1,721,345 research outputs found
Stochastic Optimization of Cognitive Networks
No full textIn this paper we aim to propose, upon a statistical modeling of the spectrum sensing energy, a stochastic joint optimization method that allows the minimization of the energy consumption of the spectrum sensing of a multi-hop secondary network subject to constraints on the detection performance and the number of network hops, in a trade-off between the overall probability of missed detection and false alarm, and the energy consumption. The optimal closed-form solution of the optimization problem is computed by means of two approaches: worst case and stochastic approach. Both theoretical analysis and numerical results show that the proposed method allows reducing the energy consumption, by showing its effectiveness with different data fusion rules. Particularly, the optimal solution outperforms the existing ones in terms of computational complexity, and of energy consumption specially for a number of hops greater than 4. The proposed technique has been finally proven in several environments that characterize different primary operative scenarios, such as wireless metropolitan area networks and satellite communications in the presence of interference with very low signal-to-noise ratio
Hierarchical Reinforcement Learning for Multi-layer Multi-Service Non-Terrestrial Vehicular Edge Computing
Full text linkVehicular Edge Computing (VEC) represents a novel advancement within the Internet of Vehicles (IoV). Despite its implementation through Road Side Units (RSUs), VEC frequently falls short of satisfying the escalating demands of Vehicle Users (VUs) for new services, necessitating supplementary computational and communication resources. Non-Terrestrial Networks (NTN) with onboard Edge Computing (EC) facilities are gaining a central place in the 6G vision, allowing one to extend future services also to uncovered areas. This scenario, composed of a multitude of VUs, terrestrial and non-terrestrial nodes, and characterized by mobility and stringent requirements, brings in a very high complexity. Machine Learning (ML) represents a perfect tool for solving these types of problems. Integrated Terrestrial and Non-terrestrial (T-NT) EC, supported by innovative intelligent solutions enabled through ML technology, can boost the VEC capacity, coverage range, and resource utilization. Therefore, by exploring the integrated T-NT EC platforms, we design a multi-EC-enabled vehicular networking platform with a heterogeneous set of services. Next, we model the latency and energy requirements for processing the VU tasks through partial computation offloading operations. We aim to optimize the overall latency and energy requirements for processing the VU data by selecting the appropriate edge nodes and the offloading amount. The problem is defined as a multi-layer sequential decision-making problem through the Markov Decision Processes (MDP). The Hierarchical Reinforcement Learning (HRL) method, implemented through a Deep Q network, is used to optimize the network selection and offloading policies. Simulation results are compared with different benchmark methods to show performance gains in terms of overall cost requirements and reliability
A Time-Continuous Federated Learning Framework for Enabling Intelligent Applications Over Latency-Critical Aerial Networks
No full textDistributed Machine Learning (DML) methods are expected to play a crucial role in the forthcoming 6G era, with the goal of enabling ubiquitous connected intelligence. Distributed intelligence enabled through distributed computing environments, 6G technology, and big data can be extremely supportive of achieving the goals of emerging intelligent IoT applications in the proximity of end users. With this in mind, we propose an advanced Federated Learning (FL) approach for efficiently enabling intelligent applications over latency-critical networks in Non-Terrestrial environments. In the proposed solution, the client and server nodes reduce idle time using a parallel processing approach with the help of a replica of the training model. Next, the proposed FL framework is tested in a Python environment to show its effectiveness with respect to the traditional FL approach
Integrated Aerial-Ground Computation Offloading for Dependency-Aware IoV Multitask Services
No full textThe Internet of Vehicles (IoV) is a fundamental paradigm for enabling intelligent transportation systems and promoting high-quality services and applications that require a tremendous amount of data processing resources. In this paper, we consider a computational offloading problem on an Integrated Aerial-Ground (IAG) Edge Computing (EC) architecture, where each task is modeled as a chain of dependent subtasks. To solve the problem, a V2X-based Computation and Communication-efficient Multitask Offloading Approach (CCMTOA) is proposed in which mutual information is exchanged between users, allowing one to effectively solve the multitask multilayer network selection problem. The parameters of vehicle mobility are estimated using a realistic intelligent mobility model. The numerical results with varying VU density show the effectiveness of the proposed method over the benchmark approaches
Collaborative Reinforcement Learning for Multi-Service Internet of Vehicles
Full text linkInternet of Vehicles (IoV) is a recently introduced paradigm aiming at extending the Internet of Things (IoT) toward the vehicular scenario in order to cope with its specific requirements. Nowadays, there are several types of vehicles, with different characteristics, requested services, and delivered data types. In order to efficiently manage such heterogeneity, Edge Computing facilities are often deployed in the urban environment, usually co-located with the Roadside Units (RSUs), for creating what is referenced as Vehicular Edge Computing (VEC). In this paper, we consider a joint network selection and computation offloading optimization problem in multi-service VEC environments, aiming at minimizing the overall latency and the consumed energy in an IoV scenario. Two novel collaborative Q-learning based approaches are proposed, where Vehicle-to-Infrastructure (V2I) and Vehicle-to-Vehicle (V2V) communication paradigms are exploited, respectively. In the first approach, we define a collaborative Q-learning method in which, through V2I communications, several vehicles participate in the training process of a centralized Q-agent. In the second approach, by exploiting the V2V communications, each vehicle is made aware of the surrounding environment and the potential offloading neighbors, leading to better decisions in terms of network selection and offloading. In addition to the tabular method, an advanced deep learning-based approach is also used for the action value estimation, allowing to handle more complex vehicular scenarios. Simulation results show that the proposed approaches improve the network performance in terms of latency and consumed energy with respect to some benchmark solutions
Multi-Time-Scale Markov Decision Process for Joint Service Placement, Network Selection, and Computation Offloading in Aerial IoV Scenarios
No full textVehicular Edge Computing (VEC) is considered a major enabler for multi-service vehicular 6G scenarios. However, limited computation, communication, and storage resources of terrestrial edge servers are becoming a bottleneck and hindering the performance of VEC-enabled Vehicular Networks (VNs). Aerial platforms are considered a viable solution allowing for extended coverage and expanding available resources. However, in such a dynamic scenario, it is important to perform a proper service placement based on the users' demands. Furthermore, with limited computing and communication resources, proper user-server assignments and offloading strategies need to be adopted. Considering their different time scales, a multi-time-scale optimization process is proposed here to address the joint service placement, network selection, and computation offloading problem effectively. With this scope in mind, we propose a multi-time-scale Markov Decision Process (MDP) based Reinforcement Learning (RL) to solve this problem and improve the latency and energy performance of VEC-enabled VNs. Given the complex nature of the joint optimization process, an advanced deep Q-learning method is considered. Comparison with various benchmark methods shows an overall improvement in latency and energy performance in different VN scenarios
Network Selection and Computation Offloading in Non-Terrestrial Network Edge Computing Environments for Vehicular Applications
No full textThe Edge Computing (EC) paradigm and the Internet of Things (IoT) have transformed the conventional vehicular network (VN) into a highly reliable, intelligent, and complex networking system serving users with heterogeneous services. However, the traditional terrestrial network-based EC facilities, usually referred to as Vehicular Edge Computing (VEC), enabled through the Road Side Units (RSU) deployments, have limited resources, higher deployment costs, limited coverage, and can rapidly become a bottleneck for the VNs performance. On the other hand, various Non-terrestrial Networking (NTN) platforms from air and space networks have gained a lot of attention in 5G and beyond studies and are expected to play a key role in the upcoming days. Integration of NTN-based EC facilities into the current VEC system can be useful for serving vehicular users with different service types. In this work, we first design a multi-EC enabled vehicular networking platform for serving vehicular users (VUs) with a heterogeneous set of services. We model the various latency and energy requirements for processing the VUs task requests through partial computation offloading operations. We further aim at minimizing the overall latency and energy requirements for processing the VUs data by selecting the proper ENs and the offloading amounts over a multi-EC enabled VN. The problem is modeled as a constrained optimization problem and an evolutionary search-based metaheuristic approach is used to solve it. The results are compared with various benchmark methods for showing the performance gain
Multi-Layer Distributed Learning for Intelligent Transportation Systems in 6G Aerial-Ground Integrated Networks
No full textFederated Learning (FL) is a widely used distributed learning (DL) method for intelligent transportation systems (ITS) in the upcoming era of 6G-enabled ITS. In this work, we present the concept of Generalized Federated Split Transfer Learning (GFSTL) as a highly efficient and secure distributed learning framework for resource-limited ITS applications. The proposed GFSTL solution performs better in terms of overall training latency and accuracy and is useful for enabling ITS services in Aerial-Ground Integrated Networks (AGIN). Through comprehensive simulations carried out in vehicular scenarios, our results validate the efficacy of GFSTL on multilayered DL using Road-Side Units (RSUs) and High-Altitude Platforms (HAPs) in AGIN, demonstrating significant improvements in addressing the demands of intelligent vehicular networks. Through the integration of advanced DL techniques and the use of HAPs, our proposed framework holds promise for paving the way for an intelligent and connected vehicular network in the future
Distributed Learning Framework for Earth Observation on Multilayer Non-Terrestrial Networks
No full textA novel Distributed Learning (DL) framework called Generalized Federated Split Transfer Learning (GFSTL) is proposed on a multilayer Non-Terrestrial Network (NTN) for Earth Observation (EO) missions. Through this, significant gaps in the literature related to the use of multilayer NTNs and Machine Learning (ML) perspectives are addressed. Multiple layers are considered to collect images and data at different sizes and resolutions, Transfer Learning (TL) to accelerate training and improve accuracy, Federated Learning (FL) to facilitate safe and secure collaboration, and Split Learning (SL) to optimize resource use and preserve privacy. The proposed framework is expected to overcome limitations in existing techniques, offering enhanced accuracy, privacy preservation, and scalability
Enabling Intelligent Vehicular Networks Through Distributed Learning in the Non-Terrestrial Networks 6G Vision
No full textThe forthcoming 6G-enabled Intelligent Transportation System (ITS) is set to redefine conventional transportation networks with advanced intelligent services and applications. These technologies, including edge computing, Machine Learning (ML), and network softwarization, pose stringent requirements for latency, energy efficiency, and user data security. Distributed Learning (DL), such as Federated Learning (FL), is essential to meet these demands by distributing the learning process at the network edge. However, traditional FL approaches often require substantial resources for satisfactory learning performance. In contrast, Transfer Learning (TL) and Split Learning (SL) have shown effectiveness in enhancing learning efficiency in resource-constrained wireless scenarios like ITS. Non-terrestrial Networks (NTNs) have recently acquired a central place in the 6G vision, especially for boosting the coverage, capacity, and resilience of traditional terrestrial facilities. Air-based NTN layers, such as High Altitude Platforms (HAPs), can have added advantages in terms of reduced transmission distances and flexible deployments and thus can be exploited to enable intelligent solutions for latency-critical vehicular scenarios. With this motivation, in this work, we introduce the concept of Federated Split Transfer Learning (FSTL) in joint air-ground networks for resource-constrained vehicular scenarios. Simulations carried out in vehicular scenarios validate the efficacy of FSTL on HAPs in NTN, demonstrating significant improvements in addressing the demands of ITS applications
- …
