多天线系统中信号发射方案与检测技术研究

【作者】 项玉玲；

【导师】 李少谦；

【作者基本信息】 电子科技大学 ， 通信与信息系统， 2008， 硕士

【摘要】 多入多出(MIMO)通信系统通过在发射端和接收端设计多个天线,能最大程度地利用无线信道的信道容量,实现高速可靠通信。正交频分复用(OFDM)技术能够有效地对抗多径传播引起的频率选择性衰落,将频率选择性衰落信道转化为平坦衰落信道。将MIMO技术和OFDM技术结合起来构成的MIMO-OFDM系统充分发挥了二者的优势,具有非常实际的意义。本文在上述的MIMO系统以及MIMO-OFDM系统的基础上,研究了接收端的检测技术和发射端的预编码技术。对于MIMO系统中接收端的检测技术虽然已经相当的成熟,但是基于快衰落信道下已编码的MIMO-OFDM系统中的软输出检测算法还缺乏研究。本论文的第三章针对现有的大多数带信道编码的系统,在原有的检测算法基础上研究了四种不同检测算法的软输出算法,包括联合最大似然检测,简化最大似然检测,迫零检测以及判决反馈检测,并对其中的判决反馈检测算法提出了抑制错误传播的改进算法。本论文第四章研究了通过在接收端进行实际的信道估计和噪声方差估计,发射端如何利用反馈回来的信道信息和噪声方差信息进行自适应的功率分配以获得系统误码率性能的提高。研究了等功率分配、等SNR功率分配以及拉格朗日功率分配三种功率分配方案下系统的误码率性能差异。通过研究发现,拉格朗日功率分配算法性能最优,等SNR功率分配算法算法简单且性能与拉格朗日功率分配算法十分接近且算法简单,具有很大的实用价值。二者相对于等功率分配算法都有相当的性能提升,因此当拉格朗日功率分配算法出现分配盲点时,可以联合等SNR功率分配算法进一步改善性能。在强相关信道下,仅通过功率分配不足以改善此时严重恶化的误码率性能,此时通过选择部分发射天线并且在已选的部分发射天线上有效分配功率能够提高发射天线上的信噪比,最终获得更大的性能增益。在第五章研究了发射端未知信道状态信息时,基于空频码的循环延迟分集技术的预编码发射方案。该方案通过在多天线的OFDM调制信号中引入不同的循环延迟,增加了信道的频率选择性,提高了频率分集增益,能够很大程度的改善系统的误码率性能。并分析比较了该发射方案和3GPP提案中CDD Precoding的循环延迟分集方案的性能差异。最后通过仿真验证了基于空频码的循环延迟分集发射方案选择合适的循环延迟,可以达到逼近经典的Alamouti空时编码方案的分集增益,实现简单且不受发射天线数的限制。更多还原

【Abstract】 By setting multiple antennas in transmitter and/or receiver, the Multi-input Multi-output (MIMO) system can exploit the best channel capacity and realize high rate and reliable communication. Orthogonal Frequency-Division Multiplexing (OFDM) is a fine technique to mitigate the effect of frequency-selective fading. Therefore, the MIMO-OFDM system, which takes advantage of these two techniques, has a very promising prospect. This thesis mainly investigates the detection techniques and the pre-coding techniques in the MIMO system and MIMO-OFDM system.Although the detection algorithms in MIMO systems have been broadly researched, the soft-output detection algorithm in coded MIMO-OFDM system under fast fading channel is still an open question. In Chaper 3, four soft-output detection algorithms are investigated, including joint maximum likelihood detection, simple maximum likelihood detection, decision feedback detection and zero forcing detection. Furthermore, an improved algorithm of decision feedback detection is proposed, which can suppress the error propagation.Chapter 4 investigates the pre-coding techniques, including the power allocation technique and the antenna selection technique with Channel State Information in the Transmitter (CSIT). Three different transmit power allocation (TPA) schemes, including equal power allocation scheme, equal SNR power allocation scheme and Lagrange power allocation scheme, are studied and compared. Simulation results show that Lagrange power allocation scheme has the best performance. However, it has the highest computational complexity. Equal SNR power allocation scheme can approach the same performance of the Lagrange scheme but has lower computational complexity. Furthermore, both the two schemes have better performance than equal power allocation scheme. In the strong correlative channel, only power allocation scheme cannot improve the deteriorative BER performance. Therefore, we investigate the joint scheme of power allocation technique and antenna selection technique. By selecting partial antennas at the transmitter, the effective SNR of the transmitting antennas is increased and the performance will be improved.Chaper 5 investigates the pre-coding scheme which combines space-frequency coding and cyclic delay diversity technology (SFC-CDD) without CSIT. In order to increase the frequency selectivity of the channel, this scheme introduces different cyclic delays into the OFDM symbols. In this approach, the frequency diversity gain and the BER performance of this scheme could be greatly improved. Furthermore, we analyse and compare SFC-CDD scheme and the CDD Pre-coding scheme proposed in 3GPP LTE proposals. Simulation results indicate that the performance of the SFC-CDD scheme is much better than CDD Pre-coding scheme. It can almost approach the performance of classical Alamouti space-time coding technique. However, compared with the Alamouti space-time coding, the SFC-CDD scheme is not limited by the number of transmit antennas and can be realized simply.更多还原