Tactical flying ad hoc networks (T-FANETs) are mission-critical networks that connect mobile airborne platforms, such as tactical unmanned aerial vehicles (TUAVs), to enhance operations at the tactical edge. These networks are vulnerable to adversarial attacks, node failures, intermittent connectivity, and frequent topology variations, which make packet routing challenging. Due to the complex nature of T-FANETs, classical routing methods, which rely on sequential processing, require longer processing times to discover optimal routes, which can hinder mission effectiveness in fast-paced military operations. In this paper, we propose hybrid quantum federated learning (HQFL) to address the high computational complexity and long processing times of classical approaches. The proposed HQFL-based routing solution uses the computational power of quantum computing (QC) and the scalability of classical neural networks to develop routing agents with hybrid quantum circuits (HQCs) in order to accelerate the learning of optimal routes. Unmanned surface vehicles (USVs) are then used as parameter servers to collaboratively train lightweight HQC models for routing in T-FANETs operating in maritime environments. In addition, we design an adversarial model that combines mobile jamming and physical attacks on TUAVs and evaluate the performance of the proposed routing scheme under these attacks. Our simulation results show that the proposed HQFL-based approach outperforms classical FL algorithms, reducing communication overhead by 95%, lowering end-to-end delay by 9.43%, and improving the packet delivery ratio and routing energy efficiency by 39.6% and 52.6%, respectively.

Reliable low Earth orbit satellite networks (LEO-SNs) should be capable of optimally adapting to dynamic environments and remaining resilient against smart jamming attacks. In this context, dynamic inter-satellite link (ISL) routing is crucial for enabling efficient and adaptive data transmission across any satellite network during smart jamming attacks. However, due to the environmental variability, ISL routing becomes a complex time-sequential optimization problem. Accordingly, in this study, we propose a digital twin-assisted federated quantum deep reinforcement learning (DT-FQDRL) framework to solve dynamic ISL routing with faster convergence and minimal-error solutions. The DT-FQDRL framework optimizes ISL routing by minimizing jamming success rate and total delay while maximizing energy efficiency. Specifically, a digital twin (DT) replicates the LEO-SN environment to simulate satellite interactions and jamming behaviors at each time step. In this virtual setting, each satellite employs quantum deep reinforcement learning (QDRL) for local training and long-term prediction. To enhance data privacy and prevent node conflicts, a hierarchical federated learning scheme aggregates local QDRL models within the DT. The optimized weights are then transferred to real satellites. Our numerical results demonstrate that the DT-FQDRL framework reduces jamming success rate by 48.16%, decreases total delay by 22.26%, and improves energy efficiency by 6.17% over existing benchmarks.

In recent years, the demand for seamless connectivity and highly efficient, reliable network services for low earth orbit (LEO) satellites has escalated. To meet these expectations, a critical issue that must be addressed is combating malicious jamming attacks on satellite networks, which occur due to the open nature of satellite-ground connections. Moreover, in the era of artificial intelligence (AI), AI-based jamming poses a severe threat to the security of satellite networks and disrupts secure communications, particularly given the dynamic movements of LEO satellites and the time-sequential complexity of such attacks. Accordingly, this paper proposes a quantum adversarial deep reinforcement learning (QADRL) approach to mitigate AI-based jamming attacks while enhancing the quality-of-service (QoS) for LEO satellite networks. Specifically, the proposed QADRL approach is based on a zero-sum Markov game utilizing two opposing learning networks: one optimizing satellite routing links to avoid jamming and improve QoS, while the other, focuses on the jammer, optimizes the trajectory, jamming nodes, and power of unmanned aerial vehicles (UAVs) to maximize jamming success. The results demonstrate that the proposed QADRL outperforms classical adversarial DRL (CADRL) by reducing the jamming success rate by 33.33% and increasing the average QoS of the satellite network by 18.4975%.

A broader coverage of low earth orbit (LEO) satellite networks has been an object of research interest in recent years and will remain to be of keen interest to both start-ups and established companies in the future. However, achieving optimal coverage remains a challenge today, as the satellites dynamically move along their orbit, requiring frequent constellation size and beamsteering adjustments. Such frequent adjustments and the number of variables that need to be optimized result in a high computational complexity. Aiming to attend to these concerns, the present study proposes a quantum adaptive learning (QAL) as a potential solution for coverage optimization of stochastic geometry-based LEO satellite networks with low computational complexity by taking advantage of quantum computing and adaptive learning. The QAL scheme utilizes quantum computing and a feedback mechanism to improve the learning network design and parameters for higher accuracy with low computational complexity. As a study case, considering the binomial point process (BPP) distribution of contact distance between satellites and user terminals, we use the QAL scheme to optimize the satellites' constellation size and their corresponding beamsteering at each time step to achieve maximum coverage probability. To evaluate performance, the simulation results of the QAL scheme are compared with those of quantum machine learning (QML), which lacks feedback and adaptive mechanisms. The results show that the proposed QAL outperforms QML by achieving higher coverage probability, higher accuracy, and faster convergence.

This study proposes a layerwise quantum-based deep reinforcement learning (LQ-DRL) method for optimizing continuous large space and time series problems using deep-layer training. The actions in LQ-DRL are optimized using a layerwise quantum embedding that leverages the advantages of quantum computing to maximize reward and reduce training loss. Moreover, this study employs a local loss to minimize the occurrence of barren plateaus phenomena and further enhance performance. As a particular case, the proposed scheme is employed to jointly optimize: 1) unmanned aerial vehicle (UAV) trajectory planning; 2) user grouping; and 3) power allocation for higher energy efficiency of a UAV as the reward. The combination of these optimized factors is referred to as action space in the presented LQ-DRL. The LQ-DRL is employed to solve the optimization problem due to its nonconvexity, continuous and large action space, and time-series domain. In a practical view, LQ-DRL aims to solve the issue of energy consumption related to limited-battery energy of a UAV base station (BS) while maintaining Quality of Service (QoS) for users, by gaining maximum energy efficiency as the reward. One of real applications, as an example, LQ-DRL can be employed to maximize the energy efficiency of a UAV BS in UAV empowered disaster recovery networks scenario. The quantum circuits of layerwise quantum embedding are presented to show the practical implementation in noisy intermediate-scale quantum computers. Based on the results, LQ-DRL outperformed the classical DRL by achieving higher effective dimension, rewards, and lower learning losses. In addition, better performances were achieved using more layers.

This study proposes a double deep reinforcement learning (D-DRL) to improve an index of rate fairness and sum-rate in UxV-enabled multi-user communication systems. In this study, a UxV-assisted multi-user communications scenario is considered. By taking into account the tradeoff between the two objectives and the time-sequential movement of the UxV, two DRL-based actor-critic networks are integrated to solve the designated problem. In the first actor-critic network, the rate fairness is maximized by jointly optimizing a hybrid precoder with a UxV trajectory. Subsequently, considering the rate fairness as a learning reward of the first network, sum-rate is maximized in the second network under the consideration of transmission power budgets, limited UxV battery capacity, and quality of service (QoS) constraints. The results show that the D-DRL which considered rate fairness outperformed DRL which did not by achieving maximum rate fairness and a higher sum-rate.

This article proposes a multidimensional search space (or directional space) with more degrees-of-freedom (DOFs) to increase the energy efficiency of limited-battery-powered unmanned aerial vehicle (UAV) in the Internet of Things (IoT) data collection scenario. In this article, the UAV navigates from the initial to the goal point while collecting data from IoT sensors on the ground. Owing to the limited battery power of UAVs, an optimized trajectory is a crucial practical problem. Based on the available directional space, the direction of the UAV related to the navigation trajectory is optimized using reinforcement learning (RL). The objective of RL is to maximize the energy efficiency of the UAV as a long-term reward by selecting the optimal direction. Moreover, a practical energy consumption model and environment are presented in this article. Simulation results verified that the proposed multidimensional trajectory for UAV achieves higher energy efficiency compared with benchmark models.