基于FPGA软件无线电平台的D2D寻呼算法设计与实现

发布时间：2018-04-29 10:29

本文选题：设备到设备 + FlashLinQ　；参考：《西安电子科技大学》2014年硕士论文

【摘要】：移动互联网的快速发展促进了设备到设备(D2D)通信技术的发展。D2D系统是一个没有控制中心的分布式系统,系统中的无线设备自行组网,设备之间直接通信。高通公司研发的FlashLinQ技术,是一种应用在D2D系统的通信协议。在FlashLinQ协议中,节点检测到干扰信号,并不是立即退避,而是通过判断SIR是否满足阈值来决定是否退避,因此相应网络具有很高的频谱空间复用度。FlashLinQ协议在进行链路调度时,随机地分配优先级,不能充分利用链路分集增益提高系统吞吐量,一定程度上增加了链路调度时延。本文介绍的DO-Fast协议采用两种改进方法,能够很好地解决这个问题。方法一是使链路依照信道质量由高到低进行排序分组,给信道质量高的链路分配较多的优先调度机会。方法二是采用组间轮询和组内轮询优先调度机制,分别从大尺度和小尺度范围保证链路的公平性。本文介绍了DO-Fast协议的整体设计思想,剖析协议的帧结构,估算协议的开销,描述协议的实现流程,给出每个时隙信息交互方式。最后在理论上分析验证了协议相对于原协议对网络性能的提升作用。本文根据DO-Fast协议描述,基于FPGA软件无线电平台设计了协议寻呼算法的实现方案。该设计方案需要完成OFDM基带系统、OFDMA基带系统和MAC层寻呼算法的设计。OFDM系统和OFDMA系统参考IEEE 802.11a标准设计。MAC层寻呼算法的主要思想是对所有链路按照信道质量进行排序分组,为业务时隙的连接调度做准备。DO-Fast是时间同步协议,全网节点需要精确的同步操作。FPGA具有并行计算的能力和确定的处理时延,非常适合实现同步协议。本文使用的FPGA开发板是Nutaq Perseus601x,射频收发模块是Microsystems Radio420。这两个器件功能强大,通过Nutaq平台方便地与PC机通信,完全满足设计要求。本文完成了物理层OFDM和OFDMA基带系统设计、DO-Fast协议MAC层整体框图设计。使用Verilog HDL语言完成了物理层程序编写、MAC层对等发现算法程序编写、MAC层寻呼算法程序编写。物理层程序通过了实际测试,数据收发正确。MAC层对等发现算法和寻呼算法程序通过了仿真验证。由于项目工作量较大,协议寻呼算法程序未完成实际测试。
[Abstract]:The rapid development of mobile Internet promotes the development of D2D) communication technology. D2D system is a distributed system without control center. Qualcomm FlashLinQ technology is a communication protocol used in D-D system. In the FlashLinQ protocol, the node detects the interference signal, but determines whether the interference signal is Backoff by judging whether the SIR meets the threshold value or not. Therefore, the corresponding network has a high spectral spatial reuse. Flash LinQ protocol is used to schedule the link. If the priority is assigned randomly, the link diversity gain can not be fully utilized to improve the system throughput, which increases the link scheduling delay to a certain extent. The DO-Fast protocol introduced in this paper adopts two improved methods, which can solve this problem well. The first method is to arrange the links according to the channel quality from high to low to assign more priority scheduling opportunities to the links with high channel quality. Secondly, inter-group polling and intra-group polling are adopted to ensure the fairness of the link from large scale to small scale, respectively. This paper introduces the whole design idea of DO-Fast protocol, analyzes the frame structure of the protocol, estimates the cost of the protocol, describes the implementation flow of the protocol, and gives the way of information exchange in each slot. Finally, the paper theoretically analyzes and verifies the effect of the protocol on the performance of the network compared with the original protocol. According to the description of DO-Fast protocol, this paper designs a protocol paging algorithm based on FPGA software radio platform. This design scheme needs to complete the design of OFDM baseband system and MAC layer paging algorithm. The main idea of .OFDM system and OFDMA system referring to IEEE 802.11a standard design .MAC layer paging algorithm is to sort all links according to channel quality. DO-Fast is a time synchronization protocol for connection scheduling of service slots. The whole network node needs accurate synchronization operation. FPGA has the ability of parallel computing and the processing delay is determined, so it is very suitable to implement synchronization protocol. The FPGA development board used in this paper is Nutaq Perseus 601x. the RF transceiver module is Microsystems Radio420. These two devices have powerful function and can communicate with PC conveniently through Nutaq platform, which fully meet the design requirements. In this paper, the design of physical layer OFDM and OFDMA baseband system is completed. The whole block diagram of MAC layer of DO-Fast protocol is designed. In this paper, Verilog HDL language is used to complete the programming of the physical layer program and the peer discovery algorithm program of the MAC layer, and the programming of the Paging algorithm program of the MAC layer is developed. The physical layer program has passed the actual test, the data sending and receiving correct .MAC layer peer-to-peer discovery algorithm and the paging algorithm program have passed the simulation verification. Due to the heavy workload of the project, the protocol paging algorithm program did not complete the actual test.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5;TN791

【相似文献】