Using traffic data to accurately predict the traffic flow at a certain time in the future can alleviate problems such as traffic congestion, which plays an important role in the healthy transportation and economic development of cities. However, current traffic flow prediction models rely on human experience and only consider the advantages of single machine learning model. Therefore, in this work, we propose a multi-layer model based on transformer and deep learning for traffic flow prediction (MTDLTFP). The MTDLTFP model first draws on the idea of transformer model, which uses multiple encoders and decoders to perform feature extraction on the initial traffic data without human experience. In addition, in the prediction stage, the MTDLTFP model using deep learning technology, which input the hidden features into the convolutional neural network (CNN) and multi-layer feedforward neural network (MFNN) to obtain the prediction score respectively. The CNN model can captures the correlation information between the hidden features, and the MFNN can captures the nonlinear relationship between the features. Finally, we use a linear model to combine the two prediction scores, which can make the final prediction value take into account the common advantages of both models. Multiple experimental results on two real datasets demonstrate the effectiveness of the MTDLTFP model. The experimental results on the $WorkDay$ dataset are as follows, with the RMSE value of 0.191, MAE value of 0.165. The experimental results on the $HoliDay$ dataset are as follows, with RMSE value of 0.227, MAE value of 0.192.

In this paper, the bit error rate (BER) performance of underwater wireless optical communication (UWOC) systems with on-off keying (OOK) modulation and an avalanche photodiode (APD) detector are analyzed in a wide range of oceanic turbulence environment, such as lognormal, Gamma, K, Weibull, and Exponentiated Weibull distributions. To achieving this goal, a new mathematical system model is established, which has not only considered the aggregated degrading phenomena, namely the absorption, scattering, and turbulence, but also been suitable for describing UWOC with APD receiver. Then the closed-form expressions of the BER performances of UWOC with APD for different turbulence distributions are derived in detail by subdividing the working situation into high and lower inter-symbol interference (ISI) scenarios. Theoretical analysis and computer simulations show that no matter what the ISI levels are, the derived closed-form expressions for system BER under various turbulences can approximate the accurate integral ones well. This is obviously helpful to the design and performance evaluation of UWOC systems.

Federated learning (FL) has been recognized as a transformative approach in vehicular networks, enabling collaborative training between vehicles and preserving data privacy. However, the high mobility and dynamic topology changes inherent in vehicular environments pose significant challenges, primarily due to the increased communication overhead associated with exchanging model parameters. To mitigate these issues, a novel framework for decentralized wireless FL with soft clustering and 1-bit compressed sensing (SC1BCS-WFL) is proposed in this article. The framework considers vehicle position, vehicle attributes, vehicle speed, and model cosine similarity when grouping vehicles. It utilizes an adaptive threshold mechanism based on 1-bit compression to reduce uplink transmission load while maintaining FL performance. In addition, an early stopping strategy is incorporated into the proposed framework to avoid unnecessary waste of computational and communication resources. Simulation results show that the SC1BCS-WFL framework can enhance the efficiency of FL in vehicular settings, particularly with nonindependent and identically distributed data. Simulation results also validate the framework’s ability to reduce communication overhead while achieving high-model accuracy, indicating its suitability for distributed Internet of Vehicles (IoV) scenarios and contributing to the development of smarter and more efficient IoV applications.

Vehicle-to-vehicle (V2V)-enabled platooning is a promising application for future intelligent transportation systems. V2V communications allow real-time sharing of vehicle status information within the platoon, enabling vehicles to dynamically adjust their control strategies and maintain spacing between platoon members. As such, the platoon stability depends on the timeliness of the control information. However, the information timeliness can be affected by factors, such as transmission period, co-channel interference and actuator lag. In this article, the performance of the V2V communication is analyzed for the platooning, from an information timeliness perspective. First, an Age-of-Information (AoI)-based joint communication and control model is proposed to determine the requirement of the AoI to maintain the stability. Then, a large-scale V2V communication model is constructed by taking account of the V2V packet transmission and resource allocation scheme. Considering the spatiotemporal correlation among the V2V transmissions, the distribution of the average AoI and average peak AoI are investigated, by using the queuing theory and stochastic geometry. The analytical results can reveal the proportions of the stable platoons in the large-scale networks. Simulation results corroborate the analytical results and demonstrate the impact of the system parameters on the platoon stability. Furthermore, simulation results showed that the requirements of the maximum transmission period and the maximum actuator lag can be analytically obtained. More importantly, it is shown that the proposed model and the analytical results could provide insightful guidelines for designing the V2V-enabled platooning systems.

With their superior maneuverability and flexible deployment options, unmanned aerial vehicles (UAVs) present a viable solution to augment cellular network capabilities by serving as mobile base stations (BSs). Yet, the 3-D, dynamic nature of UAV deployments, along with changing aerial conditions, often results in frequent handovers, adversely affecting communication quality. Moreover, the impacts of transitions between Line-of-Sight (LoS) and Non-LoS (NLoS) links on handover and coverage probabilities in mobile UAV networks have not been thoroughly investigated. This article provides an in-depth analysis of how multitier deployment altitudes and variations in LoS link probabilities influence handover and coverage probabilities. Utilizing stochastic geometry, we consider two association strategies: 1) distance-based and 2) strongest average received signal strength (RSS)-based. For both strategies, we provide semi-closed-form expressions for handover probability and subsequently derive network coverage probabilities. Through numerical simulations, we not only unearth an optimal configuration of UAV density and deployment altitude that maximizes coverage probability for ground users but also uncover that the relative benefits of the RSS-based association strategy wane as UAV density escalates compared to the distance-based association strategy. Furthermore, our results underscore the potential for enhancing coverage performance by adopting a strategy of deploying UAVs at various altitudes, contrasting with the traditional approach of uniform altitude deployment.

Wireless powered communication (WPC) is able to provide wireless devices (WDs) practically infinite energy by using the radio frequency energy harvesting technique. The electricity of WDs for the wireless information transmission (WIT) is converted from the received electromagnetic signals of wireless energy transfer (WET). Power beacon (PB) is usually deployed as the dedicated energy source in WPC networks and the performance of WET is sensitive to the path-loss effect. Thus, both WET and WIT are significantly influenced by the locations of the PBs. To enable efficient transmission in WPC, this paper analytically investigates the impact of the PB deployment strategies on the WPC performance when WDs are clustered around the access points (APs). Specifically, we propose the PB deployment strategy based on the truncated Poisson cluster process (PCP) model. The distribution of PBs is correlated to the locations of WDs by choosing APs as their common parent points. Then, we derive the semi-closed form expressions for the energy outage probability and the transmission success probability. We further present the expressions of the meta distributions of WET and WIT, and investigate the proportion of the WDs in one cluster that achieves successful transmission while satisfying the reliability constraint. Moreover, the distance distribution between the WD and the intra-cluster PBs is analytically derived for the proposed correlated PCP model. By comparing the performance with other deployment strategies, it is shown that the preference of the PB deployment strategy depends on the PB density. The analytical and numerical results can provide insightful guidelines for designing practical WPC systems.