KR20120071437A - Method and apparatus of adaptive transmission signal detection for decoding multiple stream - Google Patents

Abstract

PURPOSE: An adaptive transmission signal detection method for decoding multiple streams is provided to effectively detect transmission signals by controlling calculation rates of a sphere decoder. CONSTITUTION: A reception apparatus receives a reception signal vector(S100). The reception apparatus estimates channel matrix(S110). The reception apparatus decomposes QR(Quick Response)(S120). The reception apparatus calculates initial radius(S130). The reception apparatus calculates a plurality of space divisions of dimensional(S140).The reception apparatus detects an ML(Maximum Likehood) point within each allocation space(S150). The reception apparatus restores a transmission signal vector(S160).

Description

Adaptive transmission signal detection method and apparatus for multi-stream decoding {METHOD AND APPARATUS OF ADAPTIVE TRANSMISSION SIGNAL DETECTION FOR DECODING MULTIPLE STREAM}

 The present invention relates to signal processing, and more particularly, to an adaptive transmission signal detection method and apparatus for decoding received signals input in multiple streams into transmission signals.

When a transmitter and a receiver exchange signals in a communication system, signals may be transmitted or received through a plurality of signal streams. That is, a transmitter such as a base station, a terminal, or a repeater creates and transmits a plurality of signal streams, and the plurality of signal streams are received by the receiver through a channel. In this case, when the receiver is configured with one or more units, the plurality of transmitted signal streams are output in the form of a received signal vector for every unit time to be input to the decoder or the detection unit of the receiver.

As a method of detecting a transmission signal originally transmitted from a plurality of received signals, various methods in consideration of detection error and calculation amount have been proposed. The maximum likelihood detection method may be used as the detection method having the lowest detection error. A sphere decoding technique may be used as an algorithm for obtaining maximum likelihood detection performance. Koo Hok-ho's technique is described in M. O. Damen and H. El Gamal, “On Maximum-Likelihood Detection and the search for the closest lattice point”, IEEE Trans. Inform. Theory, vol. 49, no. 10, pp. 2389-2402, OCT. 2003 and E. Agrell, T. Errikson, A. Vardy, and K. Zeger, “Closest point search in lattices,” IEEE Trans. Inform. Theory, vol. 48, pp. 2201-2214, Aug. See 2002. Although the decoding method guarantees the maximum likelihood detection performance, the amount of computation varies depending on the condition of the channel matrix (especially the condition number) or the signal-to-noise ratio (SNR) multiplied by the transmitted signal vector. This is a big disadvantage. That is, when the state number of the channel matrix is close to 1 and the SNR is not low, the calculation amount is smaller than that of other detection methods. B. Hassibi and H. Vikalo, “On the sphere-decoding Algorithm I. Expected complexity,” IEEE Trans. Signal processing, vol. 53, pp. According to 2806-2818 and 2005, when the SNR is not low, the average calculation amount of the decoding method may have a value of three or four squares of the number of transmission signal streams. On the other hand, when the channel environment is poor or the SNR is low, the computational amount is so large that the algorithm cannot be sufficiently performed when the detection time is delayed or the system capability is limited.

Therefore, there is a need for a signal processing method in which a combination of an algorithm of the decoding method and an algorithm for adaptively controlling the computation amount of the decoding method is required.

An object of the present invention is to provide an adaptive transmission signal detection method and apparatus for multi-stream decoding.

In one aspect, an adaptive transmission signal detection method in a communication system is provided. The adaptive transmission signal detection method includes receiving a received signal vector, estimating a channel matrix of a channel through which the received signal vector is rough, performing QR decomposition based on the estimated channel matrix, and initial detection grid point And calculating an initial radius based on the initial detection lattice point, calculating dimensions of a plurality of divided spaces based on the initial radius, and based on the QR decomposition result, the grid points in each divided space. Detecting a maximum likelihood point having a maximum likelihood value, and restoring a transmission signal vector based on the detected maximum likelihood points.

을 계산하여 얻은 r²의 초기값 을 기반으로 계산될 수 있다. 단, v는 상기 채널 행렬을 기반으로 QR 분해 수행 결과 얻은 직교 행렬 Q의 복소 전치(conjugate transpose) 행렬 Q ^*와 상기 수신 신호 벡터 y의 곱, R은 상기 채널 행렬을 기반으로 QR 분해 수행 결과 얻은 상삼각 행렬,

는 상기 송신 신호 벡터의 초기 추정치이다.The dimension of the plurality of partition spaces is

It can be calculated based on the initial value of r ² obtained by calculating. Where v is the product of the conjugate transpose matrix Q ^* of the orthogonal matrix Q obtained from the QR decomposition based on the channel matrix and the received signal vector y , and R is a result of performing QR decomposition based on the channel matrix. Phase Triangle Matrix,

Is an initial estimate of the transmission signal vector.

The dimensions of the plurality of partition spaces may be calculated based on a reference calculation amount C and a modulation order M.

The received signal vector may be determined based on a transmission signal vector, a channel matrix, and a noise vector.

에 의해서 결정될 수 있다. 단, v는 상기 채널 행렬을 기반으로 QR 분해 수행 결과 얻은 직교 행렬 Q의 복소 전치(conjugate transpose) 행렬 Q ^*와 상기 수신 신호 벡터 y의 곱, R은 상기 채널 행렬을 기반으로 QR 분해 수행 결과 얻은 상삼각 행렬, x는 상기 송신 신호 벡터이다.The maximum likelihood point in each partition space is

Can be determined by Where v is the product of the conjugate transpose matrix Q ^* of the orthogonal matrix Q obtained from the QR decomposition based on the channel matrix and the received signal vector y , and R is a result of performing QR decomposition based on the channel matrix. An upper triangular matrix, x, is the transmission signal vector.

에 대하여 수행될 수 있다. 단, H는 상기 추정된 채널 행렬, 0≤θ≤3 , ρ²은 신호대 잡음비(SNR; Signal-to-Noise Ratio), I는 H의 랭크(rank)가 f일 때 f*f 크기의 항등 행렬이다.The QR decomposition is a matrix based on the estimated channel matrix

Can be performed for. Where H is the estimated channel matrix, 0≤θ≤3, ρ ² is the Signal-to-Noise Ratio (SNR), and I is the identity of f * f magnitude when the rank of H is f It is a matrix.

In another aspect, an adaptive transmission signal detection apparatus is provided. The adaptive transmission signal detecting apparatus includes a radio frequency (RF) unit for receiving a received signal vector, and a processor connected to the RF unit, wherein the processor estimates a channel matrix of a channel through which the received signal vector is rough, Perform QR decomposition based on the estimated channel matrix, calculate an initial radius based on an initial detection grid point and the initial detection grid point, calculate dimensions of a plurality of partition spaces based on the initial radius, and perform the QR The maximum likelihood point having the maximum likelihood value among the grid points in each partition space is detected based on a result of the decomposition, and the transmission signal vector is reconstructed based on the detected maximum likelihood points.

The dimensions of the plurality of partition spaces may be calculated based on a reference calculation amount C and a modulation order M.

The received signal vector may be determined based on a transmission signal vector, a channel matrix, and a noise vector.

In another aspect, an adaptive transmission signal detection apparatus is provided. The adaptive transmission signal detecting apparatus includes an RF (Radio Frequency) unit for receiving a received signal vector, an initial radius measuring unit connected to the RF unit, and measuring an initial radius based on an initial detection grid point, and the initial radius measuring unit. A partition space dimension calculation unit configured to adaptively calculate a plurality of partition space dimensions based on the measured initial radius, and is connected to the RF unit, and estimates a channel matrix of a channel through which the received signal vector is rough. A processor that performs QR decomposition based on the estimated channel matrix, is connected to the processor and the partition space dimension calculator, and the maximum likelihood of the grid points in each partition space based on the QR decomposition result. a maximum likelihood point detector for detecting a maximum likelihood point having a likelihood value, and is connected to the processor and the maximum likelihood point detector; And a memory for storing the maximum likelihood point of each detected partition space.

The amount of calculation of the decode and decode technique can be controlled to efficiently detect the transmission signal.

1 is a block diagram illustrating an embodiment of the proposed adaptive transmission signal detection method.
2 is a block diagram of a base station and a terminal in which an embodiment of the present invention is implemented.
3 is a block diagram of a receiver in which an embodiment of the present invention is implemented.

DETAILED DESCRIPTION Hereinafter, exemplary 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. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and like reference numerals designate like parts throughout the specification. In addition, even if the detailed description is omitted, descriptions of parts easily understood by those skilled in the art are omitted.

Throughout the specification and claims, when a portion is said to "include" a component, it means that it can further include other components, except to the contrary, unless otherwise stated.

In the following description, an adaptive signal processing method for multi-stream decoding is described. The present invention proposes a new signal processing method combining a decoding method used as an algorithm for the maximum likelihood detection method and an algorithm for controlling the calculation amount of the recovery method.

When the transmitter sends one or more signal streams or when a plurality of transmitters send signals, the received signal received by the receiver may be represented in the form of a received signal vector received every unit time. In this case, the received signal vector may be calculated by Equation 1.

In Equation 1, x denotes a transmission signal vector, y denotes a received signal vector, z denotes a noise vector, and H denotes a channel matrix experienced by the signal. y and z are vectors of length m, and x is a vector of length n. H is a matrix of size m * n. In this case, m≥n. The signals constituting z are independent of each other and have the same distribution, and the distribution is assumed to be a complex normal distribution of isotropic (circularly-symmetric). The measured variance of z _i, the signals forming z , may be represented by σ ² . For convenience of explanation, H and x are matrices and vectors in which the variance values of the elements representing the intensity are assumed to be 1 below. Referring to Equation 1, the received signal vector is a form in which the channel matrix is multiplied by the transmission signal vector and the noise vector is added thereto.

In general, when a plurality of transmission signal vectors form a code matrix, there is one reception signal matrix corresponding to one transmission code matrix. The received signal matrix may be calculated similarly to the shape of the received signal vector calculated in Equation 1. That is, the received signal matrix is a form in which the channel matrix is multiplied by the transmit code matrix and the noise matrix is added thereto. The calculation of the received signal matrix may be expressed as Equation 2.

를 이용하여 수학식 3과 같이 다시 나타낼 수 있다.In Equation 2, C (x ₁ , x ₂ , ..., x _s ) denotes a transmission code matrix, Y denotes a received signal matrix, z denotes a noise matrix, and H denotes a channel matrix experienced by the signal. In this case, the transmission code matrix may have a form in which conjugation, rotation, and linear transformation are applied to s signals. In addition, when Y is a matrix of m * T, Z is also a matrix of m * T, and elements constituting Z may be independent of each other, have the same distribution, and have a variance of σ ² . On the other hand, equation (2) s are signals x _1, x _2, from _{HC (x 1, x 2,} ..., x s) ... an effective channel matrix that directly multiplies, x _s

It can be represented again as in Equation (3).

와

는 실효 채널 행렬

의 행의 개수와 같은 행의 개수를 갖는 벡터들이다. 수학식 3을 참조하면 수학식 1의 형태와 유사함을 알 수 있다. 즉, 수신기에서 s개의 신호 x₁, x₂, …,x_s를 검출하기 전에 수학식 2의 Y, H 및 C(x₁, x₂, …, x_s)로부터

와 실효 채널 행렬

를 먼저 구할 수 있다. 이에 따라 복수의 송신 신호 벡터들이 코드 행렬을 이루는 경우에도, 수학식 3과 같이 실효 채널 행렬을 이용하여 수학식 1의 모델을 적용할 수 있다. 즉, 송신 신호열 사이에 코딩이 적용된 경우에도 수학식 3을 이용하여 수학식 1을 기반으로 하는 검출 방법을 사용할 수 있다. 이하에서 설명되는 적응적 신호 처리 방법 또는 검출 방법은 수학식 1의 모델을 기반으로 적용되는 것을 예시로 하나, 수학식 2의 모델을 기반으로 적용될 수도 있다.In equation (3)

Wow

Is the effective channel matrix

Are vectors with the same number of rows as. Referring to Equation 3, it can be seen that it is similar to the form of Equation 1. That is, s signals x ₁ , x ₂ ,... , equation (2) of the Y, H and C, prior to detecting the x _s from _{(x 1, x 2, ...} , x s)

And effective channel matrix

Can be obtained first. Accordingly, even when a plurality of transmission signal vectors form a code matrix, the model of Equation 1 may be applied using an effective channel matrix as shown in Equation 3. That is, even when coding is applied between transmission signal sequences, a detection method based on Equation 1 may be used using Equation 3. The adaptive signal processing method or detection method described below may be applied based on the model of Equation 2, but may be applied based on the model of Equation 1 as an example.

에 대하여 QR 분해를 수행하여

인 Q와 R을 구할 수 있다. 0≤θ≤3 이며, ρ²은 신호대 잡음비, I는 H의 랭크(rank)가 f일 때 f*f 크기의 항등 행렬이다. 그리고 수학식 1의 양변에 Q의 복소 전치(conjugate transpose) 행렬인 Q ^*을 곱하면 수학식 4와 같이 나타낼 수 있다.Again at equation (1), with respect to the H performs QR decomposition to break down any of the matrix as the product of the orthogonal matrix and an upper triangular matrix obtained by the QR and Q = H and R. Or a matrix based on H

Perform QR Decomposition on

Q and R can be obtained. 0 θ θ ³ , ρ ² is the signal-to-noise ratio, I is an identity matrix of size f * f when the rank of H is f. And multiply the complex transpose (conjugate transpose) of the matrix Q ^* Q of both sides of equation (1) can be expressed as shown in Equation (4).

In Equation 4, Q is an orthogonal matrix whose rows are orthogonal to each other, and the size of Euclidean of each row is 1, and R is an upper triangular matrix.

In addition, if v = Q ^* y and w = Q ^* z , Equation 4 may be represented by Equation 5.

The dispersion of the elements of w is σ ² as well as z . The receiver performs a process of QR decomposition of the estimated matrix H to obtain Q and R and multiplying Q ^* by the received signal vector y . At this time, since the rank of H is n, the lengths of v and w are n and R is a matrix of size n * n. i-th element of v is v i-th element w _i, i row j-th column element of the R _i, w can be represented as R _ij.

On the other hand, in the following description, from the i _row of the matrix A for convenience of description to i ₂ line, the sub-matrices (sub-matrix) consisting of elements from the j _first heat until j ₂ columns A [i _1: i ₂ ] [j ₁ : j ₂ ] From i _row of the vector b i part consisting of the elements to the _two row vectors (sub-vector) is b: denotes a [i _1, i _2]. In addition, the grid point set to which the transmission signal vector x belongs is denoted by D. Accordingly, the grid point set to which x [i ₁ : i ₂ ] belongs may be represented by D [i ₁ : i ₂ ]. When a specific value a is assigned to variable A, it may be represented as A ← a.

를 구할 수 있다.

는 최대 우도 값을 갖게 하는 격자점으로, 최대우도점이라 할 수 있다.

는 수학식 6에 의해서 구할 수 있다. In Equation 5, the most suitable transmission signal corresponding to the signal v according to the distribution characteristic of the noise vector

Can be obtained.

Is a lattice point that has a maximum likelihood value, and may be referred to as a maximum likelihood point.

Can be obtained by equation (6).

가 계산된다.

의 한시적인 값을 r²이라 한다면, 새로운 격자점 x에 대해

을 계산할 때마다 r²을 계산하고, 계산된 r²을 기준으로

인 x 중에서 최대우도점을 찾는다. 이때 R의 구조를 이용하여 ZF-SIC(Zero Forcidng Successive Interference Cancellation)을 수행하여

를 먼저 구하고,

을 계산하여 얻은 값이 r²의 초기값이 된다. 또한,

를 구하는 방법은 ZF-SIC에 제한되지 않는다. 만약 r의 초기값이 상대적으로 크다면 구복호기법 알고리듬의 계산의 대상이 되는 격자점의 개수가 많아지게 된다. 따라서 최대우도점을 찾는 데에 걸리는 시간 또는 계산량이 커지게 된다. 계산량의 변동량과 계산량을 줄이기 위하여 수학식 5 및 격자점이 속한 공간의 차원을 분할하여, 각 분할된 공간 내에서 대응되는 최대우도점을 찾을 수 있다. 수학식 5를 분할함에 있어서, r²의 초기값과 잡음의 분산 σ²을 사용하여 적응적으로 분할함으로써 계산량을 제어하면서 동시에 최대우도점

에 근접한 해를 구할 수 있다. r²과 σ²을 기반으로 수학식 5를 분할하는 구체적 과정은 아래와 같다.The denoising method calculates the maximum likelihood point at which a given input satisfies Equation 6 based on R and v . Hereinafter, the process of calculating the maximum likelihood point will be described. First, for multiple grid points x

Is calculated.

If the temporary value of is r ² , then for the new grid point x,

Each time we compute r ^2, we compute r ² ,

Find the maximum likelihood point of x . In this case, ZF-SIC (Zero Forcidng Successive Interference Cancellation) is performed by using the structure of R.

Find first,

Is the initial value of r ² . Also,

The method of obtaining is not limited to ZF-SIC. If the initial value of r is relatively large, the number of grid points that are the target of the calculation of the ordinal decoding algorithm increases. This increases the amount of time or computation it takes to find the maximum likelihood point. In order to reduce the amount of variation and the amount of calculation, the maximum likelihood point can be found in each divided space by dividing the equation 5 and the dimension of the space to which the grid points belong. In dividing Equation 5, the maximum likelihood point is controlled while controlling the amount of computation by adaptively dividing by using the initial value of r ² and the variance σ ² of noise.

A solution close to can be found. A specific process of dividing Equation 5 based on r ² and σ ² is as follows.

를 정한다.1) Set the standard calculation amount C. In addition, it is a value proportional to Mπ, which is a product of a modulation order M multiplied by the circumference π.

Determine.

을 계산하여

을 A에 대입한다.2) Let i = n,

By calculating

Is substituted into A.

3) Set b between 0 and 1, and insert P as calculated in Equation 8.

4) Compare P and C to perform step 7) if P is greater than C or i≤1, and perform step 5 if P is less than C and i> 1.

5) Decrease i value by 1. That is, i-1 → i.

를 수학식 9와 같이 계산한다.6)

Is calculated as in Equation (9).

를 기반으로 수학식 10과 같이 A를 갱신한다.Calculated in Equation 9

Based on E, A is updated as shown in Equation 10.

Return to step 3) again to perform the calculation.

에 n-i+1을 대입한다. n이

으로 나누어 떨어지면 k에 n을

으로 나눈 몫을 대입하고 n₁=n₂=…=n_k=

으로 놓는다. n이

으로 나누어 떨어지지 않으면 k에 n을

으로 나눈 몫에 1을 더한 값을 대입하고 n₁=n-(k-1)

, n₂=n₃=…=n_k=

으로 놓는다.7)

Replace n-i + 1 with. n is

Divided by k equals n to

Quotient divided by n ₁ = n ₂ =… = n _k =

Place it. n is

If n is not divided by k

Replace the quotient divided by 1 plus 1 and n ₁ = n- (k-1)

, n ₂ = n ₃ =... = n _k =

Place it.

를 구한다.8) Substitute k in l. Equation (11) is satisfied based on R [nn _k +1: n] and v [nn _k +1: n] given as input values of the quadrature decoding algorithm.

.

을 출력하고 종료한다. 그렇지 않으면 단계 10)을 수행한다.9) If ℓ = 1

Output and exit. Otherwise, perform step 10).

과

을 각각 수학식 12와 수학식 13에 의해서 계산한다.10) Substitute ℓ ← ℓ-1,

and

Are calculated by Equations 12 and 13, respectively.

을 x _ℓ로 표현하고, x _ℓ이 속하는 집합을 D_ℓ로 표시하면,

과

을 구복호기법의 입력값으로 하여 수학식 14에 의하여

을 구한다.11)

Is represented by x _ℓ , and the set to which x _ℓ belongs is represented by D _ℓ ,

and

Equation 14 as

.

Perform the process of step 9) again.

, …,

을 구하는 방법으로 구복호 기법을 가정하였으나, 본 발명은 이에 제한되지는 않으며 구복호 기법 이외의 다른 검출 방법을 사용할 수 있다.Subvectors from the description above

,… ,

Although a method of decoding is assumed as a method of obtaining the present invention, the present invention is not limited thereto, and other detection methods other than the recovery method may be used.

1 is a block diagram illustrating an embodiment of the proposed adaptive transmission signal detection method.

In step S100, the receiver receives a received signal vector. In step S110, the receiver estimates a channel matrix. In step S120, the receiver performs QR decomposition based on the estimated channel matrix. In step S130, the receiver calculates an initial detection grid point and an initial radius based on the initial detection grid point. In step S140, the receiver calculates the dimensions of the plurality of divided spaces based on the initial radius. In this case, the dimension of the partition space may be calculated based on the reference calculation amount and the modulation order. In step S150, the receiver detects the maximum likelihood point having the maximum likelihood value among the grid points in the divided spaces based on the QR decomposition performance result. In step S160, the receiver reconstructs the transmission signal vector based on the detected maximum likelihood points.

2 is a block diagram illustrating a base station and a terminal in which an embodiment of the present invention is implemented.

The base station 800 includes a processor 810, a memory 820, and a radio frequency unit (RF) 830. Processor 810 implements the proposed functions, processes, and / or methods. Layers of the air interface protocol may be implemented by the processor 810. The memory 820 is connected to the processor 810 and stores various information for driving the processor 810. The RF unit 830 is connected to the processor 810 to transmit and / or receive a radio signal.

The terminal 900 includes a processor 910, a memory 920, and an RF unit 930. The RF unit 930 is connected to the processor 910 to transmit and / or receive a radio signal. Processor 910 implements the proposed functions, processes, and / or methods. Layers of the air interface protocol may be implemented by the processor 910. The memory 920 is connected to the processor 910 and stores various information for driving the processor 910.

Processors 810 and 910 may include application-specific integrated circuits (ASICs), other chipsets, logic circuits, and / or data processing devices. The memory 820, 920 may include read-only memory (ROM), random access memory (RAM), flash memory, memory card, storage medium, and / or other storage device. The RF unit 830 and 930 may include a baseband circuit for processing a radio signal. When the embodiment is implemented in software, the above-described techniques may be implemented with modules (processes, functions, and so on) that perform the functions described above. The module may be stored in the memory 820, 920 and executed by the processor 810, 910. The memories 820 and 920 may be inside or outside the processors 810 and 910, and may be connected to the processors 810 and 910 by various well-known means.

3 is a block diagram of a receiver in which an embodiment of the present invention is implemented.

Referring to FIG. 3, the receiver 700 includes a processor 710, a memory 720, an initial radius measurer 730, a partition spatial dimension calculator 740, a maximum likelihood point detector 750, and an RF unit 790. ). The RF unit 790 receives the received signal vector. The initial radius measuring unit 730 is connected to the RF unit 790 and measures an initial radius based on the detected initial grid point. The partition space dimension calculator 740 is connected to the initial radius measurer 730 and calculates a plurality of partition space dimensions based on the measured initial radius. The processor 710 is connected to the RF unit 790, estimates a channel matrix of a channel through which the received signal vector is rough, and performs QR decomposition based on the estimated channel matrix. The maximum likelihood point detector 750 is connected to the processor 710 and the partition space dimension calculator 740 and based on the QR decomposition result, the maximum likelihood of the grid points in each partition space. Detects the maximum likelihood point with The memory 720 is connected to the processor 710 and the maximum likelihood point detector 750, and stores the maximum likelihood point of each detected partition space. The processor 710 restores the transmission signal vector based on the maximum likelihood point of each partition space stored in the memory 720.

The present invention may be implemented in hardware, software, or a combination thereof. (DSP), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microprocessor, and the like, which are designed to perform the above- , Other electronic units, or a combination thereof. In the software implementation, the module may be implemented as a module that performs the above-described function. The software may be stored in a memory unit and executed by a processor. The memory unit or processor may employ various means well known to those skilled in the art.

In the above-described exemplary system, the methods are described on the basis of a flowchart as a series of steps or blocks, but the present invention is not limited to the order of the steps, and some steps may occur in different orders or simultaneously . In addition, those skilled in the art will appreciate that the steps shown in the flowcharts are not exclusive and that other steps may be included or one or more steps in the flowcharts may be deleted without affecting the scope of the present invention.

The above-described embodiments include examples of various aspects. While it is not possible to describe every possible combination for expressing various aspects, one of ordinary skill in the art will recognize that other combinations are possible. Accordingly, it is intended that the invention include all alternatives, modifications and variations that fall within the scope of the following claims.

Claims (13)

An adaptive transmission signal detection method in a communication system,
Receiving a received signal vector;
Estimating a channel matrix of a channel through which the received signal vector is rough;
Performing QR decomposition based on the estimated channel matrix;
Calculating an initial radius based on an initial detection grid point and the initial detection grid point;
Calculating dimensions of a plurality of partition spaces based on the initial radius;
Detecting a maximum likelihood point having a maximum likelihood value among the grid points based on the QR decomposition result; And
Recovering a transmission signal vector based on the detected maximum likelihood points. The method of claim 1,
The dimension of the plurality of partition spaces is

The adaptive transmission signal detection method characterized in that it is calculated based on the initial value of r ² obtained by calculating the. Where v is the product of the conjugate transpose matrix Q ^* of the orthogonal matrix Q obtained from the QR decomposition based on the channel matrix and the received signal vector y , and R is a result of performing QR decomposition based on the channel matrix. Phase Triangle Matrix,

Is an initial estimate of the transmission signal vector. The method of claim 1,
The dimension of the plurality of partitions is calculated based on a reference calculation amount C and a modulation order (M). The method of claim 1,
And the received signal vector is determined based on a transmitted signal vector, a channel matrix and a noise vector. The method of claim 1,
The maximum likelihood point in each divided space is determined by the following equation.

Where v is the product of the conjugate transpose matrix Q ^* of the orthogonal matrix Q obtained from the QR decomposition based on the channel matrix and the received signal vector y , and R is a result of performing QR decomposition based on the channel matrix. An upper triangular matrix, x, is the transmission signal vector. The method of claim 1,
The QR decomposition is a matrix based on the estimated channel matrix

Adaptive transmission signal detection method characterized in that performed for. Where H is the estimated channel matrix, 0≤θ≤3, ρ ² is the Signal-to-Noise Ratio (SNR), and I is the identity of f * f magnitude when the rank of H is f It is a matrix. An RF unit for receiving a received signal vector; And
Including a processor connected to the RF unit,
The processor comprising:
Estimating a channel matrix of a channel through which the received signal vector is rough,
Perform QR decomposition based on the estimated channel matrix,
Calculate an initial radius based on an initial detection grid point and the initial detection grid point,
Calculate dimensions of a plurality of partition spaces based on the initial radius,
Detecting a maximum likelihood point having a maximum likelihood value among the grid points based on the QR decomposition result;
And adaptively reconstruct a transmission signal vector based on the detected maximum likelihood points. The method of claim 7, wherein
The dimension of the plurality of partition spaces is

The adaptive transmission signal detection apparatus, characterized in that calculated based on the initial value of r ² obtained by calculating the. Where v is the product of the conjugate transpose matrix Q ^* of the orthogonal matrix Q obtained from the QR decomposition based on the channel matrix and the received signal vector y , and R is a result of performing QR decomposition based on the channel matrix. Phase Triangle Matrix,

Is an initial estimate of the transmission signal vector. The method of claim 7, wherein
And the dimensions of the plurality of partition spaces are calculated based on a reference calculation amount C and a modulation order M. The method of claim 7, wherein
And the received signal vector is determined based on a transmitted signal vector, a channel matrix and a noise vector. The method of claim 7, wherein
The maximum likelihood point is determined by the following equation.

Where v is the product of the conjugate transpose matrix Q ^* of the orthogonal matrix Q obtained from the QR decomposition based on the channel matrix and the received signal vector y , and R is a result of performing QR decomposition based on the channel matrix. An upper triangular matrix, x, is the transmission signal vector. The method of claim 7, wherein
The QR decomposition is a matrix based on the estimated channel matrix

Adaptive transmission signal detection apparatus characterized in that performed for. Where H is the estimated channel matrix, 0≤θ≤3, ρ ² is the Signal-to-Noise Ratio (SNR), and I is the identity of f * f magnitude when the rank of H is f It is a matrix. An RF unit for receiving a received signal vector;
An initial radius measurement unit connected to the RF unit and measuring an initial radius based on an initial detection grid point;
A partition space dimension calculator connected to the initial radius measurer to adaptively calculate a plurality of partition space dimensions based on the measured initial radius;
A processor coupled to the RF unit, estimating a channel matrix of a channel through which the received signal vector is rough and performing QR decomposition based on the estimated channel matrix;
A maximum likelihood point detection unit connected to the processor and the partition space dimension calculation unit and detecting a maximum likelihood point having a maximum likelihood value among the grid points based on a result of performing QR decomposition ; And
And a memory coupled to the processor and the maximum likelihood point detector, the memory storing the maximum likelihood point of each detected partition space.

Images