Unmanned aerial vehicles (UAVs) have been recognized as a promising technology to be used in a wide range of civilian, public and military applications. However, given their limited payload and flight time, multiple UAVs may have to be harnessed for accomplishing complex high-level tasks, where a control center can be employed for coordinating their actions. In this paper, we consider image classification tasks in UAV-aided exploration scenarios, where the coordination of multiple UAVs is implemented by a ground fusion center (GFC) positioned in a strategic, but inaccessible location, such as a mountain top, where recharging the battery is uneconomical or may even be infeasible. On-board cameras are carried by each UAV and then, federated learning (FL) is invoked for reducing the communication cost between the UAVs and the GFC, and the computational complexity imposed on the GFC. In our proposed FL-aided classification approach, initially local training is performed by each UAV based on the locally collected images to create a local model. Then, each UAV sends its locally acquired model to the GFC via a fading wireless channel, where a global model is generated, which is then fed back to each UAV for the next round of their local training. In order to further minimize the computational complexity imposed on the GFC by the UAVs, weighted zero-forcing (WZF) transmit precoding (TPC) is used at each UAV based on realistic imperfect channel state information (CSI). The system performance attained is evaluated by simulations, showing that the proposed system is capable of attaining a high classification accuracy at relatively low communication cost.

Building on the ubiquity of electric power infrastructure, power line communications (PLC) has been successfully used in diverse application scenarios, including the smart grid and in-home broadband communications systems as well as industrial and home automation. However, the power line channel exhibits deleterious properties, one of which is its hostile noise environment. This article aims for providing a review of noise modeling and mitigation techniques in PLC. Specifically, a comprehensive review of representative noise models developed over the past fifty years is presented, including both the empirical models based on measurement campaigns and simplified mathematical models. Following this, we provide an extensive survey of the suite of noise mitigation schemes, categorizing them into mitigation at the transmitter as well as parametric and non-parametric techniques employed at the receiver. Furthermore, since the accuracy of channel estimation in PLC is affected by noise, we review the literature of joint noise mitigation and channel estimation solutions. Finally, a number of directions are outlined for future research on both noise modeling and mitigation in PLC.

A spatial modulation-aided orthogonal time frequency space (SM-OTFS) scheme is proposed for high-Doppler scenarios, which relies on a low-complexity distance-based detection algorithm. We first derive the delay-Doppler (DD) domain input-output relationship of our SM-OTFS system by exploiting an SM mapper, followed by characterizing the doubly-selective channels considered. Then we propose a distance-based ordering subspace check detector (DOSCD) exploiting the a priori information of the transmit symbol vector. Moreover, we derive the discrete-input continuous-output memoryless channel (DCMC) capacity of the system. Finally, our simulation results demonstrate that the proposed SM-OTFS system outperforms the conventional single-input-multiple-output (SIMO)-OTFS system, and that the DOSCD conceived is capable of striking an attractive bit error ratio (BER) vs. complexity trade-off.

A cloud-based terrestrial-satellite network (CTSN) is conceived for supporting ubiquitous high-speed multimedia services. In the CTSN, the satellite and terrestrial base stations are connected to a cloud-computing based centralized processor (CP), where joint user scheduling and multicast beamforming are performed based on realistic imperfect channel state information (CSI). Specifically, pilot-assisted channel estimation is assumed. Then, we propose a successive convex approximation (SCA) based algorithm for generating the beamforming vectors at the CP, where specific quality of service (QoS) constraints are considered. In the proposed algorithm, the beamforming vectors are obtained by iteratively solving a convex optimization problem subject to tight convex constraints. We demonstrate that feasible solutions can be obtained by our algorithm, even for the case when the system's dimension is large. Both analytical and numerical results are provided for characterizing the performance of the CTSN. Our results qualify the tradeoff between the cooperation-aided multiantenna gain and the pilot overhead imposed by training the beamformers. Furthermore, the achievable rate of the CTSN is shown to be substantially eroded by the accuracy of CSI.

Frequency hopping has been proved to be effective against radio jamming attacks in wireless communications. In this article, deep reinforcement learning algorithm is applied for providing frequency hopping strategies against jamming attacks in wideband communication systems. We first model the frequency hopping communication system in a dynamic jamming environment, where a two-dimensional pattern with a certain number of time slots and channels are formulated. In particular, a jammer using multi-channel blocking jamming is considered, where the communication frequency band are attacked via probabilistic jamming patterns. In this case, an intelligent frequency hopping strategy is desired, especially when perfect knowledge of the jamming patterns is not known at the transmitter and receiver sides. In order to tackle this issue, the interaction between the users and the jammer are modeled as a Markov decision process and a deep Q-learning algorithm is proposed to solve the frequency hopping decision making problem. Finally, the system performance is evaluated by simulations. Our simulation results have shown that in comparison to Q-learning assisted frequency hopping strategy, the proposed deep Q-learning assisted frequency hopping strategy is capable of attaining a better anti-jamming performance, especially for a large number of frequency bands and long transmission time. Furthermore, the proposed deep Q-learning assisted frequency hopping strategy is able to provide robust anti-jamming performance when jamming patterns are unknown.