CN101079863A

CN101079863A - Frequency domain balancer design method in orthogonal frequency division multiplexing system

Info

Publication number: CN101079863A
Application number: CN 200710042745
Authority: CN
Inventors: 王勇; 金彦亮; 陈惠民; 徐卫兴
Original assignee: SHANGHAI UNIVERSITY
Current assignee: SHANGHAI UNIVERSITY
Priority date: 2007-06-26
Filing date: 2007-06-26
Publication date: 2007-11-28

Abstract

The invention relates to a design method of a frequency domain equalizer in an orthogonal frequency division multiplexing system. Design method of the present invention is to utilize MMSE criterion to seek optimal frequency-domain equalizer tap weight coefficient expression, comprise the forward system and compensation system of frequency-domain equalizer, the frequency-domain equalizer obtained can be able to under MMSE criterion Completely eliminate the residual interference power formed by ISI and ICI, and at the same time effectively suppress the residual noise power, that is, an optimal frequency domain equalizer. The structure of the improved model of the frequency domain equalizer is not complicated, and the computational complexity is also very low. It only needs to use channel estimation and signal-to-noise ratio estimation to determine the tap weight coefficient of the frequency domain equalizer for normal operation. The designed frequency domain equalizer works very robustly when the CP length is smaller than the channel impulse response length, and can maintain good performance without increasing the CP length, improving the frequency band utilization of the system.

Description

Design Method of Frequency Domain Equalizer in OFDM System

技术领域technical field

本发明涉及一种无线通信领域的信道均衡技术，具体是一种正交频分复用(OFDM)系统中的频域均衡器设计方法。The invention relates to a channel equalization technology in the field of wireless communication, in particular to a design method of a frequency domain equalizer in an Orthogonal Frequency Division Multiplexing (OFDM) system.

背景技术Background technique

随着无线通信、数据通信和Internet的飞速发展与日益融合，移动用户对于多种业务的需求不断增加的同时，对通信速率的要求也不断提高。无线通信技术正在经历着前所未有的发展机遇。同时无线领域研究的任何进展都必须紧紧围绕着一个主题，那就是必须能够较好的解决高速无线通信自身所面临的带宽紧张、传播信道恶劣、移动环境复杂和服务受限时的质量问题。高速率引起的信道宽带频率选择性多径衰落，同时移动终端的快速移动或者说是传播环境周围散射体的不断变化使得信道具有多普勒时间选择性衰落效应。这种联合时间-频率选择性造成信道的多径-多普勒-衰落，能够严重影响系统的总体性能。而由于多径效应引起的码间干扰(ISI)是限制无线通信系统提高传输速率的主要因素。因此，在系统设计上，人们最感兴趣的问题是如何设计整个通信系统，以消除未知信道对传输信号的失真和加性高斯噪声的影响。众所周知，在接收端对接收机进行优化设计，即均衡器的设计，是解决上述问题的有效途径之一。With the rapid development and increasing integration of wireless communication, data communication and the Internet, while mobile users' demands for various services are increasing, the requirements for communication rates are also increasing. Wireless communication technology is experiencing unprecedented development opportunities. At the same time, any progress in wireless research must focus on a theme, that is, it must be able to better solve the quality problems faced by high-speed wireless communication itself, such as bandwidth constraints, poor propagation channels, complex mobile environments, and limited services. The broadband frequency selective multipath fading caused by the high rate, and the fast movement of the mobile terminal or the constant change of the scatterers around the propagation environment make the channel have Doppler time selective fading effect. This joint time-frequency selectivity causes multipath-Doppler-fading of the channel, which can seriously affect the overall performance of the system. The intersymbol interference (ISI) caused by the multipath effect is the main factor that limits the improvement of the transmission rate of the wireless communication system. Therefore, in system design, people are most interested in how to design the entire communication system to eliminate the influence of unknown channels on the transmission signal distortion and additive Gaussian noise. As we all know, optimizing the design of the receiver at the receiving end, that is, the design of the equalizer, is one of the effective ways to solve the above problems.

正交频分复用(OFDM)是一种频谱有效的多载波技术，它利用一定数量的彼此正交的窄带子载波来并行地传送低速率数据从而实现整体高速数据的传输。OFDM系统利用循环前缀能有效的抑制码间干扰(ISI)和载波间干扰(ICI)，但是由于循环前缀的插入致使系统的频带利用率下降，而且一旦信道的冲激响应长度大于CP的长度，CP将无法完全消除ISI和ICI，那么系统的性能将会由于残留的ISI和ICI的存在而急速下降。Orthogonal Frequency Division Multiplexing (OFDM) is a spectrally efficient multi-carrier technology, which uses a certain number of narrowband sub-carriers that are orthogonal to each other to transmit low-rate data in parallel to achieve overall high-speed data transmission. The OFDM system can effectively suppress inter-symbol interference (ISI) and inter-carrier interference (ICI) by using the cyclic prefix, but the frequency band utilization of the system decreases due to the insertion of the cyclic prefix, and once the impulse response length of the channel is greater than the length of the CP, CP will not be able to completely eliminate ISI and ICI, then the performance of the system will drop rapidly due to the existence of residual ISI and ICI.

为了解决这两个问题，目前出现了各种各样的均衡方案。这些方案主要分为两类，一种是缩短信道方案，另外一种是非线性干扰抵消器。缩短信道方案是利用逆滤波缩短信道冲激响应的有效长度。这种方案在时域和频域都可以实现。时域均衡器利用时域有限冲激响应滤波器来补偿信道冲激响应，而频域均衡器通过把有限冲激响应滤波器从时域搬到频域实现了进一步的改进。为了完成均衡，这种缩短信道的方案仍需增加一定的冗余度。非线性干扰抵消器是用预判决符号和最大似然检测器的输出作为抵消器的复本。虽然这种方案能获得很好的均衡性能，但是它的计算复杂度是不能接受的。In order to solve these two problems, various equalization schemes have emerged. These schemes are mainly divided into two categories, one is a shortened channel scheme, and the other is a nonlinear interference canceller. The channel shortening scheme uses inverse filtering to shorten the effective length of the channel impulse response. This scheme can be realized in both time domain and frequency domain. Time-domain equalizers use time-domain finite impulse response filters to compensate channel impulse responses, while frequency-domain equalizers achieve further improvements by moving finite impulse response filters from the time domain to the frequency domain. In order to complete the equalization, this scheme of shortening the channel still needs to increase a certain degree of redundancy. The nonlinear interference canceller is a replica of the canceller using the pre-decision symbols and the output of the maximum likelihood detector. Although this scheme can achieve good equalization performance, its computational complexity is unacceptable.

发明内容Contents of the invention

本发明的目的在于针对现有均衡技术中存在的不足，提供一种正交频分复用系统中的频域均衡器设计方法。这种方法的出发点是在误码率最小或者均方误差最小条件下，求取性能最佳的均衡器结构模型，这种方法打破了过去传统均衡器的设计思想，即先设定均衡器的结构，这由先验知识或假设得到，然后着眼于算法的研究，以图改进均衡器的性能，相反本发明完全不设定结构，即不对均衡器做任何结构有约束的先验假设，按最小均方误差准则(MMSE)求取具有最佳均衡效果的系统模型，然后根据这种具有最佳效果的均衡器系统模型进行结构实现。本发明中的均衡器分为初级模型和改进模型，由于白高斯噪声的存在使得初级模型无法完全消除ISI和ICI，而改进模型在理论上是能完全消除ISI和ICI的。The purpose of the present invention is to provide a frequency domain equalizer design method in an OFDM system aiming at the deficiencies in the existing equalization technology. The starting point of this method is to obtain an equalizer structure model with the best performance under the condition of minimum bit error rate or minimum mean square error. structure, which is obtained by prior knowledge or hypothesis, and then focuses on the research of algorithms to improve the performance of the equalizer. On the contrary, the present invention does not set the structure at all, that is, it does not make any structurally constrained prior assumptions on the equalizer, according to The minimum mean square error criterion (MMSE) seeks the system model with the best equalization effect, and then realizes the structure according to the equalizer system model with the best effect. The equalizer in the present invention is divided into a primary model and an improved model. Due to the existence of white Gaussian noise, the primary model cannot completely eliminate ISI and ICI, while the improved model can completely eliminate ISI and ICI in theory.

本发明的主要内容是依据OFDM系统接收模型，求出在MMSE准则下的频域均衡器抽头参数表达式，为了彻底消除OFDM系统中存在的ISI和ICI，在初级频域均衡器模型的基础上求出了频域均衡器改进模型中补偿系统的参数表达式。在上述的理论基础下，本发明提出了具体的频域均衡器结构。The main content of the present invention is to obtain the tap parameter expression of the frequency domain equalizer under the MMSE criterion according to the OFDM system receiving model, in order to thoroughly eliminate ISI and ICI existing in the OFDM system, on the basis of the primary frequency domain equalizer model The parameter expressions of the compensation system in the improved model of the frequency domain equalizer are obtained. Based on the above theoretical basis, the present invention proposes a specific frequency domain equalizer structure.

为达到上述目的，本发明的技术构思和原理如下：In order to achieve the above object, technical conception and principle of the present invention are as follows:

(1)初级模型的建立与分析(1) Establishment and analysis of the primary model

OFDM系统接收机接收到的频域信号即频域均衡器的输入可以表示为：The frequency domain signal received by the OFDM system receiver, that is, the input of the frequency domain equalizer can be expressed as:

X(k)＝H(k)S(k)+N(k) (1)X(k)＝H(k)S(k)+N(k)

频域均衡器的输出为：The output of the frequency domain equalizer is:

Y(k)＝C(k)X(k) (2)Y(k)＝C(k)X(k)

而频域均衡器的估计误差为：And the estimation error of the frequency domain equalizer is:

$E E. ((k k)) = = \overset{^^}{S S} ((k k)) - - Y Y ((k k)) - - - - - - ((33))$

假设 $\hat{S} (k) = S (k)$ ，并将(1)和(2)代入(3)式，可以得到均衡器的均方误差为：suppose $\hat{S} (k) = S (k)$ , and substituting (1) and (2) into (3), the mean square error of the equalizer can be obtained as:

定义ε_I(k)和ε_N(k)分别为残留的干扰(ISI和ICI)功率和残留的噪声功率，根据MMSE准则，通过求取满足以下偏导数C(k)的值使得ε(k)有最小值：Define ε _I (k) and ε _N (k) as the residual interference (ISI and ICI) power and residual noise power respectively, according to the MMSE criterion, by finding the value that satisfies the following partial derivative C(k) so that ε(k ) has a minimum value of:

$\frac{&PartialD; &PartialD; ϵ ϵ ((k k))}{&PartialD; &PartialD; C C ((k k))} = = {σ σ}_{s the s}^{22} {C C}^{* *} ((k k)) {| | H h ((k k)) | |}^{22} - - {σ σ}_{s the s}^{22} H h ((k k)) + + {σ σ}_{n no}^{22} {C C}^{* *} ((k k)) = = 00 - - - - - - ((55))$

求得频域均衡器的前向系统抽头权系数：Find the forward system tap weight coefficient of the frequency domain equalizer:

$C C ((k k)) = = {σ σ}_{s the s}^{22} {H h}^{* *} ((k k)) {[[{σ σ}_{n no}^{22} + + {σ σ}_{s the s}^{22} {| | H h ((k k)) | |}^{22}]]}^{- - 11} - - - - - - ((66))$

为了进一步简化分析，可以将求出的C(k)代入得到最小的均方误差：In order to further simplify the analysis, the calculated C(k) can be substituted to obtain the minimum mean square error:

${ϵ ϵ}_{min min} ((k k)) = = {σ σ}_{s the s}^{22} {σ σ}_{n no}^{22} {[[{σ σ}_{n no}^{22} + + {σ σ}_{s the s}^{22} {| | H h ((k k)) | |}^{22}]]}^{- - 11} - - - - - - ((77))$

残留的干扰功率：Residual interference power:

${ϵ ϵ}_{I I} ((k k)) = = {σ σ}_{s the s}^{22} {σ σ}_{n no}^{44} {[[{σ σ}_{n no}^{22} + + {σ σ}_{s the s}^{22} {| | H h ((k k)) | |}^{22}]]}^{- - 22} - - - - - - ((88))$

残留的噪声功率：Residual noise power:

${ϵ ϵ}_{N N} ((k k)) = = {σ σ}_{n no}^{22} {σ σ}_{s the s}^{44} {| | H h ((k k)) | |}^{22} {[[{σ σ}_{n no}^{22} + + {σ σ}_{s the s}^{22} {| | H h ((k k)) | |}^{22}]]}^{- - 22} - - - - - - ((99))$

下面将对频域均衡器进行性能分析：The performance analysis of the frequency domain equalizer will be performed as follows:

当 $σ_{n}^{2} &RightArrow; 0,$ C(k)＝H^-1(k)，ε_I(k)→0，ε_min(k)＝ε_N(k)→0。频域均衡器变为一个完全的逆滤波器，ISI和ICI将被全部抵消，显然噪声功率越小，频域均衡器越理想，但实际情况是噪声不但存在，而且还具有相当的强度。下面看一个反面情况当

C (k) &RightArrow; σ_{s}^{2} H^{*} (k),

ϵ_{\min} (k) = ϵ_{I} (k) &RightArrow; σ_{s}^{2},

ε_N(k)→0。频域均衡器退化为一个匹配滤波器，输出的有用信息被ISI和ICI淹没，几乎不存在正确的信息，系统不能正常的工作。从上面的分析得出噪声对频域均衡器的初级模型影响非常大。为此必须对其加以改进。when

σ_{no}^{2} &Right Arrow; 0,

C(k)=H ⁻¹ (k), ε _I (k)→0, ε _min (k)=ε _N (k)→0. The frequency domain equalizer becomes a complete inverse filter, and ISI and ICI will be completely canceled out. Obviously, the smaller the noise power, the more ideal the frequency domain equalizer is, but the actual situation is that the noise not only exists, but also has a considerable intensity. Let's look at the opposite case when

C (k) &Right Arrow; σ_{the s}^{2} h^{*} (k),

ϵ_{\min} (k) = ϵ_{I} (k) &Right Arrow; σ_{thes}^{2},

_εN (k)→0. The frequency domain equalizer degenerates into a matched filter, the useful information output is submerged by ISI and ICI, almost no correct information exists, and the system cannot work normally. From the above analysis, it is concluded that noise has a great influence on the primary model of the frequency domain equalizer. For this it must be improved.

(2)改进模型的建立与分析(2) Establishment and analysis of improved model

由于频域均衡器的初级模型在有噪声的情况下不能全部抵消ISI和ICI，所以在初级模型的基础上引入一个补偿系统B(k)，使均衡器的输出变为：Since the primary model of the frequency domain equalizer cannot completely cancel ISI and ICI in the presence of noise, a compensation system B(k) is introduced on the basis of the primary model, so that the output of the equalizer becomes:

Y(k)＝C(k)X(k)+B(k)S(k) (10)Y(k)＝C(k)X(k)+B(k)S(k)

此时均衡器的均方误差为：At this time, the mean square error of the equalizer is:

$ϵ ϵ ((k k)) = = {σ σ}_{s the s}^{22} {| | 11 - - C C ((k k)) H h ((k k)) - - B B ((k k)) | |}^{22} + + {σ σ}_{n no}^{22} {| | C C ((k k)) | |}^{22} - - - - - - ((1111))$

类似的定义 $ϵ_{I} (k) = σ_{s}^{2} {| 1 - C (k) H (k) - B (k) |}^{2},$ $ϵ_{N} (k) = σ_{n}^{2} {| C (k) |}^{2}$ 。根据MMSE准则，可以求得使均方误差取最小值时的补偿系统抽头权系数：similar definition $ϵ_{I} (k) = σ_{thes}^{2} {| 1 - C (k) h (k) - B (k) |}^{2},$ $ϵ_{N} (k) = σ_{no}^{2} {| C (k) |}^{2}$ . According to the MMSE criterion, the tap weight coefficient of the compensation system when the mean square error takes the minimum value can be obtained:

$B B ((k k)) = = 11 - - C C ((k k)) H h ((k k)) = = {σ σ}_{n no}^{22} {[[{σ σ}_{n no}^{22} + + {σ σ}_{s the s}^{22} {| | H h ((k k)) | |}^{22}]]}^{- - 11} - - - - - - ((1212))$

利用初级模型中求得的 $C (k) = σ_{s}^{2} H^{*} (k) {[σ_{n}^{2} + σ_{s}^{2} {| H (k) |}^{2}]}^{- 1}$ ，最终得到了改进模型的各项参数。将C(k)和B(k)代入误差函数表达式可得：ε_I(k)＝0， $ϵ_{\min} (k) = ϵ_{N} (k) = σ_{n}^{2} σ_{s}^{4} {| H (k) |}^{2} {[σ_{n}^{2} + σ_{s}^{2} {| H (k) |}^{2}]}^{- 2}$ 。从上述结果可知，改进的频域均衡器模型才是真正的最佳频域均衡器，它能彻底抵消ISI和ICI，使信息无衰减的通过，输出的均方误差要比初级模型小，仅由噪声产生，改进的频域均衡器模型的性能已经到达了极限，这是按MMSE准则所能得到的最理想性能。obtained from the primary model $C (k) = σ_{the s}^{2} h^{*} (k) {[σ_{no}^{2} + σ_{thes}^{2} {| h (k) |}^{2}]}^{- 1}$ , and finally get the parameters of the improved model. Substitute C(k) and B(k) into the error function expression to get: ε _I (k)=0, $ϵ_{\min} (k) = ϵ_{N} (k) = σ_{no}^{2} σ_{the s}^{4} {| h (k) |}^{2} {[σ_{no}^{2} + σ_{thes}^{2} {| h (k) |}^{2}]}^{- 2}$ . From the above results, it can be known that the improved frequency domain equalizer model is the real optimal frequency domain equalizer, it can completely cancel ISI and ICI, make the information pass through without attenuation, and the mean square error of the output is smaller than that of the primary model, only Generated by noise, the performance of the improved frequency domain equalizer model has reached the limit, which is the best performance that can be obtained according to the MMSE criterion.

为了进一步简化分析和处理，均衡器的前向系统和补偿系统可变为：In order to further simplify the analysis and processing, the equalizer's forward system and compensation system can be changed to:

C(k)＝snrH^*(k)[1+snr|H(k)|²]^-1 (13)C(k)＝snrH ^* (k)[1+snr|H(k)| ² ] ^-1 (13)

B(k)＝[1+snr|H(k)|²]^-1 (14)B(k)＝[1+snr|H(k)| ² ] ^-1 (14)

其中 $snr = σ_{s}^{2} / σ_{n}^{2}$ 是系统接收信噪比。因此，只要在OFDM系统接收端获得信道冲激响应的频域估计值和接收信噪比的估计值，就能将频域均衡器的各项参数唯一确定。这也使得信道估计和信噪比估计的精确程度将直接影响频域均衡器的均衡性能。可以根据不同系统对性能要求的不同而采用适合不同情况的信道估计算法和信噪比估计算法。in $snr = σ_{thes}^{2} / σ_{no}^{2}$ is the system receiving signal-to-noise ratio. Therefore, as long as the estimated value of the channel impulse response in the frequency domain and the estimated value of the received signal-to-noise ratio are obtained at the receiving end of the OFDM system, the parameters of the frequency domain equalizer can be uniquely determined. This also makes the accuracy of channel estimation and signal-to-noise ratio estimation directly affect the equalization performance of the frequency domain equalizer. Channel estimation algorithms and signal-to-noise ratio estimation algorithms suitable for different situations can be adopted according to different performance requirements of different systems.

根据上述的发明技术构思和原理，本发明采用下述技术方案：According to above-mentioned invention technical design and principle, the present invention adopts following technical scheme:

一种正交频分复用系统中的频域均衡器设计方法，其特征在于设计步骤为：A frequency-domain equalizer design method in an OFDM system, characterized in that the design steps are:

1)初级模型参数的求解方法及结果；1) The solution method and results of primary model parameters;

2)初级模型与补偿系统构成的改进模型参数的求解方法及结果；2) The solution method and results of the improved model parameters composed of the primary model and the compensation system;

3)各模型参数的最终实现表达式；3) The final realization expression of each model parameter;

4)改进模型的结构设计；4) Improve the structural design of the model;

5)改进模型结构在应用中的实现。5) Improve the realization of the model structure in the application.

上述的初级模型的求解方法是：OFDM系统接收机接收到的频域信号即频域均衡器的输入为X(k)＝H(k)S(k)+N(k)，频域均衡器的输出为Y(k)＝C(k)X(k)，频域均衡器的均方误差 $ϵ (k) = E [{| E (k) |}^{2}] = σ_{s}^{2} {| 1 - C (k) H (k) |}^{2} + σ_{n}^{2} {| C (k) |}^{2}$ ，定义残留的干扰功率为 $ϵ_{I} (k) = σ_{s}^{2} {| 1 - C (k) H (k) |}^{2}$ ，残留的噪声功率为 $ϵ_{N} (k) = σ_{n}^{2} {| C (k) |}^{2}$ ，在MMSE准则下，求得使均方误差取最小值时的频域均衡器初级模型抽头权系数结果为：The solution method of the above-mentioned primary model is: the frequency domain signal received by the OFDM system receiver, that is, the input of the frequency domain equalizer is X(k)=H(k)S(k)+N(k), and the frequency domain equalizer The output of Y(k)=C(k)X(k), the mean square error of the frequency domain equalizer $ϵ (k) = E. [{| E. (k) |}^{2}] = σ_{thes}^{2} {| 1 - C (k) h (k) |}^{2} + σ_{no}^{2} {| C (k) |}^{2}$ , defining the residual interference power as $ϵ_{I} (k) = σ_{thes}^{2} {| 1 - C (k) h (k) |}^{2}$ , the residual noise power is $ϵ_{N} (k) = σ_{no}^{2} {| C (k) |}^{2}$ , under the MMSE criterion, the result of the tap weight coefficient of the primary model of the frequency domain equalizer when the mean square error takes the minimum value is:

$C C ((k k)) = = {σ σ}_{s the s}^{22} {H h}^{* *} ((k k)) {[[{σ σ}_{n no}^{22} + + {σ σ}_{s the s}^{22} {| | H h ((k k)) | |}^{22}]]}^{- - 11}$

上述的初级模型与补偿系统构成的改进模型参数的求解方法是：频域均衡器的输入与初级模型中相同，输出为Y(k)＝C(k)X(k)+B(k)S(k)，频域均衡器的均方误差 $ϵ (k) = E [{| E (k) |}^{2}] = σ_{s}^{2} {| 1 - C (k) H (k) - B (k) |}^{2} + σ_{n}^{2} {| C (k) |}^{2}$ ，在MMSE准则下，求得使均方误差取最小值时的频域均衡器补偿系统抽头权系数结果为：The method for solving the improved model parameters of the above-mentioned primary model and compensation system is: the input of the frequency domain equalizer is the same as that of the primary model, and the output is Y(k)=C(k)X(k)+B(k)S (k), the mean square error of the frequency domain equalizer $ϵ (k) = E. [{| E. (k) |}^{2}] = σ_{thes}^{2} {| 1 - C (k) h (k) - B (k) |}^{2} + σ_{no}^{2} {| C (k) |}^{2}$ , under the MMSE criterion, the result of obtaining the tap weight coefficient of the frequency domain equalizer compensation system when the mean square error takes the minimum value is:

$B B ((k k)) = = 11 - - C C ((k k)) H h ((k k)) = = {σ σ}_{n no}^{22} {[[{σ σ}_{n no}^{22} + + {σ σ}_{s the s}^{22} {| | H h ((k k)) | |}^{22}]]}^{- - 11}$

改进模型的前向系统抽头权系数结果与初级模型抽头权系数C(k)相同。The tap weight coefficient result of the improved model is the same as the tap weight coefficient C(k) of the primary model.

上述的各模型参数的最终实现表达式是：将前向系统C(k)与补偿系统B(k)的表达式进一步化简，变为实际应用中能被求出的某种值的函数表达式，将C(k)与B(k)中的分子分母同时除以σ_s ²，前向系统与补偿系统的表达式变为：The final realization expressions of the above-mentioned model parameters are: further simplify the expressions of the forward system C(k) and the compensation system B(k) into a function expression of a certain value that can be obtained in practical applications Divide the numerator and denominator in C(k) and B(k) by σ _s ² at the same time, the expressions of forward system and compensation system become:

C(k)＝snrH^*(k)[1+snr|H(k)|²]^-1 C(k)＝snrH ^* (k)[1+snr|H(k)| ² ] ^-1

B(k)＝[1+snr|H(k)|²]^-1 B(k)＝[1+snr|H(k)| ² ] ^-1

其中 $snr = σ_{s}^{2} / σ_{n}^{2}$ 是系统接收信噪比，能在系统的接收端被估计出来，H(k)同样能在系统的接收端被估计出来，这样上面的C(k)与B(k)便能在现实系统中获得。in $snr = σ_{thes}^{2} / σ_{no}^{2}$ is the system receiving signal-to-noise ratio, which can be estimated at the receiving end of the system, and H(k) can also be estimated at the receiving end of the system, so that the above C(k) and B(k) can be calculated in the real system get.

上述的改进模型的结构设计方法为：在FFT处理后，经过一个具有FFT点数个单抽头的有限冲激响应滤波器组，滤波器组的输出一路送入硬判决器A，一路送入一组加法器。硬判决器的输出再经过一个具有FFT点数个单抽头的有限冲激响应滤波器组，此滤波器组的输出同样作为上述这组加法器的输入，这组加法器的输出送入硬判决器B，硬判决器B的输出也就是频域均衡器改进模型的输出；有限冲激响应滤波器组也可以用一组乘法器来代替，每个乘法器的一路输入是经过FFT处理后的一路数据，另一路输入是原有限冲激响应滤波器组的一个抽头权系数；假设FFT的点数为N，那么整个频域均衡器的初始模型结构由一个N抽头的有限冲激响应滤波器组和一个硬判决器组成，而整个频域均衡器的改进模型结构是由两个N抽头的有限冲激响应滤波器组、N个加法器和两个硬判决器共同组成。其中可以用N个乘法器替换N抽头的逆滤波器组。The structural design method of the above-mentioned improved model is as follows: after FFT processing, through a finite impulse response filter bank with FFT points and several single taps, the output of the filter bank is sent to the hard decision device A all the way, and sent to a group of adder. The output of the hard decision device passes through a finite impulse response filter bank with several single taps of FFT points. The output of this filter bank is also used as the input of the above-mentioned group of adders, and the output of this group of adders is sent to the hard decision device. B, the output of the hard decision device B is the output of the improved model of the frequency domain equalizer; the finite impulse response filter bank can also be replaced by a set of multipliers, and one input of each multiplier is the one after FFT processing The other input is a tap weight coefficient of the original finite impulse response filter bank; assuming that the number of FFT points is N, then the initial model structure of the entire frequency domain equalizer consists of an N-tap finite impulse response filter bank and The improved model structure of the whole frequency domain equalizer is composed of two N-tap finite impulse response filter banks, N adders and two hard decision devices. Wherein the inverse filter bank of N taps can be replaced by N multipliers.

上述的改进模型结构在应用中的实现方法是：频域均衡器的结构一旦确定，那么每次处理的实现就是确定有限冲激响应滤波器组的抽头系数，实际上也就是估计出信道冲激响应值和信噪比的值；信道估计的算法是采用最小二乘算法，通过在每个符号中插入导频的方法来得到导频位置的信道估计值，其余位置的估计值利用插值的方法得到；对于信噪比的估计是采用在线式信噪比估计方案，这种信噪比估计方案是基于对一组特定的可观测的接收数据的统计比率来获得信噪比估计值的，这种方案用到了萨姆斯和威尔逊设计的信噪比估计器，一组特定的可观测的接收数据是沿用了信道估计中的导频数据，其它位置的信噪比也利用插值的方法获得。The implementation method of the above-mentioned improved model structure in the application is: once the structure of the frequency domain equalizer is determined, the realization of each processing is to determine the tap coefficients of the finite impulse response filter bank, in fact, it is to estimate the channel impulse The value of the response value and the signal-to-noise ratio; the algorithm of channel estimation is to use the least squares algorithm, and the channel estimation value of the pilot position is obtained by inserting the pilot into each symbol, and the estimated value of the remaining positions uses the method of interpolation Obtained; for the estimation of SNR, an online SNR estimation scheme is adopted, which is based on the statistical ratio of a specific set of observable received data to obtain the estimated value of SNR, which This scheme uses the signal-to-noise ratio estimator designed by Sams and Wilson. A specific set of observable received data is followed by the pilot data in channel estimation, and the signal-to-noise ratio of other positions is also obtained by interpolation.

本发明与现有技术相比较，具有如下显而易见的突出实质性特点是显著优点：Compared with the prior art, the present invention has the following obvious outstanding substantive features, which are significant advantages:

本发明利用MMSE准则求解出频域均衡器的前向和补偿系统抽头系数的最优解，由信道估计和信噪比估计算法来确定最优解的参数而最终得到一个确定的数值，最优解的数值表达式将用来更新频域均衡器的抽头系数，由于补偿系统的存在，均衡器的最终输出将不会有ISI和ICI，同时噪声功率也会得到抑制，系统的性能将得到大大的提高。本发明的频域均衡器中由于不需要迭代和多维矩阵运算使计算复杂度变得非常低，适合应用在实际的通信系统中。本发明中的频域均衡器可以解决OFDM系统中CP长度小于信道冲激响应长度而导致系统性能急速下降的情况，因此，在信道条件恶劣的情况下或是频谱资源紧张需要适当缩减CP长度的情况下，本发明的长处将会得到充分的体现。在其他系统中，同其他的频域均衡器相比，本发明也有其自身的优点，例如缩减CP长度从而提高频谱利用率，能有效抑制噪声，抵消多普勒频移产生的ICI等等。The present invention uses the MMSE criterion to solve the optimal solution of the forward and compensation system tap coefficients of the frequency domain equalizer, and determines the parameters of the optimal solution by channel estimation and signal-to-noise ratio estimation algorithms to finally obtain a definite value, the optimal The numerical expression of the solution will be used to update the tap coefficients of the frequency domain equalizer. Due to the existence of the compensation system, the final output of the equalizer will not have ISI and ICI, and the noise power will be suppressed at the same time, and the performance of the system will be greatly improved. improvement. Since the frequency domain equalizer of the present invention does not need iteration and multi-dimensional matrix operation, the calculation complexity becomes very low, and is suitable for application in actual communication systems. The frequency domain equalizer in the present invention can solve the situation that the CP length in the OFDM system is smaller than the channel impulse response length and cause the system performance to decline rapidly. Therefore, in the case of poor channel conditions or the shortage of spectrum resources, it is necessary to appropriately reduce the CP length. situation, the advantages of the present invention will be fully reflected. In other systems, compared with other frequency domain equalizers, the present invention also has its own advantages, such as reducing CP length to improve spectrum utilization, effectively suppressing noise, offsetting ICI generated by Doppler frequency shift and so on.

附图说明Description of drawings

图1是本发明初级模型的OFDM基带系统框图。Fig. 1 is a block diagram of the OFDM baseband system of the primary model of the present invention.

图2是本发明改进模型的OFDM基带系统框图。Fig. 2 is a block diagram of the OFDM baseband system of the improved model of the present invention.

图3是本发明的结构示意图。Fig. 3 is a structural schematic diagram of the present invention.

图4是加性白高斯信道下的OFDM系统误码率性能。Fig. 4 is the bit error rate performance of the OFDM system under the additive white Gaussian channel.

图5是步行B(Pedestrian-B)信道下的OFDM系统误码率性能。Fig. 5 is the bit error rate performance of the OFDM system under the pedestrian B (Pedestrian-B) channel.

图6是车载B(Vehicular-B)信道下的OFDM系统误码率性能。Fig. 6 is the BER performance of the OFDM system under the Vehicular-B (Vehicular-B) channel.

具体实施方式Detailed ways

本发明的一个优选实施例结合附图详述如下：A preferred embodiment of the present invention is described in detail as follows in conjunction with accompanying drawing:

如图1所示，要传输的数据首先经过串行/并行处理，然后经过IFFT处理之后再进行并行/串行处理，这便实现了OFDM调制。在已经调制好的每个OFDM符号前加上循环前缀CP，便可以将其送入信道。通过信道的数据将会经历信道多径、衰落、多普勒频移、加性高斯白噪声等各种影响。接收到的信号首先被移除CP，然后经过串行/并行处理等待FFT处理，经过FFT处理之后的数据便实现了OFDM解调，解调后的数据中导频位置的数据将被用来进行信道估计和信噪比估计，必便获得在导频位置的信道冲激响应和信噪比的估计值，通过插值的方法便能得到其它位置上的信道冲激响应和信噪比估计值。解调后的整个数据将会作为频域均衡器的输入，频域均衡器有初级模型和改进模型两种。As shown in Figure 1, the data to be transmitted is firstly processed serially/parallelly, and then parallelly/serially processed after IFFT processing, which realizes OFDM modulation. Add the cyclic prefix CP before each OFDM symbol that has been modulated, and then it can be sent into the channel. The data passing through the channel will experience various effects such as channel multipath, fading, Doppler frequency shift, and additive Gaussian white noise. The received signal is first removed from the CP, and then undergoes serial/parallel processing and waits for FFT processing. The data after FFT processing realizes OFDM demodulation, and the data of the pilot position in the demodulated data will be used for For channel estimation and signal-to-noise ratio estimation, the channel impulse response and signal-to-noise ratio estimate at the pilot position must be obtained, and the channel impulse response and signal-to-noise ratio estimate at other positions can be obtained through interpolation. The entire demodulated data will be used as the input of the frequency domain equalizer, and the frequency domain equalizer has two types: primary model and improved model.

如图1所示，FFT处理之后便是一个频域均衡器的初级模型，图2所示的FFT处理之后是一个频域均衡器的改进模型，具体的频域均衡器改进模型结构如图3所示。OFDM解调后的并行数据将经过频域均衡器的前向系统，每一路数据将乘上频域均衡器对应位置上的抽头权系数然后输出，输出的数据一路进入判决器A，另一路作为加法器的一路输入。判决器A此处为硬判决器，当然也可以用软判决器，其性能更优，但是结构要比硬判决器复杂。判决器A的输出作为补偿系统的输入。前向系统C(k)和判决器A构成的是一个频域均衡器的初级模型，而补偿系统B(k)的输入正是初级模型的输出。补偿系统的输出与前向系统的输出共同组成加法器的输入，加法器的输出送入判决器B，同样判决器B可以是硬判决器也可以是软判决器，只是判决的门限会随输入的信号而改变。判决器B的输出就是整个频域均衡器改进模型的输出。As shown in Figure 1, after the FFT processing is a primary model of a frequency domain equalizer, after the FFT processing shown in Figure 2 is an improved model of a frequency domain equalizer, and the specific structure of the improved model of the frequency domain equalizer is shown in Figure 3 shown. The parallel data after OFDM demodulation will pass through the forward system of the frequency domain equalizer, and each channel of data will be multiplied by the tap weight coefficient at the corresponding position of the frequency domain equalizer and then output. One input of the adder. The decider A is a hard decider here, of course a soft decider can also be used, its performance is better, but the structure is more complex than the hard decider. The output of the decision device A is used as the input of the compensation system. The forward system C(k) and the decision unit A constitute a primary model of a frequency domain equalizer, and the input of the compensation system B(k) is the output of the primary model. The output of the compensation system and the output of the forward system together form the input of the adder, and the output of the adder is sent to the decision device B. Similarly, the decision device B can be a hard decision device or a soft decision device, but the decision threshold will vary with the input signal changes. The output of decision device B is the output of the improved model of the whole frequency domain equalizer.

本正交频分复用系统中的频域均衡器设计方法的步骤为：The steps of the frequency domain equalizer design method in this OFDM system are:

4)改进模型的结构设计；4) Improve the structural design of the model;

C(k)＝snrH^*(k)[1+snr|H(k)|²]^-1 C(k)＝snrH ^* (k)[1+snr|H(k)| ² ] ^-1

B(k)＝[1+snr|H(k)|²]^-1 B(k)＝[1+snr|H(k)| ² ] ^-1

本实施例的OFDM系统频域均衡器初级模型、改进模型和传统的频域均衡器的性能比较的仿真结果如图4、图5和图6所示。The simulation results of the performance comparison between the primary model of the OFDM system frequency domain equalizer, the improved model and the traditional frequency domain equalizer in this embodiment are shown in FIG. 4 , FIG. 5 and FIG. 6 .

Claims

1. the frequency domain balancer design method in the ofdm system is characterized in that design procedure is:

1) method for solving of primary mold parameter and result;

2) method for solving and the result of the improved model parameter of primary mold and bucking-out system formation;

3) the final realization expression formula of each model parameter;

4) structural design of improved model;

5) realization of improved model structure in application.

2. the frequency domain balancer design method in the ofdm system according to claim 1, the method for solving that it is characterized in that described primary mold is: the frequency-region signal that the ofdm system receiver receives be frequency-domain equalizer be input as X (k)=H (k) S (k)+N (k), frequency-domain equalizer is output as Y (k)=C (k) X (k), the mean square error of frequency-domain equalizer

ϵ (k) = E [{| E (k) |}^{2}] = σ_{s}^{2} {| 1 - C (k) H (k) |}^{2} + σ_{n}^{2} {| C (k) |}^{2},

Defining residual interference power is

ϵ_{I} (k) = σ_{s}^{2} {| 1 - C (k) H (k) |}^{2},

Residual noise power is

ϵ_{N} (k) = σ_{n}^{2} {| C (k) |}^{2},

Under the MMSE criterion, the frequency-domain equalizer primary mold tap weights coefficient results of trying to achieve when making mean square error get minimum value is:

C (k) = σ_{s}^{2} H^{*} (k) {[σ_{n}^{2} + σ_{s}^{2} {| H (k) |}^{2}]}^{- 1}

3. the frequency domain balancer design method in the ofdm system according to claim 1, the method for solving that it is characterized in that the improved model parameter that described primary mold and bucking-out system constitute is: identical in the input of frequency-domain equalizer and the primary mold, be output as Y (k)=C (k) X (k)+B (k) S (k), the mean square error of frequency-domain equalizer

ϵ (k) = E [{| E (k) |}^{2}] = σ_{s}^{2} {| 1 - C (k) H (k) - B (k) |}^{2} + σ_{n}^{2} {| C (k) |}^{2},

Under the MMSE criterion, the frequency-domain equalizer bucking-out system tap weights coefficient results of trying to achieve when making mean square error get minimum value is:

B (k) = 1 - C (k) H (k) = σ_{n}^{2} {[σ_{n}^{2} + σ_{s}^{2} {| H (k) |}^{2}]}^{- 1}

The forward direction system tap weights coefficient results of improved model is identical with primary mold tap weights coefficient C (k).

4. the frequency domain balancer design method in the ofdm system according to claim 1, the final realization expression formula that it is characterized in that described each model parameter is: with the expression formula further abbreviation of the C of forward direction system (k) with bucking-out system B (k), become the function expression of certain value that can be obtained in the practical application, with the molecule denominator among C (k) and the B (k) simultaneously divided by σ _s ², the expression formula of forward direction system and bucking-out system becomes:

C(k)＝snrH ^*(k)[1+snr|H(k)| ²] ^-1

B(k)＝[1+snr|H(k)| ²] ^-1

Wherein

snr = σ_{s}^{2} / σ_{n}^{2}

Be system's received signal to noise ratio, can be estimated at the receiving terminal of system that H (k) can be estimated at the receiving terminal of system equally, Shang Mian C (k) just can obtain in reality system with B (k) like this.

5. the frequency domain balancer design method in the ofdm system according to claim 1, the construction design method that it is characterized in that described improved model is: after FFT handles, through a finite impulse response filter group with several the single taps of FFT point, hard decision device A is sent in the output one tunnel of bank of filters, and one the tunnel sends into one group of adder.The output of hard decision device is again through a finite impulse response filter group with several the single taps of FFT point, the output of this bank of filters is organized the input of adder as above-mentioned this equally, the output of this group adder is sent into hard decision device B, the output of the hard decision device B output of frequency-domain equalizer improved model just; The finite impulse response filter group also can replace with one group of multiplier, and one tunnel input of each multiplier is the circuit-switched data after handling through FFT, and another road input is a tap weights coefficient of former limited impact response filter group; Suppose that counting of FFT is N, the initial model structure of so whole frequency-domain equalizer is made up of finite impulse response filter group and a hard decision device of a N tap, and the improved model structure of whole frequency-domain equalizer is made of jointly finite impulse response filter group, a N adder and two hard decision devices of two N taps.Wherein can replace the inverse filterbank of N tap with N multiplier.

6. the frequency domain balancer design method in the ofdm system according to claim 1, it is characterized in that the implementation method of described improved model structure in application is: the structure of frequency-domain equalizer is in case determine, so each realization of handling is exactly a tap coefficient of determining the finite impulse response filter group, in fact just estimates the value of channel impulse response value and signal to noise ratio; The algorithm of channel estimating is to adopt least-squares algorithm, obtains channel estimation values of pilot frequency positions by the method for inserting pilot tone in each symbol, and the estimated value of all the other positions utilizes the method for interpolation to obtain; Estimation for signal to noise ratio is to adopt online signal-to-noise ratio (SNR) estimation scheme, this signal-to-noise ratio (SNR) estimation scheme is based on statistics ratio to one group of specific observable reception data and obtains the signal-to-noise ratio (SNR) estimation value, this scheme has been used the SNR estimator of Sa Musi and the inferior design of Weir, one group of specific observable reception data is to have continued to use the pilot data in the channel estimating, and the signal to noise ratio of other position also utilizes the method for interpolation to obtain.