Exploring machine learning solutions for overcoming challenges in IoT-based wireless sensor network routing: a comprehensive review

Wireless Networks, Journal Year: 2024, Volume and Issue: 30(4), P. 2647 - 2673

Published: Feb. 29, 2024

Language: Английский

---

Secure Communication of Military Reconnaissance Images Over UAV-Assisted Relay Networks

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 78589 - 78610

Published: Jan. 1, 2024

This article presents a novel 2-layer image encryption cryptosystem and transmission protocol designed for secure communication of military reconnaissance images over unmanned aerial vehicle (UAV)-assisted relaying networks. The proposed scheme aims to address the growing need robust, secure, efficient sensitive imagery data across wireless ad-hoc networks, often characterized by unpredictable hostile environments. first layer employs Genetic Algorithm (GA) utilizing Mersenne Twister (MT) key, providing robust framework initial encryption. second further leverages security employing DNA coding, that is also driven MT key. encrypted are subsequently transformed into one-dimensional bit-stream, ready transmission. bit-stream then channel coded using either convolutional code or low-density parity-check (LDPC) code, offering flexibility based on specific network conditions requirements. BPSK-modulated transmitted multi-hop relay network, with relays mounted freely-moving UAVs. optimizes both efficiency, critical time-sensitive mission-critical nature operations. Extensive performance evaluation carried out, presenting bit error rate (BER) curves various metrics, demonstrating robustness, reliability, scheme. contribution expected significantly enhance images, paving way more advanced, systems in sector.

Language: Английский

---

A Fairness-Enhanced Federated Learning Scheduling Mechanism for UAV-Assisted Emergency Communication

Sensors, Journal Year: 2024, Volume and Issue: 24(5), P. 1599 - 1599

Published: Feb. 29, 2024

As the frequency of natural disasters increases, study emergency communication becomes increasingly important. The use federated learning (FL) in this scenario can facilitate collaboration between devices while protecting privacy, greatly improving system performance. Considering complex geographic environment, flexible mobility and large radius unmanned aerial vehicles (UAVs) make them ideal auxiliary for wireless communication. Using UAV as a mobile base station better provide stable signals. However, number ground-based IoT terminals is closely distributed, so if all transmit data to UAV, will not be able take on computation tasks because its limited energy. In addition, there competition spectrum resources among many terrestrial devices, transmitting bring about an extreme shortage resources, which lead degradation model This indelible damage rescue disaster area threaten life safety vulnerable injured. Therefore, we user scheduling select some participate FL process. order avoid resource waste generated by device prediction, multi-armed bandit (MAB) algorithm equipment evaluation. fairness issue selection, try replace single criterion with multiple criteria, using freshness energy consumption weighting reward functions. state art our approach demonstrated simulations datasets.

Language: Английский

---

Task Offloading Strategy for Unmanned Aerial Vehicle Power Inspection Based on Deep Reinforcement Learning

Sensors, Journal Year: 2024, Volume and Issue: 24(7), P. 2070 - 2070

Published: March 24, 2024

With the ongoing advancement of electric power Internet Things (IoT), traditional inspection methods face challenges such as low efficiency and high risk. Unmanned aerial vehicles (UAVs) have emerged a more efficient solution for inspecting facilities due to their maneuverability, excellent line-of-sight communication capabilities, strong adaptability. However, UAVs typically grapple with limited computational energy resources, which constrain effectiveness in handling computationally intensive latency-sensitive tasks. In response this issue, we propose UAV task offloading strategy based on deep reinforcement learning (DRL), is designed scenarios consisting mobile edge computing (MEC) servers multiple UAVs. Firstly, an innovative UAV-Edge server collaborative architecture fully exploit mobility high-performance capabilities MEC servers. Secondly, established model concerning consumption processing latency system, enhancing our understanding trade-offs involved strategies. Finally, formalize problem multi-objective optimization issue simultaneously it Markov Decision Process (MDP). Subsequently, proposed algorithm Deep Deterministic Policy Gradient (OTDDPG) obtain optimal The simulation results demonstrated that approach outperforms baseline significant improvements consumption.

Language: Английский

---

Unmanned Aerial Vehicles based low-altitude economy with lifecycle techno-economic-environmental analysis for Sustainable and Smart Cities

Journal of Cleaner Production, Journal Year: 2025, Volume and Issue: unknown, P. 145050 - 145050

Published: Feb. 1, 2025

Language: Английский

---

Fundamentals and Applications of Unmanned Aerial Vehicle Swarms

Apress eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 523 - 590

Published: Jan. 1, 2025

Language: Английский

---

Towards Achieving Energy Efficiency and Service Availability in 6G O-RAN via Formal Verification

Lecture notes in computer science, Journal Year: 2025, Volume and Issue: unknown, P. 137 - 157

Published: Jan. 1, 2025

Language: Английский

---

QoS-Driven Slicing Management for Vehicular Communications

Electronics, Journal Year: 2024, Volume and Issue: 13(2), P. 314 - 314

Published: Jan. 10, 2024

Network slicing is introduced for elastically instantiating logical network infrastructure isolation to support different application types with diversified quality of service (QoS) class indicators. In particular, vehicular communications are a trending area that consists massive mission-critical applications in the range safety-critical, intelligent transport systems, and on-board infotainment. Slicing management can be achieved if has computing sufficiency, dynamic control policy, elastic resource virtualization, cross-tier orchestration. To functionality management, incorporating core deep learning reinforcement become hot topic researchers practitioners analyzing traffic/resource patterns before orchestrating steering policies. this paper, we propose QoS-driven by considering (edge) block utilization, scheduling, slice instantiation three-tier placement, namely, small base stations/access points, macro stations, networks. The proposed scheme integrates recurrent neural networks trigger hidden states availability predict output QoS. agent controller, RDQ3N, gathers from observations optimizes action on allocation scheduling algorithms. Experiments conducted both physical virtual representational vehicle-to-everything (V2X) environments; furthermore, requests set thresholds rendering V2X congestion flow entries.

Language: Английский

---

Multi-Agent-Deep-Reinforcement-Learning-Enabled Offloading Scheme for Energy Minimization in Vehicle-to-Everything Communication Systems

Electronics, Journal Year: 2024, Volume and Issue: 13(3), P. 663 - 663

Published: Feb. 5, 2024

Offloading computation-intensive tasks to mobile edge computing (MEC) servers, such as road-side units (RSUs) and a base station (BS), can enhance the computation capacities of vehicle-to-everything (V2X) communication system. In this work, we study an MEC-assisted multi-vehicle V2X system in which multi-antenna RSUs with liner receivers BS zero-forcing (ZF) receiver work MEC servers jointly offload vehicles. To control energy consumption ensure delay requirement system, minimization problem under constraint is formulated. The multi-agent deep reinforcement learning (MADRL) algorithm proposed solve non-convex optimization problem, train vehicles select beneficial server association, transmit power offloading ratio intelligently according reward function related consumption. improved K-nearest neighbors (KNN) assign specific RSU, reduce action space dimensions complexity MADRL algorithm. Numerical simulation results show that scheme decrease while satisfying constraint. When adopt indirect transmission mode are equipped matched-filter (MF) receivers, joint by 56.90% 65.52% compared maximum full schemes, respectively. ZF 36.8% MF receivers.

Language: Английский

---

Survey of Advanced Nonlinear Control Strategies for UAVs: Integration of Sensors and Hybrid Techniques

Sensors, Journal Year: 2024, Volume and Issue: 24(11), P. 3286 - 3286

Published: May 21, 2024

This survey paper explores advanced nonlinear control strategies for Unmanned Aerial Vehicles (UAVs), including systems such as the Twin Rotor MIMO system (TRMS) and quadrotors. UAVs, with their high nonlinearity significant coupling effects, serve crucial benchmarks testing algorithms. Integration of sophisticated sensors enhances UAV versatility, making traditional linear techniques less effective. Advanced strategies, sensor-based adaptive controls AI, are increasingly essential. Recent years have seen development diverse sliding surface-based, sensor-driven, hybrid offering superior performance over methods. reviews significance these emphasizing role in addressing complexities outlining future research directions.

Language: Английский

---