Earticle

Distributed Hash Tables (DHT) provide a fault-tolerant and scalable means to store data blocks in P2P systems. In this thesis, we have proposed an improved version of CAN called “small-world DHT”. The key idea is to link each CAN node to a constant number of long-distance contacts, which are chosen according to the probability density function. This enables our scheme to achieve O(log2n) routing path length with O(1) routing table size per node. We have shown that the hybrid infrastructure of a structured overlay network (CAN) and a random graph (small-world model) not only preserve CAN’s simplicity, but also achieves a resilient and efficient DHT routing algorithm. We have also exploited the LookAhead-GREEDY routing algorithm, whereby each node obtains information about its neighbors’ neighbors by periodically exchanging the routing table entries with its neighbors. The routing algorithm improves the routing efficiency and enhances the network’s resilience to failure. Our scheme requires knowledge of the current network size in order to construct the long-distance links. To estimate the network size, the latter maintains a distributed binary partition tree and measures the size of the sample area by counting the number of leaves in the corresponding branch.

Cloud computing has attracted users due to high speed and bandwidth of the internet. The e-commerce systems are best utilizing the cloud computing. The cloud can be accessed by a password and username and is completely dependent upon the internet. The threats to confidentiality, integrity, authentication and other vulnerabilities that are associated with the internet are also associated with cloud. The internet and cloud can be secured from threats by ensuring proper security and authorization. The channel between user and cloud server must be secured with a proper authorization mechanism. The research has been carried out and different models have been proposed by the authors to ensure the security of clouds. In this paper, we have critically analyzed the already published literature on the security and authorization of the internet and cloud.

In order to solve this problem of cloud model, this paper presents another new collaborative filtering recommendation algorithm by combining the item classification and cloud model. Firstly the algorithm utilizes the item classification information and cloud model to compute items inner-similarity, and then gets the scores from neighbor items which have the highest similarity and uses their scores to forecast the unrated inner-class items. Secondly, the neighbors of user are obtained by computing the inner-class user similarities in the cloud model, providing the final forecast grade and carrying out the recommendation.

Recent advancement in Internet Technologies has made web browsing increasingly easy and user friendly. From the traditional method of desktop web browsing and the birth of dial up modem connection, users nowadays are able to enjoy a fast and reliable web browsing via high speed wireless Internet connection and portable mobile devices. Browsing a web has become much easier with the state of the art search engines such as Google, which provide much functionalities which could make browsing easier such as improved crawler, easy to use search interface, web personlization, Web 3.0 support and integration and many more. In order to build a robust and reliable search engine, the developer needs to integrate all the data and present them in a meaningful format for user’s viewing convenience. Integrating these data is a tedious task as data usually occur in numerous format, and layout. Furthermore, web developers usually present the data content in various languages of their choice, which made the processing of these data increasingly difficult. There is also no standard convention to represent the data format and even a standardize rule to process this data has not been developed. To resolve this issue, researchers develop data extractor which could effectively extract data from web sources, tabulate them, and used it for further processing. However, not all data are correctly extracted, they may either missed certain valuable information or contain additional unnecessary information. In the case of unnecessary information, researchers use a cleaning method to remove them such that the data extracted are free of errors. Removing these data is important as unnecessary information may affect the accuracy of subsequent extractor tools, hence may eventually prevent the tool from performing its task efficiently. In this research proposal, we embark on a data cleaning tool to clean data using ontology tools. Experimental results show that our tool is highly efficient in data cleaning and is able to outperform existing state of the art tools.

In order to solve the problem of transient oscillation of microgrid during process switching between grid-connected operation and islanded operation, the basic principle of traditional control strategy is analyzed, a novel control structure of smooth switchover for microgrid based on the double negative feedback controller is proposed. Also, to avoid the risk of voltage oscillation caused by the mutual interference of the two controllers in the traditional quasi-synchronization method, an improved quasi-synchronization controller is proposed, which is simpler and easier to implement. Finally, the simulation results verify the validity of the proposed structure. Research results show that the busbar voltage and current oscillation can be solved better with the novel structure, and lay the foundation for smooth switchover.

In this paper, achievable rate region of two-user interference channel is studied. We use the Pareto-optimal method (e.g., specially in production and consumption) into the wireless sensor nodes transmit power allocation strategies in coupled channels. The result shows that when systems in optimality conditions, minim transmit power can arrive maximum channel throughout. What’s more, achievable rate of each channel equals and it is only related the channel parameters at the moment.

The energy aware based active node decision and the sensing directions thereof become the research hotspots of DNN (Directional Sensor Network). Therefore, an EACO (Energy aware-based coverage optimal) algorithm is proposed in this paper. Specifically, each cluster head decides the number of the active nodes of the cluster members and the sensing directions thereof according to the coverage area of the sensor nodes and the energy information thereof. Meanwhile, the dormancy mechanism is adopted to minimize the overlapping area among the clusters. The simulation result shows that the proposed EACO algorithm is improved in the aspects of coverage rate and network lifetime.

Due to the normal forecasting methods for subgrade settlement using observation data have different applicabilities, and the predicting results has bigger volatility and lower accuracy. In view of the above problems, a method based on least square support vector regression (LSTSVR) and real-coded quantum evolutionary algorithm (RQEA) is proposed. Firstly, the LSTSVR parameter is chosen as a combinatorial optimization problem, and determining the objective function of the combinatorial optimization problem, then, using RQEA to solve the combinatorial optimization problem and optimize the LSTSVR parameters, Finaly, LSTSVR-RQEA is used to sovle the prediction of subgrade settlement. The simulation results show that RQEA is an effective method to select LSTSVR parameters, and has excellent performance when applied to the prediction of subgrade settlement.

Cloud computing is the technology that moves the information technology (IT) services out of the office. Unfortunately, Cloud computing has faced some challenges. The task scheduling problem is considered one of the main challenges because a good mapping between the available resources and the users' tasks is needed to reduce the execution time of the users’ tasks (i.e., reduce make-span), and increase resource utilization. The objective of this paper is to introduce and implement an enhanced task scheduling algorithm to assign the users' tasks to multiple computing resources. The aim of the proposed algorithm is to reduce the execution time, and cost, as well as, increase resource utilization. The proposed algorithm is considered an amalgamation of the Particle Swarm Optimization (PSO),the Best–Fit (BF), and Tabu-Search (TS) algorithms; called BFPSOTS. According to the proposed BFPSOTS algorithm, the BF algorithm has been used to generate the initial population of the standard PSO algorithm instead of to be random. The Tabu-Search (TS) algorithm has been used to improve the local research by avoiding the trap of the local optimality which could be occurred using the standard PSO algorithm. The proposed hybrid algorithm (i.e., BFPSOTS) has been implemented using Cloudsim. A comparative study has been done to evaluate the performance of the proposed algorithm relative to the standard PSO algorithm using five problems with different number of independent task, and Virtual Machines (VMs). The performance parameters which have been considered are the execution time (Makspan), cost, and resources utilization. The implementation results prove that the proposed hybrid algorithm (i.e., BFPSOTS) outperforms the standard PSO algorithm..

Reducing energy consumption has also become an important way to reduce expenditures and increase revenue for ISP. For solving the problem of dynamically changing network-flow of virtual network in runtime, we first analyze relations and differences between virtual network and IP network on energy-saving issues. We secondly model the makeup of energy consumption in virtual network, and determine the object of energy-saving in virtual network at. Finally we propose the solutions to this energy-saving problem. The solutions can be described as follows: (1) based on the historical data of daily network-flow, we divide one day into two time slices, and one is called idle time slice, the other is called busy time slice; (2) we design the dynamic switch-off algorithm and run it at the beginning of idle time slice, and we use former algorithms to be started at the beginning of busy time slice; (3) in the dynamic switch-off algorithm we determine the method to select virtual networks and the method to remap; (4) we apply roulette wheel to select virtual networks, in the selection method we consider both current influences on energy-saving and influences in the future. Simulation results show that the dynamic switch-off algorithm can effectively select virtual networks, and the whole solution performs well in energy aware virtual network in runtime.

Tracking the maximum power point is an appropriate solution against climatic changes like the irradiance and temperature. The classic incremental conductance algorithm is one of the most widely used methods in commercial photovoltaic MPPT’s (Maximum Power Point Tracking). A wide variation of climatic condition can degrade the produced energy. In this work we present our new approach based on fuzzy logic which can guarantee better performance. The studied system used to prove the efficiency of the proposed method is composed by a photovoltaic panel Kaneka K60 connected to a Sepic Converter which fed a resistive load. The All results quoted in this work are found using MATLAB-Simulink. We put the system in different working condition to highlight our new approach. All presented results are discussed.

Cloud computing is a new technology, instead of all computer hardware and software that used on desktop, or somewhere within company's network, it's presented as a service by cloud service providers and accessed via the Internet. Exactly where hardware and software are located and how everything works does not matter. In cloud computing there are many jobs that requires to be executed on the available resources to achieve best minimal execution time. Several optimization methods are available for cloud job scheduling. However, the job scheduling process is still need to be optimized. This paper proposes a new job scheduling mechanism using Firefly Algorithm to minimize the execution time of jobs. The proposed mechanism based on information of jobs and resources such as length of job speed of resource and identifiers. The scheduling function in the proposed job scheduling mechanism firstly creates a set of jobs and resources to generate the population by assigning the jobs to resources randomly and evaluates the population using a fitness value which represents the execution time of jobs. Secondly the function used iterations to regenerate populations based on firefly behavior to produce the best job schedule that gives the minimum execution time of jobs. Several scenarios are implemented using Java Language and CloudSim simulator. Different settings have been considered in the evaluation and experimentation phase to examine the proposed mechanism in different workloads. The first phase of the evaluation process describes how the proposed mechanism can be used to minimize the execution time of jobs. The second phase of the evaluation process compares the proposed mechanism with First Come First Serves (FCFS) algorithm. The results revealed that the proposed mechanism minimizes the execution time significantly. Furthermore, the proposed mechanism outperformed the FCFS algorithm.

Social network has become the important platform for the current social individual to exchange information and access various media. Traditional searching and locating method is faced with the characteristics of the high order correlation and implicit correlation for large-scale users, and user-association multidimensional and heterogeneity. In order to effectively deal with large-scale social network data, and improve the efficiency of user’s locating, this article introduces the peer-to-peer distributed searching mechanism with the help of the cloud computing platform. This searching mechanism assigns user a logical identifier, and matches the underlying physical address and the upper users’ logical address, so as to build the cloud logical topology structure of social network. This paper designs a K neighbor discovery algorithm. It is used to cluster the nodes according to the features of the user, so as to realize the quickly locating of social network. The performance of the algorithm is analyzed according to the user’s searching logic path length of social network, and information amount of routing state. The simulation of the algorithm is evaluated by maintenance costs of average network aggregation coverage and query time. The performance analysis and simulation results demonstrate that social network cloud has good performance and searching efficiency.

The paper Proposed Item parallel collaborative filtering recommendation algorithm (IP-CF). Through designing efficient parallel algorithm, compute-extensive procedures are distributed to different processing nodes in Hadoop platform. Taking advantage of parallel computing, we accelerate the response of recommendation. The experimental results show that our proposed algorithm IP-CF is more efficient and scalable than current parallel algorithms.

A network which consists of miniaturized wireless sensor nodes with limited memory capacity and energy source is wireless sensor network. Power consumption is very important in wireless sensor network (WSN) because replacement or recharging of sensor nodes is difficult. There are many routing protocols in wireless sensor network. In this paper, we are considering three routing protocols for different traffic types: Convergecast with WiseRoute, Flooding and ProbabilisticBroadcast. A relative examination of three routing protocols is done by series of simulations in OMNET++. This paper focuses on identifying the best routing protocol on the basis of performance parameters backoff duration, latency, the mean number of hops, the number of frames dropped with interference and number of packets received using uniform, periodic and exponential application traffic situations.

An improved ant colony algorithm based on the three constraint conditions that aiming at the problem of resource scheduling in cloud computing is proposed in this paper. this method is divided into three steps: Firstly, we describe the state transition probability by using the information heuristic factor and the expected heuristic factor. Secondly, the pheromone update strategy is used to design the scheduling process. Finally, the optimal path is based on the expected execution time, network delay and network bandwidth of three constraints. Experimental results show that the proposed method has faster execution speed than the traditional ant colony algorithm, and the load balancing of the results is more satisfactory.

Cloud stroke is the cloud platform according to the medical association theory and combing cloud computing technology, build for the hospital systems which are involved in the medical association to sharing and exchange medical information. However, at this stage, there are some problems in the data exchange of the cloud platform: Firstly, every hospital in the medical association needs to configure a front machine and the front machine needs to connect with the cloud platform database directly. So there is a hidden risk for the cloud stroke. Secondly, the data synchronization vector is Excel file, the data reading efficiency is not very well, and at the same time, the front machine memory consumption is relatively very large. This paper, improve the data interactive process of cloud stroke, so that the front machine unneeded to connect with the cloud platform database directly. Then, design data specification and synchronization process based on JSON for cloud stoke information synchronization to compress the expenses, improve the speed and analysis the security of data synchronization.

A novel nature inspired algorithm ant lion optimizer (ALO) is recently developed which is motivated from the hunting mechanism of ant lions .Inherit steps of hunting prey such as the random walk of ants, building traps, entrapment of ants in traps, catching preys, and re-building are simulated to find the optimal solution of real life problems . Intelligent and Optimization techniques based on evolutionary computing, metaheuristic,biological base,nature inspired, search method establish their applications in the area of electrical economic power dispatch planning(EEPDP) to reach global optimal solution for this multi scale, multi-decision, multi-objective combinatorial problem subjected to different constraints.An application of ALO to solve non linear elec-tric economic power dispatch problem(EEPDP) is proposed in this paper. Efficient and optimal planning of economic electrical power dispatch problem is an integral part of economic electrical energy generation planning and it is the need of time for the electrical engineers to browse this area in multi-scale planning scenarios.. The performance of s ant lion optimizer (ALO) to solve electrical economic power dispatch problem is tested on three and six unit system.Test results are compared with other techniques grey wolf optimization(GWO),cuckoo search(CS),artificial bee colony(ABC),firefly algorithm(FA),particle swarm optimization(PSO),shuffled frog leap (SFL) ,bacteria foraging algorithm(BFO),harmony search(HS) applied in literature. Simulation results proved that the ALO technique is better as compared to other nature inspired,heuristic,metaheuristic techniques to find global minima and maintain the solution quality in terms of low fuel cost.

With the development of Internet of things, especially the development of mobile Internet, electric vehicle charging device and location data and power allocation can reach the cloud through mobile Internet, providing reliable basis for public decision-making and public services. However, the provided real-time performance and cost is the key problem of the implementation of the vehicle area network cloud computing. In this paper, aiming at this problem, proposing the mobile Internet scheduling and transmission strategy based on cloud computing, the simulation results show the effectiveness of the proposed method.

Logistics from the user to pass character customization items, it has a certain degree of randomness and comprehensive, and the source node stream as nodes of the network, but how logistics resource scheduling can be more economical to transport goods, and because information technology, especially the development of things, dynamic and timely logistics scheduling becomes possible, for this problem, we integrated logistics dynamic scheduling features, applications random graph theory, based on a complex system of logistics scheduling scheme, simulation results show that the algorithm The advantages.

Target area coverage is the most fundamental goal of WSN (Wireless Sensor Network) application, so the design of energy aware based algorithm for coverage time maximization becomes a core problem of a large network. For this reason, a DECAR (Distributed-energy and Coverage Aware Routing) protocol is proposed in this paper to realize the goal of network coverage maximization. Specifically, the rest energy and the coverage overlapping degree of the nodes are considered during CH (cluster head) selection process so as to provide more opportunities for the nodes with more rest energy or higher coverage overlapping degree to become CH; like this, the nodes with less rest energy will not be selected as CH, thus to avoid shortening the network lifetime due to early failure of these nodes and balance the network energy consumption. Meanwhile, the main data transmission line composed of CHs is constructed in the data transmission stage in order to improve data transmission efficiency. The simulation result shows: compared with CPCP-ea and EEUC protocols, the proposed DECAR protocol can obtain longer network lifetime and better data coverage ratio.

Clustering is one of the effective methods which are applied in the wireless sensor networks to preserve the precious battery power of sensor nodes. Various clustering approaches have been proposed in the literature and LEACH is one of clustering and routing schemes which can be used to manage the sensor network. In this paper, we present an improved version of LEACH protocol which uses Kmeans clustering algorithm for selecting better nodes for clusterhead role. Our proposed solution tries to create symmetric clusters and by reducing the average intra-cluster communications distance, improves the network lifetime. It also, considers the sink location when the sink node is positioned far away and selects nodes as clusterhead which will consume less energy for handling cluster members. Thus, because, sensed data are transmitted to the closer clusterheads, our proposed clustering solution can prolong the network lifetime.

Micro-grid is the significant part of the distribution network in the future of smart grid, which has advanced and flexible operation and control pattern, and integrates distributed clean energy. It is a study of the direction of the grid in the future. In the paper, it gives an account of the operation and control of micro-grid. First, the concept of micro-grid, the structure and characteristics of the typical micro-grid and the significance of the development of micro-grid are analyzed. Second, it introduces typical control strategy which contains the integrated control strategy of Micro-grid and a typical control mode of Micro-power. At last, based on a comprehensive control strategy, it proposed the operation mode of Micro-grid.

Focusing on Internet’s emerging paradigm of ubiquitous content sharing, following the Broadcast-Storage concept, we study the key technologies of base station server content delivery based on broadcast-storage architecture, introduce a general architecture model of broadcast-storage network, propose a technological framework of base station server content delivery based on broadcast-storage architecture, analyze the key technologies and application algorithms included in the foundation technology layer and the key technology layer. In the key technology layer, we mainly research on Satellite-Base Station data processing technology, base station content service platform, content security management-control, topic security management-control, and user management. Finally, we summarize the nonfunctional requirements of base station service platform. Based on the above research, a base station server content delivery system based on broadcast-storage architecture is designed and implemented, which is docked with the integrated broadcast control platform, integrated management-control platform and clients, completing application demonstration in important typical business. Therefore, a complete broadcast control content delivery system has been initially established, laying solid foundation for further researches.

Vehicular ad-hoc networks (VANETs) technology has come out as an important research field over the last few years. VANETs are the likely an influencing approach to provide safety of driver and other applications for the traffic conditions as well as passengers. Being dynamic in nature, it establishes the network, according to the situation and need of the users and provides reliable communication among the vehicles. Due to its great benefits, it is highly vulnerable to various attacks and security in VANET should be taken into consideration. This paper presented the security attacks between vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I). Many research works have been done to improve the performance and security of this network. The main aim of this paper is the security using hashing and techniques to calculate the trust in VANETs.