CN113676237A

CN113676237A - Linear equalization method, device, equipment and medium

Info

Publication number: CN113676237A
Application number: CN202110886841.7A
Authority: CN
Inventors: 余昌学
Original assignee: Guangzhou Huiruisitong Technology Co Ltd
Current assignee: Guangzhou Huiruisitong Technology Co Ltd
Priority date: 2021-08-03
Filing date: 2021-08-03
Publication date: 2021-11-19

Abstract

The present disclosure relates to a linear equalization method, apparatus, device and medium, and relates to the field of communication technology, wherein the method comprises: acquiring a first noise and interference covariance matrix corresponding to a target subcarrier band according to a data symbol, wherein the target subcarrier band is a subcarrier band corresponding to a subcarrier band number to which the data symbol belongs; carrying out equalization processing on the data symbols according to a first noise and interference covariance matrix corresponding to the target subcarrier band to obtain signal-to-interference-and-noise ratio information and a first equalization signal; carrying out inverse scaling restoration on the first equalized signal according to a restoration coefficient to generate a second equalized signal, wherein the restoration coefficient is a coefficient determined according to the signal-to-interference-and-noise ratio information; and performing soft demodulation processing according to the second equalization signal to obtain soft bit information corresponding to the data symbol. According to the method and the device, the first equalization signal is subjected to inverse scaling restoration according to the restoration coefficient, and the influence caused by scaling is removed, so that the constellation amplitude of the equalization signal is more accurate.

Description

Linear equalization method, device, equipment and medium

Technical Field

The present disclosure relates to the field of communications technologies, and in particular, to a linear equalization method, apparatus, device, and medium.

Background

In a New 5th Generation mobile communication technology air-interface (5G-NR) system, a large-scale Multiple-Input Multiple-Output (MIMO) technology is proposed. For uplink reception, when the number of user equipment in other cells with the same frequency is larger than that of the user equipment in other cells with the same frequency, the interference is more serious. When the ue has multiple antennas for reception, a special reception algorithm may be used to eliminate the interference, such as Maximum Likelihood Detection (MLD) algorithm. Although the MLD algorithm has the best performance theoretically, the MLD algorithm belongs to a nonlinear algorithm, the complexity is high, and the implementation is complex.

Disclosure of Invention

The related linear equalization method mainly performs equalization by selecting a Maximum Ratio Combining (MRC) algorithm and an Interference Rejection Combining (IRC) algorithm, so as to obtain the advantages of the IRC algorithm under the condition of Interference and the MRC algorithm under the condition of no Interference; however, the method and device design cannot reach the most favorable state, and the system performance is affected.

In view of this, the present disclosure provides a linear equalization method, apparatus, device, and medium to improve the anti-interference performance of the linear equalization scheme, and reduce the bit error rate of the receiving end, thereby improving the throughput of the system.

In a first aspect, an embodiment of the present disclosure provides a linear equalization method, including:

acquiring a first noise and interference covariance matrix corresponding to a target subcarrier band according to a data symbol, wherein the target subcarrier band is a subcarrier band corresponding to a subcarrier band number to which the data symbol belongs;

carrying out equalization processing on the data symbols according to a first noise and interference covariance matrix corresponding to the target subcarrier band to obtain signal-to-interference-and-noise ratio information and a first equalization signal;

carrying out inverse scaling restoration on the first equalized signal according to a restoration coefficient to generate a second equalized signal, wherein the restoration coefficient is a coefficient determined according to the signal-to-interference-and-noise ratio information;

and performing soft demodulation processing according to the second equalization signal to obtain soft bit information corresponding to the data symbol.

Optionally, the equalizing the data symbol according to the first noise and interference covariance matrix corresponding to the target subcarrier band to obtain the signal-to-interference-and-noise ratio information and a first equalized signal includes:

calculating based on the first noise and interference covariance matrix and the channel estimation value corresponding to the data symbol, and taking a calculation result as the signal-to-interference-and-noise ratio information;

and performing channel equalization on the receiving antenna signal corresponding to the data symbol according to the SINR information, the first noise and interference covariance matrix and the channel estimation value to obtain the first equalized signal.

Optionally, the sir information is used to scale the constellation amplitude of the first equalized signal, and before performing inverse scaling restoration on the first equalized signal according to a restoration coefficient, the method further includes: and determining the repair coefficient based on the real part data of the signal-to-interference-and-noise ratio information.

Optionally, after obtaining the signal to interference plus noise ratio information and the first equalized signal, the method further includes: estimating a power ratio according to the SINR information to obtain a target SINR; and carrying out weighting processing on the soft bit information according to the target signal-to-noise ratio to obtain weighted confidence information.

Optionally, the obtaining a first noise and interference covariance matrix corresponding to a target subcarrier band according to a data symbol includes:

extracting the subcarrier band number to which the data symbol belongs;

and extracting a corresponding noise interference covariance inverse matrix according to the subcarrier band number to be used as the first noise and interference covariance matrix.

Optionally, the numbering of the subcarrier band is a numbering of a subcarrier band, and before the obtaining of the first noise and interference covariance matrix corresponding to the target subcarrier band according to the data symbol, the method further includes:

determining a noise-to-interference ratio for each sub-carrier band;

if the noise interference ratio of the subcarrier band is larger than the threshold value, carrying out inversion processing on the noise and interference covariance matrix of the subcarrier band according to a maximum ratio combining MRC algorithm to obtain a noise interference covariance inverse matrix;

and if the noise interference ratio of the subcarrier band is not larger than the threshold value, combining an IRC algorithm according to the interference rejection ratio to perform inversion processing on the noise and interference covariance matrix of the subcarrier band to obtain the noise interference covariance inverse matrix.

Optionally, the performing, according to the maximal ratio combining MRC algorithm, an inversion process on the noise and interference covariance matrix of the subcarrier band to obtain the noise and interference covariance inverse matrix includes:

and calculating the reciprocal of each main diagonal element in the noise and interference covariance matrix of the subcarrier band to obtain an inverse matrix which is used as a noise interference covariance inverse matrix corresponding to the subcarrier band.

Optionally, the obtaining the noise-interference covariance inverse matrix by performing inversion processing on the noise-interference covariance matrix of the subcarrier band according to the interference rejection ratio combining IRC algorithm includes:

determining the number of antennas;

if the number of the antennas is equal to two, inverting the noise and interference covariance matrix of the subcarrier band by using a determinant inversion method to obtain a corresponding inverse matrix which is used as a noise and interference covariance inverse matrix corresponding to the subcarrier band;

if the number of the antennas is more than two, inverting the noise and interference covariance matrix of the subcarrier band by using a cyclic index inversion method to obtain a corresponding inverse matrix, wherein the corresponding inverse matrix is used as the noise interference covariance inverse matrix corresponding to the subcarrier band.

Optionally, the determining the noise-to-interference ratio of each subcarrier band includes:

determining a ratio of an antenna noise interference energy correlation result and an antenna interference correlation result as the noise interference ratio for each subcarrier band, wherein the antenna noise interference energy correlation result is a sum of products of elements on a target diagonal line, the target diagonal line is a main diagonal line of a noise and interference covariance matrix of the subcarrier band, the antenna interference correlation result is a sum of products of elements conjugate to each other in a first triangular matrix and a second triangular matrix in the noise and interference covariance matrix of the subcarrier band, and elements of the first triangular matrix and elements of the second triangular matrix do not include elements on the main diagonal line.

Optionally, before determining the noise-to-interference ratio of each subcarrier band, the method further includes:

according to resource block information distributed to a user terminal, sub-carrier division is carried out on time-frequency domain resources of the user terminal to obtain at least two sub-carrier wave bands;

and determining a noise and interference covariance matrix of the subcarrier band according to a noise and interference covariance matrix corresponding to a target reference symbol, wherein the target reference symbol is a reference symbol contained in the subcarrier band.

Optionally, the resource block information includes resource block size information and resource block location information, and the sub-carrier division is performed on the time-frequency domain resource of the user terminal according to the resource block information allocated to the user terminal, so as to obtain at least two sub-carrier bands, including:

acquiring size information and position information of resource blocks allocated to a user terminal;

and based on the resource block size information and the resource block position information, performing subcarrier division on the time-frequency domain resources of the user terminal according to the number of subcarrier resource blocks to obtain at least two subcarrier bands, wherein the number of the subcarrier resource blocks is the number of resource blocks in a preset frequency domain bandwidth.

Optionally, the determining the noise and interference covariance matrix of the subcarrier band according to the noise and interference covariance matrix corresponding to the target reference symbol includes:

determining a target reference symbol belonging to the subcarrier band based on a subcarrier band number, the subcarrier band number of the subcarrier band to which the target reference symbol belongs being related to a subcarrier number of the target reference symbol;

accumulating the noise and interference covariance matrixes corresponding to the target reference symbols in the subcarrier band to obtain a second noise and interference covariance matrix of the subcarrier band;

and performing time domain filtering according to the second noise and interference covariance matrix of the subcarrier band to obtain the noise and interference covariance matrix of the subcarrier band.

Optionally, the performing time-domain filtering according to the second noise and interference covariance matrix of the subcarrier band to obtain the noise and interference covariance matrix of the subcarrier band includes:

acquiring a subcarrier band covariance matrix of a current time slot and a subcarrier band covariance matrix of a reference time slot, wherein the subcarrier band covariance matrix of the current time slot is a second noise and interference covariance matrix of the subcarrier band of the current time slot, and the subcarrier band covariance matrix of the reference time slot is a second noise and interference covariance matrix of the subcarrier band of a time slot before the current time slot;

and according to a preset time domain filter coefficient, carrying out weighting processing on the subcarrier band covariance matrix of the current time slot and the subcarrier band covariance matrix of the reference time slot.

In a second aspect, an embodiment of the present disclosure provides a linear equalization apparatus, including:

a covariance matrix obtaining module, configured to obtain, according to a data symbol, a first noise and interference covariance matrix corresponding to a target subcarrier band, where the target subcarrier band is a subcarrier band corresponding to a subcarrier band number to which the data symbol belongs;

the equalization processing module is used for carrying out equalization processing according to the first noise and interference covariance matrix corresponding to the target subcarrier band to obtain signal-to-interference-and-noise ratio information and a first equalization signal;

the inverse scaling restoration module is used for performing inverse scaling restoration on the first equalization signal according to a restoration coefficient to generate a second equalization signal, wherein the restoration coefficient is a coefficient determined according to the SINR information;

and the soft demodulation processing module is used for performing soft demodulation processing according to the second equalization signal to obtain soft bit information corresponding to the data symbol.

In a third aspect, an embodiment of the present disclosure provides a communication device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete mutual communication through the communication bus; a memory for storing a computer program; a processor, configured to implement the steps of the linear equalization method according to any one of the first aspect when executing a program stored in the memory.

Optionally, the equalization processing module may include the following sub-modules:

a calculation submodule, configured to perform calculation based on the first noise and interference covariance matrix and a channel estimation value corresponding to the data symbol, so as to use a calculation result as the signal-to-interference-and-noise ratio information;

and the channel equalization submodule is used for carrying out channel equalization according to the signal-to-interference-and-noise ratio information, the first noise and interference covariance matrix, the channel estimation value and the receiving antenna signal corresponding to the data symbol to obtain the first equalization signal.

Optionally, the sir signal is used to scale the constellation amplitude of the first equalized signal, and the linear equalization apparatus further includes: and the repair coefficient determining module is used for determining the repair coefficient based on the real part data of the signal to interference and noise ratio information.

Optionally, the interference estimation apparatus may further include: the device comprises a power ratio estimation module and a weighting processing module. The power ratio estimation module is used for estimating the power ratio according to the signal to interference plus noise ratio information to obtain a target signal to interference plus noise ratio; and the weighting processing is used for carrying out weighting processing on the soft bit information according to the target signal-to-noise ratio to obtain weighted confidence information.

Optionally, the covariance matrix obtaining module may include the following sub-modules:

a subcarrier band number extraction submodule for extracting a subcarrier band number to which the data symbol belongs;

and the noise interference covariance inverse matrix extraction submodule is used for extracting a corresponding noise interference covariance inverse matrix according to the subcarrier band number to be used as the first noise and interference covariance matrix.

Optionally, the number of the subcarrier band is the number of the subcarrier band, and the linear equalization apparatus further includes:

a noise-to-interference ratio determination module for determining a noise-to-interference ratio for each sub-carrier band;

the first inversion processing module is used for performing inversion processing on the noise and interference covariance matrix of the subcarrier band according to a maximum ratio combining MRC algorithm when the noise-interference ratio of the subcarrier band is larger than a threshold value to obtain the noise-interference covariance inverse matrix;

and the second inversion processing module is used for merging an IRC algorithm according to the interference rejection ratio to perform inversion processing on the noise and interference covariance matrix of the subcarrier band when the noise-interference ratio of the subcarrier band is not greater than the threshold value, so as to obtain the noise-interference covariance inverse matrix.

Optionally, the first inversion processing module is specifically configured to: and calculating the reciprocal of each main diagonal element in the noise and interference covariance matrix of the subcarrier band to obtain an inverse matrix of the noise and interference covariance matrix, wherein the inverse matrix is used as a noise interference covariance inverse matrix corresponding to the subcarrier band.

Optionally, the second inversion processing module is specifically configured to: determining the number of antennas; if the number of the antennas is equal to two, inverting the noise and interference covariance matrix of the subcarrier band by using a determinant inversion method to obtain a corresponding inverse matrix which is used as a noise and interference covariance inverse matrix corresponding to the subcarrier band; if the number of the antennas is more than two, inverting the noise and interference covariance matrix of the subcarrier band by using a cyclic index inversion method to obtain a corresponding inverse matrix, wherein the corresponding inverse matrix is used as the noise interference covariance inverse matrix corresponding to the subcarrier band.

Optionally, the noise-to-interference ratio determining module is specifically configured to: determining a ratio of an antenna noise interference energy correlation result and an antenna interference correlation result as the noise interference ratio for each subcarrier band, wherein the antenna noise interference energy correlation result is a sum of products of elements on a target diagonal line, the target diagonal line is a main diagonal line of a noise and interference covariance matrix of the subcarrier band, the antenna interference correlation result is a sum of products of elements conjugate to each other in a first triangular matrix and a second triangular matrix in the noise and interference covariance matrix of the subcarrier band, and elements of the first triangular matrix and elements of the second triangular matrix do not include elements on the main diagonal line.

Optionally, the linear equalization apparatus further includes:

the system comprises a subcarrier dividing module, a resource block allocating module and a resource block allocating module, wherein the subcarrier dividing module is used for dividing time-frequency domain resources of a user terminal according to resource block information allocated to the user terminal to obtain at least two subcarrier bands;

and the subcarrier band covariance matrix determination module is used for determining a noise and interference covariance matrix of the subcarrier band according to a noise and interference covariance matrix corresponding to a target reference symbol, wherein the target reference symbol is a reference symbol contained in the subcarrier band.

Optionally, the resource block information may include resource block size information and resource block location information, and the subcarrier dividing module may include the following sub-modules:

a resource block information obtaining submodule for obtaining size information and position information of resource blocks allocated to the user terminal;

and the subcarrier dividing submodule is used for carrying out subcarrier division on the time-frequency domain resources of the user terminal according to the number of the subcarrier resource blocks based on the size information and the position information of the resource blocks to obtain at least two subcarrier bands, wherein the number of the subcarrier resource blocks is the number of resource blocks in a preset frequency domain bandwidth.

Optionally, the subcarrier band covariance matrix determination module may include the following sub-modules:

a target reference symbol determination submodule for determining a target reference symbol belonging to the subcarrier band based on a subcarrier band number, the subcarrier band number of the subcarrier band to which the target reference symbol belongs being related to the subcarrier number of the target reference symbol;

the accumulation processing submodule is used for carrying out accumulation processing on the noise and interference covariance matrix corresponding to the target reference symbol in the subcarrier band to obtain a second noise and interference covariance matrix of the subcarrier band;

and the time domain filtering submodule is used for carrying out time domain filtering according to the second noise and interference covariance matrix of the subcarrier band to obtain the noise and interference covariance matrix of the subcarrier band.

Optionally, the time-domain filtering sub-module is specifically configured to: acquiring a subcarrier band covariance matrix of a current time slot and a subcarrier band covariance matrix of a reference time slot, wherein the subcarrier band covariance matrix of the current time slot is a second noise and interference covariance matrix of the subcarrier band of the current time slot, and the subcarrier band covariance matrix of the reference time slot is a second noise and interference covariance matrix of the subcarrier band of a time slot before the current time slot; and according to a preset time domain filter coefficient, carrying out weighting processing on the subcarrier band covariance matrix of the current time slot and the subcarrier band covariance matrix of the reference time slot.

In a fourth aspect, the disclosed embodiments provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of the linear equalization method according to any one of the first aspect.

According to the method and the device, the first noise and interference covariance matrix corresponding to the target subcarrier band is obtained according to the data symbol, and equalization processing is carried out according to the first noise and interference covariance matrix corresponding to the target subcarrier band to obtain the signal-to-interference-and-noise ratio information and the first equalization signal, so that the restoration coefficient corresponding to the data symbol can be determined according to the signal-to-interference-and-noise ratio information, the first equalization signal is restored by inverse scaling according to the restoration coefficient, the influence caused by scaling is removed, and the constellation amplitude of the equalization signal is more accurate; and then, soft demodulation processing can be carried out according to the second equalized signal generated after restoration, so that division operation in the soft demodulation process is reduced, and soft bit information corresponding to the data symbols is obtained. Furthermore, the implementation of the present disclosure may determine a signal-to-interference-and-noise ratio corresponding to the data symbol according to the signal-to-interference-and-noise ratio information, so as to perform weighting processing on the soft bit information according to the signal-to-interference ratio, and improve accuracy of a confidence level when a Hybrid Automatic Repeat Request (HARQ) is combined, thereby improving performance of soft decoding and achieving a purpose of improving system performance.

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.

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.

Fig. 1 is a flowchart illustrating steps of a linear equalization method according to an embodiment of the present disclosure;

FIG. 2 is a flowchart illustrating steps of a linear equalization method according to an alternative embodiment of the present disclosure;

fig. 3 is a schematic diagram of subcarrier band division in an embodiment of the present disclosure;

fig. 4 is a block diagram of a linear equalization apparatus according to an embodiment of the present disclosure;

fig. 5 is a schematic structural diagram of a linear equalization apparatus according to an embodiment of the present disclosure.

Detailed Description

To make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are some embodiments of the present disclosure, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.

In order to solve the problem that when a 5G-NR small cell receives a Physical Uplink Shared CHannel (PUSCH) signal sent by a User Equipment (UE), a receiving end needs to perform CHannel equalization on a received signal due to multipath CHannel fading and inter-symbol interference caused by co-CHannel interference generated by UEs in other cells.

One of the core concepts of the embodiments of the present disclosure is to provide an improved linear equalization method, so as to suppress co-channel cell interference without knowing interference parameters of other co-channel cells, improve the interference resistance of a small base station to PUSCH signals, and improve the throughput of a PUSCH receiving system.

Referring to fig. 1, a flowchart illustrating steps of a linear equalization method provided by an embodiment of the present disclosure is shown. The linear equalization method provided by the present disclosure may be applied to a channel equalization situation, and specifically may include the following steps:

step 110, obtaining a first noise and interference covariance matrix corresponding to a target subcarrier band according to a data symbol, where the target subcarrier band is a subcarrier band corresponding to a subcarrier band number to which the data symbol belongs.

Specifically, for each data symbol, the embodiment of the present disclosure may determine, according to a subcarrier band number i to which the data symbol belongs, a corresponding subcarrier band to serve as a target subcarrier band, and then may extract, according to the subcarrier band number, a covariance matrix of corresponding noise and interference as a first noise and interference covariance matrix corresponding to the target subcarrier band.

In an alternative embodiment, the covariance matrix of the noise and interference extracted according to the subcarrier band number i may be a noise interference covariance inverse matrix corresponding to the subcarrier band number. Further, the acquiring a first noise and interference covariance matrix corresponding to a target subcarrier band according to a data symbol in the embodiment of the present disclosure may specifically include: extracting the subcarrier band number to which the data symbol belongs; and extracting a corresponding noise interference covariance inverse matrix according to the subcarrier band number to be used as the first noise and interference covariance matrix. The noise interference covariance matrix may refer to an inverse matrix of a noise and interference covariance matrix of the subcarrier band.

In specific implementation, a suitable linear equalization algorithm may be selected according to the noise-to-interference ratio of the subcarrier band to serve as a target linear equalization algorithm corresponding to the subcarrier band, so that the inverse processing may be performed on the noise and interference covariance matrix of the subcarrier band according to the target linear equalization algorithm to obtain a noise-to-interference covariance inverse matrix. Further, before obtaining the first noise and interference covariance matrix corresponding to the target subcarrier band according to the data symbol, the linear equalization method provided in the embodiment of the present disclosure further includes: determining a noise-to-interference ratio for each sub-carrier band; if the noise interference ratio of the subcarrier band is larger than the threshold value, carrying out inversion processing on the noise and interference covariance matrix of the subcarrier band according to a maximum ratio combining MRC algorithm to obtain a noise interference covariance inverse matrix; and if the noise interference ratio of the subcarrier band is not larger than the threshold value, combining an IRC algorithm according to the interference rejection ratio to perform inversion processing on the noise and interference covariance matrix of the subcarrier band to obtain the noise interference covariance inverse matrix. Therefore, the target linear equalization algorithm corresponding to each sub-carrier wave band can be determined according to the noise interference ratio of each sub-carrier wave band; then, for each subcarrier band, the inverse processing may be performed on the noise and interference covariance matrix of the subcarrier band according to a target linear equalization algorithm to obtain a noise and interference covariance inverse matrix, so that the noise and interference covariance inverse matrix is extracted according to the subcarrier band number in the following procedure and is used as a first noise and interference covariance matrix corresponding to the target subcarrier band. The target subcarrier band is a subcarrier band corresponding to the subcarrier band number, and the subcarrier band number is the subcarrier band number.

And step 120, performing equalization processing according to the first noise and interference covariance matrix corresponding to the target subcarrier band to obtain signal-to-interference-and-noise ratio information and a first equalization signal.

Specifically, after the first noise and interference covariance matrix corresponding to the target subcarrier band is obtained, the equalization processing may be performed on the received data based on the first noise and interference covariance matrix to obtain an equalized signal, which is used as a first equalized signal, and meanwhile, the signal to interference and noise ratio information generated in the equalization process may be retained, so that the repair coefficient corresponding to the data symbol may be determined subsequently according to the signal to interference and noise ratio information. The sir information refers to power ratio information of a signal to noise and interference.

And step 130, performing inverse scaling restoration on the first equalized signal according to a restoration coefficient to generate a second equalized signal, where the restoration coefficient is a coefficient determined according to the signal-to-interference-and-noise ratio information.

Specifically, after the signal to interference plus noise ratio information is obtained, the embodiment of the present disclosure may determine a repair coefficient corresponding to the data symbol according to the signal to interference plus noise ratio information, and may perform inverse scaling repair on the first equalized signal according to the repair coefficient to remove an influence caused by scaling, so that the constellation amplitude of the equalized signal is more accurate. Further, the sir information in the embodiments of the present disclosure may be used to scale the constellation amplitude of the first equalized signal. Optionally, before performing inverse scaling restoration on the first equalized signal according to a restoration coefficient, the embodiment of the present disclosure may further include determining the restoration coefficient based on real part data of the signal to interference and noise ratio information.

Specifically, in the equalization process, the signal-to-interference-and-noise ratio information generated in the calculation process may be used to calculate a coefficient that each data symbol should be subjected to constellation amplitude restoration, that is, calculate a restoration coefficient corresponding to the data symbol, so that the subsequent equalized signal (that is, the first equalized signal) may be subjected to inverse scaling restoration by using the restoration coefficient corresponding to the data symbol, and a second equalized signal is generated. The second equalized signal may be an equalized signal generated by performing inverse scaling restoration on the first equalized signal according to the restoration coefficient. For example, the calculation formula of constellation amplitude restoration can be followed: x2 ═ X1 ·, β, and performing constellation amplitude restoration on the first equalized signal by using a restoration coefficient β to obtain a second equalized signal; it should be noted that the constellation amplitude repair is one of inverse scaling repairs; x2 may represent the second equalized signal generated after the repair, X1 may represent the first equalized signal before the repair, and X in the formula may represent a matrix dot product, and the repair coefficient β may be a constellation amplitude repair coefficient, which is not particularly limited in this example.

Step 140, performing soft demodulation processing according to the second equalized signal to obtain soft bit information corresponding to the data symbol.

Specifically, after obtaining the second equalized signal generated by performing inverse scaling restoration on the first equalized signal according to the restoration coefficient, the embodiment of the present disclosure may perform soft demodulation processing according to a preset soft demodulation formula based on the second equalized signal, to obtain soft bit information corresponding to the data symbol. The soft bit information may be used to represent soft bits after soft demodulation, and specifically may include soft bit information after soft demodulation of data symbols.

For example, in the case of the channel equalization output second equalized signal X2, it is possible to follow a preset soft demodulation formula:

soft demodulation processing is carried out to obtain soft bit information B_i,t. Wherein the soft bit information B_i,tMay be a confidence of a probability that an ith bit (bit) of a received signal of a v-th layer of a transmitted signal corresponding to the second equalized signal X2 is 0;

and

the ith bit of the transmission symbol of the v-th layer can be represented as a vector set with values of 1 and 0, respectively.

Therefore, the first noise and interference covariance matrix corresponding to the target subcarrier band is obtained according to the data symbol, and equalization processing is performed according to the first noise and interference covariance matrix corresponding to the target subcarrier band, so that the signal-to-interference-and-noise ratio information and the first equalization signal are obtained, and thus the restoration coefficient corresponding to the data symbol can be determined according to the signal-to-interference-and-noise ratio information, the first equalization signal is restored by inverse scaling according to the restoration coefficient, and the influence caused by scaling is removed, so that the constellation amplitude of the equalization signal is more accurate; and then, soft demodulation processing can be carried out according to the second equalized signal generated after restoration, so that division operation in the soft demodulation process is reduced, and soft bit information corresponding to the data symbols is obtained.

In specific implementation, the sir information generated in the equalization process may be used to determine the sir corresponding to the soft bit information, so that the sir may be subsequently used to perform weighting processing on the soft bits corresponding to the data symbols at different positions, thereby improving the accuracy of the confidence of the soft bits corresponding to the redundancy versions in different HARQ processes. Further, the linear equalization method provided by the embodiment of the present disclosure may further include, after obtaining the signal to interference plus noise ratio information and the first equalized signal: estimating a power ratio according to the SINR information to obtain a target SINR; and carrying out weighting processing on the soft bit information according to the target signal-to-noise ratio to obtain weighted confidence information. The target signal to interference plus noise ratio may refer to a signal to interference plus noise ratio corresponding to the soft bit information.

Specifically, the target sir calculated in the channel equalization process of this embodiment may represent an estimate of the power of the useful signal and the power of the interference and noise at each data symbol position, and may be a kind of a posteriori information. This indicates that the soft bit information after soft demodulation of the data symbols at different positions should have different confidence levels, for example, the higher the signal to interference and noise Ratio, the higher the confidence level of the corresponding Log Likelihood Ratio (LLR) at the position should be. After the demodulated soft bit information is obtained, the embodiment of the disclosure can perform weighting processing on the demodulated soft bit information according to the signal-to-interference-and-noise ratio, and use the result after the weighting processing as confidence information, thereby improving the accuracy of the confidence in HARQ combining and improving the performance of soft decoding.

Therefore, the first noise and interference covariance matrix corresponding to the target subcarrier band is obtained according to the data symbol, and equalization processing is performed according to the first noise and interference covariance matrix corresponding to the target subcarrier band, so that the signal-to-interference-and-noise ratio information and the first equalization signal are obtained, a restoration coefficient corresponding to the data symbol and a signal-to-interference-and-noise ratio corresponding to the soft bit information can be determined according to the signal-to-interference-and-noise ratio information, the first equalization signal is subjected to inverse scaling restoration according to the restoration coefficient, the influence caused by scaling is removed, and the constellation amplitude of the equalization signal is more accurate; then, soft demodulation processing can be performed according to the second equalized signal generated after restoration, division operation in the soft demodulation process is reduced, soft bit information corresponding to the data symbols is obtained, weighting processing can be performed on the soft bit information according to signal-to-interference-and-noise ratios, accuracy of confidence coefficient during HARQ merging is improved, accordingly, performance of soft decoding is improved, and the purpose of improving system performance is achieved.

In a specific implementation, the embodiment of the present disclosure may extract a channel estimation value from a channel estimation result for each data symbol in the received data, and may extract a receiving antenna signal from an Orthogonal Frequency Division Multiplexing (OFDM) demodulation processing result, so as to perform channel equalization according to the signal-to-interference-and-noise ratio information, the first noise and interference covariance matrix, the channel estimation value, and the receiving antenna signal, so as to obtain a first equalized signal. Optionally, on the basis of the foregoing embodiment, the equalizing processing is performed according to the first noise and interference covariance matrix corresponding to the target subcarrier band in the embodiment of the present disclosure to obtain the signal-to-interference-and-noise ratio information and the first equalized signal, which may specifically include: calculating based on the first noise and interference covariance matrix and the channel estimation value corresponding to the data symbol, and taking a calculation result as the signal-to-interference-and-noise ratio information; and performing channel equalization according to the SINR information, the first noise and interference covariance matrix, the channel estimation value and the receiving antenna signal corresponding to the data symbol to obtain the first equalized signal.

In the actual processing, the time-frequency domain resource of the user terminal may be sub-carrier divided according to the resource block information allocated to the user terminal to obtain one or more sub-carrier bands, so that the inversion processing may be performed according to the noise and interference covariance matrix of the sub-carrier bands to obtain a noise-interference covariance inverse matrix, which is used as the first noise and interference covariance matrix corresponding to the sub-carrier bands. Optionally, on the basis of the foregoing embodiment, before determining the noise-to-interference ratio value of each subcarrier band, the linear equalization method provided in the embodiment of the present disclosure may further include: according to resource block information distributed to a user terminal, sub-carrier division is carried out on time-frequency domain resources of the user terminal to obtain at least two sub-carrier wave bands; and determining a noise and interference covariance matrix of the subcarrier band according to a noise and interference covariance matrix corresponding to a target reference symbol, wherein the target reference symbol is a reference symbol contained in the subcarrier band. The Resource Block information allocated to the user terminal may be used to determine the size and the position of a Resource Block (RB) allocated to the user terminal, for example, the size information and the position information of the Resource Block may be included, which is not specifically limited in this disclosure. The resource block size information allocated to the user terminal may indicate the size of the RB allocated to the user terminal; the resource block location information may indicate the location of the RB allocated to the user terminal.

Referring to fig. 2, a flowchart illustrating steps of a linear equalization method according to an alternative embodiment of the present disclosure is shown. As shown in fig. 2, the linear equalization method implemented by the present disclosure may specifically include the following steps:

step 210, according to the resource block information allocated to the user terminal, performing subcarrier division on the time-frequency domain resource of the user terminal to obtain at least two subcarrier bands.

Specifically, the embodiments of the present disclosure may perform subcarrier band division on the time-frequency domain resource of the user terminal according to the size and the position of the RB allocated to the user terminal based on the resource block information allocated to the user terminal, for example, perform division in a time domain by taking one slot as a unit, so as to obtain two or more subcarrier bands.

Further, the resource block information in the embodiment of the present disclosure may include resource block size information and resource block location information, where the sub-carrier division is performed on the time-frequency domain resource of the user terminal according to the resource block information allocated to the user terminal to obtain at least two sub-carrier bands, and the sub-step specifically includes:

substep 2201, obtaining size information and position information of resource block allocated to user terminal;

and a substep 2202, based on the resource block size information and the resource block position information, performing subcarrier division on the time-frequency domain resources of the user terminal according to the number of subcarrier resource blocks to obtain at least two subcarrier bands, where the number of subcarrier resource blocks is the number of resource blocks in a preset frequency domain bandwidth.

Specifically, after the resource block size information and the resource block position information allocated to the user terminal are obtained, the time-frequency domain resource of the user terminal may be divided into the subcarrier bands according to the resource block size information and the resource block position information allocated to the user terminal and the number of resource blocks in a preset frequency domain bandwidth, so that a plurality of subcarrier bands may be obtained through division.

For example, the frequency domain resource range allocated at the current user terminal is RB_m～RB_nIn the case of (3), if the number of RBs in one subcarrier band is set to K in advance, the number of RBs in the RB band may be set to K_m～RB_nIn the frequency domain resource range of (3), each K RBs in the frequency domain and each time slot in the time domain are defined as a subcarrier band. Thus, the range in which the subcarrier band number can be calculated is

And the noise and interference covariance matrix corresponding to the subcarrier band numbered i may be labeled as R_uuI, and may correlate the noise and interference covariance matrices R for each subcarrier band_uu，iIs set to 0. Where m and n are RB offsets relative to the initial position RB0 of the current resource grid, respectively.

Taking fig. 3 as an example to illustrate the subcarrier band division, specifically, the frequency domain resource range in fig. 3 is subcarriers 0 to 47, which correspond to 4 RBs, which can be recorded as RB0, RB1, RB2, and RB3, respectively; in the case where the number K of resource blocks is set to be equal to 4, that is, when 4 RBs can be divided into one subcarrier band, the number i of the subcarrier band may be recorded as 0, that is, i is 0, and the time domain length of the subcarrier band may be 14 OFDM symbols, and the frequency domain length may be 4 RBs, so that all the subcarrier numbers may be in the range of 0 to 47, and data symbols or reference symbols having OFDM symbol numbers in the range of 0 to 13 are determined as data included in the subcarrier band. It should be noted that the gray boxes in fig. 3 may represent Reference (RS) symbols, and the white boxes in fig. 3 may represent DATA (DATA) symbols.

In the actual processing, if the number K of RBs in one subcarrier band is too large, the interference situation in each frequency domain may be different among RBs in the subcarrier band due to the presence of co-channel interference in some RBs in one subcarrier band and the absence of co-channel interference in some RBs; if the number K of RBs in one subcarrier band is too small, for example, K is 1, the number K of reference symbols in the subcarrier band may be too small, and the calculation result does not satisfy the statistical property, so that the number K of subcarrier resource blocks may be set to a positive integer greater than 1, which may give consideration to more accurate statistical properties caused by more reference symbols and different situations of intra-frequency interference of RBs in the subcarrier band.

In combination with the actual situation, preferably, the number K of RBs in one subcarrier band may be set to 4, so that more accurate statistical characteristics brought by more reference symbols and different situations of the RBs in the subcarrier band possibly suffering from the same frequency interference may be considered. In the actual use process, the number K of RBs in one subcarrier band may be adjusted according to the interference observed by other means, which is not specifically limited by the embodiment of the present disclosure.

Step 220, determining a noise and interference covariance matrix of the subcarrier band according to a noise and interference covariance matrix corresponding to a target reference symbol, where the target reference symbol is a reference symbol included in the subcarrier band.

Specifically, after one or more subcarrier bands are obtained through division, the noise and interference covariance matrices corresponding to all reference symbols included in each subcarrier band may be counted, so that the counted noise and covariance matrices are added to the noise and covariance matrices corresponding to the subcarrier bands, and the noise and covariance matrices corresponding to the subcarrier bands may be expected, so that the expected values of the noise and interference covariance matrices within the subcarrier bands are used as the noise and interference covariance matrices of the entire subcarrier bands. For example, expected values of noise and interference covariance matrices corresponding to reference symbols within a subcarrier band may be counted to form a noise and interference covariance matrix for the entire subcarrier band based on the expected values of the noise and interference covariance matrices corresponding to reference symbols contained within the subcarrier band.

Further, the determining the noise and interference covariance matrix of the subcarrier band according to the noise and interference covariance matrix corresponding to the target reference symbol in the embodiment of the present disclosure may specifically include: determining a target reference symbol belonging to the subcarrier band based on a subcarrier band number, the subcarrier band number of the subcarrier band to which the target reference symbol belongs being related to a subcarrier number of the target reference symbol; accumulating the noise and interference covariance matrixes corresponding to the target reference symbols in the subcarrier band to obtain a second noise and interference covariance matrix of the subcarrier band; and performing time domain filtering according to the second noise and interference covariance matrix of the subcarrier band to obtain the noise and interference covariance matrix of the subcarrier band. Wherein the subcarrier band number is a subcarrier band number.

In the actual process, k may be given as the formula i that divides the number of the sub-carrier bands_SCK, determining the number of the subcarrier band to which the reference symbol belongs; wherein, i can be expressed as the number of the sub-carrier band to which the reference symbol belongs; k is a radical of_SCA subcarrier number, which may be denoted as a reference symbol; k may be expressed as the number of sub-carrier resource blocks within one sub-carrier, and the number of sub-carrier resource blocks within one sub-carrier may be an algorithm parameter that is self-configured according to an observed interference law; m may be expressed as the number of subcarriers within one resource block, and M may be set to 12, as in the case where 12 subcarriers are one RB. The formula i ═ k_SCThe same applies to the calculation of the data symbolsThe number of subcarrier bands.

After determining the number of the subcarrier band to which the reference symbol belongs, all the reference symbols contained in the subcarrier band may be determined as target reference symbols based on the number of the subcarrier band to which the reference symbol belongs, and then the noise and interference covariance matrix corresponding to each target reference symbol may be added to the noise and interference covariance matrix of the subcarrier band to which the target reference symbol belongs through each target reference symbol, so that the noise and interference covariance matrix of each subcarrier band may be determined. For example, it can be according to the formula

To calculate a noise and interference covariance matrix within the sub-carrier band; wherein R is_uu，iMay be the expected value, N, of the noise and interference covariance matrix within the subcarrier band numbered i_iIs the number of reference symbols, R, within a sub-carrier band i_uu，kIs the covariance matrix of noise and interference corresponding to the kth reference symbol within subcarrier band i.

As an example of the present disclosure, when calculating a subcarrier band number corresponding to each reference symbol, the number k is given for one subcarrier_SCReference symbol of (2), which sub-carrier band number is i ═ k_SCAnd/12 × K, then, the subcarrier band number of each reference symbol may be saved, so that the reference symbol included in the subcarrier may be determined based on the subcarrier band number in the following, that is, the index reference symbol corresponding to the subcarrier band number is determined based on the subcarrier band number. In addition, i ═ k_SC12 of/12 × K are set for one RB according to 12 subcarriers, i.e., 12 REs (Resource elements) are one RB.

By traversing each reference symbol, its corresponding noise and interference covariance matrix R can be determined_uuSum of noise and interference covariance matrix accumulations R added to the subcarrier band to which it belongs_uu，i，sumIn (i) R_uu，i，sum＝R_uu，i，sum+R_uuIs equivalent to R_uu，i，sumFor all references within a sub-carrier bandR corresponding to number_uuThe result of the accumulation. Subsequently, the data can be decoded by traversing each sub-carrier band i ∈ 0,1,2, … … [ (n-m +1)/K ∈]For each subcarrier band, the sum R of the noise and interference covariance matrix accumulations is calculated_uu，i，sumTaking expectation, the formula can be embodied as

Wherein N is_iMay represent the number of reference symbols contained in the subcarrier band numbered i; r_uu，iThe expected values of the noise and interference covariance matrices for each subcarrier band. Thus, a second noise and interference covariance matrix for each subcarrier band may be determined based on the expected values of the noise and interference covariance matrices for the subcarrier band.

In a specific implementation, each subcarrier band may be traversed, and the second noise and interference covariance matrix of each subcarrier band is subjected to time domain filtering, so that the second noise and interference covariance matrix of the current subcarrier band of the previous time slot and the second noise and interference covariance matrix of the current subcarrier band of the current time slot are subjected to weighting processing through the time domain filtering to obtain a covariance matrix of the current time slot, which is an interference estimation result, so that accuracy of interference estimation may be improved. Wherein, the covariance matrix of the current time slot may represent the calculated covariance matrix of the noise interference.

It can be seen that, in the present example, expected values of noise and interference covariance matrices in a subcarrier band can be counted by flexibly dividing the subcarrier band, and the expected values are used as the noise and interference covariance matrices of the entire subcarrier band, so that time-domain filtering can be performed according to the noise and interference covariance matrices of the entire subcarrier band, and the problem of low interference estimation accuracy caused by performing interference estimation by taking one RB as a unit in a frequency domain in the related art is solved.

In summary, the embodiments of the present disclosure may perform time-domain filtering on the second noise and interference covariance matrix of each subcarrier band by traversing each subcarrier band, that is, perform filtering processing in the time domain, and perform weighting processing on the second noise and interference covariance matrix of the current subcarrier band of the previous time slot and the second noise and interference covariance matrix of the current subcarrier band of the current time slot, thereby improving accuracy of noise and interference estimation. Further, the performing time-domain filtering according to the second noise and interference covariance matrix of the subcarrier band to obtain the noise and interference covariance matrix of the subcarrier band according to the embodiment of the present disclosure includes: acquiring a subcarrier band covariance matrix of a current time slot and a subcarrier band covariance matrix of a reference time slot, wherein the subcarrier band covariance matrix of the current time slot is a second noise and interference covariance matrix of the subcarrier band of the current time slot, and the subcarrier band covariance matrix of the reference time slot is a second noise and interference covariance matrix of the subcarrier band of a time slot before the current time slot; and according to a preset time domain filter coefficient, carrying out weighting processing on the subcarrier band covariance matrix of the current time slot and the subcarrier band covariance matrix of the reference time slot.

Specifically, the noise and interference covariance matrix of the second subcarrier band in the embodiment of the present disclosure may include a noise and interference covariance matrix of at least two timeslot subcarrier bands, for example, the noise and interference covariance matrix may include a current timeslot subcarrier band covariance matrix and a reference timeslot subcarrier covariance matrix, and the reference timeslot subcarrier covariance matrix may be a noise and interference covariance matrix of a subcarrier band of a last timeslot corresponding to the current timeslot. The embodiment of the disclosure can perform weighting processing on the subcarrier band covariance matrix of the current time slot and the subcarrier band covariance matrix of the reference time slot based on the preset time domain filter coefficient a to obtain the noise and interference covariance matrix of the current time slot, which is used as the noise and interference covariance matrix of the subcarrier band.

For example, the formula R can be filtered according to the time domain_uu，i，t＝A*R_uu，i，t+(1-A)*R_uu，i，t-1And weighting the noise and interference covariance matrix of the current subcarrier band of the previous time slot and the noise and interference covariance matrix of the current subcarrier band of the current time slot to obtain the noise and interference covariance matrix of the current time slot as an interference estimation result. Wherein R is_uu，i，tIs the noise in the sub-carrier band numbered i in the t-th time slotCovariance matrix of acoustic interference, R_uu，i，t-1The covariance matrix of noise and interference in the subcarrier band numbered i in the t-1 th time slot is, a is a time-domain filter coefficient as an algorithm parameter, and if 0.8 is selected as the time-domain filter coefficient a, that is, a is 0.8, which is not limited in this example. Therefore, in the implementation of the time-domain filtering in this example, the covariance matrix of the noise and the interference is also counted in the time domain, and the distribution characteristic of the interference in the time domain is also reflected, so that the accuracy of the covariance matrix of the noise and the interference obtained by calculation is improved.

Therefore, the time-frequency domain resources of the user terminal are divided into the subcarriers according to the resource block information distributed to the user terminal, so that the subcarrier bands are flexibly divided; and determining a noise and interference covariance matrix of the subcarrier band according to the noise and interference covariance matrix corresponding to the reference symbol contained in the subcarrier band, performing time-domain filtering according to the noise and interference covariance matrix of the subcarrier band, and performing statistics on the noise and interference covariance matrix in a time domain, so that the distribution characteristic of the interference in the time domain is reflected, namely performing filtering processing in the time domain, not only considering the statistical characteristic in the frequency domain, but also considering the statistical characteristic in the time domain, thereby improving the accuracy of the calculated noise and interference covariance matrix, namely improving the accuracy of noise and interference estimation, and further solving the problem of poor receiver detection performance caused by low accuracy of the existing estimation method of the interference covariance matrix.

At step 230, a noise-to-interference ratio for each subcarrier band is determined.

Specifically, after determining the noise and interference covariance matrix of the subcarrier band, the embodiments of the present disclosure may perform calculation according to the noise and interference covariance matrix of the subcarrier band, for example, according to a formula

A calculation is performed to calculate the noise-to-interference ratio NIR within the sub-carrier band. Wherein, | [ R ]_uu，i]_m,nI can represent solving [ R ]_uu，i]_m,nComplex modulus, [ R ]_uu，i]_m,nRepresents R_uu，iM rows and n columns.

In actual processing, the noise-to-interference ratio of a subcarrier band may be the ratio of the noise to the interference power of the subcarrier band. The embodiment of the disclosure may calculate the antenna noise interference energy correlation result P1 and the antenna interference correlation result P2 according to the noise and interference covariance matrix of the sub-carrier band, so that the ratio of the antenna noise interference energy correlation result P1 to the antenna interference correlation result P2 may be determined as the noise interference ratio NIR of the sub-carrier band, that is, the ratio NIR is the ratio of the noise interference ratio of the sub-carrier band

The antenna noise interference energy correlation result P1 may represent a sum of correlation results between energies of noise and interference on different antennas, and may be specifically determined according to each element on a main diagonal in a covariance matrix of noise and interference of a subcarrier band; the antenna interference correlation result P2 may represent the sum of interference cross correlation results between different antennas, and may be determined according to each element on the upper right triangle or the lower left triangle (excluding the main diagonal) in the noise and interference covariance matrix of the subcarrier band. Optionally, the determining the noise-to-interference ratio of each subcarrier band in this embodiment may specifically include: determining a ratio of an antenna noise interference energy correlation result and an antenna interference correlation result as the noise interference ratio for each subcarrier band, wherein the antenna noise interference energy correlation result is a sum of products of elements on a target diagonal line, the target diagonal line is a main diagonal line of a noise and interference covariance matrix of the subcarrier band, the antenna interference correlation result is a sum of products of elements conjugate to each other in a first triangular matrix and a second triangular matrix in the noise and interference covariance matrix of the subcarrier band, and elements of the first triangular matrix and elements of the second triangular matrix do not include elements on the main diagonal line.

For example, the noise and interference covariance matrix R at the subcarrier band_uu，iAs an element of the first triangular matrix of the main embodiment of the present disclosureNoise and interference covariance matrix R of prime and sub-carrier bands_uu，iThe left lower triangular element of (1) as an element of the second triangular matrix of the embodiment of the present disclosure, a formula can be expressed

Computing a noise and interference covariance matrix R for a subcarrier band_uu，iAs the antenna noise interference energy correlation result P1, the sum of the products of the elements on the main diagonal line; and can be according to the formula

Calculating an interference covariance matrix R_uu，iThe sum of the products of the elements on the upper right triangle or the lower left triangle (excluding the main diagonal) as the antenna interference correlation result P2. Then, can be according to

And calculating the ratio of the antenna noise interference energy correlation result P1 to the antenna interference correlation result P2 to be used as the noise interference ratio NIR of the sub-carrier band, and comparing the relationship between the noise interference ratio NIR of the sub-carrier band and the threshold value to determine that the noise interference covariance matrix of the sub-carrier band is subjected to inversion processing by adopting an interference rejection ratio combining algorithm IRC or a maximum ratio combining algorithm MRC.

Specifically, the noise-to-interference ratio NIR may be compared with a threshold, so as to select a maximum ratio combining MRC algorithm or an interference rejection ratio combining IRC algorithm according to the comparison result, and perform inversion processing on the noise and interference covariance matrix of the subcarrier band. If the noise-to-interference ratio NIR of the sub-carrier band is greater than the threshold value, step 240 is executed to invert the noise-to-interference covariance matrix of the sub-carrier band according to the maximal ratio combining MRC algorithm; and if the noise-to-interference ratio NIR of the sub-carrier band is smaller than the threshold value, skipping to step 250 for execution, so as to perform inversion processing on the noise-to-interference covariance matrix of the sub-carrier band according to the interference rejection ratio combining IRC algorithm. If the noise-to-interference ratio NIR of the sub-carrier band is equal to the threshold, an interference rejection ratio combining IRC algorithm may be selected, and the inverse processing is performed on the noise-to-interference covariance matrix of the sub-carrier band, which is not specifically limited in the embodiment of the present disclosure.

And 240, if the noise-interference ratio of the subcarrier band is larger than the threshold value, carrying out inversion processing on the noise and interference covariance matrix of the subcarrier band according to the maximum ratio combining MRC algorithm to obtain the noise-interference covariance inverse matrix.

Specifically, when the noise-to-interference ratio NIR of a sub-carrier band is greater than the threshold value, the embodiments of the present disclosure may be based on the noise-and-interference covariance matrix R of the sub-carrier band_uu，iThe MRC combining algorithm is carried out, and the starting point is that the maximum ratio combining MRC algorithm treats the interference as white noise, so that the correlation value between different interferences can be 0, and the complexity of the inversion algorithm is greatly reduced no matter how many receiving antennas are. For example, when the noise-to-interference ratio NIR of a certain sub-carrier band is greater than the threshold value, for all data symbols in this sub-carrier band, the MRC combining algorithm is adopted to set the non-dominant diagonal elements of the covariance matrix of noise and interference to 0, that is, the covariance matrix R of noise and interference of the sub-carrier band is set to be 0_uu，iThe non-dominant diagonal element of (2) is set to 0, so that the correlation value between different interferences can be set to 0, and the complexity of the inversion algorithm is greatly reduced no matter how many receiving antennas are.

Further, for the covariance matrix of noise and interference using MRC combining algorithm, the inverse method may be used to simplify the calculation amount due to the particularity of the matrix. Optionally, in the embodiment of the present disclosure, the inverting the noise and interference covariance matrix of the subcarrier band according to the maximal ratio combining MRC algorithm to obtain the noise and interference covariance inverse matrix may specifically include: and calculating the reciprocal of each main diagonal element in the noise and interference covariance matrix of the subcarrier band to obtain an inverse matrix which is used as a noise interference covariance inverse matrix corresponding to the subcarrier band. Specifically, in performing the MRC algorithm, the disclosed embodiments directly sub-carrier band noise and interference covariance matrix regardless of the number of receive antennasR_uu，iThe noise and interference covariance matrix R is obtained by inverting each of the principal diagonal elements of (a)_uu，iInverse matrix of

Thereby inverting the matrix

As the inverse matrix of the noise interference covariance corresponding to the subcarrier band.

And step 250, if the noise interference ratio of the subcarrier band is not larger than the threshold value, inverting the noise and interference covariance matrix of the subcarrier band according to the interference rejection ratio combining IRC algorithm to obtain the noise interference covariance inverse matrix.

In particular, embodiments of the present disclosure may be based on the noise and interference covariance matrix R of the sub-carrier band when the noise-to-interference ratio NIR of the sub-carrier band is not greater than the threshold value_uu，iPerforming IRC algorithm, for example, when the noise interference ratio NIR of a certain sub-carrier band is not greater than the threshold value, for all data symbols in the sub-carrier band, adopting IRC combination algorithm, and performing different inversion calculation according to the number of antennas to obtain corresponding inverse matrix

As the inverse matrix of noise interference covariance corresponding to the subcarrier band.

It should be noted that, for the covariance matrix of noise and interference using the IRC combining algorithm, a general matrix inversion algorithm may be used, for example, when the number of antennas is 2, a formula method is used for inversion, and when the number of antennas is greater than 2, a cyclic index method is used for inversion, and the like, which is not limited in this disclosure.

Further, the embodiment of the present disclosure performs inversion processing on the noise and interference covariance matrix of the subcarrier band according to the interference rejection ratio combining IRC algorithm to obtain the noise and interference covariance inverse matrix, including: determining the number of antennas; if the number of the antennas is equal to two, the determinant inversion method is used for the sub-carrier wavesCarrying out inversion processing on the noise and interference covariance matrixes of the sub-carrier bands to obtain corresponding inverse matrixes which serve as noise and interference covariance inverse matrixes corresponding to the sub-carrier bands; if the number of the antennas is more than two, inverting the noise and interference covariance matrix of the subcarrier band by using a cyclic index inversion method to obtain a corresponding inverse matrix, wherein the corresponding inverse matrix is used as the noise interference covariance inverse matrix corresponding to the subcarrier band. The interference suppression ratio combining IRC algorithm can treat the interference as colored noise, and the correlation value between different interferences is not 0. Specifically, when the number of antennas is equal to 2 in the interference rejection ratio combining IRC algorithm, R can be obtained by using determinant inversion method_uu，iInverse matrix of

If the number of the antennas is more than 2, R can be obtained by using the cyclic index inversion method_uu，iInverse matrix of

Therefore, the embodiment of the present disclosure may perform different inversion calculations according to different selected algorithms and the number of antennas, and when the noise interference ratio NIR of the subcarrier band is selected to belong to the MRC algorithm, because R is the inverse of the number of antennas_uu，iThe non-main diagonal element of (2) is set to 0, so that the complexity of the inversion algorithm is greatly reduced no matter how many receiving antennas are, and the power of noise on each receiving antenna can be kept (the corresponding characteristic is that the matrix R can be used as the matrix R)_uuIn the form of Minimum Mean Square Error (MMSE) so that the received value processing for each antenna in the subsequent equalization operation can be more accurate, i.e., the accuracy of the received value processing for each antenna in the subsequent equalization operation is improved.

And step 260, acquiring a first noise and interference covariance matrix corresponding to the target subcarrier band according to the data symbol.

Specifically, after extracting the subcarrier band number to which the data symbol belongs, the embodiment of the present disclosure may extract a corresponding noise interference covariance inverse matrix according to the subcarrier band number, so as to serve as a first noise and interference covariance matrix corresponding to a target subcarrier band. And the target subcarrier band is a subcarrier band corresponding to the subcarrier band number to which the data symbol belongs.

And 270, performing equalization processing according to the first noise and interference covariance matrix corresponding to the target subcarrier band to obtain signal-to-interference-and-noise ratio information and a first equalization signal.

In this embodiment of the present disclosure, the signal to interference plus noise ratio information may be used to determine a signal to interference plus noise ratio and a repair coefficient corresponding to the data symbol. Specifically, the embodiment of the present disclosure may extract the channel estimation value from the channel estimation result for each data symbol in the received data, and extract the receiving antenna signal from the OFDM demodulation result, so as to perform channel equalization according to the sinr information, the first noise and interference covariance matrix, the channel estimation value, and the receiving antenna signal, to obtain the snr information and the first equalized signal, so that the repair coefficient corresponding to the data symbol may be determined according to the snr information in the following process, and the power ratio may be estimated according to the snr information, to obtain the snr corresponding to the soft bit information.

Further, the sir information in the embodiments of the present disclosure may include an identity matrix, and the identity matrix may be used to scale the constellation amplitude of the first equalized signal. In order to remove the influence of the unitary matrix in the snr information on the scaling of the constellation amplitude of the final equalization result, the embodiment of the present disclosure may determine the repair coefficient based on the snr information, perform inverse scaling repair on the first equalization signal according to the repair coefficient, for example, perform inverse scaling repair on the constellation amplitude of the first equalization signal, so as to perform soft demodulation processing subsequently according to the equalization signal generated after repair, that is, perform step 280.

As an example of the present disclosure, based on a MIMO receiving model, after whitening processing is performed on noise and interference, an MMSE receiving formula of a MIMO receiver may be derived according to a Minimum Mean Squared Error (MMSE) principle, and the MMSE receiving formula may be used for simultaneously processing an MRC algorithm and an IRC algorithmThe method can be expressed as an MMSE-MRC receiving algorithm and an MMSE-IRC receiving algorithm. For example, the specific calculation formula of the MMSE-MRC receiving algorithm and MMSE-IRC receiving algorithm is:

wherein Y is N_rxX1 dimensional received signal, H being N_rx×V_rxChannel matrix, X being V_rxX1 dimensional transmission signal, R_uu，iIs the noise and interference covariance matrix of the subcarrier band in which the data symbol is located. I is V_rx×V_rxA dimension unit matrix.

Specifically, under white noise conditions, according to the MMSE criterion, the equalization formula that can be obtained is

It should be noted that the equalization formula

Wherein Y may be represented by N_rxSignal on x 1-dimensional data symbol receiving antenna, equalization formula

Wherein H may be represented by N_rx×V_rxChannel estimates corresponding to the symbol positions of the dimensional data,

is a V_rxA channel equalization value of the x 1-dimensional data symbol,

is the noise variance. Formula for equalisation

Is a formula under white noise condition, and when same frequency interference is taken into account, the equalization formula

Derivation based on white gaussian noise is not applicable. Suppose thatThe covariance matrix of noise and interference at this time is R_uuBased on the theory of interference whitening, for this model Y ═ HX + n + I_itfIs changed into

The covariance of the noise and interference at that time is obtained

Where I is the identity matrix, i.e., the covariance of noise and interference is the identity matrix. Then, in conjunction with the MMSE criterion, the MMSE equalization formula under interference whitening can be derived as:

for the formula

If substituted into

If the MMSE-MRC equalization is performed, the MMSE-MRC equalization can be considered to be performed; if substituted into

And if the data is processed by the IRC, MMSE-IRC equalization is considered to be performed. In addition, the received data may be equalized by using a receiving formula of MMSE, so as to obtain an equalized signal. The noise and interference covariance matrix used in the MMSE receiving formula of each received data may be the noise and interference covariance matrix of the subcarrier band where the data symbol is located, so that the formula may be obtained

In the actual processing, the formula

Data symbols at different resource locations can be processed in parallel. Specifically, during the equalization process, the calculated values can be calculatedOf intermediate variables

As SINR information to utilize intermediate variables of the calculation process

According to the formula

And calculating an estimated value SINR of the power ratio of the signal to the noise and the interference at the position corresponding to each data symbol, and determining the estimated value SINR as a signal to interference and noise ratio corresponding to the soft bit information, so that the soft bits corresponding to the data symbols at different positions can be weighted by using the estimated value SINR in the following process, thereby improving the accuracy of confidence coefficients of the soft bits corresponding to the redundancy versions of different HARQ processes. For example, the corresponding soft bits may be weighted by the estimated value SINR according to the formula b ═ b × SINR; wherein b represents V_rxM soft bits corresponding to the x 1-dimensional data symbol, M being equal to the modulation order.

It can be seen that this example may be based on a first noise and interference covariance matrix for the subcarrier bands

Calculating channel estimation value H corresponding to data symbol to obtain calculation result

As signal to interference plus noise ratio information; the first noise and interference covariance matrix can then be determined based on the SINR information

And performing channel equalization on the channel estimation value H and the receiving antenna signal Y corresponding to the data symbol to obtain a first equalized signal X1.

Furthermore, in the equalization process, intermediate variables of the calculation process can be utilized

Calculating the coefficient of each symbol for constellation amplitude restoration, i.e. determining the constellation amplitude restoration coefficient, for example, according to the calculation formula of the constellation amplitude restoration coefficient

Using real part data of intermediate variables

And determining a constellation amplitude restoration coefficient beta, so that the constellation amplitude restoration can be performed on the equalized first signal by using the constellation amplitude restoration coefficient in the following process to generate a second equalized signal.

Step 280, performing soft demodulation processing according to a second equalization signal to obtain soft bit information corresponding to the data symbol, where the second equalization signal is an equalization signal generated after performing inverse scaling restoration on the first equalization signal according to the restoration coefficient.

And 290, performing weighting processing on the soft bit information according to the target signal-to-noise ratio to obtain weighted confidence information.

And the target signal-to-interference-and-noise ratio is the signal-to-interference-and-noise ratio corresponding to the soft bit information. Specifically, the greater the signal-to-interference-and-noise ratio corresponding to the soft bit information calculated in the channel equalization process, the greater the confidence of the corresponding LLR at the position should be. Meanwhile, as the 5G-NR uplink PUSCH protocol prescribes the processing of the HARQ process, soft information combination is required for the data of different redundancy versions of the same HARQ process. In the combining process, considering that the data of different redundancy versions are inconsistent in transmission time and have different signal-to-noise ratios, if the soft information can be weighted, more performance gains can be obtained in a soft decoding link, and the performance of HARQ combining can be improved, so that the error rate is reduced and the system throughput is improved. The embodiment of the disclosure performs signal to interference and noise ratio weighting on the soft bit information after soft demodulation, for example, according to formula b_m＝b_mX SINR, M is 0,1, … …, M-1, soft bit information after soft demodulation is subjected to soft bit demodulation using an estimated value SINR of the power ratio of signal to noise and interferenceAnd weighting to obtain weighted confidence information, thereby improving the accuracy of the confidence of the soft bits corresponding to the redundancy versions of different HARQ processes. Wherein, b_mRepresents V_rxThe M-th soft bit corresponding to the x 1-dimensional data symbol, wherein the value of M is equal to the modulation order

As an example of the present disclosure, to obtain more useful results from the noise and interference covariance matrix and channel estimation results, the SINR may be calculated according to a specific formula

And estimating the power ratio SINR of the signal, the interference and the noise corresponding to each data symbol position to be used as the signal-to-interference-and-noise ratio corresponding to the soft bit information, so that SINR weighting can be carried out on the soft bit information after soft demodulation.

In the actual process, the soft demodulation formula under the white gaussian noise condition is:

wherein, B_i,vIs that

The confidence of the probability that the ith bit of the received signal of the corresponding transmitted signal layer v is 0.

And

vector sets with the ith bit of the transmitted symbol of the v layer taking values of 1 and 0 respectively, different modulation modes have different sets,

is the noise variance. However, soft demodulation formula

Is based on inference in a white gaussian noise environment. For the presence of interference, i.e. coloured noiseIn case the above soft demodulation formula is no longer satisfied

The inference conditions are implemented. Because the noise and interference whitening technology is adopted in the embodiment of the disclosure, the interference is changed from colored noise to white noise, and the noise variance

That is, the channel equalization output of the embodiment of the present disclosure may be considered as output in a white noise link, and the corresponding soft demodulation formula is as follows:

compared with the derivation of Gaussian white noise, the method better accords with the statistical characteristic of soft demodulation under the condition of same frequency interference, and does not need division, namely does not need to divide by a signal-to-noise ratio in the soft demodulation link.

In summary, the equalization formula in the embodiment of the present disclosure is the same for both the MRC combining algorithm and the IRC combining algorithm, so that the uniformity in form is achieved, and the design of the channel equalization module is simpler. In addition, the equalization formula in the embodiment of the disclosure is derived based on the interference and noise whitening theory, and the starting point better conforms to the theory and the practice in the presence of interference, and division operation in the soft demodulation process is reduced, so that the soft demodulation efficiency can be improved.

It is noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the disclosed embodiments are not limited by the described order of acts, as some steps may occur in other orders or concurrently with other steps in accordance with the disclosed embodiments.

Referring to fig. 4, a block diagram of a linear equalization apparatus provided in an embodiment of the present disclosure is shown. The linear equalization apparatus 400 may include the following modules:

a covariance matrix obtaining module 410, configured to obtain, according to a data symbol, a first noise and interference covariance matrix corresponding to a target subcarrier band, where the target subcarrier band is a subcarrier band corresponding to a subcarrier band number to which the data symbol belongs;

the equalization processing module 420 is configured to perform equalization processing according to the first noise and interference covariance matrix corresponding to the target subcarrier band to obtain signal-to-interference-and-noise ratio information and a first equalization signal;

an inverse scaling repair module 430, configured to perform inverse scaling repair on the first equalized signal according to a repair coefficient, so as to generate a second equalized signal, where the repair coefficient is a coefficient determined according to the signal-to-interference-and-noise ratio information;

the soft demodulation processing module 440 is configured to perform soft demodulation processing according to the second equalized signal to obtain soft bit information corresponding to the data symbol.

Optionally, the equalization processing module 420 may include the following sub-modules:

Optionally, the signal-to-interference-and-noise ratio signal in the embodiment of the present disclosure is used to scale the constellation amplitude of the first equalization signal, and the linear equalization apparatus further includes a repair coefficient determining module, where the repair coefficient determining module is used to determine the repair coefficient based on real part data of the signal-to-interference-and-noise ratio information.

Optionally, on the basis of the foregoing embodiment, the interference estimation apparatus provided in the embodiment of the present disclosure may further include: other modules, such as an extraction module, a power ratio estimation module, a weighting processing module, etc., may also be included, which is not limited in this disclosure. The extracting module is configured to extract, for each data symbol in the received data, the channel estimation value from a channel estimation result, and extract the receiving antenna signal from an orthogonal frequency division multiplexing OFDM demodulation processing result; the power ratio estimation module is used for estimating the power ratio according to the signal to interference plus noise ratio information to obtain a target signal to interference plus noise ratio; and the weighting processing is used for carrying out weighting processing on the soft bit information according to the target signal-to-noise ratio to obtain weighted confidence information.

Optionally, the covariance matrix obtaining module 410 may include the following sub-modules:

Optionally, the number of the subcarrier band is the number of the subcarrier band, and the linear equalization apparatus 400 further includes:

Optionally, the linear equalization apparatus further includes:

Optionally, the resource block information in this embodiment of the present disclosure may include resource block size information and resource block location information, and the subcarrier dividing module may include the following sub-modules:

It should be noted that the linear equalization apparatus 400 provided above can execute the linear equalization method provided in any embodiment of the present invention, and has the corresponding functions and advantages of the execution method.

In a specific implementation, the linear equalization apparatus 400 described above may be integrated into a device, such that the device functions as a linear equalization device. As shown in fig. 5, an embodiment of the present disclosure provides a linear equalization apparatus, which includes a processor 111, a communication interface 112, a memory 113, and a communication bus 114, where the processor 111, the communication interface 112, and the memory 113 complete mutual communication through the communication bus 114, and the memory 113 is used for storing a computer program; the processor 111 is configured to implement the steps of the linear equalization method provided in any one of the foregoing method embodiments when executing the program stored in the memory 113.

The embodiments of the present disclosure also provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the linear equalization method provided by any one of the foregoing method embodiments.

It should be noted that, as for the embodiments of the apparatus, the device, and the storage medium, since they are basically similar to the embodiments of the method, the description is relatively simple, and in relevant places, reference may be made to the partial description of the embodiments of the method.

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

1. A method of linear equalization, comprising:

2. The linear equalization method according to claim 1, wherein the equalizing the data symbols according to the first noise and interference covariance matrix corresponding to the target subcarrier band to obtain the signal-to-interference-and-noise ratio information and the first equalized signal comprises:

3. The linear equalization method according to claim 1, wherein the signal-to-interference-and-noise ratio information is used for scaling the constellation amplitude of the first equalized signal, and before performing inverse scaling repair on the first equalized signal according to a repair coefficient, the method further comprises:

and determining the repair coefficient based on the real part data of the signal-to-interference-and-noise ratio information.

4. The linear equalization method of claim 1, further comprising, after obtaining the signal-to-interference-and-noise ratio information and the first equalized signal:

estimating a power ratio according to the SINR information to obtain a target SINR;

and carrying out weighting processing on the soft bit information according to the target signal-to-noise ratio to obtain weighted confidence information.

5. The linear equalization method of claim 1, wherein the obtaining a first noise and interference covariance matrix corresponding to a target subcarrier band according to a data symbol comprises:

extracting the subcarrier band number to which the data symbol belongs;

6. The linear equalization method of claim 5, wherein the subcarrier band number is a subcarrier band number, and wherein before obtaining the first noise and interference covariance matrix corresponding to the target subcarrier band according to the data symbols, the method further comprises:

determining a noise-to-interference ratio for each sub-carrier band;

7. The linear equalization method according to claim 6, wherein said inverting the noise and interference covariance matrices of the subcarrier bands according to the Maximal Ratio Combining (MRC) algorithm to obtain the noise and interference covariance inverse matrix comprises:

8. The linear equalization method according to claim 6, wherein said inverting the noise and interference covariance matrices of the subcarrier bands according to the combined IRC algorithm to obtain the noise and interference covariance inverse matrix comprises:

determining the number of antennas;

9. The linear equalization method of claim 6 wherein said determining a noise-to-interference ratio value for each subcarrier band comprises:

10. The linear equalization method of claim 9, wherein prior to determining the noise-to-interference ratio value for each subcarrier band, further comprising:

11. A linear equalization apparatus, comprising:

12. The communication equipment is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing the communication between the processor and the memory through the communication bus;

a memory for storing a computer program;

a processor for implementing the steps of the linear equalization method of any of claims 1-10 when executing a program stored in a memory.

13. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the linear equalization method according to any one of claims 1-10.