干扰场景下车辆移动通信网络性能分析及优化设计

发布时间：2018-04-30 04:14

本文选题：干扰场景 + 节点移动　；参考：《西南交通大学》2017年博士论文

【摘要】：随着无线通信技术和国家交通事业的飞速发展,车联网作为第五代(5G)移动通信系统的一个重要应用,越来越受到人们的关注。在未来车辆移动通信网络中,不仅要在“量”上要满足移动用户多样化业务需求,还要在“质”上确保移动用户以更低的时延获得更大的传输速率。随着车载用户数量不断增加,在信息传输过程中,接收节点受到干扰的可能性越来越大。针对干扰和移动这两大问题,本文考虑干扰处于高动态场景下,即每一时刻,都会有新的干扰节点产生,从理论上分析了车辆移动通信网络的性能及其相应的系统优化设计。首先,本文研究高速公路场景,即源-中继-目的节点移动场景的通信链路中断概率和最优功率分配方案。在本研究中,假设中继节点只受到噪声的影响,目的节点只受到共道干扰(co-channel interference, CCI)的影响,并且干扰处于高移动状态,即每一个时刻,干扰节点的数目和位置都有所不同。在这一模型的基础上,本文得到了中断概率的分析表达式和紧的下界,分析了两跳链路对于端到端中断概率的影响,源节点、中继节点的发送功率对于端到端中断概率的影响。此外,得到在源-中继-目的节点链路移动过程中,中断概率的变化情况。当源节点和中继节点的发送功率之和受限时,得到了中断概率最小的最优功率分配方案。接着,本文分析多跳车辆移动通信网络的端到端时延,并得到时延最小的最优跳数的分析结果。在传统的无线通信网络时延分析中,通常假设中继节点等间距分布在源节点和目的节点的连线上,并忽略了多跳链路跳数对于时延的影响。事实上,随着跳数增多,每一跳的距离减小有利于时延的减小,但同时需要处理数据的时间会增多,从而使时延变大。本文对这一场景进行了扩展,考虑中继服从三种分布场景下的端到端时延性能及最优跳数分析,包括中继等间距分布、中继服从均匀分布以及中继服从随机路径点移动(random waypoint mobility,RWPM)模型分布。假设单跳网络在干扰受限和噪声受限场景下有相同的接收信干比(signal-to-interference ratio,SIR)和信噪比(signal-to-noise ratio, SNR),论文分析了单跳通信网络在这两种场景下的时延性能。通过比较中继等间距分布和中继服从均匀分布这两种场景的时延性能,研究表明选择在源节点和目的节点连线上等间距分布的中继节点可以更有效地降低时延。中继服从均匀分布类似于静止中继节点的场景,中继服从RWPM模型可视为移动中继节点场景。通过比较这两种场景,本文的分析和仿真实验表明,移动性有利于减小多跳信息传输的时延。其次,本文分析了干扰受限场景车辆移动通信网络的连通度与移动性之间的关系。在传统的无线通信网络连通概率分析工作中,大多假设接收节点不受干扰的影响,忽略了节点密度以及移动场景下节点移动性对于连通概率的影响。当考虑节点之间的干扰时,节点密度增大,节点之间的距离减小,则接收节点接收有用信号的强度增大,从而有利于提高通信质量和连通概率。但节点密度增大,将导致接收节点受到的干扰增强,从而通信质量下降,直接影响网络的连通概率。根据这一基本想法,本文推导出干扰场景下有限区间内线性网络全连通概率的上、下界,并得到全连通概率是节点密度的拟凹函数。在此基础上,分析了移动场景下节点移动的平均速度与有限线性网络连通概率之间的关系。再次,本文分析了随机多址接入协议下,半双工(half-duplex)和全双工(full-duplex) 单跳信息传输系统的随机接入传输容量 (random access transport capacity,RATC)。假设接收节点选择离它最近的节点作为发送节点进行传输,从而节点之间的距离不是固定值,而是一个随机变量。当节点的随机接入概率增大时,意味着发送节点发送信息的机会增多,随机接入传输容量可能提高,但同时,接收节点受到的共道干扰节点数目增多,从而成功传输的概率减小,传输次数可能增多,随机接入传输容量可能减小。类似地,当SIR阈值增大时,信道的最大传输速率会增大,随机接入传输容量可能增大,但同时意味着单跳链路的成功概率减小,随机接入传输容量可能变小。通过仿真实验验证了,随机接入传输容量是节点的发送概率、SIR阈值的拟凹函数,并得到全双工系统的性能优于半双工系统的性能。最后,本文分析线性多跳译码转发(decode-and-forward,DF)通信链路的吞吐量并得到对应系统的优化设计。在传统的无线通信网络吞吐量分析中,一般仅分析某一个系统参数与吞吐量之间的关系,少有系统参数联合优化的方案分析。本文研究了多跳无线通信网络的跳数、SIR阈值与吞吐量之间的关系,揭示了固定跳数或SIR阈值,吞吐量是另一个参数的拟凹函数。根据吞吐量函数的结构特点,得到以多跳无线通信网络吞吐量最大为目标,联合优化向量(跳数、SIR阈值)的算法。
[Abstract]:With the rapid development of wireless communication technology and national transportation, the Internet of vehicles, as an important application of the fifth generation (5G) mobile communication system, has attracted more and more attention. In the future vehicle mobile communication network, it is not only to meet the needs of the mobile users in the "quantity", but also to ensure the movement on the "quality". With the increasing number of users on the vehicle, the probability of receiving nodes is more and more likely to be disturbed in the process of information transmission. Considering the two major problems of interference and movement, this paper considers that the interference is in high dynamic scene, that is, a new interference node will be generated at every moment. The performance of the vehicle mobile communication network and its corresponding system optimization design are analyzed. First, this paper studies the highway scene, the communication link interrupt probability and the optimal power allocation scheme of the source relay destination node mobile scene. In this study, the relay node is assumed to be only affected by the noise, the destination node is only shared by the common channel. The interference (co-channel interference, CCI) is affected and the interference is in a high mobility state, that is, the number and location of the interference nodes are different at every moment. On the basis of this model, the analysis expression and the tight lower bounds of the interruption probability are obtained, and the influence of the two hop links on the end to end interrupt probability is analyzed, and the source node is analyzed. Point, the transmission power of the relay node affects the end to end interrupt probability. In addition, the change of the interruption probability is obtained during the link movement of the source relay destination node. When the sum of the transmission power of the source and relay nodes is limited, the optimal power allocation scheme is obtained for the minimum interruption probability. Then, the paper analyzes the multi hop vehicle. The end to end delay of a mobile communication network is analyzed. In the traditional time delay analysis of the wireless communication network, the distance between the relay nodes is usually assumed to be distributed on the connection of the source node and the destination node, and the effect of the multiple hop number on the delay is ignored. In fact, the number of hops increases as the number of hops increases. The reduction of each jump is beneficial to the decrease of time delay, but the time for processing data is increased and the time delay becomes larger. This paper extends this scene to consider the end to end delay performance and optimal hops analysis under three distributed scenarios, including the relay equidistance distribution, the relay obeys the uniform distribution, and the relay obeys the uniform distribution, and the relay obeys the uniform distribution, and the relay obeys the uniform distribution. The relay obeys the random path point movement (random waypoint mobility, RWPM) model distribution. It is assumed that the single hop network has the same receiver signal to dry ratio (signal-to-interference ratio, SIR) and the signal to noise ratio (signal-to-noise ratio, SNR) under the restricted and noisy scene. The paper analyzes the delay of the single hop communication network in these two scenarios. Performance. By comparing the delay performance of the two scenarios with the relay equidistance distribution and the relay obeying the uniform distribution, it is shown that the relay nodes selected in the source node and the destination node is more effective in reducing the delay. The relay obeys the stationary relay node and the relay obeys the RWPM model. It is considered as a mobile relay scene. By comparing these two scenarios, the analysis and simulation experiments in this paper show that mobility is beneficial to reduce the delay of multi hop information transmission. Secondly, this paper analyzes the connection between the connectivity and mobility of the vehicle mobile communication network with limited interference scene. In the analysis, most of the work assumes that the receiving node is not affected by interference, ignoring the node density and the influence of node mobility on the connectivity probability. When considering the interference between nodes, the density of nodes increases and the distance between nodes decreases, and the intensity of receiving nodes to receive useful signals increases, which is beneficial to the improvement of the connection. But the density of the node increases, which will lead to the enhancement of the interference received by the receiving node, thus the communication quality is reduced and the connectivity probability of the network is directly affected. Based on this basic idea, the upper and lower bounds of the total connectivity probability of the linear network in the finite interval under the interference scene are derived, and the total connectivity probability is the node density. On this basis, the relationship between the average velocity of node movement and the connectivity probability of a finite linear network is analyzed on this basis. The random access capacity (random access T) of the semi duplex (half-duplex) and full duplex (full-duplex) single hop information transmission system under the random multiple access protocol (random access Protocol) is analyzed. Ransport capacity, RATC). Assuming the receiving node selects the nearest node as the transmission node, the distance between the nodes is not a fixed value, but a random variable. When the node's random access probability increases, it means that the opportunity for sending the nodes to send the information is increased, but the capacity of the random access transmission may be improved, but the capacity of the random access transmission may be improved, but the probability of the random access transmission may be increased, but the capacity of the random access transmission may be improved, but At the same time, the number of common channel interference nodes received by the receiving node is increased, thus the probability of successful transmission decreases, the number of transmission may increase, and the random access transmission capacity may be reduced. Similarly, when the SIR threshold increases, the maximum transmission rate of the channel will increase and the random access transmission capacity may increase, but it also means the single hop link is formed at the same time. The power probability decreases and the random access transmission capacity may be smaller. The simulation experiment shows that the random access transmission capacity is the node's sending probability, the quasi concave function of the SIR threshold, and the performance of the full duplex system is superior to the half duplex system. Finally, this paper analyzes the linear multi hop code forwarding (decode-and-forward, DF) communication link. In the throughput analysis of the traditional wireless communication network, the relationship between the parameters and the throughput of a certain system is analyzed, and the scheme analysis of the joint optimization of the system parameters is seldom analyzed. This paper studies the number of hop, the relationship between the SIR threshold and the throughput, and the relationship between the threshold and the throughput. The fixed hop number or SIR threshold is shown, and the throughput is a quasi concave function of another parameter. According to the structure characteristics of the throughput function, the algorithm of combining the optimization of the vector (hop, SIR threshold) is obtained for the maximum throughput of the multi hop wireless communication network.

【学位授予单位】：西南交通大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：TN929.5

【相似文献】