With the recent evolution of technology, the number of image archives has increased exponentially. In Content-Based Image Retrieval (CBIR), high-level visual information is represented in the form of low-level features. The semantic gap between the low-level features and the high-level image concepts is an open research problem. In this paper, we present a novel visual words integration of Scale Invariant Feature Transform (SIFT) and Speeded-Up Robust Features (SURF). The two local features representations are selected for image retrieval because SIFT is more robust to the change in scale and rotation, while SURF is robust to changes in illumination. The visual words integration of SIFT and SURF adds the robustness of both features to image retrieval. The qualitative and quantitative comparisons conducted on Corel-1000, Corel-1500, Corel-2000, Oliva and Torralba and Ground Truth image benchmarks demonstrate the effectiveness of the proposed visual words integration.
15 July 2021 in International Journal of Distributed Sensor Networks
Vehicular ad hoc network is a pretty research vibrant area since last decade. It has been successfully used for intelligent transportation system and entertainment purposes for realization of smart cities. However, intermittent connectivity, high routing overhead, inflexible communication infrastructure, unscalable networks, and high packet collision are the key challenges that put hindrances on the wide applications of vehicular ad hoc network. The severity of these challenges become even more intensified when deployed in urban areas. To overcome these hurdles, integrating micro unmanned aerial vehicles with vehicular ad hoc network provides a viable solution. In this article, we proposed an unmanned aerial vehicle–assisted vehicular ad hoc network communication architecture in which unmanned aerial vehicles fly over the deployed area and provide communication services to underlying coverage area. Unmanned aerial vehicle–assisted vehicular ad hoc network avails the advantages of line-of-sight communication, load balancing, flexible, and cost effective deployment. The performance of the proposed model is evaluated against a case study of vehicle collision on highway. Results show that utilization of unmanned aerial vehicles ensures the guaranteed and timely delivery of emergency messages to nearby vehicles so that a safe action can be taken to avoid further damages.
SDN and NFV collaboratively recognized as the most promising bearing for flexible programmability of network control functions and protocols with dynamic usage of network resources. SDN provides abstraction of network resources over well-defined APIs to achieve underlying topology-independent multiple tenant networks with required QoS and SLAs. NFV paradigm deploys network functions as software instances namely VNFs on commodity hardware using virtualization techniques. This way, virtual IP functions such as load balancing, routing and forwarding or firewall can operate as VNF in cloud with positive outcome in network performance. In this paper we aimed to achieve traffic load balancing by using virtual SDN controller (vSDN) as a VNF. With vSDN, when there is uneven and increased load, secondary vSDN controllers can be added to share this load. Need of secondary vSDN is determined and a copy vSDN with exactly same configurations as original vSDN is created which operates accurately and shares traffic load balancing tasks with original vSDN controller. Both vSDN controllers independently placed in cloud with transparency assuring that every client in network is familiar with the existence of the newly created secondary vSDN controller. We experimentally validated the load balancing in Fat-Tree topology using two vSDN controllers in Mininet emulator. Results showed 50% improvement in Average Load, 41% improvement in Average Delay and considerable improvements in terms of Ping Response, Bandwidth Utilization and Throughput of the system.
22 January 2024 in IEEE Transactions on Consumer Electronics
Wireless sensor networks (WSNs) play a major role in increasing the pervasiveness of Internet of Thing (IoT) with smart sensors for Consumer electronics, which are low cost and easy to install. In WSN, data is collected from sensors and transmitted to the sink for further operation which is used by IoT applications. However, resource restrictions on sensors and sensitivity of radio links in these networks causes serious routing issues and affects application performance. Routing protocols are designed to provide reliable route and effective communication between sensors while preserving energy efficiency. Numerous routing solutions that have been suggested in this context are associated with significant energy consumption, latency, and poor packet delivery ratios. Considering these problems, this work introduces the connection-Quality based Energy-Efficient routing (LQEER) protocol, which increases connection dependability through link quality estimation. Additionally, LQEER uses energy and link information in planning routes and a cost to select most appropriate for nodes for packet transmission which ultimately reduce packet loss. As a result, it improves packet delivery ratio and reduces delay. The result obtained from simulation show that LQEER gains superior performance in terms of packet delivery ratio, energy consumption, and end-to-end delay against state-of-the-art protocols.
An intrusion detection system (IDS) is designed to detect and analyze network traffic for suspicious activity. Several methods have been introduced in the literature for IDSs; however, due to a large amount of data, these models have failed to achieve high accuracy. A statistical approach is proposed in this research due to the unsatisfactory results of traditional intrusion detection methods. The features are extracted and selected using a multilayer convolutional neural network, and a softmax classifier is employed to classify the network intrusions. To perform further analysis, a multilayer deep neural network is also applied to classify network intrusions. Furthermore, the experiments are performed using two commonly used benchmark intrusion detection datasets: NSL-KDD and KDDCUP'99. The performance of the proposed model is evaluated using four performance metrics: accuracy, recall, F1-score, and precision. The experimental results show that the proposed approach achieved better accuracy (99%) compared with other IDSs.