Mobile ad hoc network is a self organized & dynamic multi hop wireless network comprised of mobile nodes that communicate with each other without any fixed infrastructure. Different solution proposed for scaling down the large size of networks, one of the most intrinsic approaches is clustering techniques which evolved an important role in mobile ad hoc network (MANETs). Cluster based routing protocols are used to reduce the routing overload and improves the system performance in the transmission delay and throughput. In this paper we have presented a brief review of state of the art scenario of routing protocols and present a new weight based hierarchical clustering algorithm for MANETs that leads to cluster formation. In our proposed approach the cluster head selection technique is based on combined weight metric that takes into account like the node degree, transmission range and mobility of the nodes. This new algorithm takes into consideration to reduce routing overhead and traverse preferred distance through selected route
22 March 2016 in Brazilian Journal of Science and Technology
As one of the most sought-after technologies in the world today, cloud computing finds a prominent place among researchers, service providers, hardware and software engineers, and consumers alike. Among the key services provided by cloud computing, the most widely used is storage on the cloud. To ensure a high degree of reliability and an almost constant state of availability, cloud storage systems rely on replication of data. Replicating data is not without its costs. In an attempt to mitigate costs of storage, designers have banked on erasure coding practices in lieu of data replication. This has resulted in development of Cloud RAID systems. An ideal Cloud RAID system strikes an optimum balance between performance, storage, and reliability. In addition to this, data security is of paramount interest. In this paper, we introduce an entirely new technique of storing data in cloud networks by file type classification. Combining RAID types with this, our paper proposes a model that is robust and can withstand instances of hard disk failures. Further, we discuss a methodology that is embedded in our proposed framework that supports user authentication for data storage and retrieval. Our Secure User Authenticated Cloud RAID model shields user data from prying eyes and even from unauthorized attempts by the cloud service provider our Secure User Authenticated Cloud RAID architecture.
Sensors are essential to collect information from the environment. Smart sensing technology can use artificial intelligence (AI) and machine learning (ML) techniques to enhance the process of information acquisition. AI and ML techniques in smart sensors can improve the process of collection of data and reduce the exchange of data within the networks significantly. The attributes of remote sensor systems bring numerous difficulties. Moreover, the number of interconnected gadgets generate huge amounts of data. These require smart processing and examination to build up a more significant level of information and outcomes in better dynamic, estimating, and solid administration of sensors. So, in this article, a comprehensive review of AI- and ML-based applications has been done to enhance the utility of smart sensor network in different aspects, and discussion about feature selection and dimension reduction with AI and ML tools is also depicted.
1 April 2020 in International Journal of Cloud Applications and Computing
Cloud load balancing has become one of the most vital aspects of Cloud computing that has captured the attention of IT organizations and business firms in recent years. Among the issues related to this particular aspect, one such issue which needs to be addressed is the issue of effectively serving the clients' requests among multiple servers using an appropriate load balancer. Previous survey papers discussed various issues of cloud load balancing and accordingly devised various methods and techniques to address those issues with the objectives of reduction of processing time and response time along with optimization of costs. In this article, we have discussed an effective load balancing technique using the weighted Round-Robin algorithm which can process the client requests among multiple servers with minimal response time. Considering all these aspects, a cloud-based dynamic load balancer is being used to solve the problem of load balancing in the cloud infrastructure.