Yuan An

Bio: Yuan An is an academic researcher from Xuzhou Institute of Technology. The author has contributed to research in topics: Cloud computing & Wireless network. The author has an hindex of 1, co-authored 7 publications receiving 2 citations.

The Design of Vehicular Voice Cloud Protocol

Abstract: This paper proposes UDP-based reliable application protocol vehicular voice cloud service. Instead of ACK message, the receiver asks for a retransmission via delivering NACK message to sender. This design is able to reduce the latency of voice-recognition-based services in vehicular environment. The simulation is performed to comparing the UDP-based voice cloud transmission with the traditional TCP-based transmission. The result the proposed application protocol obtains low latency and better user experience in unstable environment.

Channel selection observation period length analysis under different channel service rates

Abstract: In wireless communication, the sender needs to select an optimal wireless channel from several available ones. However, the instantaneous channel state is time-varying with unknown statistics. Therefore, the channel selection must be based on channel observation. Before the packet arrives, the sender needs to observe the channel state in the observation period. And then the sender transmits packets through the best channel. Observation needs cost time. We investigate the trade-off between the observation period and transmission period. A short observation period usually leads to wrong selection while long observation might waste time. Our simulation results show that there is an optimal length of observation period. The total transmission time experience a sharp decrease before the optimal point. The longer observation does not cause an obvious increase in transmission time. We analyze how the observation time affects the transmission time to explain the simulation results.

Massive MIMO for Maximal Spectral Efficiency: How Many Users and Pilots Should Be Allocated?

Abstract: Massive MIMO is a promising technique for increasing the spectral efficiency (SE) of cellular networks, by deploying antenna arrays with hundreds or thousands of active elements at the base stations and performing coherent transceiver processing. A common rule-of-thumb is that these systems should have an order of magnitude more antennas $M$ than scheduled users $K$ because the users’ channels are likely to be near-orthogonal when $M/K > 10$ . However, it has not been proved that this rule-of-thumb actually maximizes the SE. In this paper, we analyze how the optimal number of scheduled users $K^\star$ depends on $M$ and other system parameters. To this end, new SE expressions are derived to enable efficient system-level analysis with power control, arbitrary pilot reuse, and random user locations. The value of $K^\star$ in the large- $M$ regime is derived in closed form, while simulations are used to show what happens at finite $M$ , in different interference scenarios, with different pilot reuse factors, and for different processing schemes. Up to half the coherence block should be dedicated to pilots and the optimal $M/K$ is less than 10 in many cases of practical relevance. Interestingly, $K^\star$ depends strongly on the processing scheme and hence it is unfair to compare different schemes using the same $K$ .

Fractional Modeling of Non-Newtonian Casson Fluid between Two Parallel Plates

Abstract: In this manuscript, fractional modeling of non-Newtonian Casson fluid squeezed between two parallel plates is performed under the influence of magneto-hydro-dynamic and Darcian effects. The Casson fluid model is fractionally transformed through mixed similarity transformations. As a result, partial differential equations (PDEs) are transformed to a fractional ordinary differential equation (FODE). In the current modeling, the continuity equation is satisfied while the momentum equation of the integral order Casson fluid is recovered when the fractional parameter is taken as α = 1 . A modified homotopy perturbation algorithm is used for the solution and analysis of highly nonlinear and fully fractional ordinary differential equations. Obtained solutions and errors are compared with existing integral order results from the literature. Graphical analysis is also performed at normal and radial velocity components for different fluid and fractional parameters. Analysis reveals that a few parameters are showing different behavior in a fractional environment as compared to existing integer-order cases from the literature. These findings affirm the importance of fractional calculus in terms of more generalized analysis of physical phenomena.

Global-Local Spatial-Temporal Residual Correlation Network for Urban Traffic Status Prediction

Abstract: The recent proposed Spatial-Temporal Residual Network (ST-ResNet) model is an effective tool to extract both spatial and temporal characteristics and has been successfully applied to urban traffic status prediction. However, the ST-ResNet model only extracts the local spatial characteristics and ignores the very important global spatial characteristics. In this paper, a novel Global-Local Spatial-Temporal Residual Correlation Network (GL-STRCN) model is proposed for urban traffic status prediction to further improve the prediction accuracy of the existing ST-ResNet model. The GL-STRCN model firstly applies Pearson's correlation coefficient method to extract high correlation series. Then, considering both global and local spatial properties, two components consisting of 2D convolution and residual operation are used to capture spatial features. After that, based on Long Short-Term Memory (LSTM) or Gated Recurrent Unit (GRU), a novel long-term temporal feature extraction component is proposed to capture temporal features. Finally, the spatial and temporal features are aggregated together in a weighted way for final prediction. Experiments have also been performed using two datasets from TaxiCD and PEMS-BAY. The results indicated that the proposed model produces a better prediction performance compared with the results based on other baseline solutions, e.g., CNN, ST-ResNet, GL-TCN, and DGLSTNet.

AdaCB: An Adaptive Gradient Method with Convergence Range Bound of Learning Rate

Abstract: Adaptive gradient descent methods such as Adam, RMSprop, and AdaGrad achieve great success in training deep learning models. These methods adaptively change the learning rates, resulting in a faster convergence speed. Recent studies have shown their problems include extreme learning rates, non-convergence issues, as well as poor generalization. Some enhanced variants have been proposed, such as AMSGrad, and AdaBound. However, the performances of these alternatives are controversial and some drawbacks still occur. In this work, we proposed an optimizer called AdaCB, which limits the learning rates of Adam in a convergence range bound. The bound range is determined by the LR test, and then two bound functions are designed to constrain Adam, and two bound functions tend to a constant value. To evaluate our method, we carry out experiments on the image classification task, three models including Smallnet, Network IN Network, and Resnet are trained on CIFAR10 and CIFAR100 datasets. Experimental results show that our method outperforms other optimizers on CIFAR10 and CIFAR100 datasets with accuracies of (82.76%, 53.29%), (86.24%, 60.19%), and (83.24%, 55.04%) on Smallnet, Network IN Network and Resnet, respectively. The results also indicate that our method maintains a faster learning speed, like adaptive gradient methods, in the early stage and achieves considerable accuracy, like SGD (M), at the end.

Grey Wolf Optimizer with Deep Learning based Short Term Traffic Forecasting in Smart City Environment

Abstract: Intelligent Transportation System (ITS) is one of the revolutionary technologies in smart cities that aids in minimizing traffic congestion and improving traffic quality. ITS provides real-time analysis and very effective traffic management by utilizing big data and communication technology. Traffic Flow Prediction (TFP) becomes a dynamic component in smart city management and was utilized for predicting the future traffic conditions on transportation networks relevant to past data. Machine Learning (ML) and Neural Network (NN) techniques can be broadly used in resolving real-time problems as these techniques are capable of managing adaptive data for some time. Deep Learning (DL) is a sub-divison of ML methods which earns effective performance on prediction and data classification tasks. This article designs a Grey Wolf optimizer with Deep Learning Based Short Term Traffic Forecasting (GWODL-STTF) in smart city environment. The presented GWODL-STTF technique concentrates on the prediction of traffic flow in smart cities. The presented GWODL-STTF technique involves two major processes. At the initial stage, the GWODL-STTF technique employed gated recurrent unit-neural network (GRU-NN) model to forecast traffic flow. Next, in the second stage, the GWODLSTTF technique makes use of GWO algorithm as a hyperparameter optimizer. The simulation values of the GWODL-STTF method can be tested under several metrics and the outcomes show the significant performance of the GWODLSTTF method over recent approaches with minimum MSE of 105.627.