无线通信中的无缝自适应技术

发布时间：2018-06-22 22:55

本文选题：无线通信 + 编码调制　；参考：《中国科学技术大学》2014年博士论文

【摘要】：无线通信系统中的自适应传输存在着以下两类主要问题。 ·当前物理层技术无法实现对二进制数据的高效传输,主要有三个原因。第一,传统自适应技术的性能依赖于信道估计的准确性,而准确性与实时性存在着本质的矛盾。第二,由于无线通信系统中可选速率数量的有限性,速率的调整是不连续的。第三,当前系统的物理层假设上层传来的都是已压缩数据,而这个假设与实际测量相矛盾。 ·作为一类特殊的二进制数据,除了上述效率问题,视频码流的无线传输还存在着视频码率控制与物理层自适应相互独立导致的传输效率问题。另外,在多天线视频多播系统中,信道条件和接收端天线数量是各态互异的,最差的信道条件和最少的天线数量会成为系统性能的瓶颈。为解决上述两类问题,本文提出了无缝自适应的概念,即无线通信中信息的传输可以自动的随着信道情况的变化而连续的且平稳的调整。针对不同类型的数据及应用,本文研究了如何实现无缝自适应传输,主要研究成果如下： ·对于二进制数据,本文提出了随机投影RP编码,并基于此编码设计了速率可变调制RCM和压缩编码调制CCM。首先,发送端使用固定的方法递增的发送编码符号,直到接收端可以正确解码,因而不需要信道估计反馈选择速率。其次,RP编码具有无码率特性,传输速率可以连续的调整。再次,优化设计的RP编码具有压缩感知特性,可以充分利用二进制数据的冗余性从而提高系统吞吐量。 ·对于视频数据,本文提出了基于图像处理的线性视频传输技术Cactus和AirScale。去掉量化、熵编码和信道编码这些非线性处理,视频经过预测、变换及能量分配等线性处理后直接通过幅度调制传输。信道噪声直接作用于线性能量缩放后的数据,通过MMSE恢复的视频质量可以连续且平稳的随着接收端信道条件的变化而改变,避免了码率控制和物理层速率自适应带来的一系列问题。 ·为解决多天线视频多播系统中接收端天线数量各态互异性问题,本文提出了多相似描述编码MSD和复用的空时块编码M-STBC。MSD是M-STBC的基础,生成多个对视频的相似描述。基于传统STBC, M-STBC在不同的时隙中传输相似描述的不同形式。少于发送天线数量的接收端可以恢复相似描述中的相同部分,具有较多天线的接收端则还可以恢复不同的部分。从而实现了接收端的视频质量不仅可以随着信噪比的变化而改变,还可以随着有效天线数量的增加而提升。 ·本文还研究了线性视频传输在快速衰落信道下的鲁棒性。研究发现快速衰落信道中的资源分配是一个NP-hard问题。通过对快速衰落信道中视频传输的性能分析,本文提出了一个快速算法,优化了能量和带宽分配从而提高系统的鲁棒性。
[Abstract]:There are two main problems in adaptive transmission in wireless communication systems. There are three main reasons why physical layer technology can not efficiently transmit binary data. First, the performance of traditional adaptive techniques depends on the accuracy of channel estimation, and there is an essential contradiction between accuracy and real-time. Second, due to the limited number of optional rates in wireless communication systems, the rate adjustment is discontinuous. Third, the physical layer assumption of the current system is full of compressed data, and this assumption contradicts the actual measurements. As a special class of binary data, except for the efficiency problems mentioned above, The wireless transmission of video stream also has the problem of transmission efficiency caused by the independence of video bit rate control and physical layer adaptation. In addition, in multi-antenna video multicast systems, the channel conditions and the number of antennas at the receiver end are different from each other, and the worst channel conditions and the least number of antennas will become the bottleneck of the system performance. In order to solve the above two kinds of problems, this paper proposes the concept of seamless adaptation, that is, the information transmission in wireless communication can automatically adjust continuously and smoothly with the change of the channel situation. For different types of data and applications, this paper studies how to realize seamless adaptive transmission, the main research results are as follows: for binary data, this paper proposes random projection RP coding. Based on this coding, the rate variable modulation RCM and the compression coding modulation CCM are designed. First the sender incrementally increments the transmission code symbols using a fixed method until the receiver can decode correctly so that the channel estimation feedback selection rate is not required. Secondly, RP coding has no bit rate characteristic, and the transmission rate can be adjusted continuously. Thirdly, the optimized RP coding has the characteristics of compression sensing, which can make full use of the redundancy of binary data to improve the system throughput. In this paper, the linear video transmission techniques based on image processing, Cactus and AirScale, are proposed. The nonlinear processing such as quantization entropy coding and channel coding are removed and the video is transmitted directly through amplitude modulation after linear processing such as prediction transformation and energy distribution. The channel noise acts directly on the data after linear energy scaling. The video quality recovered by MMSE can be changed continuously and steadily with the change of channel conditions at the receiving end. A series of problems caused by bit rate control and physical layer rate adaptation are avoided. In this paper, multiple similarity description codes MSD and multiplex space-time block codes M-STBC.MSD are proposed as the basis of M-STBC, which generates multiple similar descriptions of video. Based on traditional STBCand M-STBC, different forms of similar description are transmitted in different time slots. The receiver with less than the number of transmitting antennas can recover the same part of the similar description, while the receiver with more antennas can also recover different parts. Thus, the video quality of the receiver can be improved not only with the change of SNR, but also with the increase of the number of effective antennas. The robustness of linear video transmission in fast fading channel is also studied in this paper. It is found that resource allocation in fast fading channels is a NP-hard problem. By analyzing the performance of video transmission in fast fading channels, a fast algorithm is proposed, which optimizes the allocation of energy and bandwidth to improve the robustness of the system.
【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN92

【共引文献】