KR101878417B1 - Phase Noise Estimation and Compensation Method and Apparatus in Orthogonal Frequency-Division Multiplexing Communication - Google Patents

Abstract

개의 심볼조각으로 나누어 각 심볼조각이 가질 수 있는 위상 잡음(PHN) 값을 코드북으로 미리 저장한다. 코드북에 미리 저장된 모든 위상 잡음에 대해 파일럿 심볼을 복호하여, 복호 오차가 가장 작은 위상잡음을 각 심볼조각의 위상잡음 추정값으로 정한다. 그리고 각 심볼조각 별로 독립적으로 추정된 위상잡음의 조합을 이용하여 부반송파 간 간섭(ICI)을 일으키는 위상잡음을 보상한다. 이렇게 하나의 OFDM 심볼을 구성하는 전체 심볼조각들에 대하여 각 심볼조각 별 PHN 값을 그 심볼조각의 이전 심볼조각과는 무관하게 독립적으로 추정함으로써, 파일럿 심볼 복호의 수행횟수를 줄여 전체 계산량을 크게 감축할 수 있다.A phase noise estimation and compensation method and an apparatus therefor in an orthogonal frequency division multiplexing communication system are disclosed. OFDM symbol

And stores the phase noise (PHN) value that each symbol segment can have in the codebook in advance. The pilot symbols are decoded for all the phase noises stored in advance in the codebook, and the phase noise with the smallest decoding error is determined as the phase noise estimation value of each symbol fragments. Then, phase noise that causes inter-subcarrier interference (ICI) is compensated by using a combination of phase noise independently estimated for each symbol piece. By independently estimating the PHN value of each symbol fragment with respect to all symbol fragments constituting one OFDM symbol independently of the previous symbol fragments of the symbol fragments, the number of times of performing the pilot symbol decoding is reduced, can do.

Description

[0001] The present invention relates to a phase noise estimation and compensation method in an orthogonal frequency division multiplexing (OFDM) communication system,

The present invention relates to estimating and compensating for phase noise causing inter-carrier interference (ICI) in an Orthogonal Frequency-Division Multiplexing (OFDM) communication system.

An orthogonal frequency division multiplexing (OFDM) system is known as one of wireless communication systems. The OFDM communication system is widely adopted and used in current digital communication systems because of its strong characteristics in a multipath environment and high frequency efficiency. In an OFDM system, a frequency oscillation of a local oscillator used in a transmitting apparatus and a receiving apparatus for modulating and demodulating a signal appears as a phase noise (PHN) in a received signal. If the local oscillator can not accurately generate the intended frequency, it means a noise that changes in the amplitude and phase of the transmission signal on the time axis due to the error of the oscillation frequency.

In the OFDM communication system, the effect of the PHN is divided into two components. One is Common Phase Error (CPE) and the other is Inter-Carrier Interference (ICI). The CPE is a noise that rotates (rotates) the phase of all subcarriers by the same amount in one OFDM symbol, and ICI is noise caused by interference between each other due to loss of orthogonality between subcarriers. Since the OFDM scheme uses orthogonality between subcarriers to enhance frequency efficiency, ICI degrades OFDM performance.

Recent outcomes of studies to remove ICI include "Phase-noise mitigation in OFDM by best match trajectories," IEEE Transactions on Communications, vol. 63, no. 5, May 2015, there is a Best Match Trajectory (BMT) technique. The BMT technique removes ICI based on a codebook.

PHN can be classified into two types. PHNs that follow the Weiner process and PHNs that follow the Gaussian process. The Winner processor model is a model in which the transition from the current state to the next state follows the Gaussian distribution. That is, a value obtained by subtracting PHN at time N-1 from PHN at time N follows a Gaussian distribution. The BMT technique can be effectively applied when specifically following the Winner processor model.

개의 심볼조각(segments)으로 나뉘고, 각 심볼조각 별 평균 PHN이 추정된다. 그리고 PHN의 통계적 특성을 바탕으로, 심볼조각과 심볼조각 사이의 PHN 증가량이 가질 수 있는 값이

가지로 근사화 된다. 이를 바탕으로 각 심볼조각이 가질 수 있는 PHN 값을 조합하여 '한 심볼 동안 PHN의 경로'가 코드북으로 저장된다. 그런 다음, 코드북에 저장된 PHN 경로마다 파일럿 심볼을 복호하여, 복호 오차가 가장 작은 'PHN 경로' 즉, 파일럿 심볼의 복호가 가장 양호하게 이루어지는 경로가 '추정 PHN'으로 결정된다. BMT 기법은 위와 같이 추정된 PHN을 이용하여 ICI를 제거한다. According to the BMT technique, the PHN path with the smallest decoding error is estimated, and the ICI is removed using the estimated PHN path. Specifically, if we had a certain PHN in a particular time sample, then the probability distribution that the PHN could have in the next time sample would follow the Gaussian distribution. If one of the samples with the highest probability is selected, the next sample will also follow the Gaussian distribution. Consider PHN that proceeds in this way. However, since it is difficult to predict all the samples, one OFDM symbol is transmitted in the time domain

Symbols, and an average PHN for each symbol fragment is estimated. Based on the statistical properties of the PHN, the value that the PHN increase between the symbol fragments and the symbol fragments can have

. Based on this, the 'PHN path for one symbol' is stored as a codebook by combining the PHN values that each symbol fragment can have. Then, the pilot symbol is decoded for each PHN path stored in the codebook, and the path with the smallest decoding error, i.e., the path where the decoding of the pilot symbol is performed the best, is determined as the 'estimated PHN'. The BMT scheme removes ICI using the estimated PHN.

가 증가함에 따라 계산량이 기하급수적으로 늘어나는 단점이 있다. The conventional BMT scheme has an advantage in that it has superior performance and computation efficiency as compared with other ICI cancellation schemes. However, since the BMT technique performs pilot symbol decoding by combining all of the PHN values that can be included in each symbol fragment,

There is a disadvantage that the amount of computation increases exponentially.

&Quot; Phase-noise mitigation in OFDM by best match trajectories, " IEEE transactions on communications, vol. 63, no. 5, May 2015

In order to solve the problems of the prior art, the present invention reduces the number of PHN paths to be considered in comparison with the conventional BMT scheme in estimating and compensating PHN caused by instability of the oscillation frequency of the transmitting / receiving end in the OFDM communication system The present invention aims to provide a novel PHN estimation and compensation method which can greatly reduce the calculation amount for estimating and compensating the PHN, and does not cause degradation of the system performance, and an apparatus therefor.

According to an aspect of the present invention, there is provided a method for processing a symbol segment of a single OFDM symbol, the method comprising: By independently estimating, the number of pilot symbol decoding operations can be reduced to greatly reduce the total amount of computation.

According to an embodiment of the present invention, an apparatus for estimating and compensating a phase noise of an orthogonal frequency division multiplexing (OFDM) communication system is provided. The channel and CPE compensation unit of this apparatus estimates and compensates for the common phase noise (CPE) and channel of the received OFDM symbol. The symbol segment phase noise estimator estimates a phase noise (PHN) by dividing a channel and a CPE compensated symbol into a plurality of symbol fragments in a time domain. The ICI compensator compensates for the phase noise causing the inter-subcarrier interference (ICI) of the symbol from the estimated phase noise.

According to an embodiment of the present invention, the apparatus may further include a codebook storing an inverse matrix of an ICI matrix corresponding to phase noise varying during an OFDM symbol time. The value of the phase noise may be unchanged within a single symbol segment.

According to an embodiment of the present invention, the codebook may further include information determined by phase noise to quantize the phase noise values that the symbol fragments may have, into a finite number according to statistical characteristics of phase noise.

According to the embodiment of the present invention, the phase noise value to be examined in one symbol block may be independently determined information regardless of the phase noise value of other symbol fragments.

Also, according to the embodiment of the present invention, the range of the value of the phase noise to be examined can be set to be wider as the symbol fragments far from the center of the symbol.

According to the embodiment of the present invention, the codebook may include inverse matrix information of an ICI matrix corresponding to a case where the phase noise value is 0 in other symbol fragments excluding one symbol fragment.

According to an embodiment of the present invention, the codebook may include inverse matrix information of an ICI matrix corresponding to a combination of phase noise values to be examined for each symbol piece.

According to the embodiment of the present invention, the symbol segment phase noise estimation unit may estimate the phase noise by decoding the pilot symbols using the codebook for each symbol fragment.

According to an embodiment of the present invention, an inverse matrix of an ICI matrix corresponding to a combination of estimated phase noise for each symbol fragments can be found in the codebook and used for phase noise estimation.

According to an embodiment of the present invention, the apparatus may further include a discrete Fourier transformer performing a discrete Fourier transform only once on the received OFDM symbol.

개의 심볼조각으로 나누어 각 심볼조각이 가질 수 있는 위상 잡음(PHN) 값을 코드북으로 미리 저장한다. 코드북에 미리 저장된 모든 위상 잡음에 대해 파일럿 심볼을 복호하여, 복호 오차가 가장 작은 위상잡음을 각 심볼조각의 위상잡음 추정값으로 정한다. 그리고 각 심볼조각 별로 독립적으로 추정된 위상잡음의 조합을 이용하여 부반송파 간 간섭(ICI)을 일으키는 위상잡음을 보상한다. According to another aspect of the present invention, there is provided a phase noise estimation and compensation method of an OFDM communication system. According to this method, an OFDM symbol

And stores the phase noise (PHN) value that each symbol segment can have in the codebook in advance. The pilot symbols are decoded for all the phase noises stored in advance in the codebook, and the phase noise with the smallest decoding error is determined as the phase noise estimation value of each symbol fragments. Then, phase noise that causes inter-subcarrier interference (ICI) is compensated by using a combination of phase noise independently estimated for each symbol piece.

According to another embodiment of the present invention, it is assumed that the 'phase noise value that each symbol segment can have' has a phase noise value of 0 in the middle of the time domain, and a value that can have phase noise Can be determined according to statistical characteristics.

According to another embodiment of the present invention, the pilot signal may be restored for each phase noise value of each symbol block, and the phase noise with the smallest error may be selected and compensated for as the estimation value.

According to another embodiment of the present invention, the phase noise value to be examined in one symbol segment may be independently determined information regardless of the phase noise values of other symbol fragments.

According to another embodiment of the present invention, the range of the value of the phase noise to be examined may be set to be wider as the symbol fragments far from the center of the symbol.

According to another embodiment of the present invention, the codebook may include inverse matrix information of an ICI matrix corresponding to a case where the phase noise value is 0 in other symbol fragments excluding one symbol fragment.

According to another embodiment of the present invention, the method may further include estimating phase noise by decoding a pilot symbol using the codebook for each symbol fragment.

According to another embodiment of the present invention, the codebook stores an inverse matrix of an ICI matrix corresponding to phase noise varying during an OFDM symbol time, and an inverse matrix of an ICI matrix corresponding to a combination of phase noise estimated for each symbol fragments And can be used in the phase noise estimation by searching in the codebook.

According to another embodiment of the present invention, the codebook further includes information determined by phase noise to quantize the phase noise values that the symbol fragments the symbol fragments can have, into a finite number according to the statistical characteristics of phase noise can do.

에서 CPE

가 제거된 위상 잡음 매트릭스의 성분

, 정확하게는 이것의 역행렬

을 미리 '코드북'에 저장해두고, OFDM 통신이 실제로 수행될 때 PHN의 추정에 사용하되, 그 OFDM 통신을 실제로 수행할 때 심볼을 여러 개의 심볼조각으로 구분하고 각 심볼조각별 PHN 값을 '독립적으로' 추정한다. 파일럿 심볼 복호의 반복 수행 횟수(즉, 코드북에서 상기 위상 잡음 매트릭스의 성분

을 꺼내봐야 할 횟수)를 크게 줄일 수 있어 계산량을 크게 줄일 수 있는 효과가 있다. 또한,

을 코드북으로 미리 만들어 둠으로써, 실시간 계산량을 크게 줄일 수 있어, 시스템의 전반적인 성능을 좋게 할 수 있다. According to the present invention, when the PHN value of symbol fragments is estimated by the BMT technique, the amount of calculation can be greatly reduced as compared with the prior art. Distortion component caused by phase noise

CPE

Of the phase noise matrix

, Exactly the inverse of its

Is stored in advance in the 'codebook', and is used to estimate the PHN when OFDM communication is actually performed. When the OFDM communication is actually performed, the symbol is divided into a plurality of symbol fragments and the PHN values of the symbol fragments are ''. The number of iterations of pilot symbol decoding (i.e., the number of components of the phase noise matrix

The number of times to take out the battery) can be greatly reduced, and the amount of calculation can be greatly reduced. Also,

Is prepared in advance as a codebook, the amount of real-time calculation can be greatly reduced, and the overall performance of the system can be improved.

In the above-mentioned conventional art, the number of paths that can be included in one symbol interval is K = Q ^(J-1) (where Q is the number of quantization and J is the number of symbol fragments) J + J (J + 2) (Q-1) where the phase number of the phase noise to be considered in this case is Kp = J + J / 4, which is much smaller than K. In particular, since the number of repetitions of pilot decoding exceeds 9, the amount of computation to be performed is greatly reduced compared with the conventional BMT, and a remarkable effect is obtained.

Although the present invention can greatly reduce the amount of computation compared to the prior art, it is never far behind the PHN path estimation and compensation performance.

,

,

인 경우를 예로 하여 각 심볼조각마다

이 가질 수 있는 값들을 그린 것이다.
도 4는 본 발명에 따른 OFDM 통신방식의 PHN 추정 및 보상 방법의 전반적인 절차를 나타낸 흐름도이다.
도 5는 본 발명의 바람직한 실시예에 따라 심볼조각 별로 독립적으로 위상잡음을 추정하는 알고리즘을 나타낸 흐름도이다.
도 6은 본 발명과 종래기술 간의 PHN 추정 성능의 비교를 위한 첫 번째 시뮬레이션 결과를 예시적으로 나타내는 그래프이다.
도 7은 본 발명과 종래기술 간의 PHN 추정 성능의 비교를 위한 두 번째 시뮬레이션 결과를 예시적으로 나타내는 그래프이다.
도 8은 도 6의 시뮬레이션에서 본 발명과 종래기술에 따른 PHN 추정 시 수행하는 반복 연산 횟수를 예시적으로 대비하는 그래프이다.
도 9는 도 7의 시뮬레이션에서 본 발명과 종래기술에 따른 PHN 추정 시 수행하는 반복 연산 횟수를 예시적으로 대비하는 그래프이다. 1 is a block diagram illustrating an OFDM communication system used in the present invention.
2 is a block diagram showing a configuration of a decoding unit of the OFDM receiving apparatus shown in FIG.
3,

,

,

For each symbol piece

I have drawn the possible values.
4 is a flowchart illustrating an overall procedure of a PHN estimation and compensation method of an OFDM communication method according to the present invention.
5 is a flowchart illustrating an algorithm for independently estimating phase noise for each symbol fragment according to a preferred embodiment of the present invention.
FIG. 6 is a graph illustrating a result of a first simulation for comparing the PHN estimation performance between the present invention and the prior art.
FIG. 7 is a graph illustrating a result of a second simulation for comparing the PHN estimation performance between the present invention and the prior art.
FIG. 8 is a graph illustrating an exemplary comparison between the number of iterative operations performed in the PHN estimation according to the present invention and the prior art in the simulation of FIG. 6;
FIG. 9 is a graph illustrating an exemplary comparison between the number of iterative operations performed in the PHN estimation according to the present invention and the prior art in the simulation of FIG. 7;

For the embodiments of the invention disclosed herein, specific structural and functional descriptions are set forth for the purpose of describing an embodiment of the invention only, and it is to be understood that the embodiments of the invention may be practiced in various forms, The present invention should not be construed as limited to the embodiments described in Figs.

The present invention is capable of various modifications and various forms, and specific embodiments are illustrated in the drawings and described in detail in the text. It is to be understood, however, that the invention is not intended to be limited to the particular forms disclosed, but on the contrary, is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms may be used for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between. Other expressions that describe the relationship between components, such as "between" and "between" or "neighboring to" and "directly adjacent to" should be interpreted as well.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprise", "having", and the like are intended to specify the presence of stated features, integers, steps, operations, elements, components, or combinations thereof, , Steps, operations, components, parts, or combinations thereof, as a matter of principle.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries should be construed as meaning consistent with meaning in the context of the relevant art and are not to be construed as ideal or overly formal in meaning unless expressly defined in the present application .

Hereinafter, the present invention will be described in detail to enable implementation of the present invention with reference to the accompanying drawings.

1 is a block diagram illustrating an OFDM communication system 10 in accordance with an embodiment of the present invention.

The OFDM communication system 10 includes a transmitting apparatus 20 for transmitting an OFDM symbol and a receiving apparatus 50 for receiving an OFDM symbol transmitted through a channel 90. The transmission apparatus 20 includes an Inverse Discrete Fourier Transform Unit (IDFT) 25, a Cyclic Prefix Unit (CP) 30, a Parallel to Serial Converting unit A PS converter 35), a mixer 40, and a local oscillator 45. [

The IDFT 25 inversely transforms the information symbol vector X into the N-point discrete Fourier transform so that the information symbol vector X is normalized. By this inverse transformation, the information word symbol vector X is converted into a time domain information word vector. The inverse-transformed information symbol vector x is provided to the CP 30. The CP 30 adds CP to each information symbol vector x input from the IDFT 25. [ The information symbol vectors to which the CP is added are supplied to the mixer 40 through the parallel-to-serial conversion process in the PS conversion unit 35, where they are mixed with the oscillation signal provided by the local oscillator 45 and modulated. It is then transmitted to the receiving device 50 via the forwarding channel 90 with the receiving device 50. A noise w (t) such as phase noise and / or additive white Gaussian noise (AWGN) may be generated in the symbol vector x transmitted to the receiving apparatus 50 through the channel 90.

The receiving apparatus 50 includes a mixer 55, a local oscillator 60, a serial to parallel conversion unit (SP conversion unit) 65, a cyclic prefix (CP) removing unit 70, A Fourier transform unit (DFT) 75, and a decoding unit 80. The receiving apparatus 50 performs signal processing in the transmitting apparatus 20 in reverse to recover the transmitted symbols.

For example, the transmission device 20 may be a base station, a broadcasting station, or the like that transmits OFDM symbols, and the receiving device 50 may be a communication terminal (e.g., a cellular phone, a smart phone, Portable terminals such as a laptop computer, a digital broadcasting terminal, a PDA (Personal Digital Assistants), a PMP (Portable Multimedia Player), a navigation terminal and the like, as well as fixed terminals such as a digital TV and a desktop computer).

이며,

개의 부반송파로 이루어져 있고, 그 중

개는 파일럿에 해당한다. k번째 부반송파에 실린 정보어 심볼을

라고 나타내고, 하나의 OFDM 심볼 안에 실리는 정보어 심볼 벡터

는

로 나타낼 수 있다. 이 때,

는 행렬의 전치(Transpose) 연산을 나타낸다.One OFDM symbol has a symbol time

Lt;

Subcarriers, among which

Dogs correspond to pilots. The information word symbol on the kth subcarrier

And an information symbol vector < RTI ID = 0.0 >

The

. At this time,

Represents a transpose operation of the matrix.

를

-point 푸리에 역변환 하여 시간 영역 심볼 벡터

를 만든다. 여기서,

은

번째 시간 샘플을 나타낸다.The IDFT 25 receives the information symbol vector

To

-point Fourier inverse transform to generate a time domain symbol vector

. here,

silver

Lt; th > time sample.

에 순환전치(Cyclic Prefix: CP)를 붙인 다음, 병렬-직렬 변환을 거친 심볼 벡터들을 믹서(40)에서 주파수

인 반송파에 실어 만들어진 아날로그 신호

를 송신한다. 여기서, 반송 주파수

를 생성해내는 국부발진기(45)의 주파수 흔들림으로 인하여 전송단 PHN

이 발생할 수 있다. The transmission apparatus 20 receives the time-domain symbol vector < RTI ID = 0.0 >

(CP), and then the parallel-to-serial converted symbol vectors are supplied to the mixer 40 in the frequency

An analog signal built on a carrier wave

. Here,

Due to the frequency fluctuation of the local oscillator 45 generating the PHN

Can occur.

는 채널

(90)를 거치고 열잡음

가 더해져 수신장치(50)에서 수신신호

로 수신된다. 수신 신호

는 수신장치(50)의 국부발진기(60)의 발진신호와 혼합되는 믹서(55)를 지나면서 기저대역 신호로 바뀌는데, 이 때 수신장치(50)의 국부발진기(60)의 주파수 흔들림으로 인하여 수신단 PHN

가 발생할 수 있다.The analog signal transmitted from the transmission apparatus 20

Channel

(90) and thermal noise

Is added to the reception signal from the reception device 50,

Lt; / RTI > Received signal

The signal is converted into a baseband signal by passing through the mixer 55 which is mixed with the oscillation signal of the local oscillator 60 of the receiver 50. At this time, due to the frequency fluctuation of the local oscillator 60 of the receiver 50, PHN

May occur.

를 얻는다. The receiving apparatus 50 samples the baseband signal, performs serial-to-parallel conversion, and then separates the CP to obtain a time-domain symbol vector

.

로 변환한다.

는

번째 부반송파로 수신된 정보어 심볼어로서, 다음과 같이 나타낼 수 있다.The DFT 75 of the receiving device 50 converts the received time domain symbol vector into an information symbol vector < RTI ID = 0.0 >

.

The

Th subcarrier, which can be represented as follows.

……(1)

... ... (One)

와

는

번째 부반송파에서의 채널응답과 열잡음을 각각 나타낸다. 또한,

은 모든 부반송파에 동일하게 곱해져 위상을 회전시키는 공통 위상 잡음(CPE)이며, 그 외의

들은 인접 부반송파에 간섭을 일으키는 부반송파간 간섭(ICI)이다. 채널 추정과정에서, CPE

은 위상의 회전만 일으키는 성분으로서 대체적으로 완화시킬 수 있는 잡음인 반면, ICI인

들은 완화시키기가 쉽지 않고 부반송파들간의 직교성을 무너뜨림으로써 시스템의 성능을 열화시키는 주된 요인이다. 그러므로

들의 가능한 경우의 수가 이론적으로는 무한대이지만, 적절한 간격으로 양자화하여 코드북으로 저장한 다음, PHN의 추정 및 보상에 활용한다.here,

Wow

The

And the channel response and the thermal noise at the i < th > subcarrier, respectively. Also,

Is a common phase noise (CPE) that equally multiplies all subcarriers to rotate the phase, and the other

(ICI) that causes interference to adjacent subcarriers. In the channel estimation process, CPE

Is a noise that can be largely mitigated as a component that only generates a phase rotation, whereas ICI

Are not easy to mitigate and destroy orthogonality between subcarriers, thereby deteriorating system performance. therefore

Theoretically, the number of possible cases is infinite, but it is quantized at appropriate intervals and stored as a codebook, and then used for estimation and compensation of the PHN.

는 다음과 같이 매트릭스 형태로도 쓸 수 있다.The information word symbol vector in the frequency domain obtained by the DFT 75

Can also be written in matrix form as follows.

……(2)

... ... (2)

에 곱하면, 수신신호에 남는 항목이 A^-1Y= X^dH+WA^{- 1}와 같이 깨끗하게 정리되어 수신신호를 쉽게 복구할 수 있을 것이다. 그러므로 A를 구하는 작업을 수행할 필요가 있다. 기존의 BMT 방법은 신호의 왜곡 성분 A를 만드는 것이 위상 잡음이므로, 위상 잡음의 통계적 특성을 이용하여 그 위상 잡음이 취할 수 있는 경우의 수와, 그것에 상응하는 A^-1을 미리 구해 코드북에 저장해둔다. 그리고 k개의 A^-1을 가지고 있다면, k번 x를 복호하여 그 중에 복호에러가 가장 작은 A를 채택하여 나머지 심볼들을 복호한다. 그런데 이 방법은 판단해야 할 경우의 수 k의 값이 커지면 연산량이 기하급수적으로 늘어나는 문제가 있다. 본 발명은 이 k값을 크게 줄일 수 있고, 그에 따라 연산량도 현저히 줄이는 방법을 모색하는 방법을 제안하는 것이다. If it is possible to know the distortion component of the signal A arises by the phase noise, the received signal A ^-1

, The remaining items in the received signal will be cleaned up like A ^-1 Y = X ^d H + WA ^{- 1} and the received signal will be easily recovered. Therefore, we need to perform the task of obtaining A. In the conventional BMT method, since the phase noise is generated by generating the signal distortion component A, the number of cases in which the phase noise can be taken using the statistical characteristic of the phase noise and the corresponding A ^-1 are previously obtained and stored in the codebook . If k is A ^-1 , k is x, and the remaining symbols are decoded by adopting A with the smallest decoding error. However, this method has a problem that the amount of computation exponentially increases when the value of the number k in the case of judging becomes large. The present invention proposes a method of greatly reducing the k value and thereby significantly reducing the amount of computation.

는 전송장치(20)에서 전송될 정보어 심볼 벡터

의 대각행렬이고, 채널응답 벡터 H는

이고, 열잡음 벡터 W는

이다.The method of the present invention will be described in detail. In the above equation,

Which is transmitted from the transmission apparatus 20,

And the channel response vector H is a diagonal matrix of

, And the thermal noise vector W is

to be.

는

의 순환행렬(Circulant matrix)로 다음과 같이 나타난다.Which represents a distortion component caused by phase noise

The

The circulant matrix of

……(3)

... ... (3)

은 위에서 언급하였듯이 공통 위상 잡음(CPE)이고,

는 부반송파간 간섭(ICI)이다. here,

Is the common phase noise (CPE) as mentioned above,

Is inter-carrier-interference (ICI).

는 PHN에 의해 발생하는 왜곡성분으로서 다음과 같은 수식으로 나타낼 수 있다.Intercarrier Interference (ICI)

Can be expressed by the following equation as a distortion component generated by the PHN.

……(4)

... ... (4)

는 송신단 위상잡음

와 수신단 위상잡음

의 차이며,

은

의

번째 시간 샘플이다.here,

Lt; RTI ID = 0.0 >

And receiver phase noise

Lt; / RTI >

silver

of

Th time sample.

In the OFDM communication system 10, in the course of wirelessly transmitting / receiving a signal in the OFDM manner, the receiving apparatus 50 performs PHN estimation and compensation processing. 2 shows a configuration of the decoding unit 80 of the receiving apparatus 50. As shown in Fig. FIG. 4 shows the overall procedure of the PHN estimation and compensation method in the OFDM communication system according to the present invention.

와 CPE

를 보상한다(S10 단계).The OFDM receiving apparatus 50 receives OFDM symbols according to the OFDM communication scheme by the transmitting apparatus 20. Among the noise included in the received signal, noise caused by the response characteristic of the channel during the transmission

And CPE

(Step S10).

를 추정하고, 파일럿 심볼로부터

를 구한다. 그리고 구한 채널의 주파수 응답

와 CPE

를 수신신호

에서 떼내는 처리를 하여, CPE

와 채널의 주파수응답

가 보상된 수신 심볼 벡터

를 구한다. The receiving device 50 includes a channel and CPE compensation module 82. In this compensation module 82, a known training symbol is transmitted to obtain a frequency response

From the pilot symbol,

. And the frequency response of the obtained channel

And CPE

Lt; RTI ID =

And the CPE

And the frequency response of the channel

Lt; RTI ID = 0.0 >

.

를 구하는 과정을 좀 더 구체적으로 설명하면, PHN

는 위너과정(Wiener Process)을 따른다. 위너과정의 정의에 따라 n번째 샘플의 PHN

와 n-1번째 샘플의 PHN

간의 차

는 평균

, 분산

인 가우시안 확률변수이다. 즉, 시간간격이 커질수록 분산값도 비례해서 커진다. 여기서,

는 PHN의 변화율을 나타낸다.Received symbol vector

The process of obtaining PHN

Is followed by a Wiener process. According to the definition of the Wiener process, the PHN

And the PHN of the (n-1) th sample

A liver tea

Average

, Dispersion

Gaussian random variables. That is, the larger the time interval, the larger the variance value becomes. here,

Represents the rate of change of PHN.

번째 시간 샘플에서의 PHN을

라고 하면, PHN

은 다음과 같이 나타난다.OFDM symbol

RTI ID = 0.0 > PHN < / RTI &

PHN

Is expressed as follows.

……(5)

... ... (5)

번째 시간 샘플 즉, 시간축 상 중간의 샘플에서 멀어질수록 PHN

의 변화폭이 크다. 즉, 심볼 중앙에서 멀리 떨어진 심볼조각일수록 살펴보고자 하는 PHN 값의 범위가 넓어지도록 정해진다. 이처럼 중간의 샘플을 기준으로 PHN을 표현할 수 있다.That is, due to the nature of the Winner process

Th time sample, that is, the farther from the middle sample on the time axis, the more the PHN

. That is, the more the symbol fragments far from the center of the symbol, the wider the range of the PHN value to be examined. Thus, the PHN can be expressed based on an intermediate sample.

개의 심볼조각들로 나누어,

개의 심볼조각들을 얻는다. 하나의 심볼조각은 길이

을 가지게 되고, 여기서 L은 N/J의 값을 가진다.

번째 심볼조각에서의 PHN

은 다음과 같이 근사한다. 이 근사화는 하나의 심볼조각에 있어서, PHN

은 일정하다고 가정한 것이다.OFDM symbols can be sampled

Lt; / RTI > symbol pieces,

Lt; / RTI > One symbol piece is a length

, Where L has a value of N / J.

PHN in the < RTI ID = 0.0 >

Is approximated as follows. This approximation is based on one symbol segment, PHN

Is assumed to be constant.

……(6)

... ... (6)

이라는 확률변수로 다시 정의하고 다음과 같이 나타낸다.The amount of PHN increase between the symbol fragments of the same sample

And the following is expressed as follows.

……(7)

... ... (7)

도 샘플들 간의 PHN

과 마찬가지로 평균 0인 가우시안 확률분포를 따르게 되며, 분산은 다음과 같이 계산된다.PHN increase between symbol fragments

PHN between samples

, And the variance is calculated as follows.

……(8)

... ... (8)

과 마찬가지로, 특정 OFDM 샘플 내의 심볼조각

에서의 PHN

은 다음과 같이 다시 쓸 수 있다.Then, PHN (n) related to the OFDM sample shown in equation (5)

, A symbol fragment in a particular OFDM sample

PHN at

Can be rewritten as

... ... (9)

과 마찬가지로 심볼조각에서의 PHN

도

번째 시간샘플 즉, 시간축 상의 중앙에서 멀어질수록 변화폭이 커진다. CPE

는 어떤 식으로든 보상할 수 있는 것이므로, PHN

의 변화폭

를 추정할 수 있기만 하면, 위상잡음을 보상하는 것이 가능해진다. 그러므로 샘플의 심볼조각의 PHN

의 변화폭

를 구할 필요가 있다.PHN on sample

PHN < RTI ID = 0.0 >

Degree

Th time sample, that is, the distance from the center on the time axis increases. CPE

Can compensate in any way, so PHN

The change of

It is possible to compensate for the phase noise. Therefore, the PHN

The change of

.

번째 심볼조각에서,

대비

의 변화폭을

라고 하면 다음 수식과 같다.

In the second symbol fragment,

prepare

Of change

The following equation is obtained.

……(10)

... ... (10)

은 가우시안 확률변수

들의 합이기 때문에 평균이 0이고, 분산

의 가우시안 확률분포를 가지며, 변화폭의 분산

는 다음과 같다.Change width

The Gaussian random variable

The average is 0, and the variance

And the variance of the variation width

Is as follows.

……(11)

... ... (11)

의 값을 예컨대

에서

까지

개의 균등한 간격으로 양자화(Quantize)하여 q번째 양자화 값

를 얻는다. 양자화 값은 한 번에 다음 경로로 이동할 때 그 이동할 수 있는 경로의 가지 수를 의미한다. 구해진 양자화 값

를 코드북(88b)에 저장한다. 결국, 코드북(88b)에 저장되는 정보는 심볼조각들이 가질 수 있는 위상잡음 값을 위상잡음의 통계적 특성에 따라 유한한 수로 양자화 하여, 살펴보고자 하는 위상잡음으로 정한 정보로 볼 수 있다.Variation with successive values

For example,

in

Till

The quantized values of the q-th quantized values

. The quantization value means the number of paths that can be moved when moving to the next path at a time. The obtained quantization value

In the codebook 88b. As a result, the information stored in the codebook 88b can be regarded as information determined by the phase noise to be quantized by quantizing the phase noise values that the symbol fragments may have according to the statistical characteristics of the phase noise to a finite number.

을 구하는 구간은 가변적일 수 있다. 변화폭

을 구하는 구간을

~

보다 더 넓은 범위 예컨대 -3σ

~ 3σ

로 하는 것도 가능하다. 또한, 양자화 간격을 등간격으로 설정하는 방법 대신, 등면적으로 설정하는 것도 가능하다(이에 관한 자세한 사항은 후술함).Change width

May be variable. Change width

To obtain

~

A wider range, e.

~ 3σ

. Alternatively, instead of setting the quantization interval at equal intervals, it is also possible to set the equal area (the details of which will be described later).

번째 양자화 값

는 심볼조각

가 가운데 심볼조각

에서 멀어질수록 분산이 선형적으로 증가하기 때문에

에 따라 양자화 수

가 선형적으로 증가하도록 설정한다. Variable

Th quantization value

Symbol piece

Middle middle symbol piece

The dispersion increases linearly with distance from

Quantization according to

Is linearly increased.

……(12)

... ... (12)

는 심볼조각

에 따른 양자화 수 증가분이다.here,

Symbol piece

Is an increase in the number of quantizations.

, 양자화 수 증가분

, 한 심볼의 총 심볼조각 수

인 경우, 각 심볼조각마다 q번째 양자화 값

이 가질 수 있는 값들을 그린 것이다. FIG.

, The number of quantization increases

, The total number of symbol pieces of one symbol

, The q-th quantization value

I have drawn the possible values.

는 심볼 한가운데 즉, N/2번째 시간 샘플에서의 PHN 값으로서, CPE

의 위상으로 근사된다.

Is the PHN value in the middle of the symbol, i.e., the N / 2th time sample, CPE

As shown in FIG.

가 제거된 심볼조각

에서의 PHN

은 다음과 같이 나타난다.CPE

Symbol pieces removed

PHN at

Is expressed as follows.

……(13)

... ... (13)

에서 CPE

가 제거된 위상 잡음 매트릭스의 성분

는

에 의해 발생하는 왜곡성분이며, 다음 수식과 같다.Distortion component caused by phase noise

CPE

Of the phase noise matrix

The

Is a distortion component generated by the following equation.

……(14)

... ... (14)

는 다음과 같이 근사하여 다시 쓸 수 있다.According to the above analysis, the received symbol vector in the frequency domain

Can be approximated and rewritten as

……(15)

... ... (15)

와 채널 주파수 응답

가 보상된 수신 심볼 벡터를

라 정의하면, 이 수신 심볼 벡터

는 다음과 같다.CPE

And channel frequency response

Lt; RTI ID = 0.0 >

, The received symbol vector < RTI ID = 0.0 >

Is as follows.

……(16)

... ... (16)

를 구하는 것과는 별도로, 적어도 한 OFDM 심볼 시간 동안 변화하는 위상잡음에 해당하는 ICI 행렬의 역행렬

을 미리 코드북(88a)으로 만들어둘 수 있다. CPE

와 채널의 주파수 응답

은 수신 심볼 벡터

에서 떼어낼 수 있으므로,

을 코드북(88a)으로 만들어두기만 하면, CPE

와 주파수 응답

이 보상된 수신 심볼 벡터

를 구할 수 있다.

을 코드북(88a)으로 미리 만들어 둠으로써, 실시간 계산량을 크게 줄일 수 있어, 시스템의 전반적인 성능을 좋게 할 수 있다. In this way, the channel of the decoding unit 80 and the CPE compensation module 82 compute the reception symbol vector

, The inverse of the ICI matrix corresponding to the phase noise varying during at least one OFDM symbol time

Can be made into a codebook 88a in advance. CPE

And the frequency response of the channel

Lt; RTI ID = 0.0 >

And therefore,

Is made into a codebook 88a, the CPE

And frequency response

This compensated received symbol vector

Can be obtained.

Is prepared in advance as the codebook 88a, the real-time calculation amount can be greatly reduced, and the overall performance of the system can be improved.

을, 수신 심볼 벡터

와 함께, 파일럿 복호 및 PHN 추정 모듈(84)에 제공하여 PHN의 추정 및 보상에 사용한다. 각 심볼조각에 대한 PHN이 모두 정의되어 있지 않으면, 위상 잡음에 의해 생겨난 왜곡성분에 관한 매트릭스

를 만들 수 없다. 그러므로 현재 고려하고 있는 심볼조각을 제외한 나머지 심볼조각들에 대한 PHN은 모두 0으로 두고, 현재 고려하고 있는 심볼조각에 대해서만 PHN이 있다고 가정하고, 그 PHN의 값을 변화시키면서 분석하여 실제의 PHN과 가장 유사한 최적의 PHN 값을 정한다. 즉, 본 발명은 어떤 심볼조각의 PHN 경로를 살펴봄에 있어서 나머지 심볼조각들의 PHN은 모두 0으로 두고 그 어떤 심볼조각의 최적 PHN 경로를 나머지 심볼조작들의 PHN과는 상관없이 독립적으로 정한다. 이런 방식으로 모든 심볼조각에 대하여 최적 PHN 경로를 정한 다음, 그 모든 심볼조각들의 최적 PHN 경로들을 모두 조합하여 해당 심볼의 최적 PHN 경로를 정하는 방법이다. The codebook 88a

, A received symbol vector

To the pilot decoding and PHN estimation module 84 and used for estimation and compensation of the PHN. If not all of the PHNs for each symbol fragment are defined, the matrix for the distortion component caused by the phase noise

Can not be created. Therefore, it is assumed that the PHNs for the remaining symbol fragments other than the currently considered symbol fragments are all 0, the PHN is assumed to exist only for the currently considered symbol fragments, the PHN is analyzed while changing the value of the PHN, A similar optimal PHN value is determined. That is, according to the present invention, the PHN path of a certain symbol fragment is set to 0, and the PHN of the remaining symbol fragments are all set to 0, and the optimal PHN path of the symbol fragment is set independently of the PHN of the remaining symbol operations. In this way, an optimal PHN path is determined for all symbol fragments, and then an optimum PHN path of the corresponding symbol is determined by combining all the optimal PHN paths of all the symbol fragments.

을 포함하는 코드북(88a)을 생성할 때 그 코드북(88a)에는 다음 두 가지가 포함되도록 하는 점에 주요한 특징을 갖는다.In this regard,

The codebook 88a includes the following two features. The codebook 88a includes the following two codes.

번째 양자화 값

의 '모든 조합'으로 생성한

를 포함한다.1) Possible

Th quantization value

Created with "all combinations" of

.

번째 심볼조각의 PHN은

이고, 나머지 심볼조각의 PHN은

이 되도록 생성한

를 포함한다.2)

The PHN of the < RTI ID = 0.0 >

And the PHN of the remaining symbol fragments is

Generated

.

에 관한 코드북(88a)이 생성되고 나면, 그것을 이용하여 수신신호를 실제로 복호할 수 있다. 수신 신호의 복호는 다음과 같이 이루어질 수 있다.like this

It is possible to actually decode the received signal using the codebook 88a. The decoding of the received signal can be performed as follows.

번째 심볼조각을 제외한 심볼조각들의

를 0으로 두고

번째 심볼조각에서의

를

으로 하는

인

즉, 코드북(88a)에 미리 저장된

를 수신된 정보어 심볼 벡터

에 곱하여 파일럿 심볼을 복호한다.First, the receiving apparatus 50, in the pilot decoding and PHN estimation module 84,

Symbol pieces except for the < RTI ID = 0.0 >

To 0

Symbol in the

To

To

sign

That is, the codebook 88a is stored in advance

To the received information word symbol vector

And decodes the pilot symbols.

에 대해 반복한다. 이 과정에서 파일럿 심볼 복호 오차가 가장 작은

를 찾아,

번째 심볼조각에서 추정된 값

으로 정한다.This process

Lt; / RTI > In this process, the pilot symbol decoding error is the smallest

Find,

Lt; RTI ID = 0.0 >

.

에 대해 반복하여 모든 심볼조각에 대한

을 추정한다(S12 단계). 이를 통해 PHN의 경로가 구해지고, 그 PHN 경로에 상응하는

를 추정하는 것이 가능해진다. 추정된

를 미리 코드북(88b)으로 만들어 둘 수 있다. 그리고 이 코드북(88b)을 이후 수신 신호의 복호에 활용할 수 있다.And this process

Repeat for all the symbol fragments

(Step S12). Whereby the path of the PHN is obtained, and the path corresponding to the PHN path

Can be estimated. Estimated

Can be made into a codebook 88b in advance. The codebook 88b can be utilized for decoding a received signal thereafter.

에 상응하는

(이는 코드북(88b)에 미리 저장되어 있음)를 수신된 정보어 심볼 벡터

에 곱하여 ICI를 보상하면서 모든 정보어 심볼을 복호한다(S14 단계). 코드북(88a, 88b)은 오프라인에서 미리 만들어두기 때문에

를 계산하는 복잡도는 고려대상이 아니다. 도 5에 예시한 흐름도의 S20 단계부터 S40 단계까지의 절차는 위에서 설명한 심볼조각 별 위상잡음 추정(S12단계)과 ICI 보상을 통한 정보어 심볼 복호 과정을 개략적으로 나타내고 있다.The information word decoding module 86 of the decoding unit 80 decodes the estimated

Equivalent to

(Which is stored in advance in the codebook 88b) to the received information word symbol vector

And decodes all the information word symbols while compensating for ICI (step S14). Since the codebooks 88a and 88b are made offline in advance

Is not considered. The procedure from step S20 to step S40 of the flowchart illustrated in FIG. 5 schematically shows the phase noise estimation (step S12) of the symbol fragment described above and the information symbol decoding process through ICI compensation.

According to the present invention as described above, there is an effect that the number of pilot decoding repetitions required to achieve the same PHN estimation performance (that is, the number of times that related information should be retrieved from the codebook 88a, 88b) is reduced as compared with the conventional BMT. In particular, as the number of repetitions of the pilot decoding exceeds 9, the amount of computation to be performed is greatly reduced compared with the conventional BMT, and the effect is remarkable.

FIGS. 6 and 7 show simulation results for comparing the PHN path estimation performance between the present invention and the conventional technique. In both simulation results, it can be seen that although the PHN path estimated according to the method proposed by the present invention and the PHN path estimated according to the prior art are somewhat different from each other, real), the tendency of the path change is almost the same. That is, in the PHN path estimation performance, it can be seen that there is no difference between the present invention and the conventional technology in the degree of superiority.

FIG. 8 is a graph showing the number of times that each method performs an iterative operation to find an optimal PHN path when simulation is performed according to a proposed method and a conventional method proposed by the present invention. According to the simulation results, both methods exhibit similar bit error rate (BER) performance as the number of repetitions exceeds 5. When the number of repetitions exceeds 9, it can be seen that the number of pilot decodings for achieving the same BER performance is remarkably reduced in the method proposed by the present invention. In addition, a similar level of BER is obtained when the number of iterations of the method according to the present invention is 30 and the number of iterations of the conventional method is 2187, and it is confirmed that the method of the present invention can achieve a calculation gain of about 100 times . As described above, both methods show no significant difference in the number of operations when the BER is high. However, it can be seen that as the BER is lowered, the method of the present invention has a similar BER even if it performs much less repetition operations than the conventional method.

FIG. 9 shows another simulation result concerning the number of iterations according to the proposed method and the conventional method proposed by the present invention. It can be seen that this case also shows a tendency similar to that of FIG. As a result, the results of these two simulations can be regarded as having general characteristics.

의 변화폭

이 연속적인 값을 가지며 이 값을

개의 값으로 양자화(Quantize)하여

번째 양자화 값

를 얻는 것에 관해 설명한 바 있다. 대표 실시예와 같은 시스템을 고려할 때, 양자화하는 방식은 양자화 값들

이 균등한 확률을 가지도록 할 수도 있다.In the description of the exemplary embodiment of the present invention, PHN

The change of

Has a continuous value and the value

Quot;

Th quantization value

As described above. Considering the system as in the exemplary embodiment, the manner of quantization is based on quantization values

May have an equal probability.

은 다음과 같다.The quantized value in this way

Is as follows.

The remaining process is the same as that of the representative embodiment.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the following claims. It will be understood.

The present invention can be applied to any communication system using OFDM, regardless of wired or wireless.

10: OFDM communication system 20: transmission device
25: Discrete Fourier inverse transform unit 30:
35: parallel-serial converter 40: mixer
45: Local oscillator 50: Receiver
55: mixer 60: local oscillator
65: serial-parallel conversion unit 70: cyclic prefix removal
75: Discrete Fourier transform unit 80:

Claims (20)

In an Orthogonal Frequency Division Multiplexing (OFDM) communication system,
A channel and a CPE compensator for estimating and compensating for a common phase error (CPE) of a received OFDM symbol and a channel;
A symbol block phase noise estimator for estimating a phase noise (PHN) by dividing a channel and a CPE compensated symbol into a plurality of symbol fragments in a time domain;
An ICI compensator for compensating for phase noise causing intercarrier interference (ICI) of symbols from the estimated phase noise; And
And a codebook provided to store an inverse matrix of an ICI matrix corresponding to phase noise varying during an OFDM symbol time and to utilize the inverse matrix for estimation and compensation of phase noise. Compensation device. delete 2. The apparatus of claim 1, wherein the value of the phase noise is not changed within one symbol segment. 2. The OFDM receiver of claim 1, wherein the codebook further comprises information determined by phase noise to be quantized by quantizing a phase noise value that the symbol fragments can have according to a statistical characteristic of phase noise to a finite number. A phase noise estimation and compensation apparatus for communication. 5. The apparatus of claim 4, wherein the phase noise value to be examined in one symbol segment is information independently determined regardless of phase noise values of other symbol fragments. 5. The apparatus of claim 4, wherein a range of values of the phase noise to be examined is set so as to become wider as the symbol fragments far from the center of the symbol become. 2. The OFDM communication system according to claim 1, wherein the codebook includes inverse matrix information of an ICI matrix corresponding to a phase noise value of 0 other than one symbol fragments. . The apparatus of claim 1, wherein the codebook includes inverse matrix information of an ICI matrix corresponding to a combination of phase noise values to be examined for each symbol piece. The apparatus of claim 1 or 7, wherein the symbol segment phase noise estimation unit estimates phase noise by decoding the pilot symbols using the codebook for each symbol fragment. 10. The apparatus of claim 9, wherein an inverse matrix of an ICI matrix corresponding to a combination of phase noise estimated for each symbol fragments is found in the codebook and used for phase noise estimation. The apparatus of claim 1, further comprising a discrete Fourier transformer performing a discrete Fourier transform only once on the received OFDM symbol. In an Orthogonal Frequency Division Multiplexing (OFDM) communication scheme,
OFDM symbol

Storing a phase noise (PHN) value that each symbol segment can have in a codebook in advance;
Decoding the pilot symbols for all phase noise previously stored in the codebook, and determining phase noise having the smallest decoding error as phase noise estimation values of the symbol fragments; And
And compensating phase noise causing Intercarrier Interference (ICI) using a combination of phase noise independently estimated for each symbol fragments. 13. The method of claim 12, wherein the phase noise value of each symbol segment is assumed to have a value of zero in the middle of the time domain, Wherein the phase noise estimation and compensation method of the OFDM communication is determined by a method of determining the phase noise of the OFDM communication. 13. The method of claim 12, wherein the pilot signal is recovered for each phase noise value of each symbol block to compensate for phase noise with the smallest error as an estimate. 14. The OFDM receiver of claim 12, wherein the codebook further comprises information determined by phase noise to be quantized by quantizing a phase noise value that the symbol fragments can have according to a statistical characteristic of phase noise to a finite number, Phase noise estimation and compensation method of communication. 16. The phase noise estimation and compensation method of claim 15, wherein the phase noise value to be examined in one symbol segment is independently determined independently of the phase noise value of other symbol fragments. 16. The method of claim 15, wherein the range of the value of the phase noise to be examined is set to be wider as the symbol fragments far from the center of the symbol are spread. 13. The phase noise estimation and compensation method of an OFDM communication system according to claim 12, wherein the codebook includes inverse matrix information of an ICI matrix corresponding to a case where a phase noise value is 0 in other symbol fragments excluding one symbol fragment . 13. The method of claim 12, further comprising estimating phase noise by decoding the pilot symbols using the codebook for each symbol fragment. The method of claim 19, wherein the codebook stores an inverse matrix of an ICI matrix corresponding to phase noise varying during an OFDM symbol time, and calculates an inverse matrix of an ICI matrix corresponding to a combination of phase noise estimated for each symbol fragments, And estimating and compensating phase noise of the OFDM communication.

Images