Accidents often occur at sea, so effective maritime search and rescue is essential. In the current process of sea search and rescue, the operation efficiency of large search and rescue equipment is ...low and it cannot provide stable communication link. In this article, unmanned aerial vehicles (UAVs) and unmanned surface vehicles (USVs) are used to form a cognitive mobile computing network for co-operative search and rescue, and reinforcement learning (RL) is used to plan search path and improve communication throughput. Based on the scene of marine search and rescue, the grid method is used to model the search and rescue area. Meanwhile, an intragroup communication architecture based on UAVs and USVs is designed to assist intragroup communication by recognizing the link channel state between UAVs. Search and rescue path planning is carried out through the strategy iteration of Markov decision process (MDP). Furthermore, distributed RL is used to recognize the channel state and perform mobile computing, so as to optimize the data throughput in the communication group. The simulation results show that we have successfully completed the path planning task. Compared with conventional methods, RL based on different reward functions has better throughput performance under the same number of UAVs auxiliary communications.
This letter considers secrecy simultaneous wireless information and power transfer (SWIPT) in full-duplex (FD) systems. In such a system, FD capable base station (FD-BS) is designed to transmit data ...to one downlink user and concurrently receive data from one uplink user, while one idle user harvests the radio-frequency (RF) signals' energy to extend its lifetime. Moreover, to prevent eavesdropping, artificial noise (AN) is exploited by FD-BS to degrade the channel of the idle user, as well as to provide energy supply to the idle user. To maximize the weighted sum of downlink secrecy rate and uplink secrecy rate, we jointly optimize the information covariance matrix, AN covariance matrix, and receiver vector, under the constraints of the sum transmission power of FD-BS and the minimum harvested energy of the idle user. Since the problem is nonconvex, the log-exponential reformulation and sequential parametric convex approximation (SPCA) method are used. Extensive simulation results are provided and demonstrate that our proposed FD scheme extremely outperforms the half-duplex scheme.
Unmanned aerial vehicles (UAVs) are widely used in Internet-of-Things (IoT) networks, especially in remote areas where communication infrastructure is unavailable, due to flexibility and low cost. ...However, the joint optimization of locations of UAVs and relay path selection can be very challenging, especially when the numbers of IoT devices and UAVs are very large. In this paper, we formulate the joint optimization of UAV locations and relay paths in UAV-relayed IoT networks as a graph problem, and propose a graph neural network (GNN)-based approach to solve it in an efficient and scalable way. In the training procedure, we design a reinforcement learning-based relay GNN (RGNN) to select the best relay path for each user. The theoretical analysis shows that the time complexity of RGNN is two orders lower than the conventional optimization method. Then, we jointly exploit location GNN (LGNN) and RGNN trained to optimize the locations of all UAVs. Both GNNs can be trained without relying on the training data, which is usually unavailable in the context of wireless networks. In inference procedure, LGNN is first used to optimize the location of UAVs, and then RGNN is used to select the best relay path based on the output of LGNN. Simulation results show that the proposed approach can achieve comparable performance to brute-force search with much lower time complexity when the network is relatively small. Remarkably, the proposed approach is highly scalable to large-scale networks and adaptable to dynamics in the environment, which can hardly be achieved using conventional methods.
The optical
PT
-symmetric and
PT
-antisymmetric characteristics are simultaneously investigated in the four-level atoms trapped in one-dimensional and two-dimensional optical lattices. The atoms with ...Gaussian-distributed density are driven into the closed-loop configuration by a standing wave field, two microwave fields and a probe field, where the relative phase of the external fields plays an essential role in changing the effective polarizability of such a system. It is found that the relative phase could lead to the switching from absorption to gain accompanied by the larger dispersion and, meanwhile, induce the giant gain with the simultaneous presence of the positive and negative dispersion. With the aid of spatial atomic density and standing wave modulation, both the optical
PT
symmetry and
PT
antisymmetry are achieved in one-dimensional and two-dimensional optical lattices, where the
PT
antisymmetry with the gain is presented. Furthermore, changing the other parameters such as Rabi frequencies and probe detuning has an impact on the realization of
PT
symmetry and
PT
antisymmetry, which may have some important applications in quantum information processing.
Graphical abstract
Internet of remote things (IoRT) networks are regarded as an effective approach for providing services to smart devices, which are often remote and dispersed over in a wide area. Due to the fact that ...the ground base station deployment is difficult and the power consumption of smart devices is limited in IoRT networks, the hierarchical Space-Air-Ground architecture is very essential for these scenarios. This paper aims to investigate energy efficient resource allocation problem in a two-hop uplink communication for Space-Air-Ground Internet of remote things (SAG-IoRT) networks assisted with unmanned aerial vehicle (UAV) relays. In particular, the optimization goal of this paper is to maximize the system energy efficiency by jointly optimizing sub-channel selection, uplink transmission power control and UAV relays deployment. The optimization problem is a mixinteger non-linear non-convex programming, which is hard to tackle. Therefore, an iterative algorithm that combines two sub-problems is proposed to solve it. First, given UAV relays deployment position, the optimal sub-channel selection and power control policy are obtained by the Lagrangian dual decomposition method. Next, based on the obtained sub-channel allocation and power control policy, UAV relays deployment is obtained by successive convex approximation (SCA). These two sub-problems are alternatively optimized to obtain the maximum system energy efficiency. Numerical results verify that the proposed algorithm has at least 21.9% gain in system energy efficiency compared to the other benchmark scheme.
Device-to-device (D2D) communication with increased spectral efficiency and reduced communication delay has undoubtedly become a general trend in future wireless networks. However, when D2D ...communication is incorporated into small cell networks (SCNs) with large number of randomly overlapped small cells, the co-channel interference between small cell users (SUEs) and D2D users is an inevitable challenge, especially with the heterogeneous spectrum, i.e., licensed spectrum bands and unlicensed spectrum bands. In this paper, we study the downlink channel allocation in D2D-assisted small cell networks with heterogeneous spectrum bands. By taking the required data rate of users and the interference constraint of SUEs into account, we formulate a channel allocation problem integrating channel selection and channel sharing to maximize the network utility, which is the service satisfaction of all users. To derive the solution, we decompose the optimization problem into two games: a potential game and a coalition game. Then, a potential game-based scheme using an interference graph and a coalition scheme with D2D user transferring is proposed to solve these two games, respectively. Based on these schemes, a two-stage distributed channel allocation algorithm is proposed and can converge with low computational complexity. Moreover, the simulation results reveal that the proposed algorithm could achieve high system throughput and network utility.
Abstract
Interlayer decoupling plays an essential role in realizing unprecedented properties in atomically thin materials, but it remains relatively unexplored in the bulk. It is unclear how to ...realize a large crystal that behaves as its monolayer counterpart by artificial manipulation. Here, we construct a superlattice consisting of alternating layers of NbSe
2
and highly porous hydroxide, as a proof of principle for realizing interlayer decoupling in bulk materials. In (NaOH)
0.5
NbSe
2
, the electric decoupling is manifested by an ideal 1D insulating state along the interlayer direction. Vibration decoupling is demonstrated through the absence of interlayer models in the Raman spectrum, dominant local modes in heat capacity, low interlayer coupling energy and out-of-plane thermal conductivity (0.28 W/mK at RT) that are reduced to a few percent of NbSe
2
’s. Consequently, a drastic enhancement of CDW transition temperature (>110 K) and Pauling-breaking 2D superconductivity is observed, suggesting that the bulk crystal behaves similarly to an exfoliated NbSe
2
monolayer. Our findings provide a route to achieve intrinsic 2D properties on a large-scale without exfoliation.
In this paper, to deal with the heterogeneity in federated learning (FL) systems, a knowledge distillation (KD) driven training framework for FL is proposed, where each user can select its neural ...network model on demand and distill knowledge from a big teacher model using its own private dataset. To overcome the challenge of train the big teacher model in resource limited user devices, the digital twin (DT) is exploit in the way that the teacher model can be trained at DT located in the server with enough computing resources. Then, during model distillation, each user can update the parameters of its model at either the physical entity or the digital agent. The joint problem of model selection and training offloading and resource allocation for users is formulated as a mixed integer programming (MIP) problem. To solve the problem, Q-learning and optimization are jointly used, where Q-learning selects models for users and determines whether to train locally or on the server, and optimization is used to allocate resources for users based on the output of Q-learning. Simulation results show the proposed DT-assisted KD framework and joint optimization method can significantly improve the average accuracy of users while reducing the total delay.
In this paper, a downlink device-to-device (D2D)-assisted cellular networks with mobile edge caching, where most popular video files are independently cached in D2D users and cellular base station ...(BS), are studied. In the considered system model, each user may obtain the requested video from the cache of BS or/and D2D users surrounding them. According to the different collaborative schemes of BS caching and D2D caching, it can be divided into two different resource allocation schemes. In the hybrid caching transmission scheme, users could adopt the BS caching mode or alternatively the D2D caching mode. In the joint caching transmission scheme, each user may obtain the requested files from the BS server and the adjacent D2D users, simultaneously. By taking the required data rate and the interference constraint into account, we formulate two joint resource allocation problems integrating link selection, channel allocation, and power control to maximize the system energy efficiency (EE). Leveraging on the Dinkelbach method, the EE optimization problems are transformed into mixed-integer nonlinear programming problems and can be decomposed into three subproblems: link selection, channel allocation, and power control. To solve these complicated problems, we propose two optimization algorithms that consist of a modified branch and bound method as well as Lagrange dual decomposition approach. The simulation results demonstrate the superiority of these two proposed algorithms in improving system throughput and EE compared with other algorithms.
PDF
