无线协作通信系统中基于网络编码的可靠传输技术研究

发布时间：2018-03-07 14:29

本文选题：协作通信　切入点：网络编码　出处：《南京邮电大学》2014年博士论文　论文类型：学位论文

【摘要】：多输入多输出技术(Multiple-Input-Multiple-Output, MIMO)在提高无线通信系统传输可靠性以及频谱效率方面，发挥着基础性作用。然而，由于受到终端尺寸、硬件复杂度和实现成本等因素的限制，导致其在工程实际中难以部署。近年来，网络编码(Network Coding, NC)协作通信技术作为MIMO的一种有效实现方式吸引了学术界的广泛关注，这是由于：一方面协作通信能够以较低的成本，通过形成虚拟MIMO来获得分集增益；另一方面在中继进行NC可以有效提升系统传输效率，解决协作通信频谱效率较低的问题。这些优点也使得NC协作通信技术被下一代移动通信标准(LTE-A)列为后续演进的关键技术之一。然而，NC在设计之初主要针对无差错的有线信道，如果将其应用于存在衰落效应的无线信道，则需要首先解决系统的可靠性传输问题。因此，本文围绕NC协作通信系统的传输可靠性问题展开研究。本文的具体研究内容包括： 1）多址接入信道下基于NC的错误传播避免技术研究中，针对基于瞬时信道增益的链路自适应方法存在准确性差的问题，提出了一种基于接收信号对数似然比(Log-Likelihood Ratio,LLR)的改进算法。新算法采用LLR作为信号可靠性指示，将基于LLR与基于瞬时信道增益的误码率解析公式进行联立，得到基于LLR换算的等效瞬时信道增益，，用该参数代替瞬时信道增益来计算功率因子。结果表明，新算法可以有效减少错误传播的影响，改善系统的误码性能。 2）多址接入信道下基于用户公平性的NC技术研究中，针对传统指标无法有效反映协作用户整体可靠性的问题，提出了一种基于用户公平性的网络编码功率分配算法。在新算法中，首先采用共同中断概率作为优化目标，推导出中继功率分配的约束条件，根据该条件对用户信号进行叠加，然后通过最大比合并(Maximum Ratio Combination, MRC)和串行干扰消除(Serial Interference Cancellation, SIC)进行解码。结果表明，新算法能够有效维护协作用户的公平性，提高协作群组整体的传输可靠性。 3）多信源多中继协作NC策略研究中，针对传统方案未充分利用中继间链路的问题，提出了一种联合中继排序(Relay Ording, RO)与最大分割距离码(Maximum Distance Separable,MDS)的改进方案。新方案利用MDS码能够积累互信息的特点，将RO思想引入其中，采用中断链路数作为排序权重对中继发送顺序进行规划，使得原本由于链路中断而无法进行NC的中继，通过解码其他中继的信号来对信源互信息进行积累，进而恢复编码机会。理论推导了新方案的系统满分集概率和系统中断概率，证明其具有扩散效应。结果表明，新方案能够有效改善系统的中断性能。 4）异步物理层网络编码(Physical-layer Network Coding, PNC)译码算法的设计与优化研究中，分析了载波相位偏移对置信传播(Belief Propagation, BP)译码算法的影响，提出了一种基于相位预失真的改进算法。新算法通过预先在发送信号中加入特定的相位偏移，然后在中继进行相位均衡以及相应迭代公式的更新，去除了相位偏移对BP译码精度的影响，从而提高译码准确性。针对星座图扭曲、误码率解析公式难以求解的问题，提出了一种相位盈亏模型，利用该模型分析了所提出方法的相位偏移纠正性能，理论证明其性能上界，并且对算法参数进行优化。基于所得出的分析结果，提出了一种控制门限，有效避免了过度失真的问题。同时，将符号偏移作为权重因子，提出一种相位偏移估计方案，改进了估计稳定性随机飘移的问题。结果表明，新算法在非实时反馈信道和非理想相位偏移估计的情况下，可以有效改善系统的误比特性能。
[Abstract]:Multiple input multiple output (Multiple-Input-Multiple-Output, MIMO) in a wireless communication system to improve the transmission reliability and spectral efficiency, plays a fundamental role. However, due to the size of the terminal, the hardware complexity and cost constraints, leading to difficult to deploy in engineering practice. In recent years, the network encoding (Network Coding. NC) cooperative communication technology as an effective way to achieve MIMO has attracted widespread attention in academic circles, this is because: on the one hand, cooperative communication can lower the cost, through the formation of virtual MIMO to obtain diversity gain; on the other hand, NC can effectively improve the transmission efficiency of the system in relay cooperative communication spectrum the problem of low efficiency. These advantages make NC cooperative communication technology is the next generation of mobile communication standard (LTE-A) as one of the key technologies for the subsequent evolution. However, NC, at the beginning of the design for cable channel error free wireless channel, if applied to the existence of fading effect, need to first solve the problem of the reliability of transmission system. Therefore, this article launches the research on transmission reliability issues around the NC cooperative communication system. The main research contents in this paper include:
1) to avoid error propagation NC technology research based on multi access channel, aiming at the problems of poor accuracy of link adaptive method based on the instantaneous channel gain, a received signal based on the log likelihood ratio (Log-Likelihood Ratio LLR) algorithm. The new algorithm uses LLR as signal reliability indicator, based on LLR the instantaneous channel gain and bit error rate formula are simultaneously based on equivalent instantaneous channel gain based on LLR conversion, to calculate the power factor with the parameters instead of the instantaneous channel gain. The results show that the new algorithm can effectively reduce the influence of error propagation error, improve the performance of the system.
2) based on the research of NC technology user fairness in multi access channel, the traditional index cannot reflect the overall reliability of collaborative user problems, this paper puts forward an algorithm of network encoding power allocation based on user fairness. In the new algorithm, the common fault probability as the optimization objectives, constraints are derived the relay power allocation, according to the conditions of superimposed on the user signal, and then through the maximum ratio combining (Maximum Ratio, Combination, MRC) and successive interference cancellation (Serial Interference, Cancellation, SIC) for decoding. The results show that the new algorithm can effectively maintain the fairness of the cooperation of users, improve the reliability of the transmission of the whole cooperative group.
3) of multi source multi relay cooperative NC strategy, the traditional scheme does not make full use of the relay link between the problem, and propose a joint relay ordering (Relay Ording, RO) and the maximum distance code (Maximum Distance Separable segmentation, MDS) is proposed. The new solution using MDS codes can accumulate mutual information the concept of RO is introduced, which uses interrupt number as the rank weights of the relay link transmission order planning, which was due to the relay link interruption can not NC, the other relay decoded signal to the source of mutual information accumulation, and then restore the encoding opportunity. Theoretical derivation the system outage probability for full diversity the probability and system, prove its diffusion effect. The results show that the new scheme can effectively improve the outage performance system.
4) asynchronous physical layer network encoding (Physical-layer Network Coding, PNC) design and optimization of decoding algorithm, analysis the carrier phase offset on the belief propagation (Belief Propagation, BP) of decoding algorithm, an improved algorithm is proposed based on phase predistortion. The new algorithm by pretreatment with specific phase shift in the transmit signal, then the phase equilibrium and the corresponding iterative update formula in the relay, remove the effect of phase shift on BP decoding accuracy, so as to improve the decoding accuracy. According to the constellation distortion, bit error rate formula to solve the problem, proposes a phase financing model, analysis the proposed phase offset correction performance methods by using the model, the theory prove that the performance of upper bound, and the algorithm parameters were optimized. Based on the results of the analysis results, this paper proposed a control threshold, effective To avoid excessive distortion. At the same time, the symbol offset as weighting factor, proposed a phase offset estimation scheme, improved stability of random drift estimation problem. The results show that the new algorithm in non real time feedback channel and non ideal phase offset estimation, can effectively improve the BER performance of the system.

【学位授予单位】：南京邮电大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN929.5

【相似文献】