CN109691050A

CN109691050A - A kind of ofdm system intermediate frequency bias estimation, apparatus and system

Info

Publication number: CN109691050A
Application number: CN201680089218.6A
Authority: CN
Inventors: 牛凯; 张进年
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2016-09-13
Filing date: 2016-09-13
Publication date: 2019-04-26
Anticipated expiration: 2036-09-13
Also published as: WO2018049544A1; CN109691050B

Abstract

The embodiment of the invention discloses a kind of ofdm system intermediate frequency bias estimations, apparatus and system, and by receiving orthogonal frequency-division multiplex singal, the orthogonal frequency-division multiplex singal includes zero subcarrier being periodically arranged in domain space；According to the control information of the orthogonal frequency-division multiplex singal, signal matrix is established, and is calculated eventually by the observation to the signal matrix and obtains the offset estimation value.The orthogonal frequency-division multiplex singal is the signal separated by zero subcarrier group periodicity pectination, the interference between subcarrier is effectively reduced, a large amount of subcarriers can be included in the data area of offset estimation, expand observation dimension, and then calculated by the observation of big data quantity, guarantee frequency offset estimation accuracy；Moreover, using the method successively observed, every layer of sight gauge calculator has reasonable data processing complexity in being observed calculating, and successively approach high precision computation as a result, have both it is high-precision simultaneously, there is very high practicability.

Description

Method, device and system for estimating frequency offset in OFDM system

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a method, an apparatus, and a system for estimating an offset in an OFDM system.

Background

The OFDM (Orthogonal Frequency Division Multiplexing) technology not only can overcome intersymbol interference caused by multipath effects by using a set of Orthogonal narrowband subcarriers to multiplex and transmit service data, but also can support high-speed data services without requiring complex channel equalization, and thus is widely applied to wired and wireless communication systems, such as DAB (Digital Audio Broadcasting) standard, DVB (Digital Video Broadcasting) standard, IEEE (Institute of Electrical and Electronics Engineers) 802.11, 3GPP (The 3rd Generation Partner Project, 3rd Generation mobile communication cooperation plan), LTE (Long Term Evolution of universal mobile communication technology), and The like. However, the accuracy and doppler shift of the oscillator cause the carrier of the received signal and the carrier of the transmitted signal to be not completely synchronized, and there is a frequency shift, i.e., frequency offset; the presence of frequency offset not only causes inter-subcarrier interference to a single-channel data stream, but also causes non-synchronization of multiple-channel data streams, and direct combination of the multiple-channel data streams can cause reduction of a received signal-to-noise ratio and deterioration of system performance. Therefore, high-precision frequency offset estimation is key to guarantee the accuracy of frequency offset compensation and ensure the communication quality.

In order to implement frequency offset estimation, a frequency offset estimation method based on a pilot frequency is commonly adopted at present, a mapping pattern of a downlink reference signal is specified in a 3GPP TS (3GPP Technical Specification )36.211 communication protocol, a pilot frequency subcarrier is defined by a fixed subcarrier position in the mapping pattern, and since a channel is all slowly changed in time, a frequency offset value can be obtained by utilizing correlation between pilot frequency subcarriers at the same position in different OFDM symbols. However, research shows that the frequency offset estimation method based on the pilot frequency only utilizes the pilot frequency subcarriers to perform frequency offset estimation, for example, under the condition that the bandwidth is 20MHz and a single antenna is adopted, the number of effective subcarriers is 1200 for one OFDM symbol, wherein the number of the pilot frequency subcarriers is 200, and the frequency offset estimation based on the pilot frequency subcarriers mainly uses a linear estimation method and an ML (Maximum Likelihood) method; due to the inter-subcarrier interference caused by frequency offset, the accuracy of the linear estimation method tends to a fixed value along with the increase of the signal-to-noise ratio, and under the condition of high signal-to-noise ratio, the frequency offset estimation accuracy is limited so that the accuracy requirement cannot be further met; the ML method has high algorithm complexity and poor practicability, so the frequency offset estimation method based on pilot frequency cannot meet the requirements of frequency offset estimation precision and practicability.

Disclosure of Invention

In order to overcome the problems in the prior art, the invention provides a frequency offset estimation method, device and system in an OFDM system, so as to solve the technical problems of low frequency offset estimation precision and poor practicability in the prior art.

In one aspect, an embodiment of the present invention provides a method for estimating frequency offset in an OFDM system, where the method includes:

receiving an orthogonal frequency division multiplexing signal, wherein the orthogonal frequency division multiplexing signal comprises a plurality of subcarrier sets, each subcarrier set comprises at least one zero subcarrier and a plurality of data subcarriers, and the zero subcarrier positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group, and data subcarriers with the same position in all subcarrier sets form a data subcarrier group;

acquiring control information of the OFDM signal, wherein the control information comprises information of a subcarrier group, position information of a zero subcarrier group and position information of a data subcarrier group;

extracting a subcarrier signal from the orthogonal frequency division multiplexing signal according to the control information to generate a signal matrix;

and observing the signal matrix to obtain a frequency offset estimation value of the orthogonal frequency division multiplexing signal.

In one possible design, the observing the signal matrix to obtain the frequency offset estimation value of the ofdm signal includes:

establishing an observation matrix from the signal matrix according to the equivalent frequency offset range and the sampling number;

carrying out observation iterative computation on the observation matrix to obtain an observation iterative computation result, wherein the dimension of the observation iterative computation result is equal to the column number of the observation matrix;

obtaining an equivalent frequency offset result vector from the observation matrix according to an observation iterative computation result;

and calculating a frequency offset estimation value according to the equivalent frequency offset result vector.

In one possible design, observing the signal matrix to obtain a frequency offset estimation value of the ofdm signal includes:

according to the number of the first layer of samples, carrying out equivalent frequency offset sampling in a first layer of equivalent frequency offset range, and establishing a first layer of observation matrix from the signal matrix according to the sampled equivalent frequency offset;

carrying out observation iterative computation on the first layer of observation matrix and obtaining a first layer of observation iterative computation result;

obtaining a first layer of equivalent frequency offset result vectors from the first layer of observation matrixes according to the first layer of observation iterative computation results;

and calculating a frequency offset estimation value according to the first layer equivalent frequency offset result vector.

according to the sampling number of the nth layer, carrying out equivalent frequency offset sampling in the equivalent frequency offset range of the nth layer, and establishing an nth layer observation matrix from the signal matrix according to the sampled equivalent frequency offset, wherein n is greater than or equal to 2;

carrying out observation iterative computation on the nth layer of observation matrix to obtain an nth layer of observation iterative computation result;

obtaining an nth layer equivalent frequency offset result vector from the nth layer observation matrix according to the nth layer observation iterative computation result;

and calculating a frequency offset estimation value according to the final layer of equivalent frequency offset result vector.

In one possible design, the observation iteration calculation includes:

establishing a first sparse hyperparameter vector;

obtaining a diagonal matrix according to the first sparse hyperparametric vector;

calculating a first posterior parameter (Mm) and a second posterior parameter (Sigma) according to the diagonal matrix;

obtaining a second sparse hyperparameter vector according to the first posterior parameter (M) and the second posterior parameter (Sigma);

judging whether the difference value of the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a preset convergence threshold value;

if the difference value between the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a convergence threshold value, the observation iterative computation is completed, and the second sparse hyperparameter vector is used as an observation iterative computation result; alternatively, the first and second electrodes may be,

and if the difference value of the second sparse hyperparameter vector and the first sparse hyperparameter vector is larger than or equal to a convergence threshold value, taking the second sparse hyperparameter vector as the first sparse hyperparameter vector, and recovering the second sparse hyperparameter vector according to the first posterior parameter (M) and the second posterior parameter (sigma).

In one possible design, calculating the frequency offset estimation value according to the equivalent frequency offset result vector includes:

extracting an equivalent frequency offset value corresponding to each data subcarrier group from the equivalent frequency offset result vector;

calculating a group frequency offset estimation value of each data subcarrier group according to the equivalent frequency offset value;

and carrying out average calculation on the group of frequency deviation estimated values to obtain the frequency deviation estimated values.

In one possible design, the performing equivalent frequency offset sampling includes:

determining the first layer equivalent frequency deviation range corresponding to each data subcarrier group according to the number of the total subcarrier groups, wherein the formula is as follows:

k is 0, 1, …, K-1, where K is the first layer equivalent frequency offset value corresponding to the kth data subcarrier group, K is the number of data subcarrier groups, and Q is the number of total subcarrier groups;

and carrying out equivalent frequency offset sampling on the first layer of equivalent frequency offset range corresponding to each data subcarrier group according to the first layer of sampling number respectively.

calculating the equivalent frequency offset range of the nth layer corresponding to each data subcarrier group according to the equivalent frequency offset vector of the nth-1 layer, the sampling number of the nth-1 layer and the scaling factor, wherein the formula is as follows:

the equivalent frequency offset value corresponding to the kth data subcarrier group in the (n-1) th layer equivalent frequency offset result vector is the equivalent frequency offset value corresponding to the kth data subcarrier group in the nth layer, S is a scaling factor, Q is the number of the total subcarrier groups and is the sampling number of the (n-1) th layer of the k data subcarrier group;

and carrying out equivalent frequency offset sampling on the nth layer of equivalent frequency offset range corresponding to each data subcarrier group according to the nth layer of sampling number respectively.

In one possible design, the method further includes:

setting a first discrete error according to a first expected value of the equivalent frequency offset estimation error;

and calculating the number of first layer samples according to the first discrete error and the grouping information of the orthogonal frequency division multiplexing signal.

In one possible design, the method further includes:

setting a second discrete error according to a second expected value of the equivalent frequency offset estimation error;

calculating the scaling factor and the n-th layer sample number according to a formula, wherein delta₂For the second dispersion error, S is the scaling factor, N is the number of observed calculation layers, and K is the numberThe number of data subcarrier groups, Q is the total number of subcarrier groups, J⁽ⁿ⁾Is the nth layer sample number.

In one possible design, the obtaining an equivalent frequency offset result vector from the observation matrix according to the observation iterative computation result includes:

determining K maximum peak positions in the observation iteration calculation result, wherein K is the number of data subcarrier groups;

extracting a corresponding observation matrix column vector from the observation matrix according to the maximum peak position;

and extracting equivalent frequency offset values corresponding to the column vectors of the observation matrix, and establishing equivalent frequency offset result vectors.

In one possible design, the determining K maximum peak locations within the observation iteration calculation includes:

dividing elements in the observation iteration calculation result into a plurality of calculation result groups according to the sampling number, wherein the number of the calculation result groups is equal to the number of the data subcarrier groups;

finding a maximum peak within each of said sets of calculations;

and taking the position of the maximum peak in the observation iteration calculation result as the maximum peak position.

In another aspect, an embodiment of the present invention provides a method for estimating frequency offset in an OFDM system, where the method includes:

dividing the subcarriers into a plurality of subcarrier sets in a frequency domain space;

zero subcarriers are arranged at least one zero setting position of each subcarrier set, and the zero setting positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group;

setting data subcarriers at subcarrier positions other than the set-zero position in each subcarrier set; all data subcarriers with the same position in the subcarrier set form a data subcarrier group;

and generating and sending an orthogonal frequency offset multiplexing signal according to the zero subcarrier group and the data subcarrier group.

In another aspect, an embodiment of the present invention provides an apparatus for estimating frequency offset in an OFDM system, including:

the OFDM signal receiving module is used for receiving OFDM signals, the OFDM signals comprise a plurality of subcarrier sets, each subcarrier set comprises at least one zero subcarrier and a plurality of data subcarriers, and the zero subcarrier positions in each subcarrier set are the same; the zero subcarriers with the same position in all the subcarrier sets form a zero subcarrier group, and the data subcarriers with the same position in all the subcarrier sets form a data subcarrier group;

a control information obtaining module, configured to obtain control information of the ofdm signal, where the control information includes: information of subcarrier groups, position information of zero subcarrier groups, and position information of data subcarrier groups;

a signal matrix generating module, configured to extract a subcarrier signal from the ofdm signal according to the control information from the control information acquiring module, and generate a signal matrix;

and the frequency offset estimation value acquisition module is used for observing the signal matrix to obtain the frequency offset estimation value of the orthogonal frequency division multiplexing signal.

In one possible design, the frequency offset estimation obtaining module includes:

the observation matrix establishing module is used for establishing an observation matrix from the signal matrix according to the equivalent frequency offset range and the sampling number;

the observation iterative computation module is used for performing observation iterative computation on the observation matrix and obtaining an observation iterative computation result, and the dimensionality of the observation iterative computation result is equal to the column number of the observation matrix;

the equivalent frequency offset result acquisition module is used for acquiring an equivalent frequency offset result vector from an observation matrix according to the observation iterative computation result acquired by the observation iterative computation module;

and the frequency offset estimation value calculation module is used for calculating the frequency offset estimation value according to the equivalent frequency offset result vector.

the first layer observation matrix obtaining module is used for carrying out equivalent frequency offset sampling in a first layer equivalent frequency offset range according to the first layer sampling number and establishing a first layer observation matrix from the signal matrix according to the sampled equivalent frequency offset;

the first layer observation iterative computation module is used for performing observation iterative computation on the first layer observation matrix and obtaining a first layer observation iterative computation result;

a first layer equivalent frequency offset result obtaining module, configured to obtain a first layer equivalent frequency offset result vector from the first layer observation matrix according to the first layer observation iterative computation result;

and the first layer frequency offset estimation value calculation module is used for calculating a frequency offset estimation value according to the first layer equivalent frequency offset result vector.

the nth layer observation matrix establishing module is used for carrying out equivalent frequency offset sampling in the nth layer equivalent frequency offset range according to the nth layer sampling number, and establishing an nth layer observation matrix from the signal matrix according to the sampled equivalent frequency offset, wherein n is greater than or equal to 2;

the nth layer observation iterative computation module is used for performing observation iterative computation on the signal matrix and obtaining an nth layer observation iterative computation result;

an nth layer equivalent frequency offset result vector obtaining module, configured to obtain an nth layer equivalent frequency offset result vector from the nth layer observation matrix according to the nth layer observation iterative computation result;

and the final layer frequency offset estimation value calculation module is used for calculating a frequency offset estimation value according to the final layer equivalent frequency offset result vector.

In one possible design, the observation iteration calculation module includes:

the first sparse hyperparameter vector establishing module is used for establishing a first sparse hyperparameter vector;

a diagonal matrix obtaining module, configured to obtain a diagonal matrix according to the first sparse hyperparametric vector;

the posterior parameter calculation module is used for calculating a first posterior parameter (Mm) and a second posterior parameter (Sigma) according to the diagonal matrix;

a second sparse hyperparameter vector obtaining module, configured to obtain a second sparse hyperparameter vector according to the first posterior parameter m and the second posterior parameter Σ;

a convergence judging module for judging whether the difference value of the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a preset convergence threshold value;

an observation iteration calculation ending module, configured to, according to the determination result of the convergence determination module, complete the observation iteration calculation if a difference between the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a convergence threshold, and use the second sparse hyperparameter vector as an observation iteration calculation result;

and the observation iteration calculation circulating module is used for taking the current second sparse hyperparameter vector as the first sparse hyperparameter vector if the difference value of the second sparse hyperparameter vector and the first sparse hyperparameter vector is greater than or equal to a convergence threshold value according to the judgment result of the convergence judgment module, and obtaining the second sparse hyperparameter vector again according to the first posterior parameter M and the second posterior parameter sigma.

In one possible design, the frequency offset estimation calculation module includes:

an equivalent frequency offset value extraction module, configured to extract an equivalent frequency offset value corresponding to each data subcarrier group from the equivalent frequency offset result vector;

the group frequency offset estimation value calculation module is used for calculating the group frequency offset estimation value of each data subcarrier group according to the equivalent frequency offset value;

and the group frequency offset estimation value averaging module is used for carrying out average calculation on the group frequency offset estimation value to obtain the frequency offset estimation value.

In one possible design, the first-layer observation matrix obtaining module further includes:

a first layer equivalent frequency offset range determining module, configured to determine, according to the number of total subcarrier groups, a first layer equivalent frequency offset range corresponding to each data subcarrier group, where the formula is as follows:

and the first layer equivalent frequency offset sampling module is used for carrying out equivalent frequency offset sampling on the first layer equivalent frequency offset range corresponding to each data subcarrier group according to the first layer sampling number respectively.

In one possible design, the nth layer observation matrix creating module further includes:

the nth layer equivalent frequency offset range determining module is used for calculating the nth layer equivalent frequency offset ranges respectively corresponding to each data subcarrier group according to the nth-1 layer equivalent frequency offset vector, the nth-1 layer sampling number and the scaling factor, and the formula is as follows:

and the nth layer equivalent frequency offset sampling module is used for carrying out equivalent frequency offset sampling on the nth layer equivalent frequency offset range corresponding to each data subcarrier group according to the nth layer sampling number respectively.

In one possible design, the apparatus further includes a first layer sample number determination module, where the first layer sample number determination module includes:

the first discrete error setting module is used for setting a first discrete error according to a first expected value of the equivalent frequency offset estimation error;

and the first layer sampling number calculating module is used for calculating the first layer sampling number according to the first discrete error and the grouping information of the orthogonal frequency division multiplexing signal.

In one possible design, the apparatus further includes a precision parameter determination module, where the precision parameter determination module includes:

the second discrete error setting module is used for setting a second discrete error according to a second expected value of the equivalent frequency offset estimation error;

a precision parameter determination module for calculating the scaling factor, the n-th layer sampling number and the observation calculation layer number according to a formula, wherein delta₂For the second dispersion error, S is the scaling factor, N is the number of observation calculation layers, K is the number of data subcarrier groups, Q is the number of total subcarrier groups, J⁽ⁿ⁾Is the nth layer sample number.

In one possible design, the equivalent frequency offset result obtaining module includes:

a maximum peak position determining module, configured to determine K maximum peak positions in the observation iteration calculation result, where K is the number of data subcarrier groups;

the observation matrix column vector extraction module is used for extracting a corresponding observation matrix column vector from the observation matrix according to the maximum peak position;

and the equivalent frequency offset result vector establishing module is used for establishing an equivalent frequency offset result vector according to the data subcarrier group sequence by using the equivalent frequency offset value corresponding to the observation matrix column vector.

In one possible design, the maximum peak position determination module includes:

the calculation result grouping module is used for dividing the observation iteration calculation result into a plurality of calculation result groups according to the sampling number, and the number of the calculation result groups is equal to the number of the data subcarrier groups;

a group maximum peak determination module for finding a maximum peak in each of said calculation result groups;

and the group maximum peak position acquisition module is used for taking the position of the maximum peak in the observation iteration calculation result as the maximum peak position.

the subcarrier diversity module is used for dividing the subcarriers into a plurality of subcarrier sets in a frequency domain space;

a zero subcarrier setting module, configured to set a zero subcarrier at least one zero position of each subcarrier set, where the zero positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group;

the data subcarrier setting module is used for setting data subcarriers at subcarrier positions except the position where each subcarrier is set to be zero in a centralized manner; all data subcarriers with the same position in the subcarrier set form a data subcarrier group;

and the signal management module is used for generating and sending an orthogonal frequency offset multiplexing signal according to the zero subcarrier group and the data subcarrier group.

In another aspect, an embodiment of the present invention provides an offset estimation system in an OFDM system, where the system includes the terminal capable of acquiring a target resource pool and the control device.

In yet another aspect, an embodiment of the present invention provides a communication system, including a first device and a second device, wherein:

the first device is configured to receive an orthogonal frequency division multiplexing signal from a second device, where the orthogonal frequency division multiplexing signal includes multiple subcarrier sets, each subcarrier set includes at least one zero subcarrier and multiple data subcarriers, and the zero subcarriers in each subcarrier set have the same position; the zero subcarriers with the same position in all the subcarrier sets form a zero subcarrier group, and the data subcarriers with the same position in all the subcarrier sets form a data subcarrier group; acquiring control information of the OFDM signal, wherein the control information comprises information of a subcarrier group, position information of a zero subcarrier group and position information of a data subcarrier group; extracting a subcarrier signal from the orthogonal frequency division multiplexing signal according to the control information to generate a signal matrix; observing the signal matrix to obtain a frequency offset estimation value of the orthogonal frequency division multiplexing signal;

the second device is configured to divide the subcarriers into a plurality of subcarrier sets in a frequency domain space; zero subcarriers are arranged at least one zero setting position of each subcarrier set, and the zero setting positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group; setting data subcarriers at subcarrier positions other than the set-zero position in each subcarrier set; all data subcarriers with the same position in the subcarrier set form a data subcarrier group; and generating an orthogonal frequency offset multiplexing signal according to the zero subcarrier group and the data subcarrier group and sending the orthogonal frequency offset multiplexing signal to the first equipment.

The technical scheme provided by the embodiment of the invention can have the following beneficial effects:

according to the method, the device and the system for estimating the frequency offset in the OFDM system, provided by the embodiment of the invention, through receiving an orthogonal frequency division multiplexing signal, the orthogonal frequency division multiplexing signal comprises zero subcarriers periodically arranged on a frequency domain space; and establishing a signal matrix according to the control information of the orthogonal frequency division multiplexing signal, and finally obtaining the frequency offset estimation value through observation calculation of the signal matrix. The periodically arranged zero subcarriers effectively reduce the interference among subcarriers and improve the frequency offset estimation precision; moreover, all subcarriers in the orthogonal frequency division multiplexing signal are brought into an observation range, observation dimensionality is enlarged, and frequency offset estimation precision and practicability are further improved by utilizing sparsity layer by layer observation calculation of equivalent frequency offset values.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.

Fig. 1 is a schematic structural diagram of an OFDM system provided in the present invention;

fig. 2 is a schematic flow chart of a frequency offset estimation method in an OFDM system according to a first embodiment of the present invention;

FIG. 3 is a schematic diagram of a single antenna OFDM signal transmission pattern according to the present invention;

fig. 4 is a schematic diagram of a dual-antenna ofdm signal transmission pattern according to the present invention;

fig. 5 is a flowchart illustrating a frequency offset estimation method in an OFDM system according to a second embodiment of the present invention;

FIG. 6 is a schematic flow chart of an observation iterative calculation method according to the present invention;

FIG. 7 is a flowchart illustrating a first embodiment of a method for extracting an equivalent frequency offset result vector according to the present invention;

FIG. 8 is a flowchart illustrating a second embodiment of an equivalent frequency offset result vector extracting method according to the present invention;

FIG. 9 is a flowchart illustrating a method for calculating a frequency offset estimation value according to the present invention;

fig. 10 is a flowchart illustrating a frequency offset estimation method in an OFDM system according to a third embodiment of the present invention;

fig. 11 is a schematic flow chart of a first layer sampling number determining method provided in the present invention;

fig. 12 is a flowchart illustrating a fourth embodiment of a method for estimating frequency offset in an OFDM system according to the present invention;

fig. 13 is a flowchart illustrating a fifth embodiment of a frequency offset estimation method in an OFDM system according to the present invention;

fig. 14 is a schematic structural diagram of a frequency offset estimation apparatus in an OFDM system according to a first embodiment of the present invention;

fig. 15 is a schematic structural diagram of a frequency offset estimation apparatus in a second embodiment of an OFDM system according to the present invention;

FIG. 16 is a schematic structural diagram of an observation iteration calculation module provided by the present invention;

fig. 17 is a schematic structural diagram of a first embodiment of an equivalent frequency offset result obtaining module according to the present invention;

fig. 18 is a schematic structural diagram of an equivalent frequency offset result obtaining module according to a second embodiment of the present invention;

FIG. 19 is a block diagram illustrating a frequency offset estimation calculation module according to the present invention;

fig. 20 is a schematic structural diagram of a frequency offset estimation apparatus in a OFDM system according to a third embodiment of the present invention;

FIG. 21 is a schematic structural diagram of a first-layer observation matrix obtaining module according to the present invention;

FIG. 22 is a block diagram illustrating a first layer sampling number determination module according to the present invention;

fig. 23 is a schematic structural diagram of a frequency offset estimation apparatus in an OFDM system according to a fourth embodiment of the present invention;

FIG. 24 is a schematic structural diagram of an nth layer observation matrix building block according to the present invention;

FIG. 25 is a schematic structural diagram of a precision parameter determination module according to the present invention;

fig. 26 is a schematic structural diagram of a frequency offset estimation apparatus in an OFDM system according to a fifth embodiment of the present invention;

fig. 27 is a schematic structural diagram of an offset estimation system in an OFDM system according to the present invention.

Detailed Description

In order to make those skilled in the art better understand the technical solution of the present invention, the technical solution in the embodiment of the present invention will be clearly and completely described below with reference to the drawings in the embodiment of the present invention, and it is obvious that the described embodiment is only a part of the embodiment of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Before describing the embodiments of the frequency offset estimation method in the OFDM system of the present invention, the structure of the OFDM system of the embodiments of the present invention is introduced first. Referring to fig. 1, a schematic structural diagram of an OFDM system provided in the present invention is shown.

The OFDM system is based on a C-RAN (C-Radio Access Network, cloud type wireless Access Network) architecture. The C-RAN is a green Radio access network architecture (Clean system) based on Centralized Processing, cooperative Radio (cooperative Radio) and Real-time Cloud Infrastructure (Real-time Cloud Infrastructure). The essential of the method is that resource sharing and dynamic scheduling are realized by reducing the number of base station rooms and energy consumption and adopting cooperative and virtualized technologies, so that the spectrum efficiency is improved, and the operation with low cost, high bandwidth and flexibility is achieved. In fig. 1, a BBU (Building base band Unit, indoor Baseband processing Unit) in a base station is centrally located, the BBU is connected to an RRH (Radio Remote Unit) through FrontHaul, and the RRH transmits an orthogonal frequency division multiplexing signal to a UE (User Equipment). The C-RAN network can enhance the coverage rate, improve the service quality of cell edge users and simultaneously improve the spectrum utilization rate and the system throughput by deploying a large number of RRHs, and in a downlink, UE can increase the receiving signal-to-noise ratio and obtain diversity gain by receiving and combining the transmitted data of RRH1, RRH2 and RRH3 participating in cooperative transmission, so that the effectiveness and the reliability of transmission are improved. Meanwhile, in the uplink, the UE transmits data to the BBU through one or more of RRH1, RRH2, and RRH 3.

Referring to fig. 2, which is a flowchart illustrating a first embodiment of a frequency offset estimation method in an OFDM system according to the present invention, in a downlink, a UE receives an orthogonal frequency division multiplexing signal and performs frequency offset estimation; in an uplink, a BBU in a base station receives an orthogonal frequency division multiplexing signal and performs frequency offset estimation. In the embodiment of the present invention, detailed description is given in the case of downlink, and the frequency offset estimation method includes the following steps:

step S101: receiving an orthogonal frequency division multiplexing signal, wherein the orthogonal frequency division multiplexing signal comprises a plurality of subcarrier sets, each subcarrier set comprises at least one zero subcarrier and a plurality of data subcarriers, and the zero subcarrier positions in each subcarrier set are the same; the zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group, and the data subcarriers with the same position in all subcarrier sets form a data subcarrier group.

In the downlink, the BBU may send orthogonal frequency division multiplexing signals to the UE using one or more RRHs.

When the BBU uses one RRH to send an orthogonal frequency division multiplexing signal to the UE, refer to fig. 3, which is a schematic diagram of a single-antenna orthogonal frequency division multiplexing signal transmission pattern provided by the present invention. In fig. 3, N subcarriers are included in 1 ofdm signal, and P subcarrier sets are provided in the N subcarriers, each subcarrier set including Q subcarriers, where N is P × Q. In a specific embodiment, for example, in the case of a 20MHz bandwidth, 1 ofdm signal includes 2048 total subcarriers, where there are 1200 active subcarriers, i.e., N — 1200, and if Q is set to 16, the ofdm signal includes P — 75 subcarrier sets; in fig. 3, the 1 st subcarrier, the 2 nd subcarrier, the 3rd subcarrier to the 16 th subcarrier, which are Q-16 subcarriers, constitute a 1 st subcarrier set, the 17 th subcarrier, the 18 th subcarrier, the 19 th subcarrier to the 32 th subcarrier, which are Q-16 subcarriers, constitute a 2 nd subcarrier set, and P-75 subcarrier sets are sequentially constituted according to the above rule. Moreover, subcarriers at the same position in each subcarrier set form a subcarrier group, for example, the 1 st subcarrier, the 17 th subcarrier and other subcarriers with a period of Q ═ 16 form a 1 st subcarrier group, the 2 nd subcarrier, the 18 th subcarrier and other subcarriers with a period of Q ═ 16 form a 2 nd subcarrier group, and in turn form a total of Q ═ 16 subcarrier groups, which may be understood as subchannels, and users may occupy one or more of the subchannels for transmitting data. It should be noted that the above embodiments are only exemplary, and Q may be set to any other value, such as 32, 64, etc., and P may be set accordingly.

Each subcarrier set comprises at least one zero subcarrier and a plurality of data subcarriers, and when the subcarrier set comprises one zero subcarrier, referring to fig. 3 as well, in a specific embodiment, each subcarrier set comprises 1 zero subcarrier and 15 data subcarriers; the 16 th subcarrier in the 1 st subcarrier set is a zero subcarrier, that is, the 16 th subcarrier in the whole frequency domain space is a zero subcarrier, the 16 th subcarrier in the 2 nd subcarrier set is a zero subcarrier, that is, the 32 th subcarrier in the whole frequency domain space is a zero subcarrier, and accordingly, the 16 th subcarrier in each subcarrier set is a zero subcarrier; therefore, the zero subcarriers are located at the same position in each subcarrier set and are arranged at intervals of a period Q of 16 in the frequency domain space of 1 ofdm symbol, and P of 75 zero subcarriers form one zero subcarrier group. For the data subcarriers, the other 15 subcarriers except the 16 th subcarrier in the 1 st subcarrier set are data subcarriers, that is, the 1 st subcarrier to the 15 th subcarrier are all data subcarriers, and the other 15 subcarriers except the 32 th data subcarrier in the 2 nd subcarrier set are also data subcarriers, that is, the 17 th subcarrier to the 32 th subcarrier are all data subcarriers; the data subcarriers at the same position in each subcarrier set form a data subcarrier group, for example, the 1 st subcarrier belongs to the 1 st subcarrier set, the 17 th subcarrier belongs to the 2 nd subcarrier set, the 1 st subcarrier is located at the first position in the 1 st subcarrier set, and the 17 th subcarrier is also located at the first position in the 2 nd subcarrier set, so that the 1 st subcarrier, the 17 th subcarrier and other subcarriers sequentially spaced by the period Q ═ 16 form the 1 st data subcarrier group; similarly, the 2 nd subcarrier, the 18 th subcarrier, and other data subcarriers sequentially spaced at a period of Q16 constitute a 2 nd data subcarrier group, and a total of 15 data subcarrier groups are configured according to the above rule. Certainly, in specific implementation, the subcarriers at other positions in the subcarrier set may be set as zero subcarriers, and through the above description, in the frequency domain space of 1 ofdm signal, the zero subcarriers are distributed with Q as a period and are interleaved with the data subcarriers to form an ofdm signal with a comb structure.

When the subcarrier set includes a plurality of zero subcarriers, that is, 1 ofdm signal includes a plurality of zero subcarrier groups, the zero subcarrier groups may be a continuous zero subcarrier group or a scattered zero subcarrier group. For the continuous zero subcarrier group, for example, when Q is 16, 3 continuous zero subcarrier groups are provided, the 7 th subcarrier, the 23 th subcarrier and other subcarriers spaced at a period of Q16 may be set as zero subcarriers to form a 1 st zero subcarrier group, the 8 th subcarrier, the 24 th subcarrier and other subcarriers spaced at a period of Q16 may be set as zero subcarriers to form a 2 nd zero subcarrier group, the 9 th subcarrier, the 25 th subcarrier and other subcarriers spaced at a period of Q16 may be set as zero subcarriers to form a 3rd zero subcarrier group, and the zero subcarriers in the 3 th zero subcarrier groups are continuously distributed in one subcarrier set. For the distributed zero subcarrier group, for example, in the case of Q16, 2 distributed zero subcarrier groups are provided, the 3rd subcarrier, the 19 th subcarrier and other subcarriers periodically spaced with Q16 may be set as the 1 st zero subcarrier group, the 14 th subcarrier, the 30 th subcarrier and other subcarriers periodically spaced with Q16 may be set as the 2 nd zero subcarrier group, the zero subcarriers in the 2 th zero subcarrier groups are mutually spaced and distributed in one subcarrier set, and the interval of the zero subcarrier in each zero subcarrier group in the same subcarrier set may be flexibly set to any value within 1 to Q-1. In addition, in a specific implementation, the zero subcarrier group may be set according to the strength of the subcarrier group signal, for example, subcarrier 1, subcarrier 17, and subcarrier group signals corresponding to subcarriers with a Q equal to 16 period interval are weak, and the subcarrier group with the weak signal may be preferentially set as the zero subcarrier group, so as to improve the utilization efficiency of subcarriers and ensure communication quality.

In the 3GPP TS 36.211 communication protocol, a downlink transmission pattern of an orthogonal frequency division multiplexing signal is defined, and also referring to fig. 3, in the case of a single antenna, subcarriers 6, 12, 18, and 24, and other subcarriers with a period interval of 6 are pilot subcarriers according to the communication protocol. In order to ensure the compatibility with the original communication protocol, one or more zero subcarriers and a plurality of data subcarriers can be arranged at other positions except for pilot subcarriers in a subcarrier set; specifically, when the zero subcarrier position coincides with the pilot subcarrier position, a zero subcarrier is set at the position, for example, the 12 th subcarrier is a pilot subcarrier, and if the zero subcarrier is set at the 12 th subcarrier position, the 12 th subcarrier is set to be a zero subcarrier; setting data subcarriers at positions other than the pilot subcarriers and the zero subcarriers; the process of constructing the zero subcarrier group and the data subcarrier group may refer to the above description, and is not described herein again.

In the case of a single antenna, in the frequency domain space of the ofdm signal, the zero subcarrier groups periodically separate the subcarriers to form a comb-shaped transmission pattern, which has high flexibility, and the ofdm signal of the comb-shaped transmission pattern has strong compatibility with the transmission signal specified by the communication protocol.

When the BBU uses 2 RRHs to send an orthogonal frequency division multiplexing signal to the UE, refer to fig. 4, which is a schematic diagram of a dual-antenna orthogonal frequency division multiplexing signal transmission pattern provided by the present invention. Taking a 20MHz bandwidth as an example, when Q is 16, according to the description of a single-antenna ofdm signal, the ofdm signal corresponding to the 1 st antenna includes N2048 subcarriers, has a P128 subcarrier set, and is divided into Q16 subcarrier groups; the ofdm signal corresponding to the 2 nd antenna includes N2048 subcarriers, has P128 subcarrier sets, and is divided into Q16 subcarrier groups.

The zero subcarrier group corresponding to the 1 st antenna and the zero subcarrier group corresponding to the 2 nd antenna are the same in number and distribution. When 1 zero subcarrier group is set, for example, the ofdm signal of the 1 st antenna sets the 5 th subcarrier group as the zero subcarrier group, and similarly, the ofdm signal of the 2 nd antenna also sets the 5 th subcarrier group as the zero subcarrier group. When a plurality of zero subcarrier groups are set, for consecutive zero subcarrier groups, for example, a 3rd subcarrier group, a 4 th subcarrier group and a 5 th subcarrier group which are consecutive in the orthogonal frequency division multiplexing signal of the 1 st antenna are set as the zero subcarrier group, and a 3rd subcarrier group, a 4 th subcarrier group and a 5 th subcarrier group which are consecutive in the orthogonal frequency division multiplexing signal of the 2 nd antenna are also set as the zero subcarrier group; for the scattered zero subcarrier group, for example, the scattered 1 st subcarrier group and 7 th subcarrier group are set as a zero subcarrier group in the orthogonal frequency division multiplexing signal of the 1 st antenna, and the scattered 1 st subcarrier group and 7 th subcarrier group are also set as a zero subcarrier group in the orthogonal frequency division multiplexing signal of the 2 nd antenna.

In the 3GPP TS 36.211 communication protocol, a dual-antenna ofdm signal transmission pattern is also defined, and in order to satisfy that the pilot subcarriers of the 1 st antenna and the 2 nd antenna satisfy orthogonality, in the ofdm signal corresponding to the 1 st antenna, the 6 th subcarrier, the 12 th subcarrier, the 18 th subcarrier, and other subcarriers with a period interval of 6 are used pilot subcarriers, and the 3rd subcarrier, the 9 th subcarrier, the 15 th subcarrier, and other subcarriers with a period interval of 6 are unused pilot subcarriers; in the orthogonal frequency division multiplexing signal corresponding to the 2 nd antenna, the 3rd subcarrier, the 9 th subcarrier, the 15 th subcarrier and other subcarriers with a period interval of 6 are used pilot subcarriers, and the 6 th subcarrier, the 12 th subcarrier, the 18 th subcarrier and other subcarriers with a period interval of 6 are unused pilot subcarriers. When the zero subcarrier is overlapped with the unused pilot subcarrier or the used pilot subcarrier, setting the subcarrier at the corresponding position as the zero subcarrier, for example, in the orthogonal frequency division multiplexing signal of the 1 st antenna, the 48 th subcarrier is the used pilot subcarrier, if the 16 th subcarrier group is set as the zero subcarrier group, the 48 th subcarrier belongs to the element in the zero subcarrier group, setting the 48 th subcarrier as the zero subcarrier, also in the orthogonal frequency division multiplexing signal of the 2 nd antenna, the 48 th subcarrier is the unused pilot subcarrier, the 16 th subcarrier group is also the zero subcarrier group of the 2 nd antenna, and setting the 48 th subcarrier as the zero subcarrier; since the subcarrier signal of the zero subcarrier is 0, the orthogonality of the pilot subcarriers of the two antennas is not affected, and the method is compatible with the original transmission pattern fixed by the protocol. In the orthogonal frequency division multiplexing signals of the 1 st antenna and the 2 nd antenna, a zero subcarrier group, subcarriers except for the positions corresponding to the used pilot subcarriers and unused pilot subcarriers are set as data subcarriers, and a corresponding data subcarrier group is formed.

Based on the descriptions of the single-antenna and dual-antenna ofdm signal transmission patterns, the ofdm signal transmission patterns can be extended to any other multiple antennas, such as 4 antennas, 8 antennas, etc.; for other multi-antenna ofdm signal transmission patterns, reference may be made to the above description of single-antenna and dual-antenna ofdm signal transmission patterns, which are not described herein again. In a transmission signal pattern specified by a 3GPP communication protocol, in order to ensure orthogonality of pilot signals of different antennas, a large number of used pilot subcarriers and unused pilot subcarriers need to be set, resource occupancy of the pilot subcarriers increases with the increase of the number of antennas, while the orthogonal frequency division multiplexing signals of a zero subcarrier group set in each antenna are the same, and the resource occupancy of the zero subcarrier does not increase with the increase of the number of antennas, so that the orthogonal frequency division multiplexing signals in the embodiment of the present invention have lower resource occupancy under the condition of multiple antennas; moreover, the occupancy rate of the pilot frequency sub-carrier can be further reduced and the transmitting power can be saved by setting the zero part of the pilot frequency sub-carrier.

Step S102: and acquiring control information of the OFDM signal, wherein the control information comprises information of a subcarrier group, position information of a zero subcarrier group and position information of a data subcarrier group.

In the control information, the information of the subcarrier group is used to describe the grouping situation of the subcarrier group, specifically, for example, the value of the total subcarrier group Q, the value of the subcarrier set P, and the like; the position information of the zero subcarrier group is used for describing the subcarrier position of one or more zero subcarrier groups in the orthogonal frequency division multiplexing signal; the position information of the data subcarrier group is used for describing the subcarrier position of the data subcarrier group carrying the data information in the orthogonal frequency division multiplexing signal.

The control information may be control information preset by a user in real time according to communication needs, or control information fixedly set by the user. In a first implementation manner, the control information may be stored in the memories of the base station and the UE, respectively, and it is ensured that the control information in the base station and the control information in the UE are matched consistently; and when the base station sends the orthogonal frequency division multiplexing signal, the control information is called to generate the orthogonal frequency division multiplexing signal, and the UE carries out subsequent frequency offset estimation according to the control information after receiving the orthogonal frequency division multiplexing signal. In a second implementation manner, the Control information may be stored in a Control Channel of the ofdm signal, for example, a PDCCH (Physical Downlink Control Channel) or a PUCCH (Physical Uplink Control Channel), and the UE or the base station extracts the Control information and performs subsequent frequency offset estimation when analyzing the ofdm signal.

Step S103: and extracting subcarrier signals from the orthogonal frequency division multiplexing signals according to the control information to generate a signal matrix.

Without considering the time offset, the subcarrier signal received by the receiving end, e.g., a UE in the downlink or a base station in the uplink, can be represented in the time domain as:

wherein, K is the number of data subcarrier groups, the number of data subcarrier groups can be understood as the number of data subcarrier groups occupied by a user and used for transmitting user data, and if one user occupies one data subcarrier group, the number of data subcarrier groups can be understood as the number of users; and N is the number of subcarriers in one orthogonal frequency division multiplexing signal.

r_k(n) represents the nth subcarrier signal of the kth data subcarrier group, and the specific expression is as follows:

where P is the number of subcarrier sets, ζ_kIs the normalized frequency offset corresponding to the kth data subcarrier group, Q is the number of subcarrier groups, H_k,pAnd X_k,pRespectively representing the channel frequency domain response and data of the p-th sub-carrier in the k-th data sub-carrier group.

According to the received control information, taking the first P subcarrier signals as a first row of a signal matrix, and then sequentially taking the P subcarrier signals as one row of the signal matrix, thereby obtaining a signal matrix with Q rows and P columns, wherein the representation of the signal matrix R is as follows:

step S104: and observing the signal matrix to obtain a frequency offset estimation value of the orthogonal frequency division multiplexing signal.

The signal matrix R may be further represented in the form:

R＝ΨS+Z＝Ψ[U⊙(BW)]+Z

furthermore, the matrix S can be expressed as an operation expression composed of a frequency domain signal matrix B, IFFT (Inverse Fast Fourier Transform) matrix W and a matrix U, wherein an operation symbol ⊙ is a Schur product, and the frequency domain signal matrix B is expressed as follows:

the matrix U is represented as follows:

the matrix Ψ can be represented as a vandermonde matrix of K rows and Q columns with equivalent frequency offsets, and the specific formula is as follows:

in the above expression, θ₀For equivalent frequency offset, theta, corresponding to the first data subcarrier group₁Equivalent frequency deviation corresponding to the second data subcarrier group, and then theta is carried out in sequence_K-1And equivalent frequency offset corresponding to the Kth data subcarrier group.

And observing the matrix psi obtained by decomposition in the signal matrix to obtain a frequency offset estimation value of the orthogonal frequency division multiplexing signal. Specifically, the matrix Ψ is expanded and dimensioned into a multi-column sparse observation matrix in a column dimension, and the number of columns of the observation matrix is far greater than the number K of data subcarrier groups; and carrying out observation iterative computation on the observation matrix to finally obtain the frequency offset estimation value.

It can be seen from the above embodiments that, a receiving end receives a comb-like orthogonal frequency division multiplexing signal periodically separated by one or more zero subcarrier groups, and interference between subcarriers is reduced by flexibly setting the zero subcarrier groups, so that all subcarriers are brought into a calculation range, data dimensions are enlarged, estimation errors are reduced, and estimation accuracy under a high signal-to-noise ratio condition is ensured; furthermore, all subcarrier signals in the orthogonal frequency division multiplexing signals are constructed into a signal matrix according to the control information, and a frequency offset estimation value is obtained by observing the signal matrix, so that the complexity of the algorithm is reduced, and the requirement of practicability is met.

Referring to fig. 5, a schematic flow chart of a second embodiment of the frequency offset estimation method in the OFDM system according to the present invention is shown, and the embodiment shown in fig. 5 shows a process of observing a signal matrix to obtain a frequency offset estimation value of an OFDM signal, including the following steps:

step S201: and establishing an observation matrix from the signal matrix according to the equivalent frequency offset range and the sampling number, wherein the column vector of the observation matrix corresponds to the equivalent frequency offset of the data subcarrier group.

In the matrix Ψ, each column element corresponds to the equivalent frequency offset of one data subcarrier group, e.g., the equivalent frequency offset θ of the first column corresponding to the first data subcarrier group₀The second column corresponds to the equivalent frequency deviation theta of the second data subcarrier group₁Equivalent frequency deviation theta of the Kth data subcarrier group corresponding to the Kth row in sequence_K-1。

And sampling the equivalent frequency deviation within the equivalent frequency deviation range according to the sampling number, so as to acquire and obtain the new equivalent frequency deviation consistent with the sampling number. The equivalent frequency deviation range can be a total equivalent frequency deviation range, the sampling number is a total sampling number, and equivalent frequency deviation consistent with the total sampling number is collected as new equivalent frequency deviation in the total equivalent frequency deviation range; or the equivalent frequency deviation range is a group equivalent frequency deviation range corresponding to each data subcarrier group and a group sampling number corresponding to each data subcarrier group, equivalent frequency deviations consistent with the group sampling number are collected in the group equivalent frequency deviation range, and a set of new equivalent frequency deviations collected in all the group equivalent frequency deviation ranges is used as the new equivalent frequency deviations. It should be noted that the total number of samples or the sum of the group number of samples is much larger than the number K of data subcarrier groups, for example, if K is 8, the total number of samples or the sum of the group number of samples may be 50 or 100, etc.

And finally, organizing all the observation matrix column vectors into an observation matrix according to the sequence of the data subcarrier groups, wherein the number of rows of the observation matrix is Q rows, and the number of columns is the sum of the total sampling number or the group sampling number.

Step S202: and carrying out observation iterative computation on the observation matrix to obtain an observation iterative computation result, wherein the dimension of the observation iterative computation result is equal to the column number of the observation matrix.

Referring to fig. 6, a schematic flow chart of an observation iterative computation method provided by the present invention is shown, where the observation iterative computation method includes the following steps:

step S2021: a first sparse hyperparametric vector is established.

The dimensionality of the first sparse hyperparameter vector γ is equal to the number of columns of an observation matrix, i.e. if the number of columns of the observation matrix is 150, the first coefficient hyperparameter vector contains 150 elements; when the first iteration is performed, all elements of the first coefficient hyperparametric vector are initialized to 1.

Step S2022: and obtaining a diagonal matrix according to the first sparse hyperparametric vector.

The diagonal matrix Γ ═ diag (γ), that is, the number of rows and columns of the diagonal matrix Γ are both equal to the dimensions of the first sparse hyperparametric vector; in the diagonal matrix Γ, the element in the ith row and the ith column is the ith element in the first sparse hyperparametric vector, and the elements at other positions are all 0.

Step S2023: according to the diagonal matrix, a first posterior parameter M and a second posterior parameter Sigma are calculated.

The calculation formula of the first posterior parameter m is as follows:

where is the observation matrix, is the conjugate transpose matrix, N₀For noise power constants, I is an identity matrix with dimensions Q × Q, and R is a signal matrix. The first posterior parameter m calculated by the above formula is a matrix with the row number being the dimension of the first sparse hyperparametric vector gamma and the column number being P.

The calculation formula of the second posterior parameter Σ is as follows:

the second posterior parameter Σ calculated by the above formula is a matrix in which the number of rows and the number of columns are both the dimensions of the first sparse hyperparametric vector γ.

Step S2024: and obtaining a second sparse hyperparametric vector according to the first posterior parameter (Mm) and the second posterior parameter (Sigma).

The calculation formula of the second sparse hyperparameter vector is as follows:

wherein is the ith element of a second sparse hyperparametric vector having dimensions equal to those of the first sparse hyperparametric vector, being the square of the two norms of the A posteriori parameters M, column l, Σ_llElement of row l and column l of a posteriori parameter ∑。

Step S2025: and judging whether the difference value of the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a preset convergence threshold value.

The convergence threshold may be a preset convergence threshold, for example, the convergence threshold is preset to be 1 e-5. And subtracting the first coefficient hyperparametric vector from the second coefficient hyperparametric vector to obtain a difference value of the two vectors, and judging whether each element in the difference value is smaller than a preset convergence threshold value or not.

Step S2026: and if the difference value between the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a convergence threshold value, finishing the observation iterative computation, and taking the second sparse hyperparameter vector as an observation iterative computation result.

Step S2027: and if the difference value between the second sparse hyperparameter vector and the first sparse hyperparameter vector is greater than or equal to a convergence threshold value, taking the current second sparse hyperparameter vector as the first sparse hyperparameter vector, and recovering the second sparse hyperparameter vector according to the first posterior parameter M and the second posterior parameter sigma.

If the element in the difference value obtained by subtracting the second sparse hyperparametric vector and the first sparse hyperparametric vector is larger than or equal to the convergence threshold value, the second sparse hyperparametric vector is used as the first hyperparametric vector, the step S2022 is returned, the diagonal matrix gamma is obtained again, then in the step S2023, the first posterior parameter Mm and the second posterior parameter Sigma are calculated according to the obtained diagonal matrix gamma, in the step S2024, the second sparse hyperparametric vector is obtained again according to the calculated first posterior parameter Mm and the second posterior parameter Sigma, whether the iterative calculation is converged or not is judged in the step S2025, if the iterative calculation is converged, the step S2026 is finished, if the iterative calculation is not converged, the newly obtained second sparse hyperparametric vector is used as the first hyperparametric vector and returned to the step S2022, and a new iteration is carried out until the iterative calculation is converged.

Step S203: and obtaining an equivalent frequency offset result vector from the observation matrix according to the observation iterative computation result.

Referring to fig. 7, a schematic flow chart of a first embodiment of a method for extracting an equivalent frequency offset result vector according to the present invention is shown, where the method includes the following steps:

step S2031: and determining K maximum peak positions from the observation iteration calculation result, wherein K is the number of the data subcarrier groups.

For the elements in the observation iteration calculation result, a bubble method, a dichotomy method, or the like may be adopted to select K maximum values from all the elements, and a position of the maximum value in the observation iteration calculation result may be used as the maximum peak position, for example, if a 23 th element in the observation iteration calculation result is one of the K maximum values, a position number 23 thereof may be used as the maximum peak position.

Step S2032: and extracting a corresponding observation matrix column vector from the observation matrix according to the maximum peak position.

And extracting a column vector corresponding to the maximum peak position from an observation matrix to be used as the column vector of the observation matrix. In a specific embodiment, for example, it is determined in step S2031 that K is 5 maximum peak positions, and the K corresponds to the 3rd element, the 5 th element, the 9 th element, the 10 th element, and the 13 th element in the observation iteration calculation result, respectively, then the 3rd column vector is extracted from the observation matrix as the observation matrix column vector corresponding to the 1 st data subcarrier group, the 5 th column vector is extracted as the observation matrix column vector corresponding to the 2 nd data subcarrier group, the 9 th column vector is extracted as the observation matrix column vector corresponding to the 3rd data subcarrier group, the 10 th column vector is extracted as the observation matrix column vector corresponding to the 4 th data subcarrier group, and the 13 th column vector is extracted as the observation matrix column vector corresponding to the 5 th data subcarrier group.

Step S2033: and extracting equivalent frequency offset values corresponding to the column vectors of the observation matrix, and establishing equivalent frequency offset result vectors.

In practical implementation, referring to the example of step S2032, since the column vector in each observation matrix corresponds to a new equivalent frequency offset, therefore, the equivalent frequency offset corresponding to each data subcarrier group can be determined separately, for example, 3 columns of vectors in the observation matrix correspond to the 3rd new equivalent frequency offset, thereby determining the equivalent frequency offset result corresponding to the 1 st data subcarrier group as the 3rd new equivalent frequency offset, and then sequentially determining that the equivalent frequency offset result corresponding to the 2 nd data subcarrier group is the 5 th new equivalent frequency offset, the equivalent frequency offset result corresponding to the 3rd data subcarrier group is the 9 th new equivalent frequency offset, the equivalent frequency offset result corresponding to the 4 th data subcarrier group is the 10 th new equivalent frequency offset and the equivalent frequency offset result corresponding to the 5 th data subcarrier group is the 13 th new equivalent frequency offset, and forming an equivalent frequency offset result vector by using the K-5 new equivalent frequency offsets according to the sequence of the data subcarrier groups.

Optionally, in order to improve the accuracy and efficiency of determining the maximum peak, referring to fig. 8, which is a schematic flow chart of a second embodiment of the equivalent frequency offset result vector extraction method provided by the present invention, on the basis of the method shown in fig. 7, the method further provides that the method for determining the maximum peak includes the following steps:

step S2034: and dividing the observation iteration calculation result into a plurality of calculation result groups according to the sampling number, wherein the number of the calculation result groups is equal to the number of the data subcarrier groups.

In the construction process of the observation matrix, when the sampling number is a group sampling number, each data subcarrier group corresponds to a column vector in the observation matrix, which is consistent with the group sampling number, and because the dimension of the observation iterative computation result is equal to the column number of the observation matrix, and the observation iterative computation result and the column vector of the observation matrix have the same organization sequence, each data subcarrier group corresponds to an element in the observation iterative computation result, which is consistent with the group sampling number, and the elements in the observation iterative computation result are divided into K computation result groups according to the group sampling number. In specific implementation, if the observation iteration calculation result includes 600 elements and the K value is 10, and the group sampling number of each data subcarrier group is equal to 60, the 1 st data subcarrier group corresponds to 1 to 60 elements in the observation iteration calculation result and serves as a first calculation result group, the 2 nd data subcarrier group corresponds to 61 to 120 elements in the observation iteration calculation result and serves as a 2 nd calculation result group, and the data subcarrier groups are sequentially divided into 10 calculation result groups according to the above manner.

Step S2035: a maximum peak is found within each of the sets of calculations.

And searching for a maximum peak value from each calculation result group, wherein the mode of searching for the maximum peak value can also use a bubbling method or a bisection method and the like, then determining the corresponding maximum peak values in all the calculation result groups, and finally obtaining K maximum peak values.

Step S2036: and taking the position of the maximum peak in the observation iteration calculation result as the maximum peak position.

After the maximum peak of each calculation result group is determined, the position of the maximum peak in the entire observation iteration calculation result is recorded as the maximum peak position, for example, the 3rd element in the 2 nd calculation result group is determined as the maximum peak, that is, the 63 th element in the entire observation iteration calculation result corresponds to the 63 rd element in the entire observation iteration calculation result, and the position 63 in the entire observation iteration calculation result is taken as the maximum peak position. Then, a total of K maximum peak positions are determined in turn.

Step S204: and calculating a frequency offset estimation value according to the equivalent frequency offset result vector.

Referring to fig. 9, a schematic flow chart of a method for calculating a frequency offset estimation value according to the present invention is shown, where the method includes the following steps:

step S2041: and extracting an equivalent frequency offset value corresponding to each data subcarrier group from the equivalent frequency offset result vector.

The equivalent frequency offset result vector comprises K elements, and corresponding equivalent frequency offset results are extracted respectively corresponding to the equivalent frequency offset results of the data subcarrier groups.

Step S2042: and calculating a group frequency offset estimation value of each data subcarrier group according to the equivalent frequency offset value.

The calculation formula of the group frequency offset estimation value is as follows:

the estimated value of the group frequency offset corresponding to the (k + 1) th data subcarrier group is the estimated value of the group frequency offset corresponding to the 1 st data subcarrier group calculated by the above method.

Step S2043: and carrying out average calculation on the group of frequency deviation estimated values to obtain the frequency deviation estimated values.

The formula for averaging the set of frequency offset estimates is as follows:

wherein, the estimated value is the frequency offset, K is the number of data subcarrier groups, and the group frequency offset estimated value corresponding to the kth data subcarrier group.

In addition, in the process of obtaining a frequency offset estimation value of an orthogonal frequency division multiplexing signal by observing a matrix Ψ in a signal matrix shown in the embodiment of the present invention, when a single-layer observation is performed to calculate the frequency offset estimation value, sampling is performed in a first layer equivalent frequency offset range according to a first layer sampling number, so as to obtain a first layer new equivalent frequency offset; obtaining a first layer of observation matrix column vectors according to the first layer of new equivalent frequency offset, and finally organizing the first layer of observation matrix column vectors into a first layer of observation matrix; and carrying out observation iterative computation on the first layer of observation matrix to obtain a first layer of equivalent frequency offset result vector, and calculating a frequency offset estimation value through the first layer of equivalent frequency offset result vector.

When multilayer observation calculation frequency offset estimation values are carried out, if the current observation calculation layer is the nth layer, wherein n is greater than or equal to 2, determining the equivalent frequency offset range of the nth layer according to the equivalent frequency offset result vector of the (n-1) th layer, and sampling in the equivalent frequency offset range of the nth layer according to the sampling number of the nth layer to obtain new equivalent frequency offset of the nth layer; according to the new equivalent frequency offset of the nth layer, neglecting the column vector of the nth layer of observation matrix, and finally organizing the column vector of the nth layer of observation matrix into the nth layer of observation matrix; carrying out observation iterative computation on the nth layer of observation matrix to obtain an nth layer of equivalent frequency offset result vector; and finally, calculating a frequency offset estimation value by using the final layer of equivalent frequency offset result vector.

According to the embodiment, the receiving end establishes the sparse observation matrix according to the signal matrix in a sampling mode, observes the observation matrix in a layered observation mode and recovers the frequency offset with high precision, the complexity of the algorithm is reduced, and the practical requirement and the precision requirement of frequency offset estimation are effectively met.

Referring to fig. 10, a flowchart of a frequency offset estimation method in an OFDM system according to a third embodiment of the present invention is provided, where the embodiment shows a process of calculating a frequency offset estimation value through single-layer observation, and includes the following steps:

step S301: and according to the sampling number of the first layer, carrying out equivalent frequency offset sampling in the range of the equivalent frequency offset of the first layer, and establishing a first layer observation matrix from the signal matrix according to the sampled equivalent frequency offset.

In the embodiment of the present invention, the first sampling manner is: the first layer of sampling number is a first layer of total sampling number, the first layer of equivalent frequency deviation range is a total equivalent frequency deviation range, and the equivalent frequency deviation satisfies the following formula, and data points are sampled in the first layer of equivalent frequency deviation range to form a new equivalent frequency deviation value set

The second sampling mode is as follows: according to the number Q of the total subcarrier groups, setting a corresponding first layer equivalent frequency deviation range corresponding to each data subcarrier group, wherein the first equivalent frequency deviation range is a group equivalent frequency deviation range, and the formula of the first equivalent frequency deviation range is as follows:

and the equivalent frequency offset value corresponding to the kth data subcarrier group, K is the number of the data subcarrier groups, and Q is the number of the total subcarrier groups.

Setting a first layer sampling number corresponding to each data subcarrier group, where the first layer sampling number is a group sampling number, for example, setting a first layer sampling number corresponding to a 1 st data subcarrier group as a sum of first layer sampling numbers corresponding to all data subcarrier groups, where the first layer sampling number corresponding to a 2 nd data subcarrier group is set as follows:

and sampling in a first layer equivalent frequency offset range corresponding to each data subcarrier group according to the first layer sampling number corresponding to each data subcarrier group. Of course, in practical implementation, the number of first layer samples corresponding to each data subcarrier group may be set to be equal, i.e., where J is⁽¹⁾A first layer number of samples corresponding to each data subcarrier group; or, according to the actual sampling requirement, different first layer sampling numbers are set for different data subcarrier groups.

In addition, it should be noted that, in the embodiment of the present invention, sampling in the range of the first layer equivalent frequency offset based on the number of first layer samples may be uniform sampling or random sampling, which is not limited in the embodiment of the present invention.

Obtaining a total of sampling points of equivalent frequency deviation according to the first sampling mode or the second sampling mode, arranging and representing the sampling points into a size sequence, then extracting a matrix psi from a signal matrix, establishing an observation matrix column vector according to an expression form of a column vector in the matrix psi and according to the equivalent frequency deviation after sampling, wherein an expression of the observation matrix column vector expressed as a Q dimension is J ═ 1,2, …, J ═ 1_total(ii) a Organizing the array vectors of the observation array matrix into a matrix form to obtain the first layer observation matrix as the dimension of the first layer observation matrix as Q rows J_totalColumn, and due to J_totalThe numerical value is large, and the first layer of observation matrix is guaranteed to have strong sparsity.

Step S302: and carrying out observation iterative computation on the first layer of observation matrix and obtaining a first layer of observation iterative computation result.

The process of the observation iterative computation may refer to the description in step S202, which is not described herein again, and finally the process including J is obtained_totalAnd (4) iteratively calculating the result of the first-layer observation of each element.

Step S303: and obtaining a first layer equivalent frequency offset result vector from the first layer observation matrix according to the first layer observation iterative computation result.

Determining K maximum peak positions from the first layer observation iteration calculation result; determining a corresponding observation matrix column vector from a first layer observation matrix according to the maximum peak position; and extracting equivalent frequency offset values corresponding to the observation matrix column vectors to form the first layer of equivalent frequency offset result vectors. The above process of obtaining the first-layer equivalent frequency offset result vector may refer to step S203, which is not described herein again.

Step S304: and calculating a frequency offset estimation value according to the first layer equivalent frequency offset result vector.

In order to further control the frequency offset estimation accuracy, on the basis of the frequency offset estimation value observation algorithm shown in fig. 10, referring to fig. 11, it is a flowchart of the first layer sampling number determining method provided by the present invention, and the method includes the following steps:

step S305: a first dispersion error is set based on a first desired value of the equivalent frequency offset estimation error.

In a specific implementation, the first expected value may be preset, and the first discrete error may be set according to the first expected value, so that a square of the first discrete error is two orders of magnitude smaller than an expected value of an equivalent frequency offset estimation error.

Step S306: and calculating the number of first layer samples according to the first discrete error and the grouping information of the orthogonal frequency division multiplexing signal.

When the first layer sampling number is the total sampling number, calculating the first layer sampling number according to the following formula

Wherein, delta₁K is the number of data subcarrier groups and Q is the total number of subcarrier groups for the first dispersion error.

When the number of first layer samples is the group sample number and the number of first layer samples corresponding to each data subcarrier group is equal, the number of first layer samples J may be calculated by using a formula⁽¹⁾。

According to the embodiment, the receiving end establishes the first layer observation matrix with sparsity according to the signal matrix in a sampling mode, single-layer observation calculation processing is carried out on the first layer observation matrix, high-precision recovery is carried out on the frequency offset, algorithm complexity is reduced, and practicability and precision requirements of frequency offset estimation are met.

In order to obtain a frequency offset estimation value with higher accuracy, referring to fig. 12, a flowchart of a fourth embodiment of the frequency offset estimation method in the OFDM system provided in the present invention is shown, where the embodiment shows a process of calculating the frequency offset estimation value through multi-layer observation, and includes the following steps:

step S401: and according to the number of samples of the nth layer, carrying out equivalent frequency offset sampling in the equivalent frequency offset range of the nth layer, and establishing an nth layer observation matrix from the signal matrix according to the sampled new equivalent frequency offset, wherein n is greater than or equal to 2.

Determining an n-th layer equivalent frequency offset range according to the n-1 layer equivalent frequency offset result vector, the n-1 layer sampling number and the scaling factor, wherein the n-th layer equivalent frequency offset range is determined according to the following formula:

wherein, the equivalent frequency offset value corresponding to the kth data subcarrier group in the n-1 layer equivalent frequency offset result vector is the equivalent frequency offset value corresponding to the kth data subcarrier group in the nth layer, S is a scaling factor, Q is the number of the total subcarrier groups, K is the number of the data subcarrier groups, J is^(n-1)Is the n-1 th layer sample number. Note that J is^(n-1)The sampling number corresponding to each data subcarrier group in the n-1 layer meets the relation, wherein the sampling number is the total sampling number of the n-1 layer; when n is 2, if the number of samples in the first layer is the total number of samples, it may be considered here and the number of samples corresponding to each data subcarrier group in the n-1 th layer may be the same or different.

In the first case, the total number of sampling points in each layer is equal, i.e. the total number of sampling points is the same value, and the user can flexibly preset the total number of sampling points in each layer to be any value, but in order to ensure the sparsity of establishing the observation matrix, the value of the total number of sampling points should be much larger than the number K of data subcarrier groups.

In the second case, the total number of sampling points in each layer is not equal and gradually decreases as the number of observation layers N increases; for example, the user may set the total number of sampling points corresponding to the first layer observation to be 300, the total number of sampling points corresponding to the second layer observation to be 200, and the total number of sampling points corresponding to the third layer observation to be 100, where the total number of sampling points decreases progressively, that is, certainly, in specific implementation, the total number of sampling points decreases progressively, and may decrease progressively by a manner of reducing a preset reduction ratio or by a manner of reducing a difference.

And carrying out equivalent frequency offset sampling on the nth layer of equivalent frequency offset range corresponding to each data subcarrier group according to the nth layer of sampling number respectively. In specific implementation, for the 1 st data subcarrier group, in the nth layer equivalent frequency offset range corresponding to the 1 st data subcarrier group, performing equivalent frequency offset sampling according to the nth layer sampling number corresponding to the 1 st data subcarrier group; and similarly, the equivalent frequency offset is sampled within the equivalent frequency offset range of the nth layer corresponding to the kth data subcarrier group according to the sampling number of the nth layer corresponding to the kth data subcarrier group.

Extracting a matrix psi from the signal matrix according to the sampled new equivalent frequency offsets, organizing each new equivalent frequency offset into an observation matrix column vector according to a column vector form in the matrix psi, and constructing an nth layer observation matrix by the observation matrix column vectors, wherein the dimension of the nth layer observation matrix is related to the nth layer sampling number, and specifically, the sampling dimension of the nth layer observation matrix is Q rows and columns.

Step S402: and carrying out observation iterative computation on the nth layer of observation matrix, and obtaining an nth layer of observation iterative computation result.

Step S403: and obtaining an nth layer equivalent frequency offset result vector from the nth layer observation matrix according to the nth layer observation iterative computation result.

Step S404: and calculating a frequency offset estimation value according to the final layer of equivalent frequency offset result vector.

In a specific implementation, in the first case, a user may set in advance a layer number N for performing observation calculation, for example, the layer number N is set to 3 or 4; and when the frequency offset estimation value reaches the Nth layer, the frequency offset estimation value of the Nth layer is taken as a final calculation result, and the frequency offset estimation is finished.

Or, under the second condition, calculating the difference value between the N-1 layer frequency offset estimation value and the N layer frequency offset estimation value without setting the number N of the observation layers, and judging whether the difference value is smaller than a preset threshold value, if the difference value is smaller than the preset threshold value, judging that the precision of the frequency offset estimation value is reached without next layer of observation, and taking the N layer frequency offset estimation value as the result of observation calculation to finish frequency offset estimation; and if the difference is larger than or equal to a preset threshold, judging that the precision of the frequency offset estimation value is not enough, and continuing to perform observation calculation of the (n + 1) th layer if the next layer of observation is needed.

To further control the frequency offset estimation essenceIn the method for determining the number of samples for multi-layer frequency offset estimation provided by the embodiment of the invention, in an observation calculation layer n, the number of samples J corresponding to each data subcarrier group⁽ⁿ⁾In the case of equality, on the basis of the multi-layer frequency offset estimation method shown in fig. 12, the method further includes the following steps:

step S405: and setting a second discrete error according to a second expected value of the equivalent frequency offset estimation error.

Presetting a second expected value of the equivalent frequency offset estimation error; the second discrete error has a square that is two orders of magnitude less than a second desired value of the equivalent frequency offset estimation error.

Step S406: calculating the scaling factor and the n-th layer sample number according to a formula, wherein delta₂For the second dispersion error, S is the scaling factor, N is the number of observation calculation layers, K is the number of data subcarrier groups, Q is the number of total subcarrier groups, J⁽ⁿ⁾Is the nth layer sample number.

Because the scaling factor, the sampling number and the number of observation calculation layers jointly determine the frequency offset estimation precision, through the formula, a user can flexibly and reasonably set the scaling factor, the sampling number and the number of observation calculation layers according to specific calculation resources such as CPU processing speed, memory and the like, so that the balance between the calculation resources and the calculation precision is realized. In the case that the number of samples in each layer of equivalent frequency deviation observation calculation is equal, namely J⁽¹⁾＝J⁽²⁾＝…＝J^(N)Where N is the number of observation calculation layers, the above expression can be simplified as follows:

after the scaling factor and the number N of the observation calculation layers are determined, the number of the nth layer of samples can be calculated; of course, the user may first set the nth layer sampling number and the number of observation calculation layers, so as to determine the scaling factor; or calculating the number of observation calculation layers according to the scaling factor and the nth layer sampling number. Therefore, in the actual calculation process, the user can flexibly set according to the formula, so that higher frequency offset estimation precision is obtained under the condition of optimal calculation overhead.

As can be seen from the foregoing embodiments, in the frequency offset estimation method in the OFDM system provided in the embodiments of the present invention, by receiving an orthogonal frequency division multiplexing signal, which is a signal periodically comb-separated by a zero subcarrier group, interference between subcarriers is effectively reduced, a large number of subcarriers can be brought into a data range of frequency offset estimation, an observation dimension is enlarged, and further, by observation calculation of a large amount of data, frequency offset estimation accuracy is ensured; in addition, in the observation calculation, the frequency offset is recovered with high precision by using a layer-by-layer observation method, each layer of observation calculation has reasonable data processing complexity, and approaches the high-precision calculation result layer by layer, so that the method has high precision and strong practicability.

Referring to fig. 13, which is a schematic flow chart of a fifth embodiment of the frequency offset estimation method in the OFDM system according to the present invention, in a downlink, a base station sends out an OFDM signal, so that a UE can receive the OFDM signal and perform frequency offset estimation; in the uplink, the UE sends out an orthogonal frequency division multiplexing signal, which is convenient for the base station to receive the orthogonal frequency division multiplexing signal and perform frequency offset estimation. In the embodiment of the present invention, the following type of link is described in detail, and the embodiment shows a process for generating a comb-shaped orthogonal frequency division multiplexing signal, including the following steps:

step S501: the subcarriers are divided into a plurality of subcarrier sets in a frequency domain space.

In the frequency domain space, the subcarriers corresponding to one ofdm signal are divided into a plurality of subcarrier sets, the number of the subcarrier sets is represented by P, the number of the subcarriers corresponding to one ofdm signal is represented by N, so that each subcarrier set corresponds to Q subcarriers, and Q is equal to N/P.

Step S502: zero subcarriers are arranged at least one zero setting position of each subcarrier set, and the zero setting positions in each subcarrier set are the same; the zero subcarriers in the same position in all subcarrier sets constitute a zero subcarrier group.

Null subcarriers are set at one or more null positions in one subcarrier set, and the null positions are the same in each subcarrier set for each subcarrier set. If only one zero-setting position in one subcarrier set is provided with a zero subcarrier, the zero subcarriers in all P subcarrier sets form a zero subcarrier group; if a plurality of zero setting positions are arranged in one subcarrier set, and the zero setting positions can be adjacent or mutually spaced, zero subcarriers at the same positions in all P subcarrier sets form a zero subcarrier group, and the number of the zero subcarrier group is equal to the number of the zero setting positions in one subcarrier set.

In specific implementation, in order to ensure compatibility with the current 3GPP TS 36.211 communication protocol, when the zero position coincides with the position of the pilot subcarrier, since the signal of the zero subcarrier is 0, the orthogonality of the pilot subcarrier is not affected, and the pilot subcarrier may be set as the zero subcarrier.

Step S503: setting data subcarriers at subcarrier positions other than the set-zero position in each subcarrier set; the data subcarriers at the same position in all subcarrier sets constitute a data subcarrier group.

In each subcarrier set, if the subcarrier set does not contain pilot subcarriers, data subcarriers can be set at positions other than the zero position; setting subcarriers other than the pilot subcarrier and the null subcarrier as data subcarriers if for compatibility with the current 3GPP TS 36.211 communication protocol; all the data subcarriers at the same position in the P data subcarrier sets form a data subcarrier group for carrying data signals, and one ofdm signal may include a plurality of data subcarrier groups.

Step S504: and generating and sending an orthogonal frequency division multiplexing signal according to the zero subcarrier group and the data subcarrier group.

Generating and sending an orthogonal frequency division multiplexing signal according to the zero subcarrier group and the data subcarrier group corresponding to the step; of course, the ofdm signal may further include pilot subcarriers.

It should be noted that the ofdm signal generated by the above method may include an ofdm signal transmitted by a single antenna or multiple antennas; based on the OFDM signal, in the downlink, the UE receives the OFDM signal and performs frequency offset estimation, and in the uplink, the base station receives the OFDM signal and performs frequency offset estimation.

As can be seen from the foregoing embodiments, in the frequency offset estimation method in the OFDM system provided in the embodiments of the present invention, the zero subcarrier group is periodically arranged, and the orthogonal frequency division multiplexing signal is periodically comb-separated in the frequency domain space, so that the interference between subcarriers is effectively reduced, and thus, high-dimensional observation data is used in the frequency offset hierarchical estimation algorithm to perform frequency offset estimation, and the frequency offset estimation accuracy is improved.

Through the above description of the method embodiments, those skilled in the art can clearly understand that the present invention can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods according to the embodiments of the present invention. And the aforementioned storage medium includes: various media that can store program codes, such as Read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and so on.

Corresponding to the embodiment of the frequency offset estimation method in the OFDM system, the invention also provides a frequency offset estimation device in the OFDM system.

Referring to fig. 14, a schematic structural diagram of a first embodiment of an apparatus for frequency offset estimation in an OFDM system provided in the present invention is a receiving end, for example, the apparatus may be disposed in a base station in an uplink or a UE in a downlink, and performs frequency offset estimation on a received OFDM signal by using the apparatus, where the apparatus includes:

an orthogonal frequency division multiplexing signal receiving module 11, configured to receive an orthogonal frequency division multiplexing signal, where the orthogonal frequency division multiplexing signal includes multiple subcarrier sets, each subcarrier set includes at least one zero subcarrier and multiple data subcarriers, and the zero subcarriers in each subcarrier set have the same position; the zero subcarriers with the same position in all the subcarrier sets form a zero subcarrier group, and the data subcarriers with the same position in all the subcarrier sets form a data subcarrier group; the orthogonal frequency division multiplexing signal is a signal which is separated into a comb structure by the zero subcarrier group, and the anti-interference requirement of frequency offset estimation is met;

a control information obtaining module 12, configured to obtain control information of the ofdm signal, where the control information includes: information of subcarrier groups, position information of zero subcarrier groups, and position information of data subcarrier groups;

a signal matrix generating module 13, configured to extract a subcarrier signal from the ofdm signal according to the control information from the control information acquiring module, and generate a signal matrix;

a frequency offset estimation value obtaining module 14, configured to observe the signal matrix to obtain a frequency offset estimation value of the orthogonal frequency division multiplexing signal.

It can be seen from the above embodiments that, a frequency offset estimation apparatus at a receiving end receives a comb-like orthogonal frequency division multiplexing signal periodically separated by one or more zero subcarrier groups, and by flexibly setting the zero subcarrier groups, interference between subcarriers is reduced, so that all subcarriers are brought into a calculation range, data dimensions are enlarged, estimation errors are reduced, and estimation accuracy under a high signal-to-noise ratio condition is ensured; furthermore, all subcarrier signals in the orthogonal frequency division multiplexing signals are constructed into a signal matrix according to the control information, and a frequency offset estimation value is obtained by observing the signal matrix, so that the complexity of the algorithm is reduced, and the requirement of practicability is met.

Referring to fig. 15, which is a schematic structural diagram of a second embodiment of a frequency offset estimation apparatus in an OFDM system according to the present invention, in the frequency offset estimation apparatus based on fig. 14, the frequency offset estimation value obtaining module 14 further includes:

an observation matrix establishing module 21, configured to establish an observation matrix from the signal matrix according to the equivalent frequency offset range and the sampling number; the observation matrix has strong sparsity because the sampling number is far greater than the number of data subcarrier groups;

the observation iterative computation module 22 is configured to perform observation iterative computation on the observation matrix and obtain an observation iterative computation result, where a dimension of the observation iterative computation result is equal to a column number of the observation matrix;

an equivalent frequency offset result obtaining module 23, configured to obtain an equivalent frequency offset result vector from an observation matrix according to the observation iterative computation result obtained by the observation iterative computation module;

and a frequency offset estimation value calculation module 24, configured to calculate a frequency offset estimation value according to the equivalent frequency offset result vector.

Optionally, referring to fig. 16, which is a schematic structural diagram of the observation iteration calculating module provided in the present invention, the observation iteration calculating module 22 further includes:

a first sparse hyperparametric vector establishing module 221, configured to establish a first sparse hyperparametric vector;

a diagonal matrix obtaining module 222, configured to obtain a diagonal matrix according to the first sparse hyperparametric vector;

a posterior parameter calculation module 223 for calculating a first posterior parameter m and a second posterior parameter Σ from the diagonal matrix;

a second sparse hyperparameter vector obtaining module 224, configured to obtain a second sparse hyperparameter vector according to the first a posteriori parameter m and the second a posteriori parameter Σ;

a convergence judging module 225, configured to judge whether a difference between the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a preset convergence threshold;

an observation iterative computation ending module 226, configured to, according to the determination result of the convergence determining module, complete the observation iterative computation if a difference between the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a convergence threshold, and use the second sparse hyperparameter vector as an observation iterative computation result;

and the observation iteration calculation circulating module 227 is used for taking the current second sparse hyperparameter vector as the first sparse hyperparameter vector if the difference value of the second sparse hyperparameter vector and the first sparse hyperparameter vector is greater than or equal to a convergence threshold value according to the judgment result of the convergence judgment module, and obtaining the second sparse hyperparameter vector again according to the first posterior parameter M and the second posterior parameter sigma.

Optionally, referring to fig. 17, which is a schematic structural diagram of a first embodiment of the equivalent frequency offset result obtaining module provided in the present invention, the equivalent frequency offset result obtaining module 23 further includes:

a maximum peak position determining module 231, configured to determine K maximum peak positions in the observation iteration calculation result, where K is the number of data subcarrier groups;

an observation matrix column vector extraction module 232, configured to extract a corresponding observation matrix column vector from the observation matrix according to the maximum peak position;

an equivalent frequency offset result vector establishing module 233, configured to establish an equivalent frequency offset result vector according to the data subcarrier group sequence with respect to the equivalent frequency offset value corresponding to the observation matrix column vector.

Optionally, referring to fig. 18, which is a schematic structural diagram of a second embodiment of the equivalent frequency offset result obtaining module according to the present invention, in order to improve the maximum peak value determination accuracy and efficiency, the equivalent frequency offset result obtaining module 23 further includes:

a calculation result grouping module 234, configured to divide the observation iterative calculation result into a plurality of calculation result groups according to the number of samples, where the number of the calculation result groups is equal to the number of the data subcarrier groups;

a group maximum peak determination module 235 for finding a maximum peak within each of said calculation result groups;

a group maximum peak position obtaining module 236, configured to use a position of the maximum peak in the observation iteration calculation result as the maximum peak position.

Optionally, referring to fig. 19, which is a schematic structural diagram of a frequency offset estimation value calculation module provided in the present invention, the frequency offset estimation value calculation module 24 may further include:

an equivalent frequency offset value extracting module 241, configured to extract an equivalent frequency offset value corresponding to each data subcarrier group from the equivalent frequency offset result vector;

a group frequency offset estimation value calculation module 242, configured to calculate a group frequency offset estimation value of each data subcarrier group according to the equivalent frequency offset value;

and a group frequency offset estimation value averaging module 243, configured to perform average calculation on the group frequency offset estimation value to obtain the frequency offset estimation value.

It can be seen from the above embodiments that the frequency offset estimation apparatus disposed at the receiving end establishes a sparse observation matrix according to the signal matrix in a sampling manner, observes the observation matrix in a layered observation manner, and performs high-precision frequency offset recovery, thereby reducing the complexity of the algorithm and effectively satisfying the practical requirement and precision requirement of frequency offset estimation.

Referring to fig. 20, which is a schematic structural diagram of a frequency offset estimation apparatus in an OFDM system according to a third embodiment of the present invention, the frequency offset estimation value obtaining module 14 calculates a frequency offset estimation value through single-layer observation, including:

a first layer observation matrix obtaining module 31, configured to perform equivalent frequency offset sampling in a first layer equivalent frequency offset range according to the first layer sampling number, and establish a first layer observation matrix from the signal matrix according to the sampled equivalent frequency offset;

the first-layer observation iterative computation module 32 is configured to perform observation iterative computation on the first-layer observation matrix and obtain a first-layer observation iterative computation result;

a first layer equivalent frequency offset result obtaining module 33, configured to obtain a first layer equivalent frequency offset result vector from the first layer observation matrix according to the first layer observation iterative computation result;

and a first layer frequency offset estimation value calculating module 34, configured to calculate a frequency offset estimation value according to the first layer equivalent frequency offset result vector.

Optionally, referring to fig. 21, which is a schematic structural diagram of the first-layer observation matrix obtaining module provided in the present invention, the first-layer observation matrix obtaining module 31 further includes:

a first layer equivalent frequency offset range determining module 311, configured to determine, according to the number of total subcarrier groups, first layer equivalent frequency offset ranges corresponding to each data subcarrier group, where the formula is as follows:

the first-layer equivalent frequency offset sampling module 312 is configured to perform equivalent frequency offset sampling on the first-layer equivalent frequency offset range corresponding to each data subcarrier group according to the first-layer sampling number, respectively.

In order to further control the frequency offset estimation precision, referring to fig. 22, which is a schematic structural diagram of a first layer sampling number determining module provided in the present invention, in an embodiment of the present invention, the frequency offset estimation apparatus may further include the first layer sampling number determining module, where the first layer sampling number determining module further includes:

a first discrete error setting module 35, configured to set a first discrete error according to a first expected value of the equivalent frequency offset estimation error; the square of the first discrete error is two orders of magnitude smaller than the expected value of the equivalent frequency offset estimation error;

a first layer sampling number calculating module 36, configured to calculate a first layer sampling number according to the first discrete error and packet information of the ofdm signal; when the first layer sampling number is the total sampling number, calculating the first layer sampling number according to the following formula

According to the embodiment, the frequency offset estimation device at the receiving end establishes the first layer observation matrix with sparsity in a sampling mode according to the signal matrix, carries out single-layer observation calculation processing on the first layer observation matrix and carries out high-precision recovery on the frequency offset, reduces algorithm complexity, and meets the requirements of practicability and precision of frequency offset estimation.

Referring to fig. 23, which is a schematic structural diagram of a fourth embodiment of a frequency offset estimation apparatus in an OFDM system according to the present invention, the frequency offset estimation value obtaining module 14 calculates a frequency offset estimation value through multilayer observation, including:

an nth layer observation matrix establishing module 41, configured to perform equivalent frequency offset sampling in an nth layer equivalent frequency offset range according to the nth layer sampling number, and establish an nth layer observation matrix from the signal matrix according to the sampled equivalent frequency offset, where n is greater than or equal to 2; the n layer equivalent frequency offset range is determined according to the n-1 layer equivalent frequency offset result vector, the n-1 layer sampling number and the scaling factor;

the nth layer observation iterative computation module 42 is used for performing observation iterative computation on the signal matrix and obtaining an nth layer observation iterative computation result;

an nth layer equivalent frequency offset result vector obtaining module 43, configured to obtain an nth layer equivalent frequency offset result vector from the nth layer observation matrix according to the nth layer observation iterative computation result;

and a final layer frequency offset estimation value calculation module 44, configured to calculate a frequency offset estimation value according to the last layer equivalent frequency offset result vector.

Optionally, referring to fig. 24, for a schematic structural diagram of the nth layer observation matrix establishing module provided by the present invention, the nth layer observation matrix establishing module 41 further includes:

the nth layer equivalent frequency offset range determining module 411 calculates nth layer equivalent frequency offset ranges respectively corresponding to each data subcarrier group according to the nth-1 layer equivalent frequency offset vector, the nth-1 layer sampling number and the scaling factor, and the formula is as follows:

the nth layer equivalent frequency offset sampling module 412 is configured to perform equivalent frequency offset sampling on the nth layer equivalent frequency offset range corresponding to each data subcarrier group according to the nth layer sampling number.

Alternatively, referring to fig. 25, which is a schematic structural diagram of the accuracy parameter determining module provided in the present invention, in the observation calculation layer n, the number of samples J corresponding to each data subcarrier group⁽ⁿ⁾In order to further control the frequency offset estimation precision, the frequency offset estimation apparatus further includes a precision parameter determination module, where the precision parameter determination module includes:

a second discrete error setting module 45, configured to set a second discrete error according to a second expected value of the equivalent frequency offset estimation error; the square of the second discrete error is two orders of magnitude smaller than a second expected value of the equivalent frequency offset estimation error;

a precision parameter determination module 46 for calculating the scaling factor, the nth layer sampling number and the number of observation calculation layers according to a formula, wherein δ₂For the second dispersion error, S is the scaling factor, N is the number of observation calculation layers, K is the number of data subcarrier groups, Q is the number of total subcarrier groups, J⁽ⁿ⁾Is the nth layer sample number.

As can be seen from the above embodiments, the frequency offset estimation apparatus at the receiving end receives the ofdm signal, which is a signal separated by a zero subcarrier group in a periodic comb shape, so as to effectively reduce interference between subcarriers, and can bring a large number of subcarriers into a data range of frequency offset estimation, thereby enlarging observation dimensions, and further ensuring the frequency offset estimation accuracy through observation calculation of a large amount of data; in addition, in the observation calculation, the frequency offset is recovered with high precision by using a layer-by-layer observation method, each layer of observation calculation has reasonable data processing complexity, and approaches the high-precision calculation result layer by layer, so that the method has high precision and strong practicability.

Referring to fig. 26, a schematic structural diagram of a fifth embodiment of an offset estimation apparatus in an OFDM system provided in a transmitting end, for example, in an uplink UE or a downlink base station, the apparatus includes:

a subcarrier diversity module 51, configured to divide subcarriers into a plurality of subcarrier sets in a frequency domain space;

a zero subcarrier setting module 52, configured to set a zero subcarrier at least one zero position of each subcarrier set, where the zero positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group;

a data subcarrier setting module 53 for setting data subcarriers at subcarrier positions other than the set-to-zero position in each subcarrier set; all data subcarriers with the same position in the subcarrier set form a data subcarrier group;

and a signal management module 54, configured to generate and send an orthogonal frequency offset multiplexing signal according to the zero subcarrier group and the data subcarrier group.

It can be seen from the above embodiments that, by periodically setting the zero subcarrier group, the frequency offset estimation apparatus at the transmitting end performs periodic comb-like separation on the ofdm signal in the frequency domain space, thereby effectively reducing interference between subcarriers, and thus performing frequency offset estimation using high-dimensional observation data in the frequency offset hierarchical estimation algorithm, and improving the frequency offset estimation accuracy.

Referring to fig. 27, a schematic diagram of a structure of a frequency offset estimation system in an OFDM system provided in the present invention is shown, where the frequency offset estimation system includes a first device 110 and a second device 120:

a first device 110, configured to receive an orthogonal frequency division multiplexing signal from a second device 120, where the orthogonal frequency division multiplexing signal includes multiple subcarrier sets, each subcarrier set includes at least one zero subcarrier and multiple data subcarriers, and the zero subcarriers in each subcarrier set have the same position; the zero subcarriers with the same position in all the subcarrier sets form a zero subcarrier group, and the data subcarriers with the same position in all the subcarrier sets form a data subcarrier group; acquiring control information of the OFDM signal, wherein the control information comprises information of a subcarrier group, position information of a zero subcarrier group and position information of a data subcarrier group; extracting a subcarrier signal from the orthogonal frequency division multiplexing signal according to the control information to generate a signal matrix; and observing the signal matrix to obtain a frequency offset estimation value of the orthogonal frequency division multiplexing signal.

A second device 120 for dividing the subcarriers into a plurality of subcarrier sets in a frequency domain space; zero subcarriers are arranged at least one zero setting position of each subcarrier set, and the zero setting positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group; setting data subcarriers at subcarrier positions other than the set-zero position in each subcarrier set; all data subcarriers with the same position in the subcarrier set form a data subcarrier group; and generating an orthogonal frequency offset multiplexing signal according to the zero subcarrier group and the data subcarrier group, and sending the orthogonal frequency offset multiplexing signal to the first device 110.

As can be seen from the foregoing embodiments, in the frequency offset estimation system in the OFDM system provided in the embodiments of the present invention, the second device 120 sends an orthogonal frequency division multiplexing signal to the first device 110, and the first device 110 receives the orthogonal frequency division multiplexing signal from the second device 120, where the orthogonal frequency division multiplexing signal is a signal separated by a zero subcarrier group periodic comb, so as to effectively reduce interference between subcarriers, and a large number of subcarriers can be brought into a data range of frequency offset estimation, thereby enlarging an observation dimension, and further ensuring precision of frequency offset estimation through observation calculation of a large data volume; in addition, in the observation calculation, the frequency offset is recovered with high precision by using a layer-by-layer observation method, each layer of observation calculation has reasonable data processing complexity, and approaches the high-precision calculation result layer by layer, so that the method has high precision and strong practicability.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functions of the units may be implemented in the same software and/or hardware or in a plurality of software and/or hardware when implementing the invention.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for apparatus or system embodiments, since they are substantially similar to method embodiments, they are described in relative terms, as long as they are described in partial descriptions of method embodiments. The above-described embodiments of the apparatus and system are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The foregoing are merely exemplary embodiments of the present invention, which enable those skilled in the art to understand or practice the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

A method for estimating frequency offset in an OFDM system, comprising:

receiving an orthogonal frequency division multiplexing signal, wherein the orthogonal frequency division multiplexing signal comprises a plurality of subcarrier sets, each subcarrier set comprises at least one zero subcarrier and a plurality of data subcarriers, and the zero subcarrier positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group, and data subcarriers with the same position in all subcarrier sets form a data subcarrier group;

acquiring control information of the OFDM signal, wherein the control information comprises information of a subcarrier group, position information of a zero subcarrier group and position information of a data subcarrier group;

extracting a subcarrier signal from the orthogonal frequency division multiplexing signal according to the control information to generate a signal matrix;

and observing the signal matrix to obtain a frequency offset estimation value of the orthogonal frequency division multiplexing signal.
The method of claim 1, wherein the observing the signal matrix to obtain the frequency offset estimation value of the OFDM signal comprises:

establishing an observation matrix from the signal matrix according to the equivalent frequency offset range and the sampling number;

carrying out observation iterative computation on the observation matrix to obtain an observation iterative computation result, wherein the dimension of the observation iterative computation result is equal to the column number of the observation matrix;

obtaining an equivalent frequency offset result vector from the observation matrix according to an observation iterative computation result;

and calculating a frequency offset estimation value according to the equivalent frequency offset result vector.
The method of claim 1, wherein observing the signal matrix to obtain the estimated frequency offset of the OFDM signal comprises:

according to the number of the first layer of samples, carrying out equivalent frequency offset sampling in a first layer of equivalent frequency offset range, and establishing a first layer of observation matrix from the signal matrix according to the sampled equivalent frequency offset;

carrying out observation iterative computation on the first layer of observation matrix and obtaining a first layer of observation iterative computation result;

obtaining a first layer of equivalent frequency offset result vectors from the first layer of observation matrixes according to the first layer of observation iterative computation results;

and calculating a frequency offset estimation value according to the first layer equivalent frequency offset result vector.
The method of claim 3, wherein observing the signal matrix to obtain the estimated frequency offset of the OFDM signal comprises:

according to the sampling number of the nth layer, carrying out equivalent frequency offset sampling in the equivalent frequency offset range of the nth layer, and establishing an nth layer observation matrix from the signal matrix according to the sampled equivalent frequency offset, wherein n is greater than or equal to 2;

carrying out observation iterative computation on the nth layer of observation matrix to obtain an nth layer of observation iterative computation result;

obtaining an nth layer equivalent frequency offset result vector from the nth layer observation matrix according to the nth layer observation iterative computation result;

and calculating a frequency offset estimation value according to the final layer of equivalent frequency offset result vector.
The method of claim 2, wherein the iterative observation calculation comprises:

establishing a first sparse hyperparameter vector;

obtaining a diagonal matrix according to the first sparse hyperparametric vector;

calculating a first posterior parameter (Mm) and a second posterior parameter (Sigma) according to the diagonal matrix;

obtaining a second sparse hyperparameter vector according to the first posterior parameter (Mm) and the second posterior parameter (Sigma);

judging whether the difference value of the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a preset convergence threshold value;

if the difference value between the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a convergence threshold value, the observation iterative computation is completed, and the second sparse hyperparameter vector is used as an observation iterative computation result; alternatively, the first and second electrodes may be,

and if the difference value between the second sparse hyperparameter vector and the first sparse hyperparameter vector is greater than or equal to a convergence threshold value, taking the second sparse hyperparameter vector as the first sparse hyperparameter vector, and obtaining the second sparse hyperparameter vector again according to the first posterior parameter (M) and the second posterior parameter (Sigma).
The method of claim 2, wherein calculating the frequency offset estimation value according to the equivalent frequency offset result vector comprises:

extracting an equivalent frequency offset value corresponding to each data subcarrier group from the equivalent frequency offset result vector;

calculating a group frequency offset estimation value of each data subcarrier group according to the equivalent frequency offset value;

and carrying out average calculation on the group of frequency deviation estimated values to obtain the frequency deviation estimated values.
The method of claim 3, wherein said performing equivalent frequency offset sampling comprises:

determining the first layer equivalent frequency deviation range corresponding to each data subcarrier group according to the number of the total subcarrier groups, wherein the formula is as follows:

the first layer equivalent frequency offset value corresponding to the kth data subcarrier group is shown, K is the number of the data subcarrier groups, and Q is the number of the total subcarrier groups;

and carrying out equivalent frequency offset sampling on the first layer of equivalent frequency offset range corresponding to each data subcarrier group according to the first layer of sampling number respectively.
The method of claim 4, wherein the performing equivalent frequency offset sampling comprises:

calculating the equivalent frequency offset range of the nth layer corresponding to each data subcarrier group according to the equivalent frequency offset vector of the nth-1 layer, the sampling number of the nth-1 layer and the scaling factor, wherein the formula is as follows:

the equivalent frequency offset value corresponding to the kth data subcarrier group in the (n-1) th layer equivalent frequency offset result vector is the equivalent frequency offset value corresponding to the kth data subcarrier group in the nth layer, S is a scaling factor, Q is the number of the total subcarrier groups and is the sampling number of the (n-1) th layer of the k data subcarrier group;

and carrying out equivalent frequency offset sampling on the nth layer of equivalent frequency offset range corresponding to each data subcarrier group according to the nth layer of sampling number respectively.
The method of frequency offset estimation in an OFDM system according to claim 3, further comprising:

setting a first discrete error according to a first expected value of the equivalent frequency offset estimation error;

and calculating the number of first layer samples according to the first discrete error and the grouping information of the orthogonal frequency division multiplexing signal.
The method of frequency offset estimation in an OFDM system according to claim 4, further comprising:

setting a second discrete error according to a second expected value of the equivalent frequency offset estimation error;

calculating the scaling factor and the n-th layer sample number according to a formula, wherein delta₂For the second dispersion error, S is the scaling factor, N is the number of observation calculation layers, K is the number of data subcarrier groups, Q is the number of total subcarrier groups, J⁽ⁿ⁾Is the nth layer sample number.
The method of claim 2, wherein obtaining an equivalent frequency offset result vector from the observation matrix according to the observation iterative computation result comprises:

determining K maximum peak positions in the observation iteration calculation result, wherein K is the number of data subcarrier groups;

extracting a corresponding observation matrix column vector from the observation matrix according to the maximum peak position;

and extracting equivalent frequency offset values corresponding to the column vectors of the observation matrix, and establishing equivalent frequency offset result vectors.
The method of claim 11, wherein determining K maximum peak positions within the observation iteration calculation comprises:

dividing elements in the observation iteration calculation result into a plurality of calculation result groups according to the sampling number, wherein the number of the calculation result groups is equal to the number of the data subcarrier groups;

finding a maximum peak within each of said sets of calculations;

and taking the position of the maximum peak in the observation iteration calculation result as the maximum peak position.
A method for estimating frequency offset in an OFDM system, comprising:

dividing the subcarriers into a plurality of subcarrier sets in a frequency domain space;

zero subcarriers are arranged at least one zero setting position of each subcarrier set, and the zero setting positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group;

setting data subcarriers at subcarrier positions other than the set-zero position in each subcarrier set; all data subcarriers with the same position in the subcarrier set form a data subcarrier group;

and generating and sending an orthogonal frequency offset multiplexing signal according to the zero subcarrier group and the data subcarrier group.
An offset estimation apparatus in an OFDM system, comprising:

the OFDM signal receiving module is used for receiving OFDM signals, the OFDM signals comprise a plurality of subcarrier sets, each subcarrier set comprises at least one zero subcarrier and a plurality of data subcarriers, and the zero subcarrier positions in each subcarrier set are the same; the zero subcarriers with the same position in all the subcarrier sets form a zero subcarrier group, and the data subcarriers with the same position in all the subcarrier sets form a data subcarrier group;

a control information obtaining module, configured to obtain control information of the ofdm signal, where the control information includes: information of subcarrier groups, position information of zero subcarrier groups, and position information of data subcarrier groups;

a signal matrix generating module, configured to extract a subcarrier signal from the ofdm signal according to the control information from the control information acquiring module, and generate a signal matrix;

and the frequency offset estimation value acquisition module is used for observing the signal matrix to obtain the frequency offset estimation value of the orthogonal frequency division multiplexing signal.
The apparatus for estimating frequency offset in an OFDM system according to claim 14, wherein the means for obtaining frequency offset estimation value comprises:

the observation matrix establishing module is used for establishing an observation matrix from the signal matrix according to the equivalent frequency offset range and the sampling number;

the observation iterative computation module is used for performing observation iterative computation on the observation matrix and obtaining an observation iterative computation result, and the dimensionality of the observation iterative computation result is equal to the column number of the observation matrix;

the equivalent frequency offset result acquisition module is used for acquiring an equivalent frequency offset result vector from an observation matrix according to the observation iterative computation result acquired by the observation iterative computation module;

and the frequency offset estimation value calculation module is used for calculating the frequency offset estimation value according to the equivalent frequency offset result vector.
The apparatus for estimating frequency offset in an OFDM system according to claim 14, wherein the means for obtaining frequency offset estimation value comprises:

the first layer observation matrix obtaining module is used for carrying out equivalent frequency offset sampling in a first layer equivalent frequency offset range according to the first layer sampling number and establishing a first layer observation matrix from the signal matrix according to the sampled equivalent frequency offset;

the first layer observation iterative computation module is used for performing observation iterative computation on the first layer observation matrix and obtaining a first layer observation iterative computation result;

a first layer equivalent frequency offset result obtaining module, configured to obtain a first layer equivalent frequency offset result vector from the first layer observation matrix according to the first layer observation iterative computation result;

and the first layer frequency offset estimation value calculation module is used for calculating a frequency offset estimation value according to the first layer equivalent frequency offset result vector.
The apparatus for estimating frequency offset in an OFDM system according to claim 14, wherein the means for obtaining frequency offset estimation value comprises:

the nth layer observation matrix establishing module is used for carrying out equivalent frequency offset sampling in the nth layer equivalent frequency offset range according to the nth layer sampling number, and establishing an nth layer observation matrix from the signal matrix according to the sampled equivalent frequency offset, wherein n is greater than or equal to 2;

the nth layer observation iterative computation module is used for performing observation iterative computation on the signal matrix and obtaining an nth layer observation iterative computation result;

an nth layer equivalent frequency offset result vector obtaining module, configured to obtain an nth layer equivalent frequency offset result vector from the nth layer observation matrix according to the nth layer observation iterative computation result;

and the final layer frequency offset estimation value calculation module is used for calculating a frequency offset estimation value according to the final layer equivalent frequency offset result vector.
The offset estimation apparatus in the OFDM system according to claim 15, wherein the observation iteration calculating module comprises:

the first sparse hyperparameter vector establishing module is used for establishing a first sparse hyperparameter vector;

a diagonal matrix obtaining module, configured to obtain a diagonal matrix according to the first sparse hyperparametric vector;

the posterior parameter calculation module is used for calculating a first posterior parameter (Mm) and a second posterior parameter (Sigma) according to the diagonal matrix;

a second sparse hyperparametric vector obtaining module for obtaining a second sparse hyperparametric vector according to the first posterior parameter (Mm) and the second posterior parameter (Sigma);

a convergence judging module for judging whether the difference value of the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a preset convergence threshold value;

an observation iteration calculation ending module, configured to, according to the determination result of the convergence determination module, complete the observation iteration calculation if a difference between the second sparse hyperparameter vector and the first sparse hyperparameter vector is smaller than a convergence threshold, and use the second sparse hyperparameter vector as an observation iteration calculation result;

and the observation iteration calculation circulating module is used for taking the current second sparse hyperparametric vector as the first sparse hyperparametric vector and obtaining the second sparse hyperparametric vector again according to the first posterior parameter (M) and the second posterior parameter (Sigma) if the difference value between the second sparse hyperparametric vector and the first sparse hyperparametric vector is greater than or equal to a convergence threshold value according to the judgment result of the convergence judgment module.
The apparatus of claim 15, wherein the means for calculating the frequency offset estimate comprises:

an equivalent frequency offset value extraction module, configured to extract an equivalent frequency offset value corresponding to each data subcarrier group from the equivalent frequency offset result vector;

the group frequency offset estimation value calculation module is used for calculating the group frequency offset estimation value of each data subcarrier group according to the equivalent frequency offset value;

and the group frequency offset estimation value averaging module is used for carrying out average calculation on the group frequency offset estimation value to obtain the frequency offset estimation value.
The apparatus for frequency offset estimation in an OFDM system according to claim 16, wherein the first layer observation matrix obtaining module further comprises:

a first layer equivalent frequency offset range determining module, configured to determine, according to the number of total subcarrier groups, a first layer equivalent frequency offset range corresponding to each data subcarrier group, where the formula is as follows:

the first layer equivalent frequency offset value corresponding to the kth data subcarrier group is shown, K is the number of the data subcarrier groups, and Q is the number of the total subcarrier groups;

and the first layer equivalent frequency offset sampling module is used for carrying out equivalent frequency offset sampling on the first layer equivalent frequency offset range corresponding to each data subcarrier group according to the first layer sampling number respectively.
The apparatus for frequency offset estimation in an OFDM system according to claim 17, wherein the nth layer observation matrix establishing module further comprises:

the nth layer equivalent frequency offset range determining module is used for calculating the nth layer equivalent frequency offset ranges respectively corresponding to each data subcarrier group according to the nth-1 layer equivalent frequency offset vector, the nth-1 layer sampling number and the scaling factor, and the formula is as follows:

the equivalent frequency offset value corresponding to the kth data subcarrier group in the (n-1) th layer equivalent frequency offset result vector is the equivalent frequency offset value corresponding to the kth data subcarrier group in the nth layer, S is a scaling factor, Q is the number of the total subcarrier groups and is the sampling number of the (n-1) th layer of the k data subcarrier group;

and the nth layer equivalent frequency offset sampling module is used for carrying out equivalent frequency offset sampling on the nth layer equivalent frequency offset range corresponding to each data subcarrier group according to the nth layer sampling number respectively.
The offset estimation apparatus in an OFDM system according to claim 16, further comprising a first layer sample number determination module, wherein the first layer sample number determination module comprises:

the first discrete error setting module is used for setting a first discrete error according to a first expected value of the equivalent frequency offset estimation error;

and the first layer sampling number calculating module is used for calculating the first layer sampling number according to the first discrete error and the grouping information of the orthogonal frequency division multiplexing signal.
The apparatus for frequency offset estimation in an OFDM system according to claim 17, further comprising a precision parameter determination module, wherein the precision parameter determination module comprises:

the second discrete error setting module is used for setting a second discrete error according to a second expected value of the equivalent frequency offset estimation error;

a precision parameter determination module for calculating the scaling factor, the n-th layer sampling number and the observation calculation layer number according to a formula, wherein delta₂For the second dispersion error, S is the scaling factor, N is the number of observation calculation layers, K is the number of data subcarrier groups, Q is the number of total subcarrier groups, J⁽ⁿ⁾Is the nth layer sample number.
The apparatus of claim 15, wherein the equivalent frequency offset result obtaining module comprises:

a maximum peak position determining module, configured to determine K maximum peak positions in the observation iteration calculation result, where K is the number of data subcarrier groups;

the observation matrix column vector extraction module is used for extracting a corresponding observation matrix column vector from the observation matrix according to the maximum peak position;

and the equivalent frequency offset result vector establishing module is used for establishing an equivalent frequency offset result vector according to the data subcarrier group sequence by using the equivalent frequency offset value corresponding to the observation matrix column vector.
The offset estimation apparatus in an OFDM system according to claim 24, wherein the maximum peak position determination module comprises:

the calculation result grouping module is used for dividing the observation iteration calculation result into a plurality of calculation result groups according to the sampling number, and the number of the calculation result groups is equal to the number of the data subcarrier groups;

a group maximum peak determination module for finding a maximum peak in each of said calculation result groups;

and the group maximum peak position acquisition module is used for taking the position of the maximum peak in the observation iteration calculation result as the maximum peak position.
An offset estimation apparatus in an OFDM system, comprising:

the subcarrier diversity module is used for dividing the subcarriers into a plurality of subcarrier sets in a frequency domain space;

a zero subcarrier setting module, configured to set a zero subcarrier at least one zero position of each subcarrier set, where the zero positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group;

the data subcarrier setting module is used for setting data subcarriers at subcarrier positions except the position where each subcarrier is set to be zero in a centralized manner; all data subcarriers with the same position in the subcarrier set form a data subcarrier group;

and the signal management module is used for generating and sending an orthogonal frequency offset multiplexing signal according to the zero subcarrier group and the data subcarrier group.
An offset estimation system in an OFDM system, the system comprising a first device and a second device, wherein:

the first device is configured to receive an orthogonal frequency division multiplexing signal from a second device, where the orthogonal frequency division multiplexing signal includes multiple subcarrier sets, each subcarrier set includes at least one zero subcarrier and multiple data subcarriers, and the zero subcarriers in each subcarrier set have the same position; the zero subcarriers with the same position in all the subcarrier sets form a zero subcarrier group, and the data subcarriers with the same position in all the subcarrier sets form a data subcarrier group; acquiring control information of the OFDM signal, wherein the control information comprises information of a subcarrier group, position information of a zero subcarrier group and position information of a data subcarrier group; extracting a subcarrier signal from the orthogonal frequency division multiplexing signal according to the control information to generate a signal matrix; observing the signal matrix to obtain a frequency offset estimation value of the orthogonal frequency division multiplexing signal;

the second device is configured to divide the subcarriers into a plurality of subcarrier sets in a frequency domain space; zero subcarriers are arranged at least one zero setting position of each subcarrier set, and the zero setting positions in each subcarrier set are the same; zero subcarriers with the same position in all subcarrier sets form a zero subcarrier group; setting data subcarriers at subcarrier positions other than the set-zero position in each subcarrier set; all data subcarriers with the same position in the subcarrier set form a data subcarrier group; and generating an orthogonal frequency offset multiplexing signal according to the zero subcarrier group and the data subcarrier group and sending the orthogonal frequency offset multiplexing signal to the first equipment.