共查询到17条相似文献,搜索用时 253 毫秒
1.
2.
3.
4.
车联网对于超高可靠与低时延通信(Ultra-Reliable and Low Latency Communications,URLLC)具有严格的要求,特别对于车到基础设施(Vehicle to Infrastructure,V2I)场景,URLLC对传输管理交通状况至关重要.3GPP Cel-lular-V2X(C-V2X)作为现在支撑车联网URLLC主流的无线技术,仍存在技术挑战.为进一步提升通信性能,本文在V2I场景下,基于车载终端、路侧单元(Road Side Unit,RSU)与边缘计算车联网服务器(Internet of Vehicles Server,IoV Server)的交互,设计了一种基于C-V2I规范的智能信道估计框架.在IoV Server中,本文提出了一种基于深度学习的信道估计算法,该算法利用一维卷积神经网络(One Dimensional Convolution Neural Network,1D CNN)完成频域插值和条件循环单元(Conditional Recurrent Unit,CRU)进行时域状态预测,通过引入额外的速度编码矢量和多径编码矢量跟踪环境的变化,对不同移动环境下的信道数据进行精确训练.最后通过系统仿真与分析表明,所提算法能够通过信道参数编码追踪不同高速移动环境下的信道变化,实现对信道数据的精确训练.与车联网代表性信道估计算法相比,所提算法提升了信道估计精度,降低了误码率和增强了鲁棒性. 相似文献
5.
6.
针对车联网中高通信需求和高移动性造成的车对车链路(Vehicle to Vehicle, V2V)间的信道冲突及网络效用低下的问题,提出了一种基于并联门控循环单元(Gated Recurrent Unit, GRU)和长短期记忆网络(Long Short-Term Memory, LSTM)的组合模型的车联网信道分配算法。算法以降低V2V链路信道碰撞率和空闲率为目标,将信道分配问题建模为分布式深度强化学习问题,使每条V2V链路作为单个智能体,并通过最大化每回合平均奖励的方式进行集中训练、分布式执行。在训练过程中借助GRU训练周期短和LSTM拟合精度高的组合优势去拟合深度双重Q学习中Q函数,使V2V链路能快速地学习优化信道分配策略,合理地复用车对基础设施(Vehicle to Infrastructure, V2I)链路的信道资源,实现网络效用最大化。仿真结果表明,与单纯使用GRU或者LSTM网络模型的分配算法相比,该算法在收敛速度方面加快了5个训练回合,V2V链路间的信道碰撞率和空闲率降低了约27%,平均成功率提升了约10%。 相似文献
7.
8.
9.
针对车-车(vehicle to rechicle, V2V)通信系统对无线电信道衰落特性和模型的需求,分别在两种隧道场景中5.9 GHz和5.2 GHz频段下进行了V2V无线电信道测量活动,并对隧道外、隧道内和两者之间的连接部分场景进行了小尺度衰落特性分析. 基于近距离(close-in, CI)对数模型和ABG (α-β-γ)模型建立了基于距离的接收功率模型,对两种场景隧道内外的接收功率进行了评估和比较,路径损耗指数分别为1.83和1.9,结果表明参考距离为1 m的CI对数模型具有更高的拟合度. 此外,将测量数据幅度的衰落分布与五种典型的理论衰落分布进行比较分析,发现其特征更接近于具有最小拟合优度值的莱斯分布,且隧道内的莱斯K因子小于隧道外. 同时,给出了隧道内和隧道外之间连接处基于距离的莱斯K因子模型,发现连接处的K因子与距离无关,而隧道内的K因子随距离增大而减小. 相似文献
10.
11.
12.
车联网相关应用需要基于实时、准确的交通信息。RSU会实时进行广播,同时车辆间要进行实时通信,包括车辆的身份信息、驾驶状态及位置信息等。攻击者可以利用车联网的开放性获取实时发送的空口数据,通过破解空口数据获得车辆的身份信息和位置信息,进而可以通过伪装、篡改或者植入恶意程序的方式对车辆进行攻击。因此,车联网通信过程中的信息安全问题必须得到有效的保护。基于区块链的匿名认证,车辆在V2V及V2I通信过程中将公钥作为假名进行认证,既保证了消息来源的真实性和消息的完整性,也避免了车辆身份信息的泄露。 相似文献
13.
5G时代的到来加快了以车路协同为核心的车联网技术的发展脚步。高效,可靠,安全的通信质量将是实现智能交通管理系统,完善智慧出行的基本要求。因此,针对存在信息窃取者的车联网中继协作传输场景,利用机会式中继选择策略与最大比合并技术设计了直接传输链路与多跳中继转发链路共存的安全传输方案,旨在提高信息传输的可靠性及安全性。同时,在获得信道概率密度函数及累积分布函数的基础上,利用全概率公式等方法推导出基于DF(Decode-Forward)中继协作传输的车联网系统安全中断概率的闭合表达式,其中涉及的信息传输信道均服从Nakagami-m分布,提高了理论推导的难度。最终的仿真实验证实了理论推导的准确性及本方案的可行性。 相似文献
14.
15.
作为5G中的一种重要模型,雾无线接入网络(Fog Radio Access Network, F-RAN)通过设备到设备通信和无线中继等技术获得了显著的性能增益,而边缘设备中合适的缓存则可以让内容缓存用户(Caching Users, CUs)向内容请求用户(Requesting Users, RUs)直接发送缓存内容,有效减小前传链路的负担和下载延迟。考虑一个F-RAN模型下用户发出请求并获得交付的场景,将每个CU的内容请求队列建模为独立的M/D/1模型,分析导出CUs缓存命中率和平均下载延迟关于内容缓存与交付方案的表达式,证明CUs缓存命中率与内容统计分布之间的联系有助于实现前者的近似最优解。针对在一段时间内的期望视角下建立的优化问题,提出了基于统计分布的算法并注意了执行时的交付控制。仿真结果表明,相较于现有缓存策略,优化内容整体统计分布的方案能够最大化CUs缓存命中率,同时减小平均下载延迟。 相似文献
16.
Yaser Taj Bahador Bakhshi Hessam Zandhessami 《International Journal of Communication Systems》2023,36(8):e5459
The Internet of Vehicles (IoV) is a new framework for intelligent transportation systems. One of its goals is to improve safety and increase the quality of service for passengers. Topology changes in IoV present significant challenges to the function of safety applications. Most routing algorithms offer their solution, independent of infrastructure or based exclusively on infrastructure. Although the lack of access to roadside units disrupts infrastructure-based algorithms, methods that do not use infrastructure will also not enjoy the benefits of high-speed data transmission. This research proposes the reliable and infrastructure-independent routing (RIIR) routing algorithm to increase the reliability of real-time routing in the IoV. The RIIR, in addition to simultaneous covering capability to use intelligent infrastructure for accelerating data transmission and flexibility in conditions of lack of access to infrastructure, by introducing three new criteria entitled “history of vehicle movement,” “conformity of vehicle speed with the harmonic speed average of nearby vehicles,” and “number of reliable neighbors,” scores vehicles and then selects the most suitable vehicles as route members. The RIIR measures the conformity of the previous presence of vehicles on the current route and the stability of their link with adjacent vehicles. Furthermore, it prevents congestion by halting the spread of route request messages at intersections not leading to the destination. The efficiency of the RIIR has been theoretically proven, and extensive simulations with multiple scenarios in SUMO and NS3 show the superiority of the RIIR in increasing the packet delivery rate and reducing route failure, average delay, and control overhead. 相似文献
17.
The same direction multi-hop downloading network flow model was proposed for vehicle communication range and channel competition under forwarding.According to the model and video streaming media characteristic,the same & opposite direction cooperative downloading quality of service (SODCD-QoS) scheme was presented.First,the same direction vehicles helped the target vehicle to achieve multi-hop cooperative downloading.It can reduce the waste of time which the target vehicle has to wait for opposite direction cooperative vehicle.Secondly,the next base stations loaded data into eligible opposite direction vehicles according to vehicles option strategy based on time first or picture quality first.Then the opposite direction vehicles helped the target vehicle to achieve one-hop cooperative downloading.NS simulations show that the SODCD-QoS scheme is better than DSRelay and VCoDS algorithm in terms of download time.Further,the new scheme can maintain good picture quality PSNR. 相似文献