Most of the future intelligent transportation services will rely on onboard sensing and communication protocols used in modern vehicles for providing uninterrupted services such as lane change, on demand audio-video entertainment, and emergency services to end users. Most of these services generate a huge amount of big data used for analytics to take intelligent decisions. However, keeping in view of the complex decision making and limited resources, the deployment and use of these services has various challenges and constraints including data safety, intelligent decision making, and route planning. Specifically, handling emergency situations for the end users traveling on road can be considered as an interesting problem which requires an efficient solution resilient to the aforementioned constraints and challenges. Motivated from the above, in this paper, we propose a prioritize route selection strategy using Federated learning (FL). The proposed scheme first envisions a futuristic road network scenario in which vehicles rely on an onboard intelligent route movement algorithm for reaching to its destination. By assigning higher priority to vehicles on emergency duties, the proposed scheme provides an uninterrupted route discovery by facilitating them to reach their destination on time. The proposed scheme has been validated using simulations on benchmark data sets traces using various performance evaluation metrics in comparison to the other existing state-of-the-art proposals. Results obtained prove the efficacy of the proposed solution on comparison with other existing schemes in literature.

In recent years, advancements in vehicular communication has improved road safety along with passenger convenience for many applications. However, to take intelligent and timely decisions, a large number of complex operations need to be get executed on large amount of data base repositories which in turn generates a huge burden on the underlying network infrastructure leading to a large amount of energy consumption. Most of the existing solutions reported for the aforementioned problems are based upon the traditional monolithic solutions which may not be applicable in modern scenarios in this environment. Hence, to mitigate the aforementioned challenges and constraints, in this article, we propose BOOST, a connected dominating set (CDS)-aware energy-efficient clustering scheme for Software Defined Network by integrating V2I and V2V communications for reliable and seamless data delivery. The proposed scheme has been specifically designed for urban scenario to achieve effective data delivery with minimum energy consumption. By leveraging the benefits of CDS on roadside communication infrastructure, BOOST is able to adapt with varying traffic conditions to provide seamless scalability with minimum energy and network overheads. The proposed scheme has been evaluated using various performance evaluation metrics in comparison to the existing benchmark schemes. The results obtained demonstrate its superior performance by 3% to 4% in terms of energy-efficiency, network overhead, packet delivery rate, and network throughput in comparison to the existing benchmark schemes.

Abstract Healthcare 4.0 paradigm aims at realization of data-driven and patient-centric health systems wherein advanced sensors can be deployed to provide personalized assistance. Hence, extreme mentally affected patients from diseases like Alzheimer can be assisted using sophisticated algorithms and enabling technologies. Motivated from this fact, in this paper, DeTrAs: Deep Learning-based Internet of Health Framework for the Assistance of Alzheimer Patients is proposed. DeTrAs works in three phases: (1) A recurrent neural network-based Alzheimer prediction scheme is proposed which uses sensory movement data, (2) an ensemble approach for abnormality tracking for Alzheimer patients is designed which comprises two parts: (a) convolutional neural network-based emotion detection scheme and (b) timestamp window-based natural language processing scheme, and (3) an IoT-based assistance mechanism for the Alzheimer patients is also presented. The evaluation of DeTrAs depicts almost 10–20% improvement in terms of accuracy in contrast to the different existing machine learning algorithms.

Rapid mobility and frequent disconnection in vehicular networks makes multi-hop data delivery challenging. To address this concern, adaptive data forwarding is applied over vehicular networks in a named data networking environment. We have proposed An Interest Chain based Forwarding Mechanism (ICFM), which includes a chain based forwarding mechanism by employing fuzzy logic to evaluate the next chain member. Autonomous vehicles forward a packet to the next promising vehicle and in this way a chain is formed to satisfy interest with the requested data. The proposed fuzzy-based interest chain mechanism is used to forward a packet to destination and receive a corresponding data packet with decreased data delivery delay. Experimental analysis has been performed on ndnSIM to verify the performance of the suggested scheme. The scheme runs on different scenarios and efficiency has been observed in terms of interest satisfaction ratio, average number of interest packets, average number of data packets and average delay.

Vehicular Edge Computing (VEC) is gradually evolving into one of the most prevalent paradigms for vehicular computation. This is due to its ability for effectively handling the tasks of varied complexity. VEC based task orchestration has therefore emerged as an exciting research domain. A large number of task orchestration schemes have been proposed that exploit its technical capabilities. However, identifying the most appropriate vehicles for such edges still remain a challenge. In this work, we propose an intelligence based Task Orchestration Scheme integrated with Vehicular Cloud Edge Networks that uses Federated learning (FL) for forming vehicular edges. FL is a privacy preserving technique with no data being shared centrally. This scheme uses characteristics of vehicles such as computational capacity and their starting as well as ending point for creating the edges. Obtained results depict the improved performance of this scheme as compared to conventional schemes.

Vehicular networking systems rely on Internet Protocol to exchange information among vehicles. With the increasing number of vehicles, the communication overhead has increased significantly a nd h as b ecome m ore c ontent centric. To resolve this problem, the Named data networking-based communication model has been used. This communication is completely based upon the content rather than the location and provides better network coverage comparatively. The vehicles used for communication purposes in a network are moving in some specific p atterns, b ased o n h aving t he s ame destination, with the same speed parameters etc. These vehicles which have common interests form a platoon. This vehicular platoon helps in various fields such as safe driving, energy efficiency and road safety. This paper provides a scheme for the applicability of NDN to the vehicular platoon. Special design features are proposed for communication purposes in V-NDN-based vehicular platoons. The backbone platoon network is used for data dissemination between the vehicles on the highway. To check the efficiency of the proposed scheme, extensive simulations have been performed on the ndnSim simulator. More precisely, different scenarios have been used and analyzed their efficiency i n t erms o f d elay and throughput.

The global escalation in the road traffic density alleviates the harmful emissions and fuel bills due to congestion and misaligned traffic control. The conventional traffic monitoring schemes (camera or sensor-based) are not able to cover every nook and corner and thus miss various vital traffic parameters that can otherwise be very useful for traffic density and pattern analysis. Internet of Drones (IoD) has been widely adopted to resolve various related challenges and has strong potential in traffic monitoring even in the areas where scarcity of fixed infrastructure is witnessed. Thus, in this paper, we have proposed an IoD-based traffic monitoring system to avoid the congestion on the roads within the available infrastructure. Moreover, to deal with the dynamic network typologies, an software-defined networking (SDN)-based centralized controller is configured to generate the flow rules for end to end data transmission. However, the drones communicate with each other through an open channel (now controlled through a programmable SDN architecture). Thus, the integrity of data collected by drones must be protected through a robust security mechanism. So, we have adopted a blockchain technology to secure the proposed system against unauthorised access and maintain data integrity. However, maintaining the entire blockchain on the drones can lead to several resource bottlenecks. Thus, we have used a referenced blockchain architecture that decouples the data from the blockchain part and stores it in the off-chain. The proposed scheme has been validated using simulated environments that validates its the superiority in contrast to the existing variants.

Edge intelligence propelled federated learning as a promising technology for embedding distributed intelligence in the Internet of Things (IoT) ecosystem. The multidimensional data generated by IoT devices is enormous in volume and personalized in nature. Thus, integrating federated learning to train the learning model for performing analysis on source data can be helpful. Despite the above reasons, the current schemes are centralized and depend on the server for aggregation of local parameters. So, in this paper, we have proposed a model that enables the sensor to be part of a defined cluster (based on the type of data generated by the sensor) during the registration process. In this approach, the aggregation is performed at the edge server for sub-global aggregation, which further communicates the aggregated parameters for global aggregation. The sub-global model is trained by selecting an optimal value for local iterations, batch size, and appropriate model selection. The experimental setup based on the tensor flow federated framework is verified on MNSIT-10 datasets for the validity of the proposed methodology.

Vehicular ad hoc Networks have emerged as a viable alternative for enabling user applications on moving vehicles. However, maintaining acceptable levels of Quality of Service and message latency still remains a challenging task. Several solutions have been proposed for improving performance of these networks. Clustering has been considered as one of the important mechanism that structures vehicles into organize groups. However, high deployment overheads and lack of security are the major bottlenecks hindering its deployment. Software defined networking has been emerged as a promising solution on account of its characterstics such as dynamic access control and scalabilty. In view of this, TruClu: a trust based clustering mechanism that creates vehicular clusters for a Software Defined Vehicular Network is proposed. Cluster formation and cluster head selection in TruClu is based on vehicular mobility and trust value that alleviates the drawbacks of traditional clustering and also enabling trust based communication in the network. The performance of TruClu is evaluated through extensive simulations and obtained results indicate the comparable performance of the proposed scheme in terms of standard performance parameters.

Increased vehicular flow on roads along with proposed deployment of autonomous vehicles has necessitated the need for accurate traffic forecasting so as to achieve effective route guidance, traffic management, public safety and congestion avoidance. Although a number of traffic forecasting algorithms have been proposed but most of these algorithms perform short term traffic predictions. However future vehicular systems also defined as intelligent VANETs will require a hybrid traffic forecasting model that predicts the vehicular traffic for varying values of time. This paper proposes a time varying forecasting model that predicts vehicular flow by utilizing Long Short-Term Memory (LSTM) and Convolutional Neural Network(CNN). The model is based on large-scale, network-wide traffic with spatio-temporal features. The temporal features learned by LSTM and spatial features learned by CNNs from the matrices are further fused with external factors to derive the final forecast. Model has been implemented on the traffic data set of Chandigarh city in India, mapped onto three two-dimensional matrices of time and space. The predicted information is then forwarded by the vehicle to all the other vehicles in their vicinity using vehicular adhoc networks. Experimental results indicate that the proposed model performs significantly better than other state-of-the-art models in terms of accuracy and efficiency.

Drones are equipped with high-vision cameras, advanced sensors, and GPS receivers to deliver diverse services from high altitude thereby creating an airborne network. In this environment, physical things (drones, sensors, etc.,) are controlled using computational algorithms to form a cyber-physical system for the Internet of drones. Although the drones provide manifold benefits still there are many issues (security, privacy, and data integrity) which must be resolved before the usage of drones in smart cyber-physical systems. So, in this paper, a blockchain-based security mechanism for cyber-physical systems is proposed to ensure secure transfer of information among drones. In this mechanism, the miner node is selected using a deep learning-based approach, i.e., a deep Boltzmann machine, using features like computational resources, the available battery power, and flight time of the drone. The proposed mechanism is evaluated based on different performance metrics and the results obtained show the potential benefits of the proposed scheme.

With the surge in the demand for online services and multimedia applications, the traffic on the underlying network infrastructure has escalated (multi-folded) in recent years. To meet the strict latency requirements, Software-defined Networking (SDN) provides flexible network control (and possible intelligence) that can act as an enabler for application-oriented service industry. However, the crippling gap between the business needs and the network delivery potential necessitates the underlying network to constantly (and consistently) adapt, protect, and inform across all strands of the service-oriented landscape. Intent-based network has emerged as a recent solution to the cover the above gap by capturing business intent and thereafter activating and assuring it networkwide. Motivated from these facts, in this article, an Intent-based network control framework has been designed over the SDN architecture for data dissemination in the vehicular edge computing ecosystem. In this framework, a tensor-based mechanism is used to reduce the dimensionality of the incoming elephant-like traffic and then classifying the specific-attribute data traffic according to the defined priority requirement of the underlying applications. Here, the network policies are configured using the intent-based controller according to the application requirement and then forwarded to the SDN controller to enable intelligent data dissemination (through an optimal route) at the data plane. Convolution Neural Network is used to train the flow table to allocate the route dynamically for the classified traffic queues. The proposed framework has been evaluated through extensive simulations and the results supports the claims in terms of the quality of service requirements.

Internet of Things (IoT) has brought major changes in the way the workload is processed closer to the location of the data source. The need for near-to-real time provisioning of IoT workload has necessitated the emergence of Edge Computing. However, it is not entirely possible to shit the entire workload on to the edge layer due to the computational limitations of the edge devices. Hence, this challenge ended up with the amalgamation of IoT-Edge-Cloud ecosystem. But, one of the major challenges in this ecosystem is workload management in a self-organized manner (or according to the nature of the workload). This article tries to overcome this challenge by utilizing the benefits of Self Organized Map (SOM). This article comprises of three strands, 1) a SOM-based workload classification approach to handle the IoT workloads in a flexible manner, 2) an energy-efficient workload scheduling scheme using container-based virtualization, and 3) a workload migration mechanism based on secure caching technique. The proposed strands are evaluated using a simulated environment and the outcomes seem promising in contrast to generalized container-based workload scheduling.

In the recent years, vehicular adhoc networks (VANETs) can be an attractive choice for collecting and transferring the healthcare data of the passengers to the remote healthcare centers. In VANETs, some of the intermediate nodes may act as relay nodes in which case, these networks are called as vehicular relay networks (VRNs). However, the transmitted information in VRNs can be captured by intruders during transmission. Moreover, an attacker can launch selective forwarding, blackhole, and sinkhole attacks in the network, which may in turn degrade the network performance parameters like high end-to-end delay, low packet delivery ratio (PDR) and network throughput. Hence, to address these issues, a secure data dissemination scheme using VRNs is proposed. In the proposed scheme, first, a secure vehicular medical relay network system is designed for the users belonging to disconnected rural areas. The collected information is filtered at zonal levels before transmission to a nearby road side units, which further pass it to the incoming vehicles. Second, a secure passenger health monitoring network is designed which continuously monitors health services of the passengers traveling in different vehicles. The information collected through small body sensors installed in the vehicles act as data sets that is forwarded to the on-board monitoring unit within the vehicle. This collected data is then transmitted to centralized healthcare centers for processing by using VRNs. Lastly, a strong elliptic curve cryptography-based cryptographic solution is designed for secure communication among different vehicles. The performance of the proposed scheme is evaluated in various network scenarios with respect to different selected parameters, such as throughput, network delay, PDR, jitter, transmission and computation overheads, and key distribution overhead. The obtained results indicate that the proposed scheme provides improvement of 52% in average delay and 5% in PDR. This further indicates effective message delivery even with high mobility of the vehicles.