The Internet of Vehicles (IoV) is an evolving domain fueled by advancements in vehicular communications and networking. To enhance vehicle coverage, integrating vehicle-to-everything (V2X) networks with cellular networks has become essential, though this integration places increased demand on cellular infrastructure. To address this, we propose a two-tier stable path routing algorithm designed to improve the stability and proactiveness of V2X networks. Our approach divides the coverage area into zones, further segmented into road segments based on road structure. The first tier manages routing within a road segment, while the second tier handles routing between vehicles in adjacent segments. This method improves road awareness, stabilizes topologies, adapts to dynamic changes, and reduces routing overhead. Additionally, the incorporation of Kalman filter-based prediction model further strengthens proactive routing. To validate the proposed approach, we conduct synthetic evaluations across varying vehicular densities with different mobility and traffic scenarios. We compare the traditional centralized routing strategy with the proposed distributed two-tier mechanism to assess execution cost, end-to-end latency, network resource consumption, data rates, packet flow, and packet loss. Quantified results demonstrate that our two-tier approach reduces the average execution cost from 438.61 to 230.48, lowers average latency from 232.34 ms to 129.03 ms, and minimizes average network consumption from 231.26 MB to 129.39 MB. The proposed approach continues to significantly enhance data rates, reduce packet flow processing, decrease packet loss across various routing strategies. Overall, the proposed solution enhances stability, responsiveness, and robustness of V2X communication, making it suitable for future large-scale IoV deployments.

Internet of Vehicles (IoV) is an emerging archetype that is a distributed network of various vehicles armed with sensors, actuators, technologies, and applications to connect and exchange data with each other over the Internet. The primary goal of IoV is to provide a vehicular platform to enable better communication and Quality of Service (QoS) for vehicles, pedestrians, and roadside infrastructure in real-time, through the use of Vehicle-to-Vehicle (V2V), Vehicle-to-Pedestrian (V2P), Vehicle-to-Infrastructure (V2I), Vehicle-to-Network (V2N), and Vehicle-to-Cloud (V2C) channels. However, the increasing number of vehicular services poses serious concerns and challenges to the researchers: real-time traffic forecasting, service placement, security, reliability, and routing. The primary focus of this work is concerned with the challenge of multi-regional forecasting of multi-class traffic. These traffic forecasting models enable pro-activeness in systems by providing real-time and accurate predictions. Also, they can explore traffic densities over spatial and temporal domains for various vehicle types. However, current literature cannot provide forecasts for multi-region and multi-class vehicles at the same time. This study aims at enabling a proactive platform which could make decisions based on the integrated Graph Neural Network (GNN) and Gated Recurrent Unit (GRU) based traffic forecasting model, i.e., Spatio-Temporal GNN (STGNN) based traffic forecasting model. The STGNN-based traffic forecasting model uses the GNN and GRU models to explore spatial and temporal features of varying vehicular multi-class traffic densities. In GNN, spatial data consisting of multi-class traffic densities are utilized for the feature extraction that results in graph embeddings. In the GRU, these graph embeddings are utilized for temporal feature extraction. This approach enables the forecasting of multi-class vehicular traffic densities and the pro-activeness of an IoV platform. In addition, the performance results show that an intelligent platform can be built upon the proposed traffic forecasting model that is capable of inspecting complex and nonlinear traffic accurately.

Network ecosystems have grown to encompass multiple application domains. SDN and NFV technologies have helped pave the road for the evolution of the core and edge networking systems, allowing for numerous services to be served by the same physical infrastructure. Guaranteeing the operability of the network has become an ever-increasing requirement in order to sustain the underlying services deployed on the network. For this, Intent-Based Networking (IBN) aims to abstract network management by introducing high-level rules/policies that are translated to network configurations per service requirements. By following this principle, we proposed an anomaly detection and mitigation mechanism that exploits the characteristics of IBN for collecting and analyzing flows, using Machine Learning for interpreting traffic patterns, and automatic deployment of high-level policies for corrective actions related to anomalous traffic occurrences. The complete system is deployed on the Korea Advanced Research Network (KOREN), where the abstraction provided by IBN for network anomaly detection and mitigation is a key factor in closing the gap to achieve complete network automation.

Nowadays, intelligence is paving its ways into the IoV domain. With the need of improvement in intelligence in this area, the need of self-organizing network management systems for providing V2X communication is also of vital importance, which current systems lack. Current routing solutions for IoV are complex and require intelligence embedded in the form of closed-loop systems. To this, an intent-based system is designed, which takes high level requirements for routing in IoV.The routing approach is road-aware and widens the scope of road-awareness to vehicles from multiple edge domains. Another problem with the current routing schemes in IoV is that their network management systems are not self-organizing. A self-organizing management system is of high importance and requires a closed-loop system. To this, an intent-based approach is followed integrated with a reinforcement learning model that supports the creation of a routing policy, which is further applied to the orchestrator and enables a road-aware and hierarchical routing approach. The proposed system is shown to be efficient in terms of data rate and the number of packet flows managed per unit time, in the management of vehicular networks.

Intent-Based Networking (IBN) is a model that enables proactive network control and automation to satisfy high-level demands. It follows a closed-loop mechanism that abstracts the network complexity and dynamically allows updates to the network monitoring and intelligence. Hence, it eliminates the traditionally practiced error-prone manual network control. In contrast with the traditional reactive approach, IBN-based approach promotes the proactive and dynamic solution. To this end, IBN uses machine learning (ML) to predict the future and keeps a balance between the intended and actual state. Hence, this work considers an IBN-driven routing mechanism to ensure service provisioning with the desired quality of service (QoS). In contrast to the traditional static routing scheme, this work considers AI-driven dynamic and proactively updatable routing schemes to ensure QoS. This work demonstrates AI-driven service route control mechanism for ensuring quality of service (QoS) on top of Korea Advanced Research Network-Software Defined Infrastructure (KOREN-SDI). More precisely, an ML-based approach is used to find the best path between source and destination nodes. In addition, it incorporates IBN closed-loop process to monitor and update service routes proactively based on predicted future link utilization. This novel approach introduces proactive updates on runtime to avoid future failures and ensures seamless service provisioning with the fulfillment of QoS demands by changing traffic routes dynamically.

Object detection is a classic computer vision task, which learns the mapping between an image and object bounding boxes + class labels. Many applications of object detection involve images which are prone to degradation at capture time, notably motion blur from a moving camera like UAVs or object itself. One approach to handling this blur involves using common deblurring methods to recover the clean pixel images and then the apply vision task. This task is typically ill-posed. On top of this, application of these methods also add onto the inference time of the vision network, which can hinder performance of video inputs. To address the issues, we propose a novel plug-and-play (PnP) solution that insert deblurring features into the target vision task network without the need to retrain the task network. The deblur features are learned from a classification loss network on blur strength and directions, and the PnP scheme works well with the object detection network with minimum inference time complexity, compared with the state of the art deblur and then detection solution.

Network slicing is an important pillar of 5G networks that empowers the network operators to provide the different quality of services (QoS) to the users. It enables network operators to split the physical network into multiple logical networks to meet different QoS requirements. In this research paper, we have designed an intent-based network slicing framework that can slice and manage the core network and radio access network (RAN) resources efficiently. It is an automated system, where users just needs to provide higher-level information in the form of intents/contracts for a network slice, and in return our system deploys and configures the requested resources. Moreover, a deep learning model Generative Adversarial Neural Network (GAN) has been used for the management of network resources. Several tests have been performed by creating three slices with our system, which shows better performance in terms of bandwidth and latency.

Flying networks are resource constraints while the nature of nodes' mobility is very dynamic and unpredicted. Therefore, these networks are very prone to link failure and performance degradation. By considering the existing limitations, this work proposes a new approach consists of proactive and reactive network failure mitigation techniques that have been named as a hybrid approach. In the proposed architecture, the SDN controllers are distributed where each one controls its local domain nodes. The controller node continuously monitors the network state information and proactively adjusts the near-future changes to the topology. Each local domain also contains a sink node that directly connects to the controller. The sink node is used to forward the network state information to the controller and keep the controller defined flow rules for local domain nodes. The sink node can also request a new path in case of any link failure or any topology updates cause by nodes' movement. Besides, a distributed routing protocol also runs on domain nodes to establish connectivity toward the sink node.

5G networks are aimed at provisioning of a wide range of sophisticated services with uninterrupted user experience. In addition, it is very challenging to manage all the services due the variety in the service requirement and a very large number of users causing dynamic changes in traffic streamed over the network. Currently, the networks are managed manually and requires experts to control the behavior of network. Due to increase in network domain it is an esteem requirement to manage and control network autonomously. In this paper we have introduced and Intent-Based networking approach which on one side abstracts and automates the network configuration also it assures the network resource state stability by using machine learning. We have used an IBN abstraction layer and M-CORD as an next generation testbed for the implementation of this work.

The automatic control over the network platforms is an esteem requirement of the operators. Recently, 5G with its aim to attain the internet of everything, it challenged researchers for the achievement of automatic control over the network platform. Furthermore, the 5G system consists of multi-domain network applications and platforms which make it complex to control and configure the network. Hence, the focus of this research is to enable easy configuration through high-level instructions and the use of machine learning for automatic control over the network infrastructure. The overall system consists of Intent-Base application that includes a machine learning model and it configures and controls the M-CORD-based network slicing test-bed.

Investigation and collection of real-time data plays a very crucial part in the orchestration of network resources. Selection of the correct data is very important as it decides to auto-scale the resources. In cloud & SDN environments such as NFV, auto-scaling becomes more critical in terms of precision and accuracy. In our case, we propose a solution for auto-scaling the network resources based on the calculations made for every action's execution-time [1] of respective instances of a VNF. The instances for each VNF are auto-scaled on the basis of execution-times per time slot, and the number of cores that are assigned by the usage of weight factor [2] used for virtual/physical cores. Hence by using the proposed solution, we are able to enhance the proper resource provisioning to fulfill the dynamic demands [3] of future mobile networks.

Network Function Virtualization (VNF) is one of the main drivers for the next generation of mobile networks. It allows the creation of multiple and diverse functionalities over a single physical infrastructure thus fulfilling the main requirements for 5G mobile networks that focus on serving heterogeneous ecosystems. The research presented in this literature focuses on a Network Slice Selection Function (NSSF) for the selection of Data Plane network slices, which are created and configured with specific QoS that cater to the need of network users. As the complexity of managing multiple Virtual Network Functions (VNF) grows with the number of Data Plane Network Slices, a system that abstracts the configuration of the mobile network is needed. This paper positions the NSSF as part of a three-layer Virtual Mobile Network System that contains an Application Layer for Policy creation, a Management Layer for Orchestration of VNFs and a Physical Layer for the deployment of network functions.

Seismic sensors are widely used to detect moving targets in the ground sensor network, and can be easily employed to discriminate human and quadruped based on their footstep signatures. Because of the complex environmental conditions and the non-stationary nature of the seismic signals, footstep detection and classification is a very challenging problem. The solution to this problem has various applications such as border security, surveillance, perimeter protection and intruder detection. Previous works in the domain of seismic detection of human vs. quadruped have relied on the cadence frequency-based models. However, cadence-based detection alone results in high false alarms. In this paper, we describe a seismic footstep database and present classification results based on support vector machine (SVM). We demonstrate that in addition to applying a good classification algorithm, finding robust features are very important for seismic discrimination.

This paper describes a new kernel wavelet-based anomaly detection technique for long-wave (LW) Forward Looking Infrared (FLIR) imagery. The proposed approach called kernel wavelet-RX algorithm is essentially an extension of the wavelet-RX algorithm (combination of wavelet transform and RX anomaly detector) to a high dimensional feature space (possibly infinite) via a certain nonlinear mapping function of the input data. The wavelet-RX algorithm in this high dimensional feature space can easily be implemented in terms of kernels that implicitly compute dot products in the feature space (kernelizing the wavelet-RX algorithm). In our kernel wavelet-RX algorithm, a 2-D wavelet transform is first applied to decompose the input image into uniform subbands. A number of significant subbands (high energy subbands) are concatenated together to form a subband-image cube. The kernel RX algorithm is then applied to these subband-image cubes obtained from wavelet decomposition of the LW database images. Experimental results are presented for the proposed kernel wavelet-RX, wavelet-RX and the classical CFAR algorithm for detecting anomalies (targets) in a large database of LW imagery. The ROC plots show that the proposed kernel wavelet-RX algorithm outperforms the wavelet-RX as well as the classical CFAR detector.