Efficient routing protocols are required in opportunistic networks due to the inherent challenges of sporadic connectivity, transmission delays, and dynamic network topologies. In these networks, information is forwarded and disseminated among smart devices based on opportunistic contacts driven primarily by network dynamics and user mobility. This paper introducesLearnRouter, a dynamic reinforcement learning-based routing algorithm designed to improve the message delivery ratio while minimizing end-to-end delay in opportunistic networks. Existing routing protocols use static strategies or fixed replication schemes, leading to resource utilization and high message drop rates. LearnRouter addresses these challenges by incorporating the multi-armed bandit framework and utilizing the Upper Confidence Bound algorithm to make more intelligent and data-driven forwarding decisions. The proposed approach enables the protocol to adapt continuously to network changes while balancing the trade-off between exploration and exploitation. The simulation results show that LearnRouter surpasses end-to-end delay and message delivery ratio in comparison with Direct Delivery, Spray and Wait, CBR, SimRouter, RL-Prophet, and K-DQLR in opportunistic and dynamic environments.

Payment Channel Networks (PCNs) act as a foundational layer in blockchain and deal with the scalability issues. However, a major challenge that continues to hinder the world-wide acceptance of PCNs is the transaction failure rate caused by channel dependency. This depletion occurs due to the dependency of PCNs on network topology. Due to this, effective node placement is crucial for establishing economical channels and strengthening network robustness. On one hand, node attachment is necessary to bring back the PCNs to a balanced state, while on the other hand, the network is prone to getting trapped in a local optimum. To address these challenges, ReLEaRN is proposed as a node placement strategy that deals with the rebalancing problem of PCNs and the local optimum problem using the Soft-Actor critic (SAC) algorithm. We simulated our proposed algorithm on topologies of the Lightning Network and achieved 90 % improvement in execution time when compared with ProfitPilot. The execution time is similar to basic heuristic topologies present in the Lightning Network, but the probability of fees collection is improved to 40 % compared to these topologies.

Blockchain offers an efficient, reliable, and secure environment for performing transactions. Scalability, high transaction fees, routing services, and low throughput are some of the primary challenges of blockchain-based cryptocurrencies. Off-chain transactions are used to tackle these challenges. Payment Channel Networks (PCNs) are developed for implementing off-chain transactions, which require routing algorithms for making successful payments between users. The existing literature proposes many routing algorithms for these transactions in PCNs. However, existing routing algorithms in PCNs considered only a single objective while performing routing. Our proposed Adaptive Multi-Objective Routing algorithm (AMORA) considers multiple objectives that enhance the cost-effectiveness, throughput, and network resiliency in PCNs. Additionally, it reduces the hop count for transactions to retain long-term sustainability. AMORA is based on a local search-based memetic algorithm (MA). MA improves routing through multi-objective optimization. Rigorous simulation evaluation demonstrates that AMORA optimizes transaction time by 31.10%, 47.2%, 46.75%, 33.47%, 48.38%, and 74.3% and increases throughput by 55.34%, 40%, 29.73%, 57.69%, 50.78%, 59.81% compared to the Fence, Dijkstra, MILPA-PCN, SpeedyMurmurs, Spider, and Flash algorithms, respectively. Furthermore, we carry out a theoretical analysis of AMORA’s convergence and compute computational complexity.

The growing demand for digitization necessitates modifications to current Internet technologies to effectively support Internet of Things (IoT) applications and services. Software-defined networking (SDN) supports network abstraction and dynamic management through a central controller. The ternary content addressable memory (TCAM) memory used in the OpenFlow switches is very costly and limits the flow table size. The limited size of TCAM memory motivates us to think about the flow rule replacement in SDN switches. This paper introduces a routing approach to meet traffic flow requirements in the SDN networks, consisting of two main phases - routing path selection and flow rule replacement. In the first phase, we formulate a minimum cost routing problem to minimize the network’s active links as an integer linear program (ILP) in SDN networks. We then introduce a greedy heuristic-based solution to solve the ILP in polynomial time. In the second phase, we propose a flow rule replacement method that utilizes idle timeout settings to prevent overflow in the flow tables of SDN switches. The simulation results indicate that our proposed approach reduces active links by $\mathbf{1 5 \%}, \mathbf{2 4 \%}, \mathbf{3 4 \%}$ in the Goodnet topology, and by $\mathbf{1 1 \%}$, $\mathbf{2 3 \%}$, 33% in the Sprint topology as compared to benchmark schemes ROSA, LARAC, and SPD, respectively. Additionally, we analyze the impact of idle timeout on flow table entries within both our proposed approach and the benchmark schemes.

Healthcare is a crucial element of human lives that produces a substantial amount of medical data every year. A major difficulty faced by e-Health systems is the secure storage and sharing of this data without compromising its integrity and privacy. Being a trustless, traceable, and immutable technology, blockchain has the potential to address these issues. In this paper, we propose HierChain, a hierarchical blockchain-based data storage and sharing system for healthcare. This framework provides a trustless environment that is decentralized and tamper-resistant for efficient management of data. We first introduce an optimization problem aimed at identifying the optimal data storage solution for maximum efficiency. Next, we classify the health data based on its specific features, sensitivity, and storage requirements. This classified data is then stored on three different blockchains (Ethereum, Hyperledger Sawtooth, and MultiChain) according to their optimal attributes. We utilize fog nodes for providing computational services to the resource-constrained Internet of Things (IoT) nodes, enabling us to perform data preprocessing and eliminate redundant information from the collected data. Moreover, we use differential privacy on the fog layer to ensure that sensitive medical data remains protected throughout the analysis process. Finally, we present a comprehensive attack evaluation and performance analysis by implementing our framework on a real-world medical dataset. The simulation results indicate that HierChain outperforms existing algorithms in terms of scalability and security of health data without degrading the performance of the IoT network.

In the rapidly evolving digital world, blockchain technology is becoming the foundation for numerous applications, ranging from financial services to supply chain management. As the usage of blockchain is becoming more prevalent, the energy-intensive nature of this technology has raised concerns about its long-term sustainability and environmental footprint. To address this challenge, we explore the potential of Peer-to-Peer Federated Learning (P2P-FL), a distributed machine learning approach that allows multiple nodes to collaborate without sharing raw data. We present a novel integration of P2P-FL with blockchain technology, aimed at enhancing the sustainability and efficiency of blockchain networks. The basic idea of our approach is the use of distributed learning mechanisms to find the optimal performance parameters of blockchain without relying on centralized control. These parameters are then used by a load-balancing mechanism that prioritizes energy efficiency to distribute loads on different blockchains. Furthermore, we formulate a non-cooperative game theory model to align the individual node strategies with the collective objective of energy optimization, ensuring a balance between self-interest and overall network performance. Our work is exemplified through a case study in the renewable energy sector, demonstrating the application of our model in creating an efficient marketplace for energy trading. The experimentation and results indicate a significant improvement in the execution times and energy consumption of blockchain networks. Therefore, the overall sustainability of the network is enhanced, making our framework practical and applicable in real-world scenarios.

With the rollout of smart meters, a vast amount of energy time-series became available from homes, enabling applications such as non-intrusive load monitoring (NILM). The inconspicuous collection of this data, however, poses a risk to the privacy of customers. Federated Learning (FL) eliminates the problem of sharing raw data with a cloud service provider by allowing machine learning models to be trained in a collaborative fashion on decentralized data. Although several NILM techniques that rely on FL to train a deep neural network for identifying the energy consumption of individual appliances have been proposed in recent years, the robustness of these techniques to malicious users and their ability to fully protect the user privacy remain unexplored. In this paper, we present a robust and privacy-preserving FL-based framework to train a bidirectional transformer architecture for NILM. This framework takes advantage of a meta-learning algorithm to handle the data heterogeneity prevalent in real-world settings. The efficacy of the proposed framework is corroborated through comparative experiments using two real-world NILM datasets. The results show that this framework can attain an accuracy that is on par with a centrally-trained energy disaggregation model, while preserving user privacy.

Nowadays, Internet of Things (IoT) has become very popular due to its applications in various fields such as industry, commerce, and education. Cities become smart cities by utilizing lots of applications and services of IoT. However, these intelligent applications and services significantly threaten the environment regarding air pollution. Therefore, high accuracy in air pollution monitoring and future air quality predictions have become our primary concern to save human beings from health issues coming from air pollution. In general, deep learning (DL) and federated learning (FL) techniques are suitable for solving various forecasting problems and dealing with the high volatile air components in heterogeneous big data scenarios. This ambiance of DL and FL motivates us to exploit the DL-based Bidirectional Gated Recurrent Unit (BGRU) method for future air quality prediction using big data and federated learning (FL) to train our model in a distributed, decentral, and secure ways. This paper proposes a novel distributed and decentralized FL-based BGRU model to accurately predict air quality using the smart city's big data. The effectiveness of the FL-based BGRU Model is estimated with other machine learning (ML) models by using various evaluation metrics.

Vehicular edge computing (VEC) brings computational resources at the edge of vehicular networks (VANETs). In VEC, the roadside unit (RSU) across the road segment acts as an edge server. The vehicle having less computational capability offloads high computation tasks to its nearby RSU for processing. There is a significant energy consumption occurs at the RSU in computing each high computation task. To minimize the energy consumption, a caching technique is used at RSUs. The greatest challenge of caching in VEC is the mobility of vehicles. In this poster, we propose a Mobility-Aware Caching technique (MobiCache) in VEC. MobiCache uses an actor-critic deep reinforcement learning framework to find the best routes for migrating the popular cache contents of RSUs according to the mobility pattern of vehicles. Simulation results show that our proposed caching strategy reduces the energy consumption by an average of 39.54% as compared to other existing caching techniques.

Energy efficiency is considered a significant concern in the deployment and operation of data networks. The network devices need an enormous amount of energy to function, which leads to an increase in energy consumption in the data center networks (DCNs). Software-defined networking (SDN) solves the problem by adjusting the energy consumption proportionate to the amount of traffic. The network devices with low load can be turned into switch-OFF mode after transferring the traffic to another device. This energy-saving approach by analyzing the user’s traffic demand increases the overall network utilization. In this work, we study the energy optimization problem using multipath routing in SDN-enabled data center networks (SD-DCN). We formulate the energy optimization as an integer linear program (ILP) problem to minimize the active Open vSwitch (OVS) switches in the network. To solve the problem in polynomial time, we propose a heuristic approach to route the traffic flows in the SD-DCN. The proposed approach is tested over data center network topologies – Fat-Tree and BCube. The simulation results show that our proposed approach presents an enhancement of 24%, 16%, and 15% in average delay, throughput, and energy savings in the Fat-Tree topology compared to the benchmark schemes. Further, our proposed approach achieves 17%, 19%, and 17% enhancement in average delay, throughput, and energy savings in the BCube topology compared to the benchmark schemes.

COVID-19 is spreading at an exponential rate and declared as ‘Pandemic’ by the World Health Organization (WHO). The only way of controlling its spread is ‘social distancing’ among the human beings. Apart from human beings, it can also spread through other objects such as metals and plastics which are used in public passenger buses, cars and trains. Studies reveal that the COVID-19 virus can stay on metallic things for 2-5 days. This makes traveling more difficult for regular employees to their workplace. Also, due to increasing number of COVID-19 patients, it is very difficult for a patient to find the required resources at hospitals. In this paper, we propose a VANET-IoT framework for detection of COVID-19 exposure in various entities such as vehicles. A vehicle is said to be virus-exposed if it carries any COVID-19 positive patient. We develop an Android application ‘Corona Assistant’ for travelers to find all virus-exposed vehicles and chose virus-free vehicle for traveling. In addition, a hospital recommendation system (HRS) is developed inside the ‘Corona Assistant’ application. HRS helps the COVID-19 positive patient to select the nearby hospital based on available resources.

The burgeoning demands of the Internet of Things (IoT) applications such as video/audio streaming in surveillance, disaster recovery, multimedia, and healthcare paves the need to provide better Quality of Service (QoS) delivery. The growth in the amount of data generated by multimedia applications increases congestion in the network. Software-Defined Networking (SDN) is an emerging approach that centrally controls the network and solves the congestion problem in the network. SDN has many advantages that help in network management, traffic shaping, and routing in the IoT network to manage the data generated from a diverse range of applications. In this paper, we propose a congestion technique using Hierarchical Token Bucket (HTB) to manage the bandwidth in the Software-Defined IoT (SDIoT) network. Further, we propose a routing scheme to compute optimal routing shortest paths using Dijkstra’s algorithm by selecting the min-cost path based on the priorities of traffic flows. The results illustrate that the proposed approach achieves a reduction in end-to-end delay by 38%, 44% and higher average throughput by 29%, 43% in comparison with the benchmark schemes - SDN with HTB and the Delay Minimization method, respectively.

Automatic fall detection is a vital technology for ensuring the health and safety of people. Home-based camera systems for fall detection often put people's privacy at risk. Thermal cameras can partially or fully obfuscate facial features, thus preserving the privacy of a person. Another challenge is the less occurrence of falls in comparison to the normal activities of daily living. As fall occurs rarely, it is non-trivial to learn algorithms due to class imbalance. To handle these problems, we formulate fall detection as an anomaly detection within an adversarial framework using thermal imaging. We present a novel adversarial network that comprises of two-channel 3D convolutional autoencoders which reconstructs the thermal data and the optical flow input sequences respectively. We introduce a technique to track the region of interest, a region-based difference constraint, and a joint discriminator to compute the reconstruction error. A larger reconstruction error indicates the occurrence of a fall. The experiments on a publicly available thermal fall dataset show the superior results obtained compared to the standard baseline.

Internet of Things, delineated as a network of connected heterogeneous devices is emerging as a widely adopted technology in almost all walks of life today. The massive increase in the number of IoT devices has introduced several issues related to security and management. Blockchains can be a promising technology to make IoT systems secure and distributed for the time to come. However, current blockchain systems are not capable of scaling in accordance to the huge IoT data without a loss in speed and time efficiency. Therefore, we use the concept of sidechains and offline data storage to alleviate the scalability issue of blockchains. In this work, we propose an architectural framework for the security and maintenance of IoT systems using blockchain technology. Smart contracts are used to enforce data authentication, authorization, and keep track of all the activities. Extensive simulation and analysis results demonstrate that the proposed blockchain architecture is highly scalable (in terms of average latency, throughput and cost) and can be applied efficiently in the IoT system.

Cyber-physical systems (CPSs) comprise physical systems or objects incorporated with computing functionalities and data storage systems. Different sensor nodes in a CPS communicate with each other and devices like actuators and microcontrollers regulate the physical systems through smart algorithms. Efficient communication and networking algorithms play a critical role in supporting smooth interaction between the cyber and physical worlds. Internet of Things (IoT) belongs to the physical part of CPS. In the physical part, the sensor nodes are connected with each other and each node sends data to the root node through their respective parent node in a multi-hop manner. The physical devices in a CPS may stop sensing the environment due to logical and physical failures. To solve this, we propose a real time defective nodes detection scheme (R2D) in a cyber-physical system. The root node of the CPS gathers sensed data, detects the defective nodes and performs data analysis in the cyber part of the CPS. The results show that our proposed R2D protocol achieves a 10.2% higher packet delivery ratio and a 12.5% increase in throughput as compared to the flooding approach. Moreover, an improvement by almost 50% in throughput is observed when compared to RPL with an increase in packet loss rate of 20%.

Road-side unit (RSU) plays an important role in providing connectivity among the vehicles on the road. In rural areas, RSUs are powered using renewable energy, such as solar or wind energy. Hence, the energy consumption across such RSUs should be efficient i.e., energy consumption at RSUs should be minimized and no RSU gets over-utilized while others are under-utilized. The amount of energy consumption depends upon the scheduling of different kinds of data requests at RSU. In this paper, we propose a scheduling architecture for minimizing energy consumption at RSU and attaining uniform energy consumption across neighboring RSUs. This, in turn, increases the request fulfillment percentage at RSUs. The proposed architecture categorizes the incoming request as a Traditional (less computation) or a Smart request (high computation). Two approaches- Hard-deadline Less Computation requirement Approach (HLCA) and Soft-deadline High Computation requirement Approach (SHCA) are proposed for addressing Traditional and Smart data requests, respectively. In HLCA approach, the receiving RSU uses the scheduling metric to select the servicing RSU for request fulfillment. We prove by analysis, how scheduling metric helps in minimizing and achieving uniform energy consumption across the RSUs. In SHCA approach, Fog computing is used for handling high computation requests. Energy consumption at RSUs is further optimized by using Auction game-based relay vehicle selection mechanism. Simulation results demonstrate that our proposed approaches achieve uniform energy consumption across multiple RSUs and 10% more efficient than scheduling algorithms for single RSU model such as Nearest Fastest Set Scheduler (NFS).

The existence of link correlation has been empirically validated, and different exemplary works exploit the link correlation in the designing of various network protocols. In this work, the authors investigated the impact of this link correlation in contention mechanism of medium access control (MAC). They illustrated negative link correlation could deteriorate the contention mechanism designed to handle hidden terminal problems and, consequently, increase the packet collision rate between the neighbours. They also showed that negative link correlation could increase the chance of an exposed terminal problem. Therefore, ignoring negative link correlation could lead to overestimate overall network throughput and underestimate packet delay. Next, instead of designing a new contention mechanism exploiting the link correlation, they proposed a new routing tree which mitigates the negative effect of negative link correlation without altering the underlying MAC layer. They evaluated this routing tree on Indriya testbed with TelosB nodes and compared it to the minimum spanning tree based on link quality only. The results show improvement in end‐to‐end throughput and packet reception ratio at each node.

Internet of things (IoT) has made human life simpler through its numerous applications by interconnecting various devices and people together in a heterogeneous network. The number of users associated with IoT in different sectors such as media, intelligent transportation, and healthcare are increasing rapidly, therefore generating a large amount of data each day. Because of this, maintaining quality in the network has become an important aspect for end-to-end data delivery. Software defined networking thus provides flexibility and programmability to the network due to its global approach of network management. To adapt with the dynamic nature of QoS, blockchain technique with its encryption mechanism and distributed consensus algorithm can be deployed on SDN. In this paper, we propose the use of blockchain technology in Software-Defined Internet of Things to deliver high QoS at a low price.

Air pollution is a major cause of health problem in urban areas. Vehicles are the major sources of the current air pollution in urban cities. In this work, we proposed a pollution monitoring system for vehicles using IoTs. This system measures the real-time pollution generated by vehicles on road. This paper describes the design of the system for sensing the pollution using sensor, arduino, smart phone and mobile applications for displaying the personalize air pollution information for individual vehicle. We evaluate the proposed approach on real-data experiment and it shows some preliminary results.

In cooperative communication, a set of players forming a coalition ensures communal behavior among themselves by helping one another in message forwarding. Opportunistic mobile networks (OMNs) require multihop communications for transferring messages from the source to the destination nodes. However, noncooperative nodes only forward their own messages to others, and drop others' messages upon receiving them. So, the message delivery overhead increases in OMN. For minimizing the overhead and maximizing the delivery rate, we propose two coalition-based cooperative schemes: 1) simple coalition formation (SCF) and 2) overlapping coalition formation (OCF) game. In SCF, we consider the presence of a central information center, whereas OCF is a fully distributed scheme. In SCF, coalitions are disjoint, whereas in OCF, a node may be the member of multiple coalitions at the same time. All nodes in a coalition help each other cooperatively by forwarding group messages to the intermediate or destination nodes. The goal of the nodes is to achieve high success rate in delivering messages. The proposed SCF scheme is cohesive, in which disjoint coalitions always combine to form grand coalition. In OCF, a node reaches a stable grand coalition when all the nodes of the OMN are members of overlapping coalition of the node. No node gains by deviating from the grand coalition in SCF and OCF. Simulation results based on synthetic mobility model and real-life traces show that the message delivery ratio of OMNs increase by up to 67%, as compared to the noncooperative scenario. Moreover, the message overhead ratio using the proposed coalition-based schemes reduces by up to about (1/3)rd of that of the noncooperative communication scheme.

Opportunistic Mobile Networks (OMNs) are characterized by intermittent connectivity among nodes. In many scenarios, the nodes attempt at local decision making based on greedy approaches, which can result in getting trapped at local optimum. Moreover, for efficient routing, the nodes often collect and exchange a lot of information about others. To alleviate such issues, we present SeeR, a simulated annealing-based routing protocol for OMNs. In SeeR, each message is associated with a cost function, which is evaluated by considering its current hop-count and the average aggregated inter-contact time of the node. A node replicates a message to another node, when the latter offers a lower cost. Otherwise, the message is replicated with decreasing probability. Moreover, SeeR works based solely upon local observations. In particular, a node does not track information about other nodes, and, therefore, reduces the risk of privacy leaks unlike many other protocols. We evaluated the performance of SeeR by considering several real-life traces under plausible conditions. Experimental results show that, in the best case, SeeR can reduce the average message delivery latency by about 58 percent, when compared to other popular routing protocols.

Opportunistic mobile networks (OMNs), which are formed by mobile devices carried by human users, present an interesting communication paradigm in the absence of access to global network connectivity or any form of network infrastructure. In this work, we combine thenaturalmobility of the human users—which has been shown to resemble Levy Walk—in OMNs, together with intentionalexplorations. We consider the case where the human users in an OMN undergo explorations, i.e., occasionally visit a set of fixed point of interests (PoI), for example, shopping malls. The objective of this work is two-fold-1) Establishing that limited explorations of the users can help in enhancing the performance of OMNs, and 2) Formulating a method to decide whether or not a user should undergo exploration. In this regard, we propose two schemes based on prospect theory (PT) and expected utility theory (EUT). The results of extensive simulation-based performance evaluation indicate that limited exploration can promote the delivery ratio of messages by large levels—about$7$-$33$percent depending on the number of randomly placed PoI, and about$36$percent depending upon the terrain size. Moreover, the time spent in exploration, on an average, is negligibly small—a typical value is about$0.55$percent of the simulation duration, which indicates its feasibility in real life.

Cooperation among nodes is a fundamental necessity in opportunistic mobile networks (OMNs), where the messages are transferred using the store-carry-and-forward mechanism, due to sporadic inter-node wireless connectivity. While multiple works have addressed this issue, they are often constrained in their assumptions on solutions (e.g., requirement of central authority, and tracing the recipient nodes for providing reward or punishment). In this work, we address this research lacuna by taking an evolutionary theory-based approach. In evolutionary theory, the players analyze alternative strategies and select the best one to survive in a population. Inspired by this, in this work, we propose a Distributed Information-Based Cooperation Ushering Scheme (DISCUSS) to promote cooperation in message forwarding between nodes. In this scheme, the nodes maintain and exchange information with one another during contacts about the messages created and delivered in the network. Based on this, the nodes evaluate their own performance and compare that with the approximated network performance to adapt the most successful forwarding strategy. Simulation results show that the message delivery ratio in the network improves upto 31 percent, when the nodes dynamically switch their strategies, as compared to the case when they do not. Furthermore, the DISCUSS scheme fared closely to its variant with the nodes having complete knowledge about the network-wide performance.