KR20100042032A - Apparatus and metnod for receiving efficiently a signal for mimo-ofdm system according to channel condition - Google Patents

Abstract

PURPOSE: An apparatus and a method for receiving efficiently a signal for MIMO-OFDM(Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing) system according to a channel condition are provided to detect a signal by delaying a detection technique according to a channel state. CONSTITUTION: A channel state checking unit(70) checks a channel state of a received sub-carrier, and uses a post SNR to check the channel state. If the channel state is in a better state than a certain level, a signal detection unit(80) detects the signal in an ordered DFE(Decision Feedback Equalization) mode. Otherwise, the signal detection unit detects the signal in a QRD-M(QR Decomposition and M-algorithm) mode.

Description

Apparatus and metnod for receiving efficiently a signal for MIMO-OFDM system according to channel condition}

The present invention relates to signal detection in multiple transmission and reception, and more particularly, to a signal detector method that significantly reduces the complexity while maintaining the performance of signal detection.

Recently, as high-speed data transmission is required in a wireless communication environment, interest in MIMO-OFDM is increasing. Accordingly, many MIMO-OFDM signal detection techniques have been studied. However, the maximum likelihood detection (MLD) with optimal performance increases exponentially with increasing levels of transmit antennas and modulation schemes. To compensate for this, QRD-M has been studied, whose performance is close to MLD, and whose calculation amount is smaller than that of MLD. However, QRD-M is also complicated and expensive to use in reality.

The technical problem to be solved by the present invention relates to a digital communication system method using a multiple input multiple output orthogonal frequency division multiplexing (hereinafter referred to as 'MIMO-OFDM') method of a multi-antenna system. The present invention relates to a method of reducing complexity by detecting signals using different detection techniques.

In order to achieve the above object, the apparatus for receiving a MIMO-OFDM signal according to a channel state according to the present invention includes a channel state investigator for examining a channel state of each received subcarrier; And detecting a signal in an ordered decision feedback equalization (DFE) method when the channel state is better than a predetermined criterion according to the checked channel state, and performing a QRD-M method when the channel state is not better than a predetermined criterion. And a signal detector for detecting a signal.

In order to achieve the above object, according to the present invention, a method for receiving a MIMO-OFDM signal according to a channel state includes: examining a channel state of each received subcarrier; And detecting a signal in an ordered decision feedback equalization (DFE) method when the channel state is better than a predetermined criterion according to the irradiated channel state, and performing a QRD-M method when the channel state is not better than a predetermined criterion. Detecting the signal.

According to the present invention, by detecting different signal detection schemes according to channel conditions, compared to an optimal signal detection scheme in a conventional MIMO-OFDM system, the present invention can reduce the complexity and reduce the complexity considerably. In other words, the signal detection method for optimal performance in multiple transmission and reception has a high complexity and high cost, but according to the present invention, if the signal detection technique is detected by different channel conditions, the complexity can be reduced by more than 60% in similar performance. It is cost effective in producing MIMO-OFDM system.

Hereinafter, an apparatus for receiving a MIMO-OFDM signal according to a channel state according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a MIMO-OFDM transmission / reception system including an apparatus for receiving a MIMO-OFDM signal according to a channel state according to the present invention. As shown in FIG. 1, a MIMO-OFDM transmitter having N _t transmit antennas and a MIMO-OFDM receiver having N _r receive antennas are provided. The modulator 10, the serial / parallel converter 20, the plurality of IFFTs 30 and the plurality of digital / analog converters 40 corresponding to the components of the MIMO-OFDM transmitter correspond to the general components of the transmitter. Omit the description. Meanwhile, the plurality of analog / digital converters 50, the plurality of FFTs 60, the parallel / serial converter 90, and the demodulator 100 corresponding to the components of the MIMO-OFDM receiver correspond to the general components of the receiver. , Detailed description is omitted. However, the features of the present invention are the channel state investigator 70 and the signal detector 80 of the MIMO-OFDM receiver, which will be described in detail below with reference to these two components.

If the subcarrier of one OFDM symbol is referred to as K and the transmitted data is X, the OFDM symbol transmitted from the mth transmission antenna may be represented by Equation 1 below.

In OFDM symbol, different data are simultaneously transmitted in N _t antennas. In this case, the k-th subcarrier signal among the K subcarrier signals received by the receiver may be represented by Equation 2 below.

Here, j, i represents the transmit and receive antenna symbols, and X ^k , Y ^k , and N ^k can be represented by Equation 3 below.

X ^k is a transmission symbol transmitted in a transmission antenna having a size of N _t * 1, Y ^k is a reception symbol received at a receiving end of N _r * 1, and N ^k is a vector of noise having a Gaussian form of N _r * 1.

Meanwhile, H ^k in Equation 2 may be represented by a matrix of Equation 4 below.

Where H ^k has the size of N _r * N _t . H _{i, j} represents the channel state between the j th transmit antenna and the i th receive antenna, and it is assumed that H _{i, j} passes through the multipath channel.

Prior to the description of the signal detection in the present invention, it is assumed that time and frequency synchronization and channel estimation are optimal, and it is assumed that signal detection at the receiver is performed for each subcarrier.

The channel state investigation unit 70 examines the channel state of each subcarrier received through the N _r reception antennas, and outputs the result of the investigation to the signal detection unit 80. At this time, the channel state investigation unit 70 examines the channel state using a post signal-to-noise ratio (Post SNR). Post SNR may be represented by Equation 5 below.

은 i번째 부반송파의 송신 파워를 의미 하고, G(i)는 Moore-Penrose pseudo-inverse 행렬을 의미하고, σ² 는 노이즈 분산값을 의미한다. Moore-Penrose pseudo-inverse 행렬은 아래의 두 수학식 6 및 7의 행렬로 나타낼 수 있는데, 수학식 6은 잡음이 고려가 되지 않은 zero-forcing 방식을 나타내고, 수학식 7은 잡음이 고려되지 않은 minimum mean square error(MMSE) 방식을 나타낸다. here,

Denotes the transmit power of the i-th subcarrier, G (i) denotes a Moore-Penrose pseudo-inverse matrix, and σ ² denotes a noise variance value. The Moore-Penrose pseudo-inverse matrix can be represented by the following two equations (6) and (7), where equation 6 represents a zero-forcing method where noise is not considered, and equation 7 represents a minimum where noise is not considered. Mean square error (MMSE) method.

이다. 만약

값이 크게 되면, post SNR이 작게 되어 채널의 상태가 좋지 않다고 판단이 되고, 반대로

값이 작게 되면 post SNR이 크게되어 채널의 상태가 좋다고 판단이 된다. 따라서 모든 부판송파 각각에 대해 수학식 6과 7 중 하나를 선택해

를 구하게 된다.

를 구한 후 작은 값부터 큰 값 순서대로 정렬을 하게 된다. 정렬된 부반송파의 인덱스는 아래의 수학식 8으로 표현할 수 있다.In Equation 5 for determining the state of the channel, since the power of the transmission signal is transmitted with a total power of 1, the influence on the post SNR is very small. Most of the effects on the post SNR are the power values of the Moore-Penrose pseudo-inverse matrix obtained from Equations 6 and 7.

to be. if

If the value is large, the post SNR is small, and the state of the channel is judged to be bad.

If the value is small, the post SNR is increased to determine that the channel is in good condition. Therefore, for each subcarrier, select one of Equations 6 and 7

Will be obtained.

After getting, we sort in order from small value to big value. The index of the aligned subcarriers can be expressed by Equation 8 below.

의 값이 가장 작은 부반송파의 인덱스를 뜻하고, l_K는

의 값이 가장 큰 부반송파의 인덱스를 뜻다. Where l ₁ is

Is the index of the subcarrier with the smallest value, and l _K is

Is the index of the largest subcarrier.

The signal detector 80 detects a signal in an ordered ordered decision feedback equalization (DEF) method when the channel state is better than a predetermined criterion according to the channel state irradiated by the channel state investigator 70, and the channel state is higher than the predetermined criterion. If not, the signal is detected by the QRD-M method. The detected signal is output to the bottle / serial converter 90. Here, the predetermined criterion means a predetermined S value, and this S value is a number of signal detection methods having low complexity and poor performance among the total number of subcarriers. For example, if the value of S is 32 when the total number of subcarriers is 64 in one OFDM symbol, l ₁ to l ₃₂ in the subcarrier indexes arranged in Equation 8 are low in complexity, and according to a poor signal detection method The signal is detected and the signal is detected according to a high complexity and high performance signal detection method from l ₃₃ to l ₆₄ among the aligned subcarrier indexes.

 The ordered DFE method calculates the power of each row in the Moore-Penrose pseudo-inverse matrix obtained by Equations 6 or 7 for each subcarrier. The power of each row is sorted in small order and the index corresponding to the row is stored as shown below.

In this case, the value of P ₁ is the best state of the channel and the value of P _Nt is the state of bad channel.

라고 한다. According to this value, the original channel equation 4 can be arranged as shown in equation 10 below, and the matrix

It is called.

The matrix of the channel shown in Equation 10 is decomposed through QR decomposition as in the following equation.

는 정규직교 행렬 이고

인 성질을 만족한다.

은 상 삼각행렬이다. 수학식 2에 수학식 11에서 분해된 채널의 형태를 대입하고 양변에

을 곱하면 아래와 같이 표현을 할 수 있다.

는

행렬의 conjugate transpose 행렬이다.here

Is orthonormal matrix

Satisfies the nature of phosphorus.

Is a phase triangular matrix. Substituting the shape of the channel decomposed in Equation 11 into Equation 2,

If you multiply by, you can express

Is

The conjugate transpose matrix of the matrix.

은 역시 가우시안 분포를 가지는 백터가 된다. 그리고 보낸 신호의 추정은 아래의 수학식 13과 같다.here

Also becomes a vector with Gaussian distribution. The estimation of the sent signal is shown in Equation 13 below.

에 의해서 재 정렬되었으므로 아래의 수학식 14를 통해서 다시 정렬을 하게 된다.The estimated signal here

Since it is rearranged by Equation 14 is to be rearranged again.

Here, the rearranged signal is assumed to be a signal sent from the transmitting antenna.

QRD-M method will start from the equation (12). The QRD-M technique is based on the M algorithm. Each path selects paths with high reliability from the extended path metric values, that is, with small Euclidean distance values, and does not consider paths determined to be unlikely. The path metric calculation of the first step of the QRD-M is calculated as in Equation 15 below.

는 송신신호의 모든 가능한 성상도를 나타낸다. 만약 송신신호가 QPSK 로 변조 했을 경우 4가지의 모든 가능한 값 들을 나타내게 된다. 수학식 15에서 구한 값들 중에서 우리는 M개의 작은 값들을 남기고 나머지 값 들은 버리게 된다. 다음 단계의 path metric 계산은 아래의 수학식 16과 같이 계산되어진다.here

Represents all possible constellations of the transmitted signal. If the transmitted signal is modulated with QPSK, all four possible values are displayed. Among the values obtained in Equation 15, we leave M small values and discard the remaining values. The path metric calculation in the next step is calculated as in Equation 16 below.

의 값은 전 단계에서 남겨진 성상도의 값을 나타낸다. 수학식 16에서의 path metric 값은 M*(성상도 크기)의 metric으로 확장이 되고, 여기서도 작은 값을 가지는 metric중 M 개만 남기고 나머지는 모두 버리게 된다. 계속 위와 같은 방식으로 아래와 같은 일반적인 수학식 17로 path metric 값을 구하게 된다.here

The value of represents the constellation value left in the previous step. The path metric value in Equation 16 is extended to a metric of M * (constellation size), and only M of the metric having a small value is left and all others are discarded. In the same manner as above, the path metric value is calculated using the following general equation (17).

In the final step, the smallest metric is selected to estimate the remaining constellation values as the transmission signal. FIG. 2 is a diagram illustrating a QRD-M estimation method. In FIG. 2, an estimation method when QPSK modulation, four transmission antennas and M = 2 is shown.

Meanwhile, in the present invention, an adaptive QR decomposition and M-algorithm (hereinafter, referred to as M-algorithm) that efficiently adjusts M, the number of survivor paths in each layer, is estimated by estimating the state of a channel changing in real time in a multiple-input multiple-output (MIMO) system. ('Adaptive QRD-M') can be used. QRD-M using fixed M irrespective of channel conditions requires a large amount of M, because it must use a large value M to not miss the correct path in order to approach the performance of maximum-likelihood detection (MLD). . To compensate for this, a technique for properly adjusting M by estimating the channel state by using the components of the channel matrix has been proposed, but only using the channel gain component that changes in every frame, it is impossible to use the information about the received noise that is changing every moment. Has its drawbacks. The present invention proposes a QRD-M technique that more efficiently adjusts M by reflecting not only channel gain but also information on instantaneous received noise without measuring noise power. If the channel environment is good, the ratio of the two path metrics having the smallest value is determined by taking advantage of the fact that the smallest path metric value is significantly smaller than the other path metric values. It was estimated. Since the proposed method adjusts M appropriately, it has an optimal performance close to the MLD and significantly reduces the amount of computation compared to the conventional QRD-M method.

Adaptive QRD-M operates in a similar way to QRD-M, but uses the channel state obtained in Equation 8 to allocate less M value if the channel is good, and allocates more M value if the channel is bad. Use Therefore, when reordering the channels according to the channel state obtained in Equation 8, the ordered DFE shows the state of the channel as shown in Equation 9, but when using Adaptive QRD-M, You will sort the states.

In this case, the state of the channel in the last column indicates a bad state, so the value of M is largely reversed, and the state of the channel in the first column is a good case. Therefore, by reducing the value of M, the complexity can be reduced compared to QRD-M.

3 shows the complexity of the conventional signal detection method and the method applied to the present invention. When calculating the complexity, the addition calculation is not considered because it is easy to implement in hardware. Only the multiplication operation is considered. One product of imaginary numbers is equal to four times real and real. In addition, the transmit antenna, the receive antenna number is 4, the modulation technique uses 16 QAM, and the number left in each step left of the Adaptive QRD-M in the scheme applied to the present invention is 16,8,4 and the number of QRD-M There are 16 M left in all stages. If the subcarrier 64 and S is 32, the signal corresponding to 32 subcarriers out of 64 detects the signal using the ordered DFE method, and the rest detects the signal using the adaptive QRD-M. That is, two signals are detected in the present invention using one ordered DFE method and Adaptive QRD-M method. The complexity is 7360 + 872 = 8232 and both signals are detected using QRD-M. In this case, since the complexity is 24832 times, the complexity is reduced by about 67% by the present invention, and when all 64 subcarriers are detected, the complexity is reduced by 83%.

4 to 6 is a performance comparison graph according to the present invention. The simulations assume that time and frequency synchronization and channel estimation are perfect.

4 is a diagram comparing bit error rates of QRD-M and adaptive QRD-M. The complexity of the QRD-M is 1.5 times higher, but the bit error rate is only about 1 [dB] difference from 10 ^-5 .

5 is a graph showing how the bit error rate differs as the S value changes. It can be seen that there is almost no difference in performance until S is subcarrier / 2.

FIG. 6 shows a performance graph when the S value is fixed to subcarrier / 2 and the number of Ms left in adaptive QRD-M changes. When the number of Ms left is 16, 8, and 4, the complexity decreases by 83%, but the difference in performance is that the bit error rate is about 0.5 [dB] from 10 ^-5 . You can see that the complexity is much lower but the performance is not significantly different.

Hereinafter, a method of receiving a MIMO-OFDM signal according to a channel state according to the present invention will be described in detail with reference to the accompanying drawings.

7 is a flowchart of an embodiment for explaining a method of receiving a MIMO-OFDM signal according to a channel state according to the present invention.

First, the channel state of each received subcarrier is examined (step 200).

이다. 따라서 모든 부판송파 각각에 대해 수학식 6과 7 중 하나를 선택해

를 구하게 된다.

를 구한 후 작은 값부터 큰 값 순서대로 정렬을 하게 된다.Investigation of channel conditions is carried out using Post SNR. The post signal to noise ratio (Post SNR) is obtained using Equation 5 described above. In Equation 5, which determines the state of the channel, the power of the transmitted signal is sent with 1 as the total power. Therefore, the influence on the Post SNR is mostly the power value of the Moore-Penrose pseudo-inverse matrix obtained from Equations 6 and 7.

to be. Therefore, for each subcarrier, select one of Equations 6 and 7

Will be obtained.

After getting, we sort in order from small value to big value.

After step 200, according to the irradiated channel state, when the channel state is better than a predetermined criterion, the signal is detected by an ordered decision feedback equalization (DFE) scheme, and when the channel state is not better than the predetermined criterion, the QRD-M scheme is performed. In step 202, the signal is detected. In particular, when the channel state is good, the M value is assigned small, and when the channel state is not good, the M value is assigned large. Here, the predetermined criterion means a predetermined S value, and this S value is a number for selecting a signal detection method having low complexity and poor performance among the total number of subcarriers. For example, if the value of S is 32 when the total number of subcarriers is 64 in one OFDM symbol, l ₁ to l ₃₂ in the subcarrier indexes arranged in Equation 8 are low in complexity, and according to a poor signal detection method The signal is detected and the signal is detected according to a high complexity and high performance signal detection method from l ₃₃ to l ₆₄ among the aligned subcarrier indexes. Ordered Decision Feedback Equalization (DFE) method, QRD-M method and Adaptive QRD-M method is as described above, detailed description thereof will be omitted.

Meanwhile, the above-described method for receiving a MIMO-OFDM signal according to a channel state of the present invention may be implemented by computer-readable codes / instructions / programs. For example, it may be implemented in a general-purpose digital computer for operating the code / instructions / program using a computer-readable recording medium. The computer-readable recording medium may be a magnetic storage medium (eg, ROM, floppy disk, hard disk, magnetic tape, etc.), optical reading medium (eg, CD-ROM, DVD, etc.) and carrier wave (eg Storage media, such as through the Internet). In addition, embodiments of the present invention may be implemented as a medium (s) containing computer readable code, such that a plurality of computer systems connected through a network may be distributed and processed. Functional programs, codes and code segments for realizing the present invention can be easily inferred by programmers in the art to which the present invention belongs.

The apparatus and method for receiving a MIMO-OFDM signal according to the channel state of the present invention have been described with reference to the embodiment shown in the drawings for clarity, but this is merely an example, and those skilled in the art will appreciate It will be understood that various modifications and other equivalent embodiments are possible. Therefore, the true technical protection scope of the present invention will be defined by the appended claims.

1 is a block diagram illustrating a MIMO-OFDM transmission and reception system including an apparatus for receiving a MIMO-OFDM signal according to the present invention.

2 is a diagram illustrating a QRD-M estimation method.

3 is a table comparing the complexity of the conventional signal detection method and the method applied to the present invention.

4 is a diagram illustrating the performance of QRD-M and Adaptive QRD-M.

5 is a diagram comparing QRD-M and the signal estimation technique of the present invention with different values of S. FIG.

FIG. 6 is a view showing fixed S as a subcarrier / 2 and changing M value for each step in Adaptive QRD-M.

7 is a flowchart of an embodiment for explaining a method of receiving a MIMO-OFDM signal according to the present invention.

Claims (8)

A channel state investigator for examining a channel state of each received subcarrier; And According to the examined channel state, the signal is detected by an ordered decision feedback equalization (DFE) method when the channel state is better than a predetermined criterion, and by the QRD-M method when the channel state is not better than a predetermined criterion. MIMO-OFDM signal receiving apparatus according to the channel state characterized in that it comprises a signal detecting unit for detecting. The method of claim 1, wherein the channel state inspection unit MIMO-OFDM signal receiving apparatus according to the channel state, characterized in that for irradiating using a post signal-to-noise ratio (Post SNR). The method of claim 2, The post signal-to-noise ratio (Post SNR) is obtained by using the following equation.

Denotes the transmit power of the i-th subcarrier, G (i) denotes a Moore-Penrose pseudo-inverse matrix, and σ ² denotes a noise variance value. The method of claim 1, wherein the signal detection unit When the signal is detected by the QRD-M method, if the channel state is good, the M value is assigned to be small, and if the channel state is not good, the M value is to be assigned to the MIMO-OFDM signal according to the channel state. Receiving device. Examining a channel state of each received subcarrier; And According to the examined channel state, the signal is detected by an ordered decision feedback equalization (DFE) method when the channel state is better than a predetermined criterion, and by the QRD-M method when the channel state is not better than a predetermined criterion. And receiving the MIMO-OFDM signal according to the channel state. 6. The method of claim 5, wherein examining the channel condition MIMO-OFDM signal receiving method according to the channel state characterized in that the investigation using a post signal-to-noise ratio (Post SNR). The method of claim 6, The post-signal-to-noise ratio (Post SNR) is obtained by using the following equation.

Denotes the transmit power of the i-th subcarrier, G (i) denotes a Moore-Penrose pseudo-inverse matrix, and σ ² denotes a noise variance value. The method of claim 5, wherein detecting the signal by the QRD-M method And a small M value if the channel state is good, and a large M value if the channel state is not good.

Images