KR101193792B1 - SIGNAL DETECTING METHOD OF Multiple Input Multiple Output SYSTEM USING LATTICE REDUCTION AND APPARATUS THEREOF - Google Patents

Abstract

The present invention relates to a signal detection method and apparatus for a MIMO system using LR. In the signal detection method of the MIMO system using the LR according to the present invention, receiving a signal from the transmitting terminal having a multi-antenna in the L-QAM method, QR decomposition of the received signal to remove the orthonormal matrix components, Obtaining a Euclidean distance between a received signal and L symbols transmitted in the L-QAM scheme, selecting V candidate symbols having a short Euclidean distance as a first candidate symbol, and using a lattice reduction (LR) scheme Detecting the remaining undetected symbols forming the candidate columns with each of the V candidate symbols, and selecting a signal sequence closest to the transmission signal among the V candidate columns by using a maximum likelihood (ML) scheme.
As described above, according to the present invention, in the MIMO system, the first V candidate symbols of the transmitted signal are determined, and the remaining symbols are reduced by reducing the condition number of the channel through the LLL algorithm and detecting the signal with the modified channel. The bit error rate is much improved and the transmission signal can be detected more accurately than the basic detection method through the LLL algorithm.

Description

SIGNAL DETECTING METHOD OF Multiple Input Multiple Output SYSTEM USING LATTICE REDUCTION AND APPARATUS THEREOF}

The present invention relates to a signal detection method and apparatus for a MIMO system using LR, and more particularly, to a signal of a MIMO system capable of detecting a signal transmitted through multiple transmission antennas in a state having a lower complexity and higher accuracy. A detection method and apparatus therefor.

Recently, as high-speed data transmission is required in a wireless communication environment, an interest in a multiple input multiple output (MIMO) method, which is a next-generation wireless system transmission technology using multiple antennas, is increasing. In particular, V-BLAST (vertical Bell Lab Layered space time) technique can be applied to many high-speed communication because it can significantly increase the transmission rate. In MIMO systems using the V-BLAST technique, it is important to find and select an appropriate signal detection technique with low complexity and high performance.

Accordingly, MIMO signal detection techniques are also actively studied. Zero-forcing (ZF), minimum mean square error (MMSE), and linear detection methods are well known methods for detecting transmitted signals with low complexity.

However, the linear detection method not only has a lot of performance degradation, but the complexity increases significantly as the level of the transmission antenna and the modulation scheme increases, which is expensive to use in reality due to the high complexity.

In particular, in the ordered successive interference cancellation (OSIC) technique based on the general MIMO linear detection technique and the linear detection technique, performance degradation is caused by the noise width in the linear detection process. This is because, if the condition number of the channel gain matrix is large, the noise is amplified greatly when the weight is multiplied by the received signal.

Therefore, in order to overcome the performance of the linear detection method, a LR (lattice reduction) based detection method has been proposed. LR-based detection methods can significantly increase the performance of linear detection methods with additional low complexity, but still do not have optimal performance.

In the LR-based detector method, it is very important that the first detected signal is accurately detected. This is because the first detected signal affects the entire signal from the next detected signal to the last detected signal. Therefore, a detection technique is needed to secure this problem.

Accordingly, an object of the present invention is to provide a signal detection method and apparatus for a MIMO system using LR that can accurately detect a first signal among signals transmitted through multiple transmission antennas in a MIMO system.

In order to solve this problem, a signal detection method of a MIMO system using an LR according to an embodiment of the present invention comprises the steps of: receiving a signal from a transmitting terminal having multiple antennas in an L-QAM scheme, QR-decomposing the received signal Removing a normal orthogonal matrix component, obtaining a Euclidean distance between the received signal and L symbols transmitted in the L-QAM scheme, selecting V candidate symbols having a short Euclidean distance as a first candidate symbol, Detecting the remaining undetected symbols that form each of the V candidate symbols using the lattice reduction (LR) method and the closest to the transmission signal among the V candidate columns using the maximum likelihood (ML) method; Selecting a signal sequence.

)을 다음의 수학식을 이용하여 검출할 수 있다. In the step of detecting the remaining undetected symbols for the V candidate symbols using the LR scheme, the remaining undetected symbols for the V candidate columns (

) Can be detected using the following equation.

Z (k) is a received signal from which orthonormal matrix components are removed, r (k) is an upper triangular matrix for QR decomposition, and X (k) is the candidate symbol.

In the step of selecting the signal sequence closest to the transmission signal among the V candidate columns using the ML method, the ML method may use the following equation.

은 최종적으로 선택된 상기 V개의 후보 열들이고, y는 상기 수신 신호이다. remind

Are the finally selected V candidate columns, and y is the received signal.

The signal detection apparatus of the MIMO system using the LR according to an embodiment of the present invention, a signal receiving unit for receiving a signal in a L-QAM method from a transmitting terminal having multiple antennas, QR-decomposed the received signal to form a normal orthogonal matrix component A QR decomposer to remove, a Euclidean distance between the received signal and L symbols transmitted in the L-QAM scheme, and a candidate symbol detector for selecting the V candidate symbols having the short Euclidean distance as the first candidate symbol, LR ( an LR detector for detecting the remaining undetected symbols that form each of the V candidate symbols using a lattice reduction scheme, and a signal sequence closest to a transmission signal among the V candidate sequences using a maximum likelihood (ML) scheme. It includes a ML detection unit for selecting.

As described above, according to the present invention, in the MIMO system using the LR, the first V candidate symbols of the transmitted signal are determined, and the remaining symbols are reduced by reducing the condition number of the channel through the LLL algorithm and detecting the signal with the modified channel. In addition, the bit error rate is much improved and the transmission signal can be detected more accurately than the basic detection method through the conventional LLL algorithm.

1 is a view for explaining a MIMO communication system according to an embodiment of the present invention.
2 is a flowchart illustrating a signal detection method of a receiving terminal according to an embodiment of the present invention.
3 is a diagram illustrating a signal detection process of a receiving terminal according to an embodiment of the present invention.
4 is a graph showing bit error rate results when M = 4 in the embodiment of the present invention.
FIG. 5 is a graph showing bit error rate results when M = 16 in an embodiment of the present invention. FIG.
6 is a graph showing bit error rate results according to the number of candidate symbols in an embodiment of the present invention.

DETAILED DESCRIPTION Embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. Those skilled in the art to which the present invention pertains extend the operation according to the present invention by using a symbol by referring to the following description by means of a frame, and thus it is easily understood that a separate description thereof is omitted. do.

1 is a view for explaining a MIMO communication system according to an embodiment of the present invention.

As shown in Figure 1, the transmitting terminal 100 is N _t Has multiple transmit antennas, and the receiving terminal 200 has N _r multiple receive antennas.

Receiving terminal 200 according to an embodiment of the present invention is a signal detection apparatus, and the configuration thereof in more detail, the receiving terminal 200 is a signal receiving unit 210, QR decomposition unit 220, candidate symbol detection unit 230 ), The LR detector 240 and the ML detector 250.

The signal receiver 210 receives a signal in an L-QAM scheme from a transmitting terminal having multiple antennas. Obviously, the L-QAM scheme includes the QPSK scheme.

The QR decomposition unit 220 removes the normal quadrature matrix component (Q component) by QR decomposition of the received signal. The candidate symbol detector 230 obtains a Euclidean distance between the received signal and the L symbols transmitted in the L-QAM scheme, and selects the V candidate symbols having the short Euclidean distance as the first candidate symbols.

The LR detector 240 detects the remaining undetected symbols that form a candidate sequence with each of the V candidate symbols by using a lattice reduction (LR) method. The ML detector 250 selects a signal sequence closest to the transmission signal among the V candidate columns using a maximum likelihood (ML) scheme.

Before describing an embodiment of the present invention, the following describes an LR-based detection method applied to an embodiment of the present invention.

The LR (Lattice reduction) detection method finds a good basis of lattice and reduces the condition number of the channel matrix using an algorithm such as Lenstra-Lenstra-Lovasz (LLL). Therefore, if the linear and nonlinear detection techniques such as MMSE and OSIC are performed using the new virtual channel matrix obtained by performing Lattice reduction, noise amplification can be greatly reduced. The LR detection method according to an embodiment of the present invention also starts by using QR decomposition of the channel matrix H (k).

Multiplying both sides of the above equation by Q ^H (k) leads to the following equation (2).

Since the Q matrix is a unitary matrix, the statistical characteristics of the noise term do not change.

Hereinafter, the LLL will be described in order to understand the present invention. For convenience of explanation, the 4X4 MIMO system will be described as an example.

If matrix R (k) is 4X4, first initialize T = I ₄ , b = 2, l = 1, μ = [R (k) _{(l, b)} / R (k) _{( l , l )} ] Obtain If μ = 0, R and T are modified by Equations 3 and 4 below.

When the matrix R (k) transformed by Equation 4 does not satisfy the following Equation 5, the positions of R (k) _{(l, b-1)} and R (k) _{(l, b)} are exchanged, and the matrix Equation 6 is performed to maintain R upperangular.

Here, 1/4 ≤ δ ≤ 1, θ is a rotation matrix (equation 7).

When the following Equation 8 is satisfied, b = b + 1, and when it is not satisfied, b = b-1. Repeat this process until b = M + 1.

Hereinafter, a signal detection method of a receiving terminal according to an embodiment of the present invention using the above-described LR-based detection method will be described with reference to FIGS. 2 and 3.

2 is a flowchart illustrating a signal detection method of a receiving terminal according to an embodiment of the present invention, Figure 3 is a view showing a signal detection process of the receiving terminal according to an embodiment of the present invention.

In detail, FIG. 3 illustrates a process of an improved detection scheme using an LLL algorithm having V candidate symbols in a system using QPSK as a modulation scheme and four transmitting antennas. In the first step, V candidate symbols are obtained and the remaining symbols are obtained by applying the LR technique. Finally, a process of finally detecting a signal through ML test among V candidate symbol strings is shown.

That is, the candidate symbol detector 230 detects the first V candidate symbols among the transmission signals, and the LR detector 240 detects the remaining candidate symbols through the LR method.

First, the signal receiving unit 210 of the receiving terminal 200 is N _t A signal as shown in Equation 1 is received from the transmitting terminal 100 having two transmitting antennas in an L-QAM scheme (S210).

When the number of detected candidate symbols is determined, the QR decomposition unit 210 performs QR decomposition on the received signal through the DFE technique (S220). Then, the received signal z (k) from which the Q component is removed is expressed by Equation 9 below.

과 각 후보 심볼 사이의 square Euclidean 거리를 계산하고, 거리 값이 작은 값부터 아래와 같이 수학식 10처럼 정렬한다. The candidate symbol detector 230 uses all L symbols of the L-QAM system for performing modulation using L symbols in the first layer as candidate symbols.

Calculate the square Euclidean distance between the candidate symbols and each candidate symbol, and arrange as shown in Equation 10 below from the smallest distance value.

The candidate symbol detector 230 selects V candidate symbols starting from a small value in accordance with the performance required by the system at the above values (S230). For convenience of explanation, assuming that x ₁ ⁽¹⁾ is determined as an estimated symbol in the first layer among the above candidate symbols as shown in the following equation, it is expressed by the following equation.

이고 아래의 수학식 12와 같은 과정으로 구한다.here

And obtain the same process as in Equation 12 below.

In order to detect the remaining symbols not yet detected, the LR detector 240 detects the remaining symbols by performing a LR (lattice reduction) detection method (S240).

That is, if the LR detection unit 240 repeats the LR technique by the selected number of candidate symbols V as shown in Equation 12, the LR detector 240 may obtain a sequence of V transmission estimation symbols as shown below.

Finally, the ML detector 250 calculates V candidate columns as shown in Equation 14 below through ML (maximum likelihood) detection, and estimates a value having the largest probability value among candidate strings as a transmission symbol (S250). ).

The signal sequence closest to the received signal is selected through Equation 15 below. The ML test can be expressed as follows.

ML detection is performed using only the V signal strings detected unlike the general ML detector method. Each operation is described in detail as shown in Equation 16 below.

에 V개의 신호 열을 대입하여 에러 값이 가장 작은 신호 열을 선택하게 된다. 본 발명에서는 이렇게 선택된 신호인

이 최종적으로 검출된 신호가 된다.Of the above expression

By substituting V signal streams, the signal train having the smallest error value is selected. In the present invention, the selected signal

This finally becomes a detected signal.

4 to 6 are graphs of performance according to an embodiment of the present invention. In FIG. 4 and FIG. 5, bit error rates may be checked. The simulation assumes that time, frequency synchronization, and channel estimation are perfect. In addition, it is assumed that four transmit antennas and four receive antennas are used.

4 is a graph showing the bit error rate results when M = 4 in the embodiment of the present invention, the modulation scheme used QPSK. FIG. 5 is a graph showing bit error rate results when M = 16 in the embodiment of the present invention, and a modulation scheme uses QPSK. In addition, looking at the graph of Figure 5, when comparing the ordered DFE and the present invention, it can be seen that there is a performance gain of about 5dB at a bit error rate of 10 ^-3 . 4 and 5, it can be seen that the detection method of the present invention has better performance than the basic LR method. This confirms the importance of the accuracy of the first detection signal in the MIMO system.

Since the first detection signal was detected with several candidate signals, the performance was better than that of the basic LR detection method. Comparing the ordered DFE with the proposed signal detection method, the ordered DFE outperformed the proposed method at low SNR, but the performance is reversed when the SFE is above a certain SNR. This is because the orederd DFE is rearranged according to the state of the channel. This is because the performance is similar at low SNR when using the ZF detection technique.

6 is a graph showing bit error rate results according to the number of candidate symbols in an embodiment of the present invention. The modulation scheme uses 16-QAM, and the noise-to-signal ratio is similar to the bit error rate before 20dB, but after 20dB, the more candidates, the better the performance. Therefore, according to the present invention, if the number of candidate groups is appropriately adjusted according to the system, the performance may be better than that of the conventional detection technique even at a lower complexity.

As described above, according to an embodiment of the present invention, in the MIMO system, the first V candidate symbols of a transmitted signal are determined, and the remaining symbols are reduced by reducing the condition number of the channel through the LLL algorithm and detecting the signal with the modified channel. In addition, the bit error rate is much improved and the transmission signal can be detected more accurately than the basic detection method through the conventional LLL algorithm.

Embodiments of the present invention include a computer-readable medium having program instructions for performing various computer-implemented operations. This medium records a program for executing the signal detection method of the MIMO system described so far. The medium may include program instructions, data files, data structures, etc., alone or in combination. Examples of such media include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD and DVD, programmed instructions such as floptical disk and magneto-optical media, ROM, RAM, And a hardware device configured to store and execute the program. Or such medium may be a transmission medium, such as optical or metal lines, waveguides, etc., including a carrier wave that transmits a signal specifying a program command, data structure, or the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, Of the right.

Claims (6)

Receiving a signal from an L-QAM scheme from a transmitting terminal having multiple antennas,
QR decomposition of the received signal based on DFE to remove a normal quadrature matrix component;
Obtaining a Euclidean distance between the received signal and L symbols transmitted in the L-QAM scheme;
Selecting the V candidate symbols having the short Euclidean distance as first candidate symbols,
Detecting the remaining undetected symbols that form a candidate sequence with each of the V candidate symbols using a lattice reduction method; and
Selecting a signal sequence closest to a transmission signal among the V candidate columns using a maximum likelihood (ML) scheme,
The signal Y (k) received by the L-QAM scheme from the transmitting terminal is a signal detection method of a MIMO system using the following LR:

Where H (K) is a transmission channel, Z (k) is a received signal from which orthonormal matrix components have been removed, Q (k) is a normal orthogonal matrix, R (k) is an upper triangular matrix for QR decomposition, and X (k) is the candidate symbol. The method of claim 1,
Detecting the remaining undetected symbols for the V candidate symbols using the LR scheme,
The remaining undetected symbols for the V candidate columns (

Signal detection method of a MIMO system using LR that detects

r (k) is an upper triangular matrix for the QR decomposition. The method of claim 2,
Selecting a signal sequence closest to a transmission signal among the V candidate columns using the ML method,
The ML method is a signal detection method of a MIMO system using LR using the following equation:

remind

Are the finally selected V candidate columns, and y is the received signal. A signal receiver for receiving a signal from a transmitting terminal having multiple antennas in an L-QAM scheme,
QR decomposition unit for removing the orthonormal matrix components by QR decomposition of the received signal based on the DFE,
A candidate symbol detector which obtains a Euclidean distance between the received signal and L symbols transmitted in the L-QAM scheme, and selects V candidate symbols having a short Euclidean distance as first candidate symbols;
An LR detection unit for detecting the remaining undetected symbols forming the candidate columns with each of the V candidate symbols using a lattice reduction (LR) method, and
An ML detector configured to select a signal sequence closest to a transmission signal among the V candidate columns by using a maximum likelihood (ML) method,
The signal (Y (k)) received from the transmitting terminal in the L-QAM scheme is a signal detection apparatus of the MIMO system using the following LR:

Where H (K) is a transmission channel, Z (k) is a received signal from which orthonormal matrix components have been removed, Q (k) is a normal orthogonal matrix, R (k) is an upper triangular matrix for QR decomposition, and X (k) is the candidate symbol. The method of claim 4, wherein
The LR detection unit,
The remaining undetected symbols for the V candidate columns (

Signal detection apparatus of the MIMO system using LR for detecting

r (k) is an upper triangular matrix for the QR decomposition. The method of claim 5,
The ML detection unit,
A signal detection apparatus of a MIMO system using LR selecting a signal sequence closest to a transmission signal among the V candidate columns by using the following equation:

remind

Are the finally selected V candidate columns, and y is the received signal.

Images