本文为瑞典查尔姆斯理工大学（作者：Sheng Huang & Mengbai Tao）的硕士论文，共72页。

随着高速数据传输需求的快速增长，多输入多输出（MIMO）技术在当前的通信网络中得到了广泛的应用，如宽带码分多址系统（WCDMA）。关于WCDMA下行链路中的MIMO已经研究了很长一段时间，而在WCDMA上行链路中，如何选择预编码的规范还没有确定。预编码是多输入多输出（MIMO）系统中的一个重要组成部分，目的是减轻数字编码信号序列中不同数据流之间的干扰。目前，MIMO系统在预编码方面存在两个问题。由于预编码算法需要信道信息，因此一个问题是很难对频率选择性信道中的信道特性进行建模。另一个问题是，只有有限的反馈才适用于实际系统，因为如果需要将全信道信息发送给发送机，反馈开销的成本将远远大于预编码的好处。本文的目的是在WCDMA上行链路反馈有限的情况下，通过选择合适的预编码算法，研究给定误码率（BLER）下所需信噪比（SNR）范围内的最大潜在增益。

根据我们的仿真模拟，在反馈有限的情况下，预编码的最大潜在增益是可观的（与固定预编码相比，在0:5 BLER时为1:2 dB）。在理想信道估计的假设下，通过采用最接近SVD（CSVD）算法，可以分别获得总流和主流的0:6 dB增益和1dB增益。值得指出的是，单独预编码不能完全缓解串扰。为了进一步降低MIMO中串扰的影响，需要在预编码的同时实现适当的信道均衡。预编码的性能在很大程度上也取决于信道估计的可靠性。因此，为了在商业通信系统中实现一种良好的预编码算法，信道估计需要更加可靠，而预编码算法也会增加系统开销，增大系统复杂度。因此，应该权衡预编码所获得的收益和相应的复杂性。

With rapidly increasing demand in high-speed data transmission, Multiple-Input Multiple-Output (MIMO) techniques are widely used in current communication networks, like Wideband Code Division Multiple Access (WCDMA). MIMO in downlink of WCDMA has been studied for a long time, while in WCDMA uplink, the specification of how to choose precoder is still not defined yet. Precoding is an essential element in MIMO, aiming at mitigating interference between different data streams that are sequences of digitally coded signals. Currently, a MIMO system suffer from two problems regarding precoding. Since channel information is needed for the precoding algorithm, one problem then is that it is difficult to model the channel characteristics in a frequency selective channel. The other problem is that only limited feedback is applicable in real system, since the cost of feedback overhead is far larger than the benefits from precoding if full channel information needs to be sent to the transmitter. The objective of this thesis is to investigate the maximum potential gain in the required Signal-to-Noise Ratio (SNR) range for a given Block Error Rate (BLER) by using precoding and selecting a suitable precoding algorithm in the case of limited feedback in the WCDMA uplink.

The maximum potential gain of precoding is considerable (1:2 dB at 0:5 BLER compared to a fixed precoder) in the case of limited feedback according to our simulations. By applying the proposed closest-to-SVD (CSVD) algorithm, 0:6 dB gain and 1 dB gain can be achieved for total stream and the primary stream, respectively, when ideal channel estimation is assumed. It is worth to point out that precoding alone can not mitigate crosstalk completely. Suitable channel equalization needs to be implemented together with precoding in order to further decrease the impact of crosstalk in MIMO. The performance of precoding depends largely on the reliability of the channel estimation as well. Thus, channel estimation needs to be more reliable in order to implement a good precoding algorithm in a commercial communication system, which on the other hand will cost more overhead and increase the system complexity. Hence, a trade-off should be made regarding the gain obtained from precoding and the complexity of the algorithm.

1 引言

2 WCDMA与MIMO理论

3 WCDMA标准与仿真器描述

4 算法与相关结果

5 讨论、结论与未来研究工作

附录A 仿真器评估 

下载英文原文地址：

http://page2.dfpan.com/fs/dl4cdj02c2f14219160/ 

更多精彩文章请关注微信号：