用亚图参数与回归技术估计和预测烷烃的核磁共振碳谱

系统研究了分子建模在波谱分析中的应用.采用多元线性回归算法(MLR)估计和预测了60余种烷烃的碳谱化学位移.烷烃中碳原子由十余种对应于所谓根亚树的相嵌频率描述子所决定.这些描述子等于由2~5个碳原子组成的更小结构骨架组成.说明了所用描述子作为很有用的工具可适当地描述烷烃中碳所处微观环境.同时还比较了与神经网络的计算结果.     

Parameter estimation of continuous variable quantum key distribution system via artificial neural networks

Continuous-variable quantum key distribution(CVQKD)allows legitimate parties to extract and exchange secret keys.However,the tradeoff between the secret key rate and the accuracy of parameter estimation still around the present CVQKD system.In this paper,we suggest an approach for parameter estimation of the CVQKD system via artificial neural networks(ANN),which can be merged in post-processing with less additional devices.The ANN-based training scheme,enables key prediction without exposing any raw key.Experimental results show that the error between the predicted values and the true ones is in a reasonable range.The CVQKD system can be improved in terms of the secret key rate and the parameter estimation,which involves less additional devices than the traditional CVQKD system.     

Learnable three-dimensional Gabor convolutional network with global affinity attention for hyperspectral image classification

Benefiting from the development of hyperspectral imaging technology, hyperspectral image (HSI) classification has become a valuable direction in remote sensing image processing. Recently, researchers have found a connection between convolutional neural networks (CNNs) and Gabor filters. Therefore, some Gabor-based CNN methods have been proposed for HSI classification. However, most Gabor-based CNN methods still manually generate Gabor filters whose parameters are empirically set and remain unchanged during the CNN learning process. Moreover, these methods require patch cubes as network inputs. Such patch cubes may contain interference pixels, which will negatively affect the classification results. To address these problems, in this paper, we propose a learnable three-dimensional (3D) Gabor convolutional network with global affinity attention for HSI classification. More precisely, the learnable 3D Gabor convolution kernel is constructed by the 3D Gabor filter, which can be learned and updated during the training process. Furthermore, spatial and spectral global affinity attention modules are introduced to capture more discriminative features between spatial locations and spectral bands in the patch cube, thus alleviating the interfering pixels problem. Experimental results on three well-known HSI datasets (including two natural crop scenarios and one urban scenario) have demonstrated that the proposed network can achieve powerful classification performance and outperforms widely used machine-learning-based and deep-learning-based methods.     

Group mobility assisted network selection framework in 5G vehicular cognitive radio networks

The heterogeneity nature of networks is the most eminent characteristic in 5G vehicular cognitive radio networks across complex radio environments. Since multiple communicating radios may be in motion at the same time in a vehicle. So, group mobility is the most prominent characteristic that requires to be a deep investigation. Therefore, different communication radios that are moving on a train/bus needed to select the networks simultaneously. Without considering the group mobility feature, there is a possibility that the same network may be selected by each moving node and cause congestion in a particular network. To overcome this problem, a novel network selection technique considering the group mobility feature is proposed to improve the throughput of the network. In this work, a 5G vehicular cognitive radio network scenario is also realized using USRP-2954 and LabVIEW communications system design suite testbed. The performance metrics like transmission delay, packet loss rate, reject rate and, channel utilization for vehicular nodes, are gained to analyze the proposed technique in vehicular cognitive radio networks environment. The proposed technique demonstrates a remarkable improvement in channel utilization for vehicular nodes and outperformed conventional schemes.     

Mobile sink-based data collection in event-driven wireless sensor networks using a modified ant colony optimization

The hotspot problem is one of the primary challenges in the wireless sensor networks (WSNs) because it isolates the sink node from the remaining part of the WSN. A mobile sink (MS)-based data acquisition strategy mitigates the hotspot problem, but the traditional MS-based data gathering approaches do not resolve the issue. However, the conventional techniques follow a fixed order of visits and static traversal of the MS. In this context, this paper uses a modified version of the ant colony optimization strategy for the data collected through a MS to mitigate the hotspot problem in the WSNs while improving the energy efficiency, network lifetime, throughput by reducing the packet loss and delay. In our work, we initially construct a forwarded load spanning tree to estimate the freight of each node in the WSN. Further, we choose RPs and their path simultaneously using the modified ACO algorithm by considering the forward loads, remaining energy, distance, etc. The proposed work also adopts the virtual RP selection strategy void unnecessary data exchanges between the nodes and RPs. Hence, it reduces the burden on relay nodes and optimize the energy usage among the nodes. We compare our approach with the recent ACO-based algorithms, and our approach outperforms them.     

A lightweight cell switching and traffic offloading scheme for energy optimization in ultra-dense heterogeneous networks

One of the major capacity boosters for 5G networks is the deployment of ultra-dense heterogeneous networks (UDHNs). However, this deployment results in a tremendous increase in the energy consumption of the network due to the large number of base stations (BSs) involved. In addition to enhanced capacity, 5G networks must also be energy efficient for it to be economically viable and environmentally friendly. Dynamic cell switching is a very common way of reducing the total energy consumption of the network, but most of the proposed methods are computationally demanding, which makes them unsuitable for application in ultra-dense network deployment with massive number of BSs. To tackle this problem, we propose a lightweight cell switching scheme also known as Threshold-based Hybrid cEll swItching Scheme (THESIS) for energy optimization in UDHNs. The developed approach combines the benefits of clustering and exhaustive search (ES) algorithm to produce a solution whose optimality is close to that of the ES (which is guaranteed to be optimal), but is computationally more efficient than ES and as such can be applied for cell switching in real networks even when their dimension is large. The performance evaluation shows that THESIS significantly reduces the energy consumption of the UDHN and can reduce the complexity of finding a near-optimal solution from exponential to polynomial complexity.     

Secrecy sum rate maximization for a MIMO-NOMA uplink transmission in 6G networks

Non-orthogonal multiple access (NOMA), as a well-qualified candidate for sixth-generation (6G) mobile networks, has been attracting remarkable research interests due to high spectral efficiency and massive connectivity. The aim of this study is to maximize the secrecy sum rate (SSR) for a multiple-input multiple-output (MIMO)-NOMA uplink network under the maximum total transmit power and quality of service (QoS) constraints. Thanks to the generalized singular value decomposition method, the SSR of NOMA is compared with conventional orthogonal multiple access and other baseline algorithms in different MIMO scenarios. Due to the subtractive and non-convex nature of the SSR problem, the first-order Taylor approximation is exploited to transform the original problem into a suboptimal concave problem. Simulation results are provided and compared with some other benchmarks to evaluate the efficacy of the proposed method.     

Game theory-based radio resource allocation in NOMA vehicular communication networks supported by UAV

Vehicular communication networks are emerging as a promising technology to provide high-quality internet service such as entertainment for road users via infrastructure-to-vehicle (I2V) communication, and to guarantee road users’ safety via vehicle-to-vehicle (V2V) communication. Some technical issues that impact the performance of these networks are the lack of or poor communication paths between vehicles, and the limitation of radio resources. Unmanned aerial vehicles (UAVs) as promising solutions for supporting vehicular networks could provide communication coverage in hazardous environments and areas with no capacities for installation or maintenance of ground base stations (BSs). Also, non-orthogonal multiple access (NOMA) methods can improve spectral and energy efficiency and thereby allow more users to be connected to the desired network. In this paper, exploring the NOMA, we develop a scheme for optimum resource allocation in presence of a UAV that supports vehicular communications. Resource allocation for this scenario is formulated as a mixed-integer non-linear programming (MINLP) problem. Due to the high complexity of such problems, we propose two low-complexity near-optimal methods. First, we apply difference-of-concave-functions (DC) approximations to solve the problem in an iterative process. Next, we use Stackelberg game-based method for efficient solving, and then, closed-form expressions of optimal power allocations using KKT-conditions are derived. Simulations illustrate the effectiveness of the proposed scheme along with the Stackelberg game-based method.     

Improvement of packet transmission scheduling and delivery rate in wireless Ad-Hoc networks

The principal purpose of Ad-Hoc wireless networks is to increase service efficiency in terms of transmission scheduling and packet transfer rate. The approaches that assume frame unicity to satisfy a given set of packets minimize the end-to-end delay. However, they do not guarantee a maximum packet delivery rate due to the difficulty of establishing robust paths for packet transfer across nodes deployed in the network, especially in a three-dimensional (3D) environment. The objective is to minimize the end-to-end delay by ensuring the maximum delivery of packets to their destinations. Furthermore, the signal-to-interference-and-noise-ratio (SINR) model is considered to optimize transmission scheduling. In this paper, an optimal node coordinates optimization approach is proposed to extend two recently investigated schemes in the literature (S-RDSP and I-RDSP). The developed algorithms, named S-MPDR and I-MPDR, seek to reduce the end-to-end delay by delivering a collection of inserted packets over a 3D environment while also maximizing the delivery rate of these packets. Desirability functions are used to evaluate the network’s performance in various scenarios involving two different environments, Level 0 and Level 1. Numerical results demonstrate that the developed algorithms outperform both schemes in terms of end-to-end delay and packet delivery rate. In the Level 0 environment, the overall minimum delay and packet delivery rate scores provided by S-MPDR are increased by 28% and 88% compared to S-RDSP, respectively. In comparison, those provided by I-MPDR are increased by 24% and 16% compared to I-RDSP. Similarly, in the Level 1 environment, the scores provided by S-MPDR are increased by 25% and 100% compared to S-RDSP, respectively, while those provided by I-MPDR are increased by 23% and 25% compared to I-RDSP.     

RAT selection for IoT devices in HetNets: Reinforcement learning with hybrid SMDP algorithm

Due to the increasing deployment of heterogeneous networks (HetNets), the selection of which radio access technologies (RATs) for Internet of Things (IoT) devices such as user equipments (UEs) has recently received extensive attention in mobility management research. Most of existing RAT selection methods only optimize the selection strategies from the UE side or network side, which results in heavy network congestion, poor user experience and system utility degradation. In this paper the UE side and the network side are considered comprehensively, based on the game theory (GT) model we propose a reinforcement learning with assisted network information algorithm to overcome the crucial points. The assisted information is formulated as a semi-Markov decision process (SMDP) provided for UEs to make accurate decisions, and we adopt the iteration approach to reach the optimal policy. Moreover, we investigate the impacts of different parameters on the system utility and handover performance. Numerical results validate that our proposed algorithm can mitigate unnecessary handovers and improve system throughputs.