阅读说明：本技术 Oqam/ofdm中的时频偏联合估计方法 (Time-frequency offset joint estimation method in OQAM/OFDM ) 是由 陈西宏 刘永进 刘赞 赵宇 谢泽东 张爽 于 2019-10-18 设计创作，主要内容包括：本发明提出了一种用于OQAM/OFDM系统的联合时频偏估计的方法,主要包括以下步骤：在将接收端的接收信号进行分段处理,检测其中具有共轭对称性的区域；计算统计量并与阈值进行比较；选择合适表达式对时偏进行估计；利用导频共轭对称性对频偏进行估计。本发明所提出的时频偏估计方法能够降低系统复杂度,节约频谱资源,实现较好的时频偏估计性能。(The invention provides a method for estimating joint time-frequency offset of an OQAM/OFDM system, which mainly comprises the following steps: carrying out segmentation processing on a received signal at a receiving end, and detecting a region with conjugate symmetry; calculating statistics and comparing with a threshold value; selecting a proper expression to estimate the time offset; and estimating the frequency offset by using the pilot frequency conjugate symmetry. The time frequency offset estimation method provided by the invention can reduce the complexity of the system, save frequency spectrum resources and realize better time frequency offset estimation performance.)

1. A time-frequency offset joint estimation method in OQAM/OFDM comprises the following steps:

step 1: to N_bSegmenting the transmission symbols; a. b and c represent the region a, respectively^*、b^*And c^*Conjugate symmetric regions, c represents regions that are not used, x represents regions that do not have conjugate symmetry;

step 2: defining a first M-dimensional matrix

And step 3: defining a third M-dimensional matrix

And 4, step 4: defining an M-dimensional matrix g_nRow m and column n

and 5: according to the nature of the prototype filter g (m) and

step 6: receiving signal of systemA sequence regarded as a matrix

and 7: computing statisticsWherein θ is time offset; in order to eliminate the influence of noise on the time offset estimation, a threshold value ζ is set;

and 8: estimating the time offset:

wherein

And step 9: receiving a pilot signal matrix

Technical Field

The invention relates to communication and information processing technologies, in particular to an OFDM/OQAM time-frequency offset joint estimation method.

Background

Compared with the traditional OFDM technology, the OQAM/OFDM technology adopts a prototype filter with good time-frequency focusing performance, and can obtain good performance of resisting interference between symbols and carriers without adding cyclic prefixes. The characteristics enable the OQAM/OFDM technology to have higher spectrum utilization efficiency and become one of the main candidate technologies of future mobile communication.

The structural composition and the working process of the OQAM/OFDM system are shown in figure 1, and mainly comprise the following steps:

the transmitting terminal carries out OQAM modulation on transmission data bits, adds known pilot symbols or sequences at corresponding positions of data blocks for channel estimation, converts frequency domain symbols into time domain signals through an IFFT module, and finally superposes and transmits the time domain signals after being formed through a synthesis filter.

The processing process of the receiving end is opposite to that of the sending end, firstly, the received signal passes through an analysis filter bank, then a frequency domain symbol is obtained through FFT, channel information is extracted by using a pilot frequency symbol or sequence, the influence caused by time frequency offset is eliminated, and finally, original data are obtained through OQAM demodulation.

It can be seen that time-frequency offset estimation is a key step for implementing reliable transmission in an OQAM/OFDM system, but due to the real-number domain orthogonality of the prototype filter, the wireless fading channel causes the OQAM/OFDM symbols to be inevitably interfered by adjacent symbols and subcarriers in the imaginary number domain, and such interference makes the channel estimation method of the conventional OFDM system no longer applicable.

Currently, two methods, namely a pilot-based estimation method and a blind estimation method, are mainly used as time-frequency offset estimation methods for an OQAM/OFDM system.

The former distributes pilot symbols on time frequency grid points according to a certain rule according to the coherence bandwidth and coherence time of a channel, obtains time offset estimation and coarse frequency offset estimation by calculating the cross-correlation function of the pilot symbols at a receiving end, and then performs fine frequency offset estimation by using the pilot symbols to obtain complete time frequency offset estimation. The method can utilize pilot symbols, and the complexity of the system is low. However, when this method is used, the frequency offset estimation needs two steps to be implemented, and the estimation result is affected by the pilot frequency structure. At the same time, the presence of pilots reduces the spectral efficiency of the system. How to improve the spectrum efficiency is a problem that the pilot frequency-based time frequency offset estimation method needs to be studied deeply.

The latter estimates the time frequency offset according to the self characteristics of the transmission symbol, such as cyclic symmetry, conjugate symmetry, etc., so that the system does not need to add other symbols when transmitting data, and thus, the frequency spectrum can be fully utilized to achieve the purpose of improving the frequency spectrum efficiency. However, this method needs to derive and prove the characteristic function of the system, and uses the time offset and the frequency offset as parameters. The derivation of the system characteristic function proves that the derivation is complex, so that the blind time-frequency offset estimation complexity is high.

It can be seen that the previously proposed channel estimation methods all have corresponding disadvantages: the pilot frequency-based mode depends on a pilot frequency structure, the frequency spectrum efficiency is low, and the complexity of the blind estimation-based method is high. Therefore, a channel estimation method that is compatible with both spectrum resource consumption and complexity is required.

Disclosure of Invention

Aiming at the problems in the prior art, the invention provides a time-frequency offset joint estimation method in OQAM/OFDM, which comprises the following steps:

step 1: to N_bSegmenting the transmission symbols; a. b and c represent the region a, respectively^*、b^*And c^*Conjugate symmetric regions, c represents regions that are not used, x represents regions that do not have conjugate symmetry;

step 2: defining a first M-dimensional matrix

Is arranged at the m-th row and the n-th column

M is in the range of {0, 1, …, M-1}, N is in the range of {0, 1, …, N }, and N is the number of matrix columns, wherein R, I is the real part and imaginary part corner mark respectively; defining a second M-dimensional matrix

Row m and column n

Is expressed as

Where j is an imaginary unitM', n are

The number of rows and columns of

Wherein IDFT [. C]For inverse discrete Fourier transform, w is an M-dimensional vector, and the M term is w_m＝j^m；

And step 3: defining a third M-dimensional matrix

Row m and column nIs expressed as

Wherein n' is

The column coordinates of (a) are,

is the sampling interval, T is the symbol interval, g (-) is the prototype filter function; order to

Then

Can be expressed as

And 4, step 4: defining an M-dimensional matrix g_nRow m and column n

Since the prototype filter is defined over the interval [0, KT) and K is an overlap factor, the matrix g is only if n ∈ {0, 1, …, K-1}, then_nThe significance is given; at this time，The same can be obtained

I is an imaginary part corner mark;

and 5: according to the nature of the prototype filter g (m) and

is delayed by the sampling of

Wherein

Represents

The first segment of the sample is taken,

represents

The second section of sampling, and so on; g_k，sAnd g_k，iEach represents g_kFirst one of (1)Sampling and second

Sampling;

and

respectively represent

First one of (1)Sampling and second

Sampling; when k is more than or equal to 5,

structure of and

the structure is the same;

step 6: receiving signal of system

A sequence regarded as a matrix

Detecting a region with conjugate symmetry of a received signal, namely whether positive and negative frequency amplitude components are symmetrical or not, and phase components of the positive and negative frequency amplitude components are just opposite;

and 7: computing statistics

Wherein θ is time offset; in order to eliminate the influence of noise on the time offset estimation, a threshold value ζ is set;

and 8: estimating the time offset:

wherein

And step 9: receiving a pilot signal matrix

Is a matrix formed by receiving pilot symbols according to a receiving time sequence; suppose thatFor two adjacent received pilot signal matrices, the system frequency offset is

The invention aims at an OQAM/OFDM system, combines a blind estimation method with pilot frequency, estimates time offset by using conjugate symmetry on the basis of realizing time offset blind estimation based on conjugate symmetry, and jointly estimates time offset in the OQAM/OFDM system on the basis, thereby reducing the complexity of the system and improving the spectrum utilization rate of the system. The method provided by the invention can simultaneously consider both the frequency spectrum resource consumption and the system performance.

Drawings

FIG. 1 shows a schematic diagram of the operation of an OQAM/OFDM system;

FIG. 2 shows a schematic diagram of a received signal segmentation method;

FIG. 3 is a graph showing a comparison of time-biased STO performance simulations in an exemplary embodiment of the present invention;

fig. 4 is a diagram illustrating simulation comparison of frequency offset CFO performance in an embodiment of the present invention.

Detailed Description

Specific embodiments of the present invention are given below with reference to the accompanying drawings. The parameters in the examples are merely to illustrate the invention and do not affect the generality of the invention.

To achieve the above object, a segmentation method of transmission symbols is first proposed, as shown in fig. 2. The time offset estimation is realized by proving the conjugate symmetry of each section, and on the basis, the frequency offset estimation is realized by using the pilot frequency information and the proved conjugate symmetry. The specific technical scheme comprises the following steps:

step 1: to N_bThe transmission symbols are segmented. a, b and c represent the region a^*b^*And c^*Conjugate pairThe term "region", c denotes the region that is not used, and x denotes the region that does not have conjugate symmetry.

Step 2: defining a first M-dimensional matrixIs arranged at the m-th row and the n-th column

M is in the range of {0, 1, …, M-1}, N is in the range of {0, 1, …, N }, and N is the number of matrix columns, wherein R, I is the real part and imaginary part corner mark respectively; defining a second M-dimensional matrix

Row m and column n

Is expressed as

Wherein j is an imaginary unit, m', n are

The number of rows and columns of

Wherein IDFT [. C]For inverse discrete Fourier transform, w is an M-dimensional vector, and the M term is w_m＝j^m。

And step 3: defining a third M-dimensional matrix

Row m and column n

Is expressed as

Wherein n' isThe column coordinates of (a) are,

is the sampling interval, T is the symbol interval, g (-) is the prototype filter function; order to

ThenCan be expressed as

And 4, step 4: defining an M-dimensional matrix g_nRow m and column nSince the prototype filter is defined over the interval [0, KT) (K is an overlap factor), the matrix g is only if n ∈ {0, 1, …, K-1}, then_nThe significance is given; at this time, the process of the present invention,the same can be obtained

I is an imaginary part corner mark;

and 5: according to the nature of the prototype filter g (m) and

is delayed by the sampling of

Wherein

Represents

The first segment of the sample is taken,

represents

The second section of sampling, and so on; g_k，sAnd g_k，lEach represents g_kFirst one of (1)

Sampling and second

Sampling;and

respectively representFirst one of (1)

Sampling and second

Sampling; when k is more than or equal to 5,

structure of and

the structure is the same;

step 6: receiving signal of system

A sequence regarded as a matrix

Detecting a region with conjugate symmetry of a received signal, namely whether positive and negative frequency amplitude components are symmetrical or not, and phase components of the positive and negative frequency amplitude components are just opposite;

and 7: computing statistics

Where θ is the time offset. In order to eliminate the influence of noise on the time offset estimation, a threshold value ζ is set;

and 8: estimating the time offset:

wherein

And step 9: receiving a pilot signal matrix

Is a matrix of received pilot symbols in accordance with the received timing. Suppose that

For two adjacent received pilot signal matrices, the system frequency offset is