International Journal of Computer Networks and Applications

151 - 187

Authors: Pallavi Joshi, Ajay Singh Raghuvanshi

Since the last decade, wireless sensor network (WSN) and Internet of Things (IoT) has proved itself a versatile technology in many real-time applications. The scalability, cost-effectiveness, and self-configuring nature of WSN make it the fittest technology for many network designs and scenarios. The traditional WSN algorithms are programmed for fixed parameters without any touch of Artificial Intelligence as well as the optimization technique. So, they suffer from a trade-off between various QoS parameters like network lifetime, energy efficiency, and others. To conquer the limitations of traditional WSN algorithms, machine learning has been introduced in wireless technology. But machine learning approaches also cannot solve all the problems in WSN solely. Some of the applications like target tracking, congestion control, and many more, do not give desired results even after applying the machine learning techniques. So, there is a need to introduce optimization in such cases. The paper gives an extensive survey on various optimization methods employed to solve many WSN issues from 2005 till 2020. It also gives a brief description of the usage of various machine learning techniques in WSNs from 2002 till 2020. The paper discusses the advantages, limitations, effects of these methods on various WSN techniques like topology, coverage, localization, network and node connectivity, routing, clustering, cluster head selection, cross-layer issues, intrusion detection, etc. This paper gives a lucid comparison of many state-of-the-art optimization algorithms and descriptive and statistical analysis for discussed issues and algorithms associated with them. It also elucidates some open issues for WSNs/IoT networks that can be solved using these approaches.

203 - 213

Authors: Hussein Khayou, Margarita A. Orlova, Leonid I. Abrosimov

Requirements in data centers to meet the increasing demands on traffic have evolved. There is a need for a simple, scalable routing protocol, which has the flexibility and ease of management to support large networks. Distance vector routing protocols are very simple and easy to implement but they suffer from routing loops. Link state protocols, on the other hand, have the advantages of fast convergence, area division of the routing domain, at the expense of the added complexity of implementation, configuration, and troubleshooting. A new loop free protocol is proposed in this paper that combines the simplicity of the distance vector protocols, loop freedom, and the ability to be used in large scale mesh networks as in link state protocols. The protocol uses a hybrid distance vector link state algorithm. It employs techniques from Enhanced Interior Gateway Routing Protocol (EIGRP), Babel, and Open Shortest Path First Protocol (OSPF). Simplicity, ease of implementation, and scalability make the proposed solution appropriate for large scale networks. Additionally, it can be used to perform the underlay routing in SDN (Software Defined Networks) overlay networks in place of IS-IS (Intermediate-System to Intermediate-System) protocol, which is usually used in these solutions. The combination of distance vector and link state helps to reduce the size of information in the database that is needed to be maintained by each node. It also helps to reduce the overhead and computing load after topology changes.

228 - 237

Authors: S. A. Gunasekaran, M. Senthilkumar

Minimizing energy consumption is a significant issue in cloud computing. Nodes present in cloud computing are heterogeneous in nature. Traditional routing protocols fit best for homogenous networks and while using in heterogeneous network it will never give its better performance. Accessing big data in cloud is a challenging task because more stable route is necessary for the access of big data. Routes failures are unexpected and if a route gets failed in cloud computing while accessing big data, then it will affect the network performance drastically. In this paper, Lucid Firefly based Routing Protocol (LFRP) is proposed to identify the optimized route to access the big data and to minimize the energy consumption. LFRP utilizes the natural characteristics of firefly to identify the best route and to share the identified best route with others. LFRP finds the route based on the size of data where the fitness function plays a major role in identifying the best route. The simulation results make an indication that the proposed routing protocol LFRP has consumed less amount of energy i.e., 3.95J in accessing the big data than other routing protocols which makes an indication that the routing protocol has found the better route to destination which faces low delay (65ms) and packet delivery ratio as 94.20%.

258 - 267

Authors: M. Lingaraj, T. N. Sugumar, C. Stanly Felix , J. Ramkumar

Mobility Enabled Wireless Sensor Network (MEWSN) plays a significant role in different fields including environmental control, traffic control and healthcare. The performance of MEWSN is dependent not only on sensing but also on routing. Multiple research works are carried out by different researchers in the domain of routing in MEWSN, but still the performance of MEWSN gets lacked. Poor routing is the root cause for the performance degradation of MEWSN. In this paper, a new routing protocol namely Query Aware Routing Protocol (QARP) is proposed to balance the load in MEWSN to prevent congestion and exhausted power utilization. Normal routing protocols either seek to match load or route, but both are considered in QARP. Also, identified routes are classified based on an enhanced relevant vector machine classification algorithm which assists in minimizing the delay and energy consumption. Using NS2, QARP is evaluated against previous routing protocols with standard performance metrics namely throughput, delay, packet delivery ratio and energy consumption. The packet delivery ratio achieved by QARP is 92.6%, where the existing routing protocols IFLIP and PARP has achieved 62.8% and 75.4% respectively.

90 - 107

Authors: S. Sumithra, R. Vadivel

Vehicular Ad-Hoc Network is an emerging technology, mainly developed for road safety applications, entertainment applications, and effective traffic conditions. VANET applications work based on the accurate mobility information shared among the vehicles. Sometimes attackers manipulate the mobility information shared by the adjacent vehicle or neighboring vehicle, which results in terrible consequences. To deal with the illusion-based type of attacks, researchers have proposed enormous solutions. Unfortunately, those solutions could not deal with the dynamic vehicle conditions and variable cyber malfunctions, which reduces the misbehavior detection accurateness and increases the false-positive rate. In this paper, the dynamic vehicle context is taken into account to propose a two solutions such as Miscellaneous VANET Classifiers based Misbehavior Detection Model (MVC-MDM) and Ensemble Miscellaneous VANET Classifiers based Misbehavior Detection Model (EMVC-MDM). This model is constructed based on the Mobility Data Gathering phase, Mobility Context Feature Extraction phase, Mobility Context Feature Level Fixing phase, Hampel Filter based Context Reference Building phase, Constructing Miscellaneous VANET Classifiers based Misbehavior Detection model and Ensemble Miscellaneous VANET Classifiers based Misbehavior Detection phase. Vehicle context is prepared using the data-centric features and the behavior-based features of the vehicles. The Nonparametric Hampel filter and Kalman filter are used to building the context reference model. These filters discover the temporal and spatial correlation of the uniformity in the current mobility information. Vehicle features are extracted locally according to the stability, likelihood, and performance of the vehicles' mobility information. A random forest based learning algorithm is used to train and test the classifiers. The proposed MVC-MDM and EMVC-MDM has been simulated in various context scenarios and the presence of misbehaving vehicles. NGSIM dataset has been used for extensive simulation. The results prove that the effectiveness and the reliability of the proposed MVC-MDM and EMVC-MDM are higher than the existing misbehavior detection systems.

119 - 129

Authors: Fouziah Hamza, S. Maria Celestin Vigila

The Mobile Ad-hoc Network (MANET) is a decentralized system that consists of mobile nodes. Wireless connections are used to connect these nodes. The primary issues of concern of MANETs are mobility and limited battery lifetime. Advanced techniques for improving MANET energy efficiency and extending network lifespan are critical. Clustering is one of the tried-and-true methods for increasing network lifetime by lowering and balancing energy consumption. Choosing a suitable cluster head from the cluster improves the network’s energy efficiency even further. Because of the additional workloads, the cluster heads (CHs) utilize more energy than non-cluster heads. A novel algorithm for CH selection with a Hybrid Particle Swarm Optimization-Genetic Algorithm (PSO-GA) is proposed to improve the MANET network’s energy efficiency and lifetime. The proposed method is implemented using the NS-2 platform for the analysis. The proposed model outperforms the existing OSCA, EP-MBO, GBTC, SM-WCA, CM-BCA, and FCO methods in terms of network performance. The model’s performance has achieved a low Bit Error Rate (BER) of 7% for 100 nodes with 99.38% Packet Delivery Ratio (PDR) with minimized delay in the range 2.01sec with the energy efficiency of 99.03%. The validation indicates that the Hybrid PSO-GA approach is more efficient than the other methods.

11 - 27

Authors: Karan Singh, Rajeev Gupta

In recent years, the role of telecommunications in Under Water Mobile Ad-hoc Network (UWMANET) has emerged as a significant field during disaster prevention and rescue operations. Various disaster prevention and rescue supported applications are introduced in these years for flood, tsunamis, and underwater earthquakes. While communication in UWMANET, the existing communication system has some limitations like high energy utilization, tremendous packet loss rate, and delay. Sensor nodes can be deployed for data collection from the dense underwater environment. In UWMANET, security is another critical aspect of secure data transmission. In this paper, a new UWMANET based routing protocol, i.e., E2S-AODV (Energy Efficient Secure Ad-hoc On-demand Distance Vector) is designed and tested for Under Water Disaster Response Network (UWDRN) in a controlled environment. The optimum route for data transmission is selected by Pigeons Swarm Optimization (PiSO). PiSO reduces the hop count in the chosen shortest path. Hello, messages are broadcasted to inform their neighbors that the connection to the host is active. LDW technique is used to authenticate these hello messages. For security purposes, original event message encrypted with CST (Ciphertext Stealing Technique) and qu-Vanstone ECC based public-key cryptography. To utilize energy efficiently, E2S-AODV introduced two energy concepts drains rate finder and residual energy finder. Results that are compared with existing disaster-based protocols; are pro-motive and assure an improved quality of service (QoS) achievement in terms of many multipronged metrics like energy efficiency, reliability, security, scalability, delay, and Throughput, etc. E2S-AODV achieved a 2% improvement in PDR, 5% enhancement in Throughput, 8% reduction in end-to-end delay, and 11% reduction in energy utilization compared to its near existing competent.

44 - 53

Authors: S. Boopalan, S. Jayasankari

Underwater communication is still carried out using communication cables because of the minimum development that is established in underwater wireless communications. The utilization of wires to make sure the connectivity of sensor nodes that are located at the bottom of the sea is highly expensive. Finding the best route to send the sensed data to the destination in minimum duration has become a primary challenge in underwater wireless sensor networks (UWSN). Feasible routing protocols available for general sensor networks are not feasible for UWSN because of the difficult communication medium. Existing routing protocol face the problem of consuming more energy to deliver the data packet and also due to selecting the unfit route it faces more delay. To overcome the routing challenges present in UWSN, Dolphin Swarm Inspired Protocol (DSIP) is proposed in this paper. DSIP is inspired by the swarming nature of dolphins towards finding their food. Four significant phases involved in DSIP to find the best route in UWSN are searching, calling, reception, and predation. NS3 is used to evaluate the performance of DSIP against previous routing protocols with benchmark metrics namely packet delivery ratio, end-to-end delay, node death rate, and energy consumption. Results indicate that DSIP has consumed 1.43 times less energy than other previous protocols.

67 - 78

Authors: Rekha, Rajeev Gupta

Wireless Sensor Networks (WSN) is arising as a potential computing platform in diverse zones such as weather forecasting, modern robotization, medical health care, and military systems, etc. Since the sensors are constantly gathering information from the actual world and communicate with one another through remote connections, keeping up the security and protection of WSN communication is a prerequisite. In this paper, safe confirmation and key organization scheme dependent on Elliptic Curve Cryptography (ECC) has been suggested to make sure about information/picture transmission in WSNs. The scheme proposed in this paper is protected, competent, and appropriate for providing sensor technology based IoT services and applications. The protocol provides all the security features such as mutual authentication, confidentiality, data integrity, perfect forward secrecy, fair key agreement, etc. and is secure against hello flood attack, DoS attack, man-in-middle attack, etc. Simulation software AVISPA has confirmed the safety of the protocol for the known assaults. The performance analysis ensures the superiority of the projected proposal over the existing schemes.

178 - 192

Authors: Rekha, Rajeev Gupta

Due to the advancement of wireless communication interchanges, electronic technology, and micro-electro-mechanical devices, Wireless Sensor Network (WSN) has got advanced as a promising zone of research. WSN consists of a collection of sensor nodes having a little calculative capability, limited memory, and constrained energy assets. Clusters are formed from the collection of sensor nodes whose leader node (Cluster head) can send the sensed information from hubs to the BS. To condense the power consumption and boost group longevity, the cluster head executes data accumulation. This paper discusses many algorithms based on deterministic, probabilistic, adaptive, fuzzy logic, and Multi-attribute decision-making techniques for clustering and cluster head election. Existing algorithms enhance the network lifetime and energy efficiency but fail to provide a better quality of service and security. So many issues and challenges have been laid down and it is concluded that when computational intelligence is combined with network intelligence then QoS and security both can be provided along with the network longevity and energy efficiency in homogeneous as well as a heterogeneous environment.

208 - 230

Authors: Denis R, Madhubala P

The exponential growth of communication technologies and related application environments has broadened the cloud computing ecosystem horizon to meet major communication needs. However, in-parallel upsurge in online attacks, security breaches or allied intrusion events has alarmed industries to ensure optimal data security. Unlike text data transmission, image, or other multimedia communication over the cloud requires computational efficiency, imperceptibility, etc. to meet attack-resilient transmission. Amongst the major available security systems, the combination of cryptosystems and steganography has been identified as an augmented security model for data transmission. However, it demands enhancement in both stages to meet cloud-specific communication efficiency. Considering it as motivation, in this paper an efficient Visually Imperceptible Hybrid Crypto-Steganography (VIHCS) model is developed using Hybrid Cryptosystems followed by Adaptive Genetic Algorithm assisted Least Significant Bit (LSB) embedding process. We developed a novel Hybrid Cryptosystem by strategically applying Advanced Encryption Standard (AES) and Rivest–Shamir–Adleman (RSA) algorithms to secure secret data to be embedded in a cover image. In addition, the use of the Adaptive Genetic Algorithm based Optimal Pixel Adjustment (AGA-OPAP) strengthened the Least Significant Bit embedding while retaining the best possible image quality and visual imperceptibility. The proposed model achieved higher security, high embedding capacity as well as image quality, which is vital for cloud communication. To perform LSB embedding we applied 2D-Discrete Wavelet Transform (2D-DWT-2L) method with 8×8 dimensional block-wise embedding. It helps to achieve better embedding efficiency in conjunction with the AGA-OPAP model. Simulation results and respective visio-statistical assessment revealed that the proposed VIHCS model can accomplish better performance and reliable secure data communication over the cloud environment. Thus, the VIHCS model achieves maximum possible imperceptibility and hence can avoid attacks- such as steganalysis based attack or RS-attacks.

137 - 145

Authors: G. Kiruthiga, S. Mary Vennila

With the advancement in internet technology, everyone can able to utilize resources with low cost using cloud resources. There will be numerous requests for task scheduling to share resources in the cloud environment. When the task request is received by the cloud technology it should have the ability to distribute the workload among sharable resources in a balanced manner and effective utilization of resources. Machine learning and metaheuristic algorithms provide a dynamic part in balanced task assignments in the cloud paradigm. Existing unsupervised models-based load balancing, centroid selection is done randomly and imprecise job requests are not well handled by them. This paper aims to develop a clustering model-based task scheduling with the knowledge of behavioural inspired optimization algorithm in a highly balanced manner. A robust Intuitionistic Fuzzy C-means empowered grasshopper optimization has been anticipated in this work, which utilizes the merits of the Intuitionistic fuzzy and Grass Hopper algorithm for prominent task scheduling among virtual servers in a cloud environment. The results proved that IFCM-GOA reduces the makespan, execution time and, high balance load scheduling with improved cloud resource utilization.

157 - 166

Authors: Gurpreet Singh Shahi, Ranbir Singh Batth, Simon Egerton

With the rise in the economy and population of various developing and developed nations leads to an increase in the number of vehicles on the road to manifolds recently. In the smart vehicular networks, there are numerous vehicle transportation problems (VTP) exists. The main or the most common problem is the congestion occurrence due to traffic jams and road accidents, especially in urban areas. So route planning to find the optimal route for a vehicle that is moving from source to destination is an important and challenging task as the route conditions change with the traffic conditions, hence the shortest and optimal route needs to be re-evaluated. This manuscript presents a study and comparison of various pathfinding algorithms which include Dijkstra, A Star, CH, and Floyd Warshall algorithms which are available as optimal route finding algorithms. The simulation of working algorithms with comparison has been performed using SUMO with TRACI and using python coding. The simulation is performed on real maps of urban areas. The parameters used for the comparative study are travel time, distance travelled, and speeds of vehicles.