KR102042042B1 - Method of estimating carrier frequency offset and detecting user equipment information in D2D communication - Google Patents

Abstract

In the method of estimating carrier frequency offset and detecting terminal information in a device-to-device (D2D) according to the present invention, a PSSS receiving step of receiving respective primary sidelink synchronization signals (PSSS) ; And a frequency offset / terminal information detection step of detecting the quantized frequency offset q and the terminal information i using the metric T (q, i, ξ) obtained through channel estimation independent from the PSSS. , Q, i, ξ represents the quantized frequency offset (q), residual error (ξ) and the terminal information (i), the residual error (ξ) is a random frequency offset and the quantized frequency offset ( q), and when the occurrence range of carrier frequency offset is wide as shown in the terminal-to-terminal communication or vehicle communication, a method of effectively estimating a wide range of carrier frequency offset using a synchronization signal such as PSSS and a synchronization signal It is possible to provide a method for obtaining included transmission terminal information.

Description

Method of estimating carrier frequency offset and detecting user equipment information in D2D communication}

The present invention relates to a carrier frequency offset estimation and terminal information detection method in a terminal-to-terminal communication and a terminal performing the same. More specifically, the present invention relates to a carrier frequency offset estimation and terminal information detection method when supporting device-to-device (D2D) communication or vehicle (vehicular-to-everything V2X) communication, and a terminal performing the same.

The biggest disadvantage of the wireless communication system based on Orthogonal frequency division multiplexing is that if the frequency synchronization is not correct, the orthogonality between subcarriers is broken and it is difficult to reliably recover information. Therefore, it is very important to accurately estimate the carrier frequency offset in an orthogonal frequency division multiplexing system.

      In the terminal-to-device (D2D) communication, especially when the terminal is out-of-coverage, the synchronization criterion needs to be matched to a terminal other than the base station. When the terminal-to-device (D2D) communication is extended to vehicle communication, a range of carrier frequency offsets that can occur as the transmitter and the receiver move simultaneously in a high speed mobile environment becomes larger.

      In the prior art, (US 20160227496A1, Synchronization of device to device communication filed on Sep 26, 2014), a synchronization procedure and resource allocation based on LTE sidelink channel to support inter-device communication and a method of transmitting a sync signal are also described. Providing. In another prior patent (US2016 / 0174174 A1, Method and apparatus for detecting synchronization signal in wireless communication system, Pub. Date June 16, 2016) LTE by recognizing the problem that the initial carrier frequency offset generation range in the terminal-to-device communication increases By using the PSSS signal characteristics provided by the standard, a method of reducing the metric calculation complexity required for simultaneously acquiring integer frequency offset (IFO) terminal information is provided. However, this method has a problem in that it uses no matched filtering-based metric and does not consider the intersymbol interference problem caused by time delay spreading in the multipath fading channel without phase information at an early stage.

As a performance improvement method for estimating integral frequency offset (IFO) in multipath fading channels, a method of performing channel estimation and integer frequency offset estimation using the maximum likelihood (ML) method has been described in the existing paper [Z. Z. Lu, L. L. Zhao, and J. J. Li, "Further results on the maximum likelihood IFO estimation for OFDM systems," IEEE Transactions on Consumer Electronics, vol. 2, no. 53, pp. 286-288, 2007.]. However, this method is performed using only one symbol under the assumption that there is no fractional frequency offset (FFO), and does not consider performance degradation due to fractional frequency offset.

US published patent US2016 / 0174174A1 (2016.06.16)

Accordingly, an object of the present invention is to provide a method for efficiently estimating a carrier frequency offset and a terminal information detection method in terminal-to-terminal communication or vehicle communication.

In addition, the problem to be solved by the present invention is to provide a frequency offset estimation that can be robust to the estimation error occurring in the preceding frequency offset estimation in the terminal-to-vehicle communication or vehicle communication and can utilize the characteristics of the PSSS signal transmitted repeatedly Shall be.

According to an aspect of the present invention, a carrier frequency offset estimation method and a terminal information detection method in a device-to-device (D2D) according to an aspect of the present invention may include respective primary sidelink synchronization signals (PSSS). PSSS receiving step of receiving a Primary Sidelink Synchronization Signal; And frequency offset / terminal information detection using the metric T (q, i, ξ) obtained through channel estimation independent from the PSSS to detect the quantized frequency offset q and the terminal information i included in the PSSS. Wherein q, i, ξ represents the quantized frequency offset q, residual error ξ and the terminal information i, and the residual error ξ represents an arbitrary frequency offset and the quantization. When the frequency range of the carrier frequency offset is determined by the difference between the frequency offsets (q) and the occurrence of the carrier frequency offset is wide, as in the inter-terminal communication or the vehicle communication, a method of effectively estimating a wide range of carrier frequency offsets using a synchronization signal such as PSSS And a method for acquiring transmission terminal information included in a synchronization signal.

According to an embodiment, in the frequency offset / terminal information detection step, 2K + 1 candidates are selected within a range in which the residual error ξ may occur, and a second metric for the selected 2K + 1 candidates [ The hypothesis H _{( q, i )} maximizing Equation 18 may be selected, and the frequency offset / terminal information detection step may be performed based on the selected H _{( q, i )} .

로 선택하여, [수학식 19] 또는 [수학식 20]을 최대로 하는 가설 H_{( q,i )}를 선택하고, 상기 선택된 H_{( q,i )}에 기반하여 상기 주파수 옵셋/단말 정보 검출 단계를 수행할 수 있다.According to one embodiment, in the second metric

Select the hypothesis H _{( q, i )} that maximizes Equation 19 or Equation 20 _, and detect the frequency offset / terminal information based on the selected H _{( q, i )} . Can be done.

According to an embodiment, a first error threshold associated with a detection error probability of the quantized frequency offset (QFO), a second error threshold associated with a detection error probability of a group ID associated with the terminal information, and a signal per symbol One of the J ₀ (q, i) or the J ₁ (q, i) may be selected according to the interference ratio, and the multipath fading environment type.

을 이용하여 최대 우도 추정 값을 추정하는 최대 우도 추정 단계를 포함할 수 있다. 한편, 상기 추정된 최대 우도 추정 값에 기반하여, 오류 벡터를 최소로 하는, 상기 메트릭 T(q,i,ξ)는 [수학식 17]과 같이 주어진다.According to one embodiment, in the hypothesis selection step, each PSSS is represented by zm,

The method may include a maximum likelihood estimating step of estimating a maximum likelihood estimation value using. Meanwhile, based on the estimated maximum likelihood estimation value, the metric T (q, i, ξ), which minimizes an error vector, is given by Equation 17.

According to an embodiment, in the frequency offset / terminal information detection step, 2K + 1 candidates are selected within a range in which the residual error ξ may occur, and a second metric for the selected 2K + 1 candidates [ The hypothesis H _{( q, i )} maximizing Equation 11 may be selected, and the frequency offset / terminal information detection step may be performed based on the selected H _{( q, i )} .

According to another aspect of the present invention, a carrier frequency offset estimation and terminal information detection method in a device-to-device (D2D) receives a plurality of primary sidelink synchronization signals (PSSS). PSSS receiving step; And detecting frequency offset / terminal information for detecting the quantized frequency offset q and the terminal information i by using the metric M (q, i, ξ) obtained through channel characteristic estimation common to the plurality of PSSSs. Wherein q, i, ξ represents the quantized frequency offset q, residual error ξ and the terminal information i, and the residual error ξ represents an arbitrary frequency offset and the quantized It is determined by the difference of the frequency offset q.

로 선택하여, [수학식 12] 또는 [수학식 13]을 최대로 하는 가설 H_{( q,i )}를 선택하고, 상기 선택된 H_{( q,i )}에 기반하여 상기 주파수 옵셋/단말 정보 검출 단계를 수행할 수 있다.According to one embodiment, in the second metric

Select the hypothesis H _{( q, i )} that maximizes Equation 12 or Equation 13 _, and detect the frequency offset / terminal information based on the selected H _{( q, i )} . Can be done.

According to an embodiment, a first error threshold associated with a detection error probability of the quantized frequency offset (QFO), a second error threshold associated with the terminal information detection error probability, a signal-to-interference ratio per symbol, and One of the F ₀ (q, i) or the F ₁ (q, i) may be selected according to the multipath fading environment type.

According to an embodiment, in the PSSS receiving step, a D2D sidelink subframe including the PSSS is received, and before the frequency offset / terminal information detection step, the received PSSS in a multipath fading environment. Obtaining time synchronization of the PSSS by using an autocorrelation function; And estimating an integer frequency offset according to the obtained time synchronization.

According to an embodiment, each PSSS is represented by z _m , and the frequency offset / terminal information detection step includes a maximum likelihood estimation step of estimating a maximum likelihood estimation value using Equation 8, Based on the estimated maximum likelihood estimation value, the metric M (q, i, ξ), which minimizes the error vector, is given by Equation (10).

According to at least one embodiment of the present invention, when the generation range of the carrier frequency offset is wide as shown in the terminal-to-terminal communication or vehicle communication, the method and the synchronization signal to effectively estimate a wide range of carrier frequency offset using a synchronization signal such as PSSS There is an advantage in that it can provide a method for obtaining the transmission terminal information included in.

In addition, according to at least one embodiment of the present invention, to provide a frequency offset estimation range expansion method and a terminal information acquisition method that is robust to the estimation error occurring in the preceding frequency offset estimation and can utilize the characteristics of the PSSS signal transmitted repeatedly. There is an advantage that it can.

1 illustrates a subframe structure including a primary sidelink synchronization signal (PSSS) for D2D synchronization.
2 is a block diagram illustrating a configuration of a receiving terminal that performs time and frequency synchronization according to the present invention.
3 is a block diagram illustrating a detailed configuration of a terminal of a terminal, in particular, a modem of a receiving terminal performing a time and frequency synchronization method.
4 shows a simulation result for the quantization frequency offset detection probability according to the present invention.
5 shows a simulation result of the terminal information detection probability according to the present invention.
6 shows a flowchart of a carrier frequency offset estimation method according to the present invention.

The above-described features and effects of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, and thus, those skilled in the art to which the present invention pertains may easily implement the technical idea of the present invention. Could be. As the inventive concept allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the text. However, this is not intended to limit the present invention to a specific disclosure, it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

In describing each drawing, like reference numerals are used for like elements.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. The terms are used only 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 the second component, and similarly, the second component may also be referred to as the first component. The term “and / or” includes any combination of a plurality of related items or any item of a plurality of related items.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art.

Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art, and are not construed in ideal or excessively formal meanings unless expressly defined in this application. Should not.

The suffixes "module", "block", and "unit" for components used in the following description are given or mixed in consideration of ease of specification, and do not have distinct meanings or roles by themselves. .

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described to be easily carried out by those of ordinary skill in the art. In the following description of the embodiments of the present invention, when it is determined that the detailed description of the related known functions or known configurations may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

Hereinafter, a method of estimating a carrier frequency offset in a device-to-device (D2D) according to the present invention and a terminal performing the same will be described.

샘플로 구성된다. 여기서 은 FFT 크기로 유효 데이터 샘플 수이고,

는 유효 데이터 샘플 앞에 삽입되는 순환 전치 (Cyclic Prefix) 길이이다. First, FIG. 1 illustrates a subframe structure including a primary sidelink synchronization signal (PSSS) for D2D synchronization in relation to the present invention. As shown in FIG. 1, when one terminal serves as a reference for synchronization between other terminals in a terminal-to-device (D2D) communication, the reference terminal follows the LTE D2D standard, that is, to provide a synchronization criterion as shown in FIG. 1. The D2D Sidelink subframe including the PSSS signal is transmitted. At this time, the PS2 signal is repeatedly transmitted twice in the D2D sidelink subframe, and one PSSS signal is transmitted in the time domain.

It consists of samples. Where is the number of valid data samples in FFT size,

Is the cyclic prefix length inserted before the valid data sample.

On the other hand, the valid data portion of the PSSS is shown in [Equation 1].

은 표준에서 정의한 ZC (Zadoff-Chu) 수열로, ZC 수열의 근 인자 ui는 단말 정보 i=0,1에 따라 u₀=26 또는 u₁=37가 된다. 단말 정보 i는 Sidelink ID 집합 정보로 각 단말은 네트워크 내에 있는지 (i=0), 네트워크 밖에 있는지 (i=1)를 나타낸다. 단말은 네트워크 내에 있으면 0 ~ 167사이의 정수값을 Sidelink ID로 배정받고, 네트워크 밖에 있으면 168~335 사이의 정수값을 Sidelink ID로 배정받는다. 부연하면, 송신 단말의 Sidelink ID가 0~ 167 가운데 하나의 값을 가지면 단말 정보는 i=0이고, 상기 송신 단말이 전송하는 PSSS는 근 인자가 u₀=26인 ZC 수열로 구성된다. 송신 단말의 Sidelink ID가 168~335 사이의 한 정수 값을 가지면 단말 정보는 i=1이고, 상기 송신 단말이 전송하는 PSSS는 근 인자가 u₁=37인 ZC 수열로 구성된다. here

Is a ZC (Zadoff-Chu) sequence defined in the standard, and the root factor ui of the ZC sequence is u ₀ = 26 or u ₁ = 37 according to the terminal information i = 0,1. The terminal information i is Sidelink ID set information and indicates whether each terminal is in the network (i = 0) or outside the network (i = 1). The terminal receives an integer value between 0 and 167 as a Sidelink ID when it is in a network, and receives an integer value between 168 and 335 as a Sidelink ID when it is outside a network. In other words, if the sidelink ID of the transmitting terminal has one of 0 to 167, the terminal information is i = 0 and the PSSS transmitted by the transmitting terminal is composed of a ZC sequence having a root factor of u ₀ = 26. If the sidelink ID of the transmitting terminal has an integer value between 168 and 335, the terminal information is i = 1, and the PSSS transmitted by the transmitting terminal is composed of a ZC sequence having a root factor of u ₁ = 37.

와 같이, 하나의 PSSS 신호는 Ncp 길이의 순환 전치와 N개의 유휴 데이터로 구성된다.

As shown, one PSSS signal consists of a cyclic prefix of Ncp length and N idle data.

Other terminals that receive the LTE Sidelink subframe including the PSSS adjust time and frequency synchronization from the received signal. In this case, it is assumed that there is no channel change during two PSSS signal times, and a received signal sampled for each Ts may be expressed as Equation 2 below.

Where x (n) is the sidelink transmit signal including PSSS, L is the maximum time length of multipath fading, h (l) is the lth multipath fading amplitude, w (n) is independent of mean 0 and variance σ ₂ Normal noise, ε = f ₀ NTs is a frequency offset value obtained by normalizing the actual carrier frequency offset f ₀ to subcarrier spacing 1 / NTs.

When the synchronization signal is repeatedly transmitted, the receiving terminal may use autocorrelation as a method of adjusting time and frequency synchronization. Equation 3 uses an autocorrelation function that is not normalized to the received signal.

The receiving terminal can find the starting point of the PSSS by finding n '= n ₀ which maximizes the metric including the above autocorrelation function, and calculates the frequency offset normalized by the autocorrelation value at the starting point [Equation 4]. Can be estimated as

이다. Where ∠ is a function representing the phase of a complex number

to be.

이다. 곧, 부반송파 간격 1/NTs이 15kHz인 LTE 시스템에서, N=1024, Ncp=78이면, R=0.9344이고 정규화된 주파수 옵셋 추정 범위는

이다. 결국, 추정 가능한 반송파 주파수 옵셋 범위는 |f₀|≤6.858kHz이다. 한편, 단말간 (D2D) 통신에서는 환경에 따라 반송파 주파수 옵셋이 ±50 kHz까지도 발생할 수 있다. 따라서 [수학식 4]의 주파수 옵셋 추정 방법으로는 단말간 통신에서 발생하는 넓은 범위의 반송파 주파수 옵셋 추정이 가능하지 않다. The range of ε = f ₀ NTs that can be estimated by Equation 4 is

to be. In other words, in an LTE system with subcarrier spacing 1 / NTs of 15 kHz, if N = 1024 and Ncp = 78, then R = 0.9344 and the normalized frequency offset estimation range is

to be. As a result, the estimated carrier frequency offset range is | f ₀ | ≦ 6.858 kHz. Meanwhile, in D2D communication, carrier frequency offset may occur up to ± 50 kHz depending on the environment. Therefore, in the frequency offset estimation method of Equation 4, a wide range of carrier frequency offset estimation that occurs in terminal-to-terminal communication is not possible.

In the present invention, when the normalized frequency offset range estimated in the initial synchronization process is denoted by, the normalized frequency offset is divided into two parts as shown in [Equation 5].

로 결정되는 자연수이다. Where Q is the maximum value fmax of the carrier frequency offset experienced by the receiving terminal.

Is a natural number determined by.

For example, when fmax = 50 kHz, N = 1024, Ncp = 78, 1 / Ts = 15.36 × 10 ⁶ , the frequency offset is estimated by Equation 4 during the initial synchronization process, and R = 0.9344 and Q = 3. to be.

On the other hand, when ε _q = q R is an integer, it is as follows. In this regard, the case where R = 2 and ε _q = 2q has an even value is applied by applying a structure in which an effective portion having a length of N in one OFDM symbol is repeated N / 2 times. In this regard, ε _q = 2q may be referred to as an integral frequency offset (IFO), and thus, −1 ≦ r ≦ 1 remaining as a fractional frequency offset (FFO). In this study, assuming that the fractional frequency offset was perfectly estimated, we obtained ε _q = 2q through the frequency domain correlation of the synchronous signal shifted by ε _q = 2q subcarriers in the frequency domain by the integer fraction frequency offset ε _q = 2q. . In addition, an integer frequency offset was estimated with one OFDM symbol.

On the other hand, the present invention is not limited to the integer part frequency offset, and in consideration of the case where R is any positive value that is not an integer, ε _{q may} also have a fractional part which is not an integer. Therefore, in the present invention, when R is defined as a quantization step in Equation 5 and ε _q and r are respectively quantized to quantization step R, a quantized frequency offset (QFO) and the remaining frequency offset ( remainder of CFO (RFO).

로 쓰면, 잔류 오류

는 |ξ|<R 범위에서 발생한다. 나머지 주파수 옵셋이 없거나 추정이 완벽할 경우 ξ=0이다. The remaining frequency offset r may be estimated by Equation 4 or another method, but the estimation may not be perfect in an actual system. Remaining frequency offset estimate

, Residual error

Occurs in the | ξ | <R range. Ξ = 0 if there is no remaining frequency offset or the estimate is perfect.

2 is a block diagram illustrating a configuration of a receiving terminal that performs time and frequency synchronization according to the present invention. On the other hand, Figure 3 is a block diagram showing a detailed configuration of a terminal of the terminal according to the present invention, in particular the receiving terminal performing the time and frequency synchronization method. Referring to FIG. 2, the receiving terminal 200 receives a primary sidelink synchronization signal (PSSS) from a transmitting terminal (not shown) to perform time / frequency synchronization. On the other hand, the receiving terminal 200 includes a modem 210, and the transceiver 220. As shown in FIG. 3, the modem 210 includes an initial time synchronization performer 211, a remaining frequency offset estimator and compensator 212, and a quantized frequency offset and terminal information detector 213. As illustrated in FIG. 3, the method of performing time synchronization and frequency synchronization may be performed in the order of initial time synchronization → remaining frequency offset estimation and compensation → quantization frequency offset and terminal information detection, but is not limited thereto.

로 타나낸다. 이러한 나머지 주파수 옵셋 추정값을 보상한 신호에서 순환 전치를 제거한 m번째 PSSS 수신 신호 벡터는

로 나타낼 수 있다. 송신 단말 정보가 i이고 수신 단말의 양자화된 주파수 옵셋이 ε_q=qR이고, 잔류 오류 ξ가 있을 때 m번째 PSSS 수신 벡터는 [수학식 6]과 같이 표현될 수 있다.In the present invention, as shown in FIG. 3, a method of obtaining quantized frequency offset and terminal information after performing initial time synchronization and remaining frequency offset estimation process and compensation is described. To describe the proposed method, a signal that compensates for the remaining frequency offset estimate in the received sample of [Equation 2]

To appear. The m-th PSSS received signal vector from which the cyclic prefix is removed from the signal that compensates for the remaining frequency offset estimate is

It can be represented as. When the transmitting terminal information is i, the quantized frequency offset of the receiving terminal is? _Q = qR, and there is a residual error ξ, the m th PSSS reception vector may be expressed as Equation 6 below.

,

로 둘 수 있다.Where θm is the common phase shift experienced by the mth PSSS due to the frequency offset, without loss of generality

,

Can be placed.

는 주파수 옵셋으로 인한 수신 샘플마다의 위상 천이를 반영한 대각 행렬,

Is a diagonal matrix reflecting the phase shift for each received sample due to the frequency offset,

는 첫 열 벡터가 시간 영역 PSSS 송신 신호이고, 다른 열은 첫 열의 순환 이동으로 구성되는 circulant 행렬,

Is a circulant matrix in which the first column vector is a time-domain PSSS transmission signal and the other columns consist of cyclic shifts of the first column,

는 길이가 Ncp인 채널응답 벡터,

Is a channel response vector of length Ncp,

는

에 해당하는 의 평균이 0이고 분산이 σ²인 독립 정규 잡음이다.

Is

Is an independent normal noise with an average of 0 and a variance of σ ² .

     In Equation 6, the number of columns of X (i) is limited to the maximum time length L of the multipath fading channel, but L is often unknown when initial synchronization is acquired. In this case, since the cyclic prefix length is likely to be smaller than Ncp, the channel vector length is assumed to be Ncp and the number of columns is also assumed to be Ncp. If L is known in advance, the channel vector length is limited to L, and X (i) may be considered as N × L, and the following content may be applied as it is.

In Equation 6, the quantized frequency offset q may have one of Sq = -Q, -Q + 1, ..., Q, and the terminal information may have one of I _D = 0,1. have.

을 수립하고 이 가운데 하나의 가설을 선택함으로써 양자화된 주파수 옵셋

와 단말 정보

를 결정한다. Accordingly, the receiving terminal 200 needs to restore the quantized frequency offset and the terminal information regardless of whether it is requested. Thus, multiple hypotheses

And establish one of these hypotheses to quantize the frequency offset

And terminal information

Determine.

중 하나의 가설

를 검정하는 방법으로 각 가설이 올바르다는 가정하에 두 수신 신호 벡터 z₀, z₁로부터 공통 채널 h의 최대우도 (maximum likelihood) 추정 값

을 얻고, 해당 가설 하에 확률적으로 가능성이 가장 큰 (최대 우도) 가설을 선택하는 것이다. 이를 위해 두 PSSS 수신 신호를 한 벡터로 나타내면 [수학식 7]과 같다.In one embodiment of the present invention, multiple hypotheses

One hypothesis of

The maximum likelihood estimate of the common channel h from two received signal vectors z ₀ and z ₁ under the assumption that each hypothesis is correct.

And then choose the most likely (maximum likelihood) hypothesis under that hypothesis. To this end, two PSSS received signals are represented by one vector.

Under hypothesis H _{(q, i)} , assuming that the residual error ζ is known, the maximum likelihood channel estimate is given by Equation (8).

    The maximum likelihood hypothesis test using the channel estimate is given by Equation 9 under normal noise.

In this case, the hypothesis test according to Equation 9 is the same as selecting a hypothesis that maximizes the metric of Equation 10.

이다. here,

to be.

In the present invention, the hypothesis of maximizing the metric of Equation 11 calculated by selecting 2K + 1 candidates from the range of (-R, R) where ζ can occur with respect to the residual error ζ included in Equation 10. H _{(q, i)} is to be selected.

      Where ζk is the residual error candidate

와 같이 동시에 얻을 수 있고, 양자화된 주파수 옵셋과 D2D 단말 정보를

와

로 독립적으로 얻을 수 있다. Quantized frequency offset and D2D terminal information by using the metric.

Can be obtained simultaneously, and the quantized frequency offset and D2D terminal information

Wow

Can be obtained independently.

In one exemplary embodiment of the present invention, by setting in Equation 11, metrics such as Equation 12 and Equation 13 may be utilized for hypothesis testing.

에서의 메트릭을 효율적으로 계산하는 방법으로

를 이용하여 복잡도를 줄일 수 있다. 예를 들어, [수학식 12]의 메트릭을 얻기 위해 총 |S_D|×|I_D|개의 메트릭을 계산하여야 하나, [수학식 13]의 메트릭을 적용하는 [수학식 12]의 3배가 아닌 2배의 메트릭 계산만이 필요하다.Multiple hypothesis

To efficiently calculate metrics from

Complexity can be reduced. For example, to obtain the metric of Equation 12, the total | S _D | × | I _D | metrics must be calculated, but not three times that of Equation 12, which applies the metric of Equation 13. Only two metric calculations are needed.

In addition, in calculating the metrics of Equation 12 and Equation 13, applying a simplified expression such as Equation 14 and Equation 15 reduces the complexity of calculation.

중 하나의 가설

를 검정하는 방법으로 각 가설이 올바르다는 가정하에 각 수신 신호 벡터 z_m에 대한 채널을 독립적으로 추정하는 것이다. 이를 위해 [수학식 6]의 수신 신호를 [수학식 16]으로 모형화한다. In another embodiment of the present invention multiple hypotheses

One hypothesis of

The test method is to independently estimate the channel for each received signal vector z _m under the assumption that each hypothesis is correct. To this end, the received signal of [Equation 6] is modeled by [Equation 16].

이다.here,

to be.

와 같이 독립적으로 최대우도 추정한 뒤, 가설을 검증하기 위해 필요한 [수학식 17]의 메트릭을 얻는다. In independent channel estimation, the channel is hypothesized for each PSSS received signal.

After estimating the maximum likelihood independently, we obtain the metric of Equation 17 needed to test the hypothesis.

In the present invention, the hypothesis of maximizing the metric of [Equation 15] calculated by selecting 2K + 1 candidates in the range [-R, R] where ξ can occur with respect to residual error ξ included in [Equation 14]. H _{(q, i)} is to be selected.

로 설정하여 [수학식 19]과 [수학식 20]의 메트릭을 가설 검정에 활용할 수 있다. In one embodiment of the present invention,

By setting to, we can use the metrics of Equation 19 and Equation 20 to test the hypothesis.

에서의 메트릭을 효율적으로 계산하는 방법으로

를 이용하여 복잡도를 줄일 수 있다. 즉, 모든 가설에 대해 [수학식 17]의 메트릭을 계산할 때

임을 이용하여 계산 복잡도를 낮출 수 있다. Multiple hypothesis also in this embodiment

To efficiently calculate metrics from

Complexity can be reduced. In other words, when calculating the metric in Equation 17 for all hypotheses,

Can be used to reduce the computational complexity.

In the above, the carrier frequency offset estimation and the terminal information detection method according to the present invention has been described. Hereinafter, the operation of each component of the receiving terminal for performing such carrier frequency offset estimation will be described.

In this regard, referring to Figures 2 and 3, the carrier frequency offset estimation method in the receiving terminal 200 according to the present invention will be described as follows.

The transceiver 220 receives a plurality of primary sidelink synchronization signals (PSSS).

The modem 210 detects the quantized frequency offset q and the terminal information i included in the PSSS by using a metric obtained through channel characteristic estimation common to the plurality of PSSSs. Here, the modem 210 may use the metric M (q, i, ξ) obtained through channel characteristic estimation common to the plurality of PSSS. In this case, q, i, ξ represents a quantized frequency offset (q), residual error (ξ), and terminal information (i). On the other hand, the residual error ξ is determined by the difference between an arbitrary frequency offset and the quantized frequency offset q.

Meanwhile, referring to FIG. 3, the initial time synchronization performing unit 211 may acquire the time synchronization of the PSSS using the received autocorrelation function of the PSSS in a multipath fading environment. In addition, the remaining frequency offset estimation and compensator 212 estimates common channel characteristics from the plurality of received PSSSs and maximizes the sum of 2K + 1 metrics M _k (q, i, ξ). ) H _{( q, i )} can be selected. In addition, the quantized frequency offset and the terminal information detector 213 may detect the quantized frequency offset q and the terminal information i based on the selected H _{( q, i )} .

Meanwhile, the modem 210 selects 2K + 1 candidates within a range in which the residual error ξ may occur, and maximizes Equation 11, which is a second metric for the selected 2K + 1 candidates. The hypothesis H _{( q, i )} can be selected, and the frequency offset / terminal information detection can be performed based on the selected H _{(q, i)} .

로 선택하여, [수학식 12] 또는 [수학식 13]을 최대로 하는 가설 H_{( q,i )}를 선택하고, 상기 선택된 H_{( q,i )}에 기반하여 상기 주파수 옵셋/단말 정보 검출을 수행할 수 있다.On the other hand, the modem 210, in the second metric

Is selected, the hypothesis H _{( q, i )} maximizing [Equation 12] or [Equation 13] is selected _, and the frequency offset / terminal information detection is performed based on the selected H _{( q, i )} . can do.

According to another embodiment of the present invention, the transceiver 220 receives each primary sidelink synchronization signal (PSSS).

The modem 210 detects the quantized frequency offset q and the terminal information i using the metric T (q, i, ξ) obtained through channel estimation independent from each PSSS. Here, the modem 210 may use the metric T (q, i, ξ) obtained through channel estimation independent from the PSSS. In this case, q, i, ξ represents a quantized frequency offset (q), residual error (ξ), and terminal information (i). On the other hand, the residual error ξ is determined by the difference between an arbitrary frequency offset and the quantized frequency offset q.

Meanwhile, the modem 210 selects 2K + 1 candidates within a range where the residual error ξ may occur, and maximizes Equation 18, which is a second metric for the selected 2K + 1 candidates. The hypothesis H _{( q, i )} may be selected and the frequency offset / terminal information detection may be performed based on the selected H _{(q, i)} .

로 선택하여, [수학식 19] 또는 [수학식 20]을 최대로 하는 가설 H_{( q,i )}를 선택하고, 상기 선택된 H_{( q,i )}에 기반하여 상기 주파수 옵셋/단말 정보 검출을 수행할 수 있다.On the other hand, the modem 210 in the second metric

Is selected, the hypothesis H _{( q, i )} maximizing [Equation 19] or [Equation 20] is selected _, and the frequency offset / terminal information detection is performed based on the selected H _{( q, i )} . can do.

The effect of the present invention can be seen as a simulation result for the quantization frequency offset detection probability shown in FIG. 4 and the terminal information detection probability shown in FIG. 5. That is, FIG. 4 shows a simulation result for the quantization frequency offset detection probability according to the present invention. In addition, Figure 5 shows the simulation results for the detection probability of the terminal information according to the present invention. Simulations were performed for LTE systems with N = 1024, N _G = 78, 1 / NTs = 15 kHz, considering the case where the D2D group is i = 0 and the normalized frequency offset is ε = 1.2 and Q = 1 It was. The flat fading channel and the ITU-R Extended Vehicular A (EVA) fading channel were considered. On the other hand, the results according to the present invention were compared with the existing Zhang's method and Lu's method technique.

As a technique corresponding to the present invention, Proposed F ₀ by [Equation 12], Proposed F ₁ by [Equation 12], Proposed J ₀ by [Equation 19], Proposed J ₁ by [Equation 20] Were compared together. As a result, it can be seen that the proposed technique provides improved performance in the same estimation range as the conventional method. In particular, the performance of the Proposed F ₁ and Proposed J ₁ consideration of the residual error is not perfect to see the performance better than the Proposed F ₀ and Proposed F _1, the Proposed J ₁ estimates the channels independently when the residual error You can see that the performance is slightly better.

Meanwhile, considering the results of FIGS. 4 and 5, the modem 210 may perform the following operations. That is, the modem 210 may include: a first error threshold associated with a detection error probability of the quantized frequency offset (QFO), a second error threshold associated with the terminal information detection error probability, and a signal-to-interference per symbol Depending on the ratio, and the type of multipath fading environment, one of the F ₀ (q, i) or the F ₁ (q, i) may be selected.

Specifically, the modem 210 has a first error threshold of 10 ⁻³ in a flat environment and a signal-to-interference ratio (E _s / N ₀ ) per symbol of 10 dB or more. For convenience of operation, F ₀ (q, i ) Can be selected. On the other hand, the modem 210 has F ₁ (q, i) to guarantee performance if the first error threshold is 10 ⁻³ in a flat environment and the signal-to-interference ratio E _s / N ₀ per symbol is 10 dB or less. Can be selected.

According to the first embodiment (see FIG. 4), the modem 210 has a first error threshold of 10 ⁻³ in a flat environment and a signal-to-interference ratio E _s / N ₀ per symbol of 10 dB or more. For convenience, F ₀ (q, i) can be selected. On the other hand, the modem 210 has F ₁ (q, i) to guarantee performance if the first error threshold is 10 ⁻³ in a flat environment and the signal-to-interference ratio E _s / N ₀ per symbol is 10 dB or less. Can be selected.

According to a second embodiment (see FIG. 5), the modem 210 has a second error threshold of 10 ⁻³ in a flat environment and a signal-to-interference ratio E _s / N ₀ per symbol of 10 dB or more. For convenience, F ₀ (q, i) can be selected. On the other hand, the modem 210 is the second threshold if the error is 10 ^-3, and the signal-to-interference ratio (E _s / N ₀₎ per symbol than 10dB, F ₁ (q, i) to ensure the performance on the environment Flat Can be selected.

According to the third embodiment (see FIG. 4), the modem 210 has a first error threshold of 10 ⁻³ in a flat environment and a signal-to-interference ratio E _s / N ₀ per symbol of 10 dB or more. For convenience, you can select J ₀ (q, i). On the other hand, the modem 210 and the first error threshold is ^10-3, the signal-to-interference ratio (E _s / N ₀₎ is equal to or less than 10dB, J ₁ (q, i) to ensure the performance per symbol in the environment Flat Can be selected.

According to the fourth embodiment (see FIG. 5), the modem 210 has a second error threshold of 10 ⁻³ in a flat environment and the signal-to-interference ratio E _s / N ₀ per symbol is 10 dB or more. For convenience, you can select J ₀ (q, i). On the other hand, the modem 210 has a second error threshold of 10 ⁻³ in a flat environment and a signal-to-interference ratio (E _s / N ₀ ) per symbol of 10 dB or less, so that J ₁ (q, i) to guarantee performance Can be selected.

Meanwhile, in relation to the carrier frequency offset estimation, the first embodiment selecting F _k (q, i) and the third embodiment selecting J _k (q, i) are not mutually exclusive. That is, J _k (q, i) can be selected for channel estimation (accurate initial frequency offset estimation) independent from each PSSS. On the other hand, the frequency offset estimation may then select F _k (q, i) for common channel characteristic estimation from the plurality of PSSS to reduce the number of operations. On the other hand, whether to select k = 0 or 1 (or more) can be selected in consideration of the error threshold requirement and E _s / N ₀ as described above. Alternatively, after calculating F ₀ (q, i) or J ₀ (q, i), F ₁ (q, i) or J ₁ (q, i) may be calculated through performance / error estimation. In this case, when calculating F ₁ (q, i) or J ₁ (q, i), the result of F ₀ (q, i) or J ₀ (q, i) may be used to improve the calculation speed.

In addition, in relation to group ID estimation, the second embodiment of selecting F _k (q, i) and the fourth embodiment of selecting J _k (q, i) are not mutually exclusive. That is, J _k (q, i) can be selected for channel estimation (accurate group ID estimation) independent from each PSSS. On the other hand, the group ID estimation may then select F _k (q, i) for common channel characteristic estimation from the plurality of PSSS to reduce the number of operations.

On the other hand, the carrier frequency offset estimation method according to another aspect of the present invention in detail as follows. In this regard, Figure 6 shows a flowchart of a carrier frequency offset estimation method according to the present invention. As shown in FIG. 6, the carrier frequency offset estimation method includes a D2D sidelink subframe receiving step S110, a time synchronization obtaining step S120, a remaining frequency offset estimation step S130, and a metric selection step according to a hypothesis S140. ), And frequency offset / terminal information detection step (S150).

In the step of receiving a D2D sidelink subframe (S110), a D2D sidelink subframe including an SSS is received. In the time synchronization acquisition step (S120), the time synchronization of the PSSS is obtained using the autocorrelation function of the received PSSS in a multipath fading environment. In the remaining frequency offset estimating step (S130), an integer frequency offset may be estimated according to the obtained time synchronization.

Meanwhile, according to an aspect of the present invention, in the metric selection step (S140) according to the hypothesis, the metric M (q, i) obtained through common channel characteristics from a plurality of primary sidelink synchronization signals (PSSS) , ξ). Next, in the terminal information detection step S150, the quantized frequency offset q and the terminal information i are detected using the metric M (q, i, ξ). In this case, q, i, ξ represents a quantized frequency offset (q), residual error (ξ), and terminal information (i). On the other hand, the residual error ξ is determined by the difference between an arbitrary frequency offset and the quantized frequency offset q.

On the other hand, the plurality of PSSS includes z _T = z ₀ , z ₁ . In addition, the metric selection step (S140) according to the hypothesis may include a maximum likelihood estimation step of estimating a maximum likelihood estimation value by Equation (8). Meanwhile, based on the estimated maximum likelihood estimation value, the metric M (q, i, ξ), which minimizes the error vector of Equation 9, is given by Equation 10.

 On the other hand, according to another aspect of the present invention, in the metric selection step (S140) according to the hypothesis, the metric T (q, i) through the channel estimation independent from each primary sidelink synchronization signal (PSSS) , ξ). Next, in the step S150 of detecting the frequency offset / terminal information, the quantized frequency offset q and the terminal information i using metric T (q, i, ξ) obtained through channel estimation independent from the PSSS. Detect. The q, i, ξ represents the quantized frequency offset (q), the residual error (ξ) and the terminal information (i), the residual error (ξ) is a random frequency offset and the quantized frequency offset (q) ) Is determined by the difference.

In this case, each PSSS is represented by z _m . In addition, the metric selection step S140 according to the hypothesis may include a maximum likelihood estimation step of estimating a maximum likelihood estimation value using Equation 16. Meanwhile, based on the estimated maximum likelihood estimation value, the metric T (q, i, ξ), which minimizes an error vector, is given by Equation 17.

In the step of detecting frequency offset / terminal information (S150), based on the selected H _{( q, i )} , the quantized frequency offset q and the terminal information i are detected.

According to at least one embodiment of the present invention, when the generation range of the carrier frequency offset is wide as shown in the terminal-to-terminal communication or vehicle communication, the method and the synchronization signal to effectively estimate a wide range of carrier frequency offset using a synchronization signal such as PSSS There is an advantage in that it can provide a method for obtaining the transmission terminal information included in.

In addition, according to at least one embodiment of the present invention, to provide a frequency offset estimation range expansion method and a terminal information acquisition method that is robust to the estimation error occurring in the preceding frequency offset estimation and can utilize the characteristics of the PSSS signal transmitted repeatedly. There is an advantage that it can.

In addition, the field of application of the present invention can be utilized in hardware implementation for frequency offset estimation and terminal information acquisition when designing a terminal modem in LTE-based terminal-to-terminal (D2D) communication. It can also be used for vehicle communications (V2X) based on the LTE standard. The metric, especially robust to residual errors, can be extended to wireless LAN modems with two iterations.

In addition, with respect to the expected effect of the present invention, in extending the existing carrier frequency offset estimation range, it is possible to reduce the residual error effect to improve the performance and improve the terminal information acquisition performance. This performance improvement enables the terminal to acquire fast synchronization, thereby delaying data transmission delay, thereby enabling ultra low delay services in 5G mobile communication systems and V2V communication systems. In addition, it is possible to reduce the terminal power consumption by preventing signal processing time and signal retransmission of the terminal for synchronization acquisition.

In addition, with regard to the commercialization strategy of the present invention, the core technology for modem design to be applied to the disaster communication network and the vehicle communication network applying the LTE Sidelink (D2D) standard, the technology can be transferred to the relevant modem design company and standardization company It is expected to be.

According to the software implementation, each component as well as the procedures and functions described herein may be implemented as separate software modules. Each of the software modules may perform one or more functions and operations described herein. Software code may be implemented in software applications written in a suitable programming language. The software code is stored in a memory and can be executed by a modem or a processor.

200: receiving terminal 210: modem
220: transceiver

Claims (20)

In a method of estimating carrier frequency offset and detecting terminal information in a device-to-device (D2D) communication,
A PSSS receiving step of receiving each Primary Sidelink Synchronization Signal (PSSS); And
Quantized frequency offset (q) and terminal information (i) included in the PSSS using the metric T (q, i, ξ) obtained for estimating the maximum likelihood after estimating the maximum likelihood through the independent channel estimation from the PSSS. Detecting a frequency offset / terminal information for detecting
The q, i, ξ represents the quantized frequency offset (q), the residual error (ξ) and the terminal information (i), the residual error (ξ) is a random frequency offset and the quantized frequency offset (q) Is determined by the difference between
In the frequency offset / terminal information detection step,
2K + 1 candidates are selected in the range [-R, R] where the residual error ξ can occur, and a second metric for the selected 2K + 1 candidates is selected.

Select a hypothesis H _{(q, i)} that maximizes and perform the frequency offset / terminal information detection step based on the selected H _{(q, i)} , carrier frequency offset estimation and terminal information detection Way. delete The method of claim 1,
In the second metric

Select to

or

Select a hypothesis H _{(q, i)} that maximizes and perform the frequency offset / terminal information detection step based on the selected H _{(q, i)} , carrier frequency offset estimation and terminal information detection Way. The method of claim 3,
A first error threshold associated with a detection error probability of the quantized frequency offset (QFO), a second error threshold associated with a detection error probability of a group ID associated with the terminal information, a signal-to-interference ratio per symbol, and multiplexing And selecting one of the J ₀ (q, i) or the J ₁ (q, i) according to a path fading environment type. The method of claim 1,
In the PSSS receiving step, receiving a D2D sidelink subframe including the PSSS,
Before the frequency offset / terminal information detection step,
Obtaining time synchronization of the PSSS by using the autocorrelation function of the received PSSS in a multipath fading environment;
And a remaining frequency offset estimating step of estimating an integer multiple frequency offset according to the obtained time synchronization. The method of claim 1,
The PSSS is represented by z _m ,
The frequency offset / terminal information detection step,

A maximum likelihood estimation step of estimating a maximum likelihood estimation value for each PSSS reception by
Based on the estimated maximum likelihood estimate value, the metric T (q, i, ξ), which minimizes the error vector, is

The carrier frequency offset estimation and terminal information detection method, characterized in that given.
Here, the quantized frequency offset ε _q = q R of the receiving terminal,

Is a diagonal matrix reflecting the phase shift for each received sample due to the frequency offset,

Is the circulant matrix whose first column vector is the time-domain PSSS transmission signal, and the other columns are cyclic shifts of the first column,

being. In a method of estimating carrier frequency offset and detecting terminal information in a device-to-device (D2D) communication,
A PSSS receiving step of receiving a plurality of Primary Sidelink Synchronization Signals (PSSS); And
The quantized frequency offset (q) and the terminal information (i) are obtained using the metric M (q, i, ξ) obtained for estimating the maximum likelihood after estimating the maximum likelihood through the channel characteristic estimation common to the plurality of PSSS. Detecting frequency offset / terminal information detection;
The q, i, ξ represents the quantized frequency offset (q), the residual error (ξ) and the terminal information (i), the residual error (ξ) is a random frequency offset and the quantized frequency offset (q) Is determined by the difference between
In the frequency offset / terminal information detection step,
2K + 1 candidates are selected in the range [-R, R] where the residual error ξ can occur, and a second metric for the selected 2K + 1 candidates is selected.

Select a hypothesis H _{(q, i)} that maximizes and perform the frequency offset / terminal information detection step based on the selected H _{(q, i)} , carrier frequency offset estimation and terminal information detection Way. delete The method of claim 7, wherein
In the second metric

Select to

or

Select a hypothesis H _{(q, i)} that maximizes and perform the frequency offset / terminal information detection step based on the selected H _{(q, i)} , carrier frequency offset estimation and terminal information detection Way. The method of claim 9,
A first error threshold associated with a detection error probability of the quantized frequency offset (QFO), a second error threshold associated with the terminal information detection error probability, a signal-to-interference ratio per symbol, and a multipath fading environment type. And selecting one of the F ₀ (q, i) or the F ₁ (q, i) according to the carrier frequency offset estimation and terminal information detection method. The method of claim 7, wherein
In the PSSS receiving step, receiving a D2D sidelink subframe including the PSSS,
Before the frequency offset / terminal information detection step,
Obtaining time synchronization of the PSSS by using the autocorrelation function of the received PSSS in a multipath fading environment;
And a remaining frequency offset estimating step of estimating an integer multiple frequency offset according to the obtained time synchronization. The method of claim 7, wherein
The plurality of PSSS includes z _T = z ₀ , z ₁ ,
The frequency offset / terminal information detection step,

A maximum likelihood estimating step of estimating a maximum likelihood estimated value by
An error vector, based on the estimated maximum likelihood estimate value

To minimize the metric M (q, i, ξ),

Is given by

The carrier frequency offset estimation and terminal information detection method, characterized in that.
Here, the quantized frequency offset ε _q = qR of the receiving terminal,

Is a diagonal matrix reflecting the phase shift for each received sample due to the frequency offset,

Is the circulant matrix whose first column vector is the time-domain PSSS transmission signal, and the other columns are cyclic shifts of the first column,

being. A terminal performing carrier frequency offset estimation in a device-to-device (D2D) communication,
A transceiver for receiving a primary sidelink synchronization signal (PSSS); And
The quantized frequency offset (q) and the terminal information (i) included in the PSSS using the metric T (q, i, ξ) obtained for estimating the maximum likelihood after estimating the maximum likelihood through the independent channel estimation from the PSSS. Includes a modem that detects
The q, i, ξ represents the quantized frequency offset (q), the residual error (ξ) and the terminal information (i), the residual error (ξ) is a random frequency offset and the quantized frequency offset (q) Is determined by the difference between
The modem,
2K + 1 candidates are selected in the range [-R, R] where the residual error ξ can occur, and a second metric for the selected 2K + 1 candidates is selected.

And selecting a hypothesis H _{(q, i)} that maximizes and performing the frequency offset and terminal information detection based on the selected H _{(q, i)} . delete The method of claim 13,
The modem,
In the second metric

Select to

or

And selecting a hypothesis H _{(q, i)} that maximizes and performing the frequency offset and terminal information detection based on the selected H _{(q, i)} . The method of claim 15,
The modem,
A first error threshold associated with a detection error probability of the quantized frequency offset (QFO), a second error threshold associated with a terminal information detection error probability, a signal-to-interference ratio per symbol, and a multipath fading environment type. The terminal, characterized in that to select one of the J ₀ (q, i) or the J ₁ (q, i). A terminal performing carrier frequency offset estimation in a device-to-device (D2D) communication,
A transceiver for receiving a plurality of primary sidelink synchronization signals (PSSS); And
A UE included in the PSSS and the quantized frequency offset q using the metric M (q, i, ξ) obtained for estimating the maximum likelihood after estimating the maximum likelihood through channel characteristic estimation common to the plurality of PSSS. A modem for detecting information (i),
The q, i, ξ represents the quantized frequency offset (q), the residual error (ξ) and the terminal information (i), the residual error (ξ) is a random frequency offset and the quantized frequency offset (q) Is determined by the difference between
The modem,
2K + 1 candidates are selected in the range [-R, R] where the residual error ξ can occur, and a second metric for the selected 2K + 1 candidates is selected.

And selecting a hypothesis H _{(q, i)} that maximizes and performing the frequency offset and terminal information detection based on the selected H _{(q, i)} . delete The method of claim 17,
The modem,
In the second metric

Select to

or

And selecting a hypothesis H _{(q, i)} that maximizes and performing the frequency offset and terminal information detection based on the selected H _{(q, i)} . The method of claim 19,
The modem,
A first error threshold associated with a detection error probability of the quantized frequency offset (QFO), a second error threshold associated with a terminal information detection error probability, a signal-to-interference ratio per symbol, and a multipath fading environment type. The terminal, characterized in that to select one of the F ₀ (q, i) or the F ₁ (q, i).

Images