providing medical management also.
TL;DR: This research aims to build a model to predict ambient air quality using Amplified Recurrent Neural Networks for a particular geographical area and develops deep learning-based recommenders/personal assistance/advisory systems to exploit the information systems and to capture its user’s preferences.
Abstract: Increasing volumes of offline/online data has thrown diverse challenges Availability of the data has made the researchers to urge for new paradigms to analyze data to gain useful insights from it Such analysis can be helpful for predictions in various domains such as medical informatics, cyber security, fraud detection etc Further, problem of overloaded information is that it leads to over-choice, especially for users in web applications Machine learning algorithms construct models that can learn from the data without being explicitly programmed, to perform predictions/decisions However, the limitations of machine learning have paved a way to deep learning Unlike the task specific algorithms of machine learning, deep learning learns feature representations from data itself These traits enabled deep learning to achieve futuristic results for exceptionally complex tasks of various domains Particularly in health care, natural language processing, information retrieval, computer vision etc Area of research in this work focused on processing sequential data in the domains of ambient air quality management and recommender systems Air quality data analysis is one of the domains where deep learning models have outperformed the traditional machine learning models Hence, this research aims to build a model to predict ambient air quality using Amplified Recurrent Neural Networks for a particular geographical area Recommender Systems (RS) have become a part of all the web applications these days The main objective of RS is to asist users by avoiding information overload through generating personalized suggestions Deep learning-based RS models have gained more prominence these days Consequently, this research aims to develop deep learning-based recommenders/personal assistance/advisory systems to exploit the information systems and to capture its user’s preferences The primary goals of the objectives were successfully accomplished with promising results
TL;DR: This paper presents an efficient key distribution scheme for critical infrastructure such as SCADA (Supervisory Control and Data Acquisition) running on constraint environment.
Abstract: The strength of cryptographic techniques depend on the security of keys and on the strength of encryption algorithms. In symmetric key encryption, two communicating parties share a secret key that should not be known to others. Furthermore, frequent key changes are needed to avoid the attackers from learning the key. Most of the networks are now connected to the Internet and all the information, including exchange of keys, flow through it. Therefore, key distribution is one of the major challenges in communication and network security. This problem becomes more complex when the systems have certain constraints like limited computation power, low bandwidth etc. This paper presents an efficient key distribution scheme for critical infrastructure such as SCADA (Supervisory Control and Data Acquisition) running on constraint environment.
TL;DR: The aim of this proposed research is the performance enhancement of available nanobiosensors in terms of sensitivity and selectivity for the early and rapid diagnosis and prognosis of viral infections.
Abstract: The idea is to develop nanostructure based biosensing platforms as commercial diagnostic tools for early stage non-invasive virus detection to enhance the selectivity from the existing electrical biosensors by introducing additional electrical signatures and correlating electrical characterization data with other proteomics characterization results. The aim of this proposed research is the performance enhancement of available nanobiosensors in terms of sensitivity and selectivity for the early and rapid diagnosis and prognosis of viral infections.
TL;DR: One of the key elements of engineering simplicity is building autonomous networks, which will create engineering simplicity, and there is a difference between automated and autonomous networks.
Abstract: The anchors in today’s networks, are complex. It is so complex that while building new architectures like 5G, distributed data centers, software defined wide area networks, we have to contain the complexity. In an open framework, the complexity has to be simplified. Every vendor has been innovating on these technologies, like network management where hundreds of devices could be managed from a single site or protocols defined over a period of time like SNMP, SSH. There are several protocols which were used to mine data or to command the network. Those were industry standards. The management platform never looked at which vendor equipment is really on the site as long as a common protocol to interact is used. That still created a whole level of complexity. We moved SNMP to something like telemetry now. The hardware is purpose built so that it can give proactive vectors or health indicators to the monitoring device or a big data analytics engine. This is called as telemetry. Not all hardware can do it because of the number of very vital information. For example, in a human body, say, three or four vital parameters are tracked per millisecond or per second and streamed into a device or to a doctor who can analyze all things. That is the level of information needed. But, if we multiply it with a number of locations or the number of devices in the network, that becomes very complex. That is why it has to be contained. The moment we have the complexity, the order of magnitude, the number of parameters etc. we will have a major technical debt. One of the key elements of engineering simplicity is building autonomous networks. There is a difference between automated and autonomous networks. Automated networks are available today. Even in software industry, many processes can be automated. Exactly the same way in networking domain, a lot of processes can be automated, based on certain static rules and parameters. A lot of investment is happening in building an autonomous system which will create engineering simplicity. The other area is the journey to multi-cloud. Cloud is emerging as a de-facto standard. There is a term known as cloud first companies. There are a number of companies who are born today are cloud first. These companies check whether they can do it in a cloud before doing anything themselves on-premise. At the same time, can we restrict ourselves to a single cloud? Probably, multiple clouds are available, and applications may be running across multiple clouds. Can a concept called network as a service be driven? Whether it is one cloud, two clouds, own data center, own premises, DR and DC, can all be put together to look like a multi-cloud infrastructure and network as a service? It has to be open and evolvable. & Sajan Paul skpaul@juniper.net
TL;DR: Energy efficient design techniques for clock recovery in multilane receivers, receiver frontend in digital CDRs, reconfigurable voltage-mode transmitter, and PAM4 equalizer in full-duplex transceivers are discussed.
Abstract: This paper reviews the need of a high performance wireline communication in the background of wirelessly connected billions of sensor nodes by 2020s. It compares the performance of the state-of-the-art wireline transceivers and underlines the challenges in improving the performance in the midst of tapering in CMOS technology scaling. This paper elaborates on the ongoing research to track the increasing bandwidth requirements in processing platforms with an affordable power budget. Energy efficient design techniques for clock recovery in multilane receivers, receiver frontend in digital CDRs, reconfigurable voltage-mode transmitter, and PAM4 equalizer in full-duplex transceivers are discussed.
TL;DR: FACTS incorporated power flow model is used for the determination of flow of real and reactive power by using the power flow tracing principle, which involves matrix properties for the real and proactive power allocation and cost allocation for different participants.
Abstract: In the past few decades, the demand for electricity has increased substantially. A number of factors affect this rapid rise in electricity demand. With this increased demand the role of transmission network becomes crucial for suitable operation, especially under restructured environment. The different limitations such as stability limit, voltage limit and thermal limits enforce restrictions on power wheeling capability of transmission lines. The major issues associated with power industry especially in the pool-based market are the recovery of transmission costs apportioned between the diverse. Restructuring of the electricity supply industry has taken place around the world. This restructuring consists of various new aspects such as transmission embedded cost allocation, reactive power cost allocation, transmission loss allocation, the impact of power flow regulating devices etc. Many methodologies and algorithms were proposed for addressing these issues. In this paper, FACTS incorporated power flow model is used for the determination of flow of real and reactive power by using the power flow tracing principle, which involves matrix properties for the real and reactive power allocation and cost allocation for different participants. FACTS devices which have the capability of regulating the power flow and voltage support at the buses. Hence UPFC incorporated in the transmission system have been considered in this paper. Comparative results with and without FACT controller are present for the sample 5 bus system and methodology also tested on IEEE 14 bus system.