International Journal of Networking and Virtual Organisations (22 papers in press)
Special Issue on: ICICCT 2020 Sustainable Computing and Wireless Networks
PRIORITY-BASED ENERGY EFFICIENT MULTI QUEUE HEURISTIC SCHEDULING FOR INTENSIVE DATAIN CLOUD COMPUTING (PB-EES)
by Vignesh V., Santhosh R
Abstract: In general the problem-solving strategies used in heuristic approaches is different compared to conventional algorithms. They differ in handling the issues in different operating conditions thereby it enhances the overall performance of the system. In this research paper we have also implemented a heuristic scheduling approach in cloud computing environment to improve the energy efficiency. Since cloud needs a better scheduling module for intense data applications this proposed priority-based energy efficient multi queue heuristic scheduling model is suitable and provides efficient scheduling with minimum energy. Experimental results are compared with greedy and load balancing scheduler algorithms to prove that proposed model achieves 93.45% of energy efficiency.
Keywords: Cloud computing; Heuristic model; energy efficiency; scheduling.
Identifying DDOS Attacks in 4G Networks using Artificial Neural Networks and Principal Component Analysis
by Nagesha A G, Mahesh G, Gowrishankar S
Abstract: Abstract: Denial-of-Service (DoS) attack is one in which attackers make certain queries by sending messages to the remote or target servers with an intention to stop or shutdown the servers. Those messages causes such an impact to the servers that it makes no response for the users. When this DoS attack is performed using number of systems that are compromised for attacking a single system, then it is called as Distributed Denial-of-Service (DDoS) attack. In this paper an Artificial Neural Network (ANN) combined with Principal Component Analysis (PCA) is used to identify the traffic as normal or a DDoS attack in 4G networks. The feature space dimension is reduced using PCA and the dimensionally reduced features are given as input to the feed forward neural network for training. The experiment is conducted using KDD dataset. The recognition accuracy of the proposed system is improved when compared to the existing systems using RBF Networks, Naive Bayes and Random Forest.
Keywords: Artificial Neural Network; Distributed Denial-of-Service; Principal Component Analysis.
FORECASTING INTRADAY STOCK PRICE USING ANFIS AND BIO-INSPIRED ALGORITHMS
by Kumar Chandar S.
Abstract: Forecasting in financial markets is to estimate the future behaviour of stock price.The main focus of this study is to explore the predictability of stock price with variants of Adaptive Neuro Fuzzy Inference System (ANFIS) and suggests a hybrid model to enhance the prediction accuracy.Two variants of ANFIS model is designed which include Genetic Algorithm-ANFIS(GA-ANFIS) and Particle Swarm Optimization-ANFIS (PSO-ANFIS) to forecast stock price more accurately.The standard ANFIS is tuned employing GA and PSO algorithm. The experimental data used in this investigation are stocks traded per minute price of four companies from NSE. Sixteen technical indicators were calculated from the historical prices and used as inputs to the developed models.Using the designed models, experiments were conducted for the period from January 2018 to February 2018 utilizing intraday stock price.Prediction ability of the developed models are analyzed by varying number of samples such as one day, five days and ten days data as an input. Prediction errors are measured and compared to find the suitable model.Numerical results obtained from the simulation confirmed that the proposed PSO-ANFIS model has the potential to predict the future stock price more precisely than GA-ANFIS as well as other earlier models taken for comparison.
Keywords: Adaptive neuro fuzzy inference system; bio inspired algorithm; genetic algorithm; intraday day; minute price and particle swarm optimization.
A SECURE ENERGY EFFICIENT EVENT DETERMINATION ALGORITHM FOR UNDERWATER WIRELESS SENSOR NETWORKS
by INDULEKSHMI S. KAIMAL, BINU G S
Abstract: Effective routing in Under Water Sensor Networks (UWSNs) is a highly demanding task because of the weak radio channels in water and the changing topology of sensors that maneuver indifferently with water. Energy efficient Derivative Based Prediction (DBP) approach has been proposed to reduce the amount of messages required for transferring the data samples from a wireless sensor node to a base station. However, predicting sensor data is not effortless for underwater channels as they are prone to wormholes attack due to the inconsistent propagation delays. In this paper, an enhanced empirical data predicting method that not only predicts sensor data but also the topographical movements of the sensor nodes is presented. Further, a security aware locally restrained algorithm RDV-HOP is proposed to assess the influence of paths to other broadcaster nodes and to transmit the collected information to the network. The Advanced Encryption Standard (AES) algorithm is used to protect the confidentiality of data. The proposed method demonstrates a significant benefit in managing the dynamic networks with better energy efficiency even with extensive number of attackers. Extensive simulations were performed with variable number of wormhole attackers and the results show higher packet delivery ratio with reduced delay.
Keywords: Underwater Wireless Sensor Network; Derivative based Prediction; Depth Based Routing; Encryption; and Prevention.
Social wireless network user big data mining based on Python platform and hierarchical clustering computing
by Kun Wang, Xiangbo Liang
Abstract: Human behavior, because of its complexity, makes it very important and interesting to explore the law of human behavior. In recent years, the online social network represented by online personal community, online dating network and social network makes the amount of data of network users surge. The era of big data online social network gives us unprecedented opportunities to study human behavior. The development of information science, the emergence of computer and the development of modern data storage technology provide us with a new objective material basis for the study of human behavior. Data mining is an interdisciplinary subject, involving statistics, pattern recognition, information retrieval, machine learning and other disciplines. Data mining has been paid more and more attention by domestic and foreign academic circles, and has become a research hotspot. Therefore, this paper studies social wireless network user big data mining based on hierarchical clustering computing, the system is implemented via Python and compared with the latest models. The convincing results have proven the effectiveness.
Keywords: Python; Hierarchical Clustering; Big Data; Social Nature; Wireless Network; User Data; Data Mining.
Intelligent Building BIM Fusion Data Analysis Framework Based on Speech Recognition and Sustainable Computing
by Zhiqiang Gao
Abstract: With the development of national economy, the level of science and technology and the improvement of peoples living standard and amount of urban data expands at the geometric multiple growth rate, and more and more complex buildings. Therefore, the traditional two-dimensional plane data is difficult to meet our data needs for such a complex urban system. BIM model has been applied to many fields, such as building engineering, model visualization and indoor path planning. However, the BIM model has no description of geographic information, resulting in the limited application of the model in the space. The concept of smart city originated from "smart earth". "Smart earth" refers to the formation of Internet of things and Internet connection in various media such as hospitals, power grids, railways, etc. to achieve the integration of human society and physical systems. In practical application, speech recognition is usually combined with the natural language understanding, natural language generation and speech synthesis technology to provide a natural and smooth human-computer interaction platform based on speech. This paper constructs a BIM data fusion framework based on speech recognition model with sustainable computing. Experimental results show that this method can effectively analyze BIM data and obtain satisfactory result.
Keywords: Speech Recognition? Signal Processing? Intelligent Building? BIM? Data Fusion? Smart City.
Optimal Feature Selection in Intrusion Detection using SVM-CA
by Shinly Swarna Sugi, Raja Ratna S
Abstract: Feature selection plays a vital role in toning down the effects of the curse of dimensionality in the humungous datasets seen in Intrusion Detection. Feature Selection Algorithms are used to pick the relevant features and averts the extraneous and repeated features from the dataset to improve the efficiency. It can reduce processing time, dimension of data and enhance the performance of the system in terms of precision and training time. This paper proposes a variant of SVM, known as SVM Correlation Algorithm (SVM-CA) to choose the relevant features. The combination of Support Vector Machine with correlation algorithm enhances the classification accuracy. Our proposed SVM-CA algorithm deals with the problems faced by the existing algorithm like low accuracy and high detection time. The performance of the algorithm is appraised by five parameters including the modeling time, true positive rate (TPR), false positive rate (FPR), false negative rate (FNR) and accuracy. The experimental results show that our proposed technique decreases the false positive rate and processing time.
Keywords: Correlation Algorithm Feature Selection; Intrusion detection; support vector machine; Machine Learning.
Sustainable Data Analysis Framework of Smart City Based on Wireless Sensor Network
by Hua Wang
Abstract: Under the background of the new era, the construction of smart society has been promoted to the national level. Smart city is the foundation of smart society, which is highly valued. The development of the new generation of information and communication technology and its in-depth application in various fields of the city form a new urban development model, which is called smart city. In terms of technology, smart city applies the new generation of information and communication technology to all walks of life in the city, and transmits information through communication mechanisms, including communication network and Internet. With the continuous development of wireless sensor network technology, smart phone communication technology and Internet information technology, remote wireless mobile home monitoring system will become one of the mainstream of smart home network. Smart home technology is the use of wireless communication technology, sensor network technology, Internet technology, embedded technology and computer technology to connect all kinds of home electrical equipment through the network. Wireless sensor network and wireless communication technology monitor various devices and monitoring targets in the form of self-organizing multi hop. Therefore, this paper studies the sustainable data analysis model of intelligent city based on wireless sensor network. The performance of the model is validated throug the public datbase, and the results are satisfactory.
Keywords: Wireless Sensor Network; Sustainable Development; Smart City; Energy Saving Measures; Data Analysis; Smart home; Monitoring.
Framework Design of Urban Traffic Planning Based on Wireless Network Optimization and Cognitive Sustainable Data Retrieval
by Gang Zhang, Tiechun Li
Abstract: With the acceleration of urban development in China, the development of urban traffic system should be improved. Therefore, there are many problems such as traffic congestion, resource shortage and environmental pollution. Transportation is one of most basic needs for the survival and development of human society. Urban traffic planning is of great significance to the sustainable development of modern cities. It should be noted that urban traffic planning is only a part of urban planning, but there is a very close relationship between them. If in the process of urban traffic planning, the value of urban traffic planning can not be brought into full play if the coordination relationship between them is not clarified. In order to avoid the above problems, traffic management departments have established various traffic management information systems with the help of wireless network optimization technology. The data structure and supported data types of each system are also different, which results in the operation of "island" management information system, therefore, this paper constructs a framework of urban traffic planning based on wireless network optimization model and cognitive sustainable data retrieval. We optimize the communication model to construct the efficient system.
Keywords: Wireless Network Optimization; Cognitive Model; Data Retrieval; Sustainable Development; Urban Traffic Planning.
Painting Image Quality on Visual Art Analysis for Information Transmission of Wireless Networks
by Hanxiao Li
Abstract: As a treasure of human civilization, painting art has immeasurable artistic value. However, due to the influence of human factors, accidents and natural environment, there are different degrees of damage. Traditional image acquisition is based on RGB three channel chroma information acquisition, which can not objectively represent the color information of painting works. In the research of computational aesthetics, it is necessary to analyze the high-resolution images of paintings. However, most of the actual paintings are low resolution, so this paper studies the influence of the quality of painting images on the visual art analysis effect. Nowadays, wireless network is widely used and has become the main way of user terminal access to the network. With the rapid development of wireless network services, especially the development of mobile services, the performance requirements of data transmission under wireless networks are becoming higher and higher. Based on the development of wireless network transmission technology, this paper studies the image quality analysis system of visual art painting. The designed system is validated through different scenarios. The experimental results have proven the effectiveness.
Keywords: Wireless Network Transmission; Visual Art; Painting Image; Quality Analysis; Embedded System.
Wireless Image Transmission Network: 3D Modeling Based on 2D Hand Drawing with Sustainable Computing
by Hanxiao Li
Abstract: Nowadays, we can describe the world in a variety of ways, such as the original text description form, and then to the voice, image, video and other multimedia forms. All these make the methods of describing the world more and more abundant. The appearance of 3D model and 3D scene extends the way of describing the world to 3D space. Because it is closer to the real world, our perception is more rich and more realistic. In the two-dimensional art creation, it is the basic operation mode for art personnel to input and draw two-dimensional images by the handwriting pad. This way is close to manual drawing, in line with the operating habits of artists. In recent years, the emerging research on computational sustainability has become an effective way and a new research hotspot of image wireless transmission. The advent of big data era brings opportunities for the research of computational sustainability, as well as new challenges such as problem complexity, computational efficiency, and method scalability. Based on this background, this paper designs the 3D modeling of 2D hand drawn wireless image transmission network based on sustainable computing model.
Keywords: Sustainable Computing; Big Data; 2D Image; Hand-Drawn Graph; Wireless Transmission Network; 3D Modeling.
Wireless Network for Computer Puzzle Online Software Cloud Platform Based on CBIR and Sustainable Computing
by Bing Sun
Abstract: Wireless network for the computer puzzle online software cloud platform based on the CBIR and sustainable computing is proposed in this paper. In order to improve the ability to resist shortcut attacks, which is generally speaking, encryption algorithms, the higher the number of rounds, the better of the overall performance, because the attack from the crack encryption algorithm. On the other hand, the efficiency of shortcut attack is obviously more than that of exhaustive key search attack high. Hence, we propose two novelty aspects. (1) The CBIR is optimized using the machine learning models to enhance the computational robusness. (2) The WSN and computing models are combined to improve the robustness of the proposed methodology. Our research fins the optimal combination of the wireless systems with the CBIR system that will improve the efficiency. The experimental results have shown that the performance of the model is validated both from accuracy level and robustness level.
Keywords: CBIR Technology; Wireless Network; Sustainable Computing; Online Software; Cloud Platform.
Special Issue on: ISCV 2020 Methods and Applications of Computer Science and Information Technology
A Cluster Workload Forecasting Strategy Using A Higher Order Statistics Based ARMA Model For IaaS Cloud Services.
by Zohra AMEKRAZ, Moulay Youssef Hadi
Abstract: With the cloud computing services becoming more popular among Internet users, cloud providers are facing a challenge in allocating resources to users according to demand instantly. The delay caused by the Virtual Machines (VMs) start up time makes the reactive techniques, which allocate new resources only when a given load threshold is attained, not effective for the allocation process. An interesting alternative to the reactive technique is the proactive technique. This latter consists of predicting the future demand known as workload and allocating or releasing resources in advance to prevent any overload to occur and also to reduce any related costs. In this paper, we introduce an adaptive workload prediction method based on the use of Higher Order Statistics (HOS) and Autoregressive Moving Average (ARMA) model. The proposed method uses the HOS to make a Gaussianity checking test of the cloud workload and then decides the suitable identification method of the ARMA model to be used to forecast the workload. Furthermore, the proposed method updates the parameters of the ARMA model constantly whenever new workload data are available. We evaluate our proposal with two real workload traces extracted from cluster workloads. The results show that the proposed method has an average of 34% higher accuracy than the baseline ARMA model and presents a low overhead for forecasting incoming workload (<2 s).
Keywords: IaaS Cloud Services; Workload Prediction; Cluster Workload;
Autoregressive Moving Average; Higher Order Statistics.
Cloud Spot Price Prediction Approach Using Adaptive Neural Fuzzy Inference System With Chaos Theory
by Zohra AMEKRAZ, Moulay Youssef Hadi
Abstract: The dynamic pricing of cloud computing is a major challenge for cloud
users all over the world. This challenge was first addressed by Amazon under the
name of Amazon Spot Instance Market. Cloud users can bid for a spot instance
using this market and obtain the requested spot if their bids exceed a dynamically
changing spot price. Amazon publicizes the spot price but does not reveal how
it is determined. In this paper, we perform chaotic time series analysis over the
spot price trace.We also develop a chaos based Adaptive Neural Fuzzy Inference
System (ANFIS) model based on phase-space vectors obtained during the phase
of chaotic analysis. Next, we study the effect of chaos existence on the prediction
accuracy of the spot price by comparing the proposed chaos-ANFIS model with
the baseline ANFIS model (non-chaotic approach). Evaluation results show that
the proposed chaos-ANFIS model yields better predictions of spot price compared
to the baseline ANFIS model in terms of Root Mean Square Error (RMSE) and
Mean Absolute Percentage Error (MAPE).
Keywords: dynamic pricing; cloud computing; spot instance; spot price; chaotic time series analysis; ANFIS.
Stock Market Manipulation Detection using Feature Modelling with Hybrid Recurrent Neural Networks
by Sashank Sridhar, Siddartha Mootha
Abstract: A stock market is a potent platform which handles a large of number of transactions within a second. Keeping track of every single transaction is a daunting task for regulatory bodies. The objective of a regulatory body is to ensure a fair trading environment and to verify that the price of a stock is not being manipulated. This paper proposes a hybrid stacked artificial neural network and recurrent neural network to model the static and dynamic features of stock data. Based on the manipulated stocks, affidavits provided by the Securities and Exchange Board of India (SEBI), a daily trading dataset is created by scraping the Bombay Stock Exchange (BSE) website. The system is capable of identifying three types of manipulation scenarios. The proposed hybrid system is compared to various supervised algorithms, and various ensemble models and the system outperforms all with an accuracy of 96.06%.
Keywords: Manipulation Detection; Hybrid Neural Networks; Ensemble Learning; Recurrent Neural Networks; Fraud Detection; Long Short Term Memory; Bidirectional Long Short Term Memory; Stacked Generalization; Artificial Neural Networks; Feature Engineering;.
Special Issue on: The Impact and Importance of Networking and Virtualisation in a Post-COVID-19 Scenario
Analysis of low-cost electronic device for diagnosis of Covid-19
by Akshaya Nidhi Bhati, Himanshu Maharshi, Arun Kumar
Abstract: This paper presents the design of a low-cost electronic device that can be used to diagnose Corona Virus Disease 2019 (COVID-19) at home with the help of symptoms. The device will check whether the patient has a fever or not with the help of a thermal sensor, oxygen saturation in the blood (Spo2) with the help of a Pulse-Oximeter, and cough through artificial intelligence. The remaining symptoms will be diagnosed using a survey-based system, where respondents will be invited to self-report various symptoms. The estimate, conception and development of this device can greatly contribute to the creation, and assist in breaking off the spread of the disease, getting the timely treatments and potentially save lives.
Keywords: COVID-19; Oximeter; OLED Screen; Thermal Sensor; Spo2.
PlasmaBlock: A Plasma donation Blockchain system in COVID-19
by Riya Sapra, Parneeta Dhaliwal
Abstract: COVID-19 has brought the whole world to a still. The extensive spread of the disease has adversely affected the life of the common man and the nation as a whole. Various vaccine trials for the disease are being done by scientists and researchers in different countries. Plasma therapy is one of the treatments, medical professionals are using for treating critical patients. This treatment requires a valid plasma donor who can donate plasma after its recovery from COVID-19. In this paper, a decentralized platform for plasma donation registration and plasma matching has been proposed using blockchain technology to speed up the process of convalescent plasma therapy. It uses smart contract to find appropriate plasma matches for the critical patients to locate the plasma of required blood group. It will help in tracking the donations and the procedures afterward.
Keywords: Convalescent plasma; Plasma therapy; Blockchain; donor-recipient; plasma donors; smart contract; corona virus; COVID-19.
COVID 19 and its Impact on Global Virtual Teams: Exploring the Unexplored
by Archana Shrivastava, Pooja Misra
Abstract: Based on our understanding, most of the research related to the outbreak of pandemic focus on health sector and economy. Our research article examines how internationally disruptive events like COVID 19 pandemic influence global virtual teams, particularly those in which team members have never met in person. Research also explored participants perception of the higher education institutions towards online learning and attitude of corporate organisations towards remote working in the coming years. Results of the study are promising in throwing light on the insight created via narratives.
Keywords: COVID19; corona virus; virtual teams; higher education; online learning; remote working.
An IoT and Artificial Intelligence based Patient Care System focused on COVID-19 Pandemic
by Vishal Kumar Goar, Nagendra Singh Yadav, Chiranji Lal Chowdhary, Kumaresan P, Mohit Mittal
Abstract: World Health Organization has declared COVID-19 a pandemic. The spread rate of COVID-19 outbreak is much faster over past outbreaks of the Ebola and SARS-COV. Deaths and illnesses started to increase exponentially, and nations around the globe struggled to control the spread of the COVID-19 virus. Like many other epidemic outbreaks, COVID-19 faces significant challenges, like the identification of the epidemic's source of disease, control the spread rate, and adequate healthcare for all patients. Digital technology, particularly the IoT, could be used as an essential tool to combat and control the spread of these pandemics to minimize the economic loss and disruption. The digital technology allowed health care professionals in identification and isolation to the source of the infection to prevent community transmission of the virus by remotely monitor the COVID-19 infected patients. We proposed a machine learning prediction model using the Orange Canvas Program by creating a local instance dataset of eight suspected individuals measured body parameters. The body parameter values are extracted by the caretaker in a clinical system. Furthermore, six machine learning classifiers such as KNN, DT, SVM, Random Forest, Neural Network and Na
Keywords: Internet of Things; Artificial Intelligence; COVID-19; Pandemic; Wearablernsensors; Cloud interface; Machine learning.
AHP-based evaluation of Critical Barriers for Social Distancing in India during COVID-19
by Hemant Upadhyay, Abhinav Juneja, Sapna Juneja, Deepak Gupta
Abstract: COVID-19 has made the world realize, the uncertainty of our existence on the planet. There have been unprecedented consequences of the spread of this deadly pandemic. New methods and theories are being proposed by research community to save the humans from being affected by this disease. The aim of the presented research work is to analyse the critical barriers in social distancing in India during COVID-19 using AHP technique. Experts opinions are utilized to identify critical barriers in self-distancing in India during COVID-19. Twelve critical barriers have been compared and evaluated using Analytical Hierarchical Process Method to get ranked in terms of priorities. The results may be utilized by the policy manufacturers for nurturing adequate reforms and policies to effectively deal with current pandemic considering the relative importance of these critical barriers in social distancing management.
Keywords: COVID-19; Analytic Hierarchy Process; AHP; Critical Barriers; Social Distancing.
Students' Perspective on Online teaching in higher institutions during COVID 19
by Roseline Oluwaseun Ogundokun, Muhammad Daniyal, Sanjay Misra, Joseph Bamidele Awotunde
Abstract: Learning and teaching online is not a modern concept. It has been seen primarily as a part of face-to-face teaching and learning purposes for the past few years. Evaluation is an important aspect of teaching and learning since it determines the students' accomplishment of course learning outcomes. Today due to the coronavirus outbreak, there's a peculiar issue facing the global education community. The government and educational policymakers have to find other ways of teaching and learning to help learners during this isolation period at all locations; thus, learners need to be literate and be able to use online technologies effectively and consider ever-changing dynamics. Therefore, the study aims to explore whether students of the developing countries are contented with the innovation of online education introduced by all higher establishments worldwide as a result of the new coronavirus called COVID-19. This study evaluates the effect of online teaching during COVID-19 lockdown on students in the developing country like Nigeria. An online survey was conducted between 1st October and 8th October 2020 to achieve this evaluation. A 'Google form' structural questionnaire was forwarded to scholars via WhatsApp and e-mail. A total of 1419 students submitted complete survey information. These are evaluated to discover their thoughts on online education during the COVID-19 outbreak in a developing country as Nigeria. The study findings highlighted that an overwhelmingly greater number of the scholars are incapable of connecting to the net owing to technological and fiscal problems, thereby made online learning not to be able to yield desired results. In a group of few additional concerns emphasized by university scholars is the absence of one-on-one contact with the teacher, reaction period and nonappearance of typical teaching space socialization.
Keywords: COVID-19; Online teaching and learning; Higher institution; Internet; Developing countries.
Stock Movement Prediction Using Neuro Genetic Hybrid Approach and Impact on Growth Trend due to COVID-19
by Pradeepta Kumar Sarangi, Kalpna Guleria, Devendra Prasad, Deepak Kumar Verma
Abstract: Knowing the future perspective is a matter of great concern for every business organization. Its importance increases when it comes to the matter of financial data related to the stock market. Researchers apply various methods to predict the stock trend but still there is no method that can guarantee the accurate stock movement whereas a nearly accurate approach could be achieved in case of short-term stock movements. Artificial Neural Network (ANN) is one of the most popular methods to analyze the stock trend and forecast the future direction of the stock market and the most challenging task in using the neural network is the selection of a proper architecture, input parameters and training of the network. This can be overcome by analyzing the relationship between independent and dependent factors and also by implementing hybrid models such as ANN with Genetic Algorithm (GA) for network training. This work has addressed this issue for short term stock prediction and is divided into two phases. In phase-I, experiments have been done to implement a hybrid ANN-GA model for short term stock prediction and in phase-II, a study has been carried out to analyze the impact of the COVID-19 on the share prices of selective major banks in India.
Keywords: Time Series Analysis; Neural Network(NN); Genetic Algorithm(GA); Financial Forecasting; COVID impact on banking sector.