CN117439844A

CN117439844A - Channel estimation method and system with noise elimination

Info

Publication number: CN117439844A
Application number: CN202311580549.8A
Authority: CN
Inventors: 王霄峻; 汤飞; 戚子越; 张在琛
Original assignee: Southeast University
Current assignee: Southeast University
Priority date: 2023-11-24
Filing date: 2023-11-24
Publication date: 2024-01-23

Abstract

The invention discloses a channel estimation method and a system with noise elimination, which relate to the technical field of channel estimation and comprise the following steps: the method comprises the steps of receiving a signal to be processed, carrying out LS estimation on the signal to be processed to obtain an estimation result, carrying out low-pass filtering on the estimation result, removing noise to obtain a denoising result, carrying out interpolation algorithm on the denoising result, and calculating to obtain channel estimation values of all subcarrier positions of the whole frequency domain; the invention can improve the accuracy of the channel estimation result and reduce the overall processing time delay of the channel estimation: the direct filtering method directly carries out filtering treatment on the channel estimation result in the frequency domain, so that the conversion from the frequency domain to the time domain is avoided; in addition, the invention can realize low-complexity operation by means of special hardware structures such as a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA) and the like, thereby further reducing the processing time delay.

Description

Channel estimation method and system with noise elimination

Technical Field

The invention relates to the technical field of channel estimation, in particular to a channel estimation method and system with noise elimination.

Background

The existing channel estimation method is an LS estimation method, and although the implementation is simple, the method does not consider the influence of noise on an estimation result. The main noise canceling method adopted in the current mobile communication system is the transform domain denoising method, and the flow thereof is shown in fig. 12. Firstly, obtaining a noisy channel estimation value through LS estimation, then mirror image expanding to a length suitable for IFFT conversion, converting to a time domain through IFFT, performing time domain windowing, then converting back to a frequency domain through FFT, and finally obtaining the channel estimation value of all subcarrier positions of the whole frequency domain through a difference algorithm. The transform domain denoising method can effectively eliminate the influence of noise on a channel estimation result, but the two FFT (IFFT) transformation calculation complexity is higher, and the overall processing time delay of the method is larger.

Disclosure of Invention

To solve the above-mentioned drawbacks of the related art, an object of the present invention is to provide a channel estimation method and system with noise cancellation.

The aim of the invention can be achieved by the following technical scheme: a channel estimation method with noise cancellation, the method comprising the steps of:

receiving a signal to be processed, and performing LS estimation on the signal to be processed to obtain a preliminary estimation result;

directly performing low-pass filtering on the estimation result (frequency domain signal) to remove noise components and obtain an estimation result subjected to noise removal processing;

and carrying out interpolation algorithm on the denoising result, and calculating to obtain channel estimation values of all subcarrier positions of the whole frequency domain.

Preferably, the LS estimation calculation process for the signal to be processed is as follows:

g _LS ＝(X) ^H y

wherein X is a reference signal of a transmitting end, H is a conjugate transpose of a matrix, y is a received reference signal, g _LS Is the estimation result.

Preferably, the signal to be processed includes a frequency domain received signal and an ideal reference signal.

Preferably, the interpolation algorithm determines the requirement for channel estimation accuracy.

Preferably, the low-pass filtering is performed on the estimation result to remove noise, and the process of obtaining the denoising result includes the following steps:

an appropriate window function is selected for filtering and filter parameters, including order and cut-off frequency, are determined. Wherein the filter order affects the filter performance and computational complexity; the cut-off frequency determines the transition bandwidth of the filter in the frequency domain and is influenced by the maximum multipath time delay;

performing value supplementing operation on the original sequence, intercepting the number of M/2 points at two ends of the sequence, supplementing the number of M/2 points to two ends of the original sequence by a mirror image continuation method, and adopting the following formula:

wherein,for the channel estimation at the reference signal, N is the input sequence length, M is the filter order,/->Representation pair->Before->Dot inversion, +_>Indicate->Inverting the dots;

and filtering the N+M channel estimated values after the complement values by using the generated filter coefficients, wherein the filtered output is as follows:

wherein the method comprises the steps ofIs->The output after passing through the filter; />Estimating a value sequence for the LS algorithm after the complement, wherein the value range of N is 0-N+M-1; f (F) _cof Is a filter coefficient;

extracting effective channel estimation value from the filtered output sequence, wherein the effective output value is thatStarting from the M-th value there are N points, i.e. +.>

Preferably, the process of performing interpolation algorithm on the denoising result comprises the following steps:

performing interpolation processing on the data subcarriers between the reference signals by using the selected interpolation algorithm to obtain channel estimation values of the positions of the data subcarriers;

and obtaining channel estimation values of all available subcarriers before the first reference signal subcarrier and all available subcarriers after the last reference signal subcarrier by using an extrapolation method.

In a second aspect, to achieve the above object, the present invention discloses a channel estimation system with noise cancellation, including:

a receiving estimation module: the method comprises the steps of receiving a signal to be processed, and carrying out LS estimation on the signal to be processed to obtain an estimation result;

and a denoising module: the method comprises the steps of performing low-pass filtering on an estimation result, removing noise and obtaining a denoising result;

interpolation calculation module: and the method is used for carrying out interpolation algorithm on the denoising result to obtain channel estimation values of all subcarrier positions of the whole frequency domain.

In another aspect of the present invention, in order to achieve the above object, there is disclosed an apparatus comprising:

one or more processors;

a memory for storing one or more programs;

when one or more of the programs are executed by one or more of the processors, the one or more of the processors are caused to implement a channel estimation method with noise cancellation as described above.

In a further aspect of the present invention, in order to achieve the above object, a storage medium containing computer executable instructions for performing a channel estimation method with noise cancellation as described above when executed by a computer processor is disclosed.

The invention has the beneficial effects that:

the invention can improve the accuracy of the channel estimation result and reduce the overall processing time delay of the channel estimation: the direct filtering method directly carries out filtering processing on the channel estimation result in the frequency domain, and does not need to carry out processing from IFFT conversion to the time domain, thereby effectively reducing the processing time delay.

Drawings

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, and it will be obvious to those skilled in the art that other drawings can be obtained according to these drawings without inventive effort;

FIG. 1 is a schematic flow chart of the method of the present invention;

FIG. 2 is a schematic of the workflow of the present invention;

fig. 3 is a frequency response diagram of the frequency domain filtering method and other channel estimation methods used in the present invention when the EPA channel has a signal to noise ratio of 5 dB.

Fig. 4 is a frequency response diagram of the frequency domain filtering method and other channel estimation methods used in the present invention when the EVA channel has a signal-to-noise ratio of 5 dB.

FIG. 5 is an EPA channel time domain delay power spectrum;

FIG. 6 is an EVA channel time domain delay power spectrum;

FIG. 7 is a graph showing MSE performance of the frequency domain filtering method of the present invention compared to other channel estimation methods for EVA channels with a Doppler shift of 70 Hz.

FIG. 8 is a graph showing MSE performance of the frequency domain filtering method of the present invention compared to other channel estimation methods for ETU channels with a Doppler shift of 300 Hz.

Fig. 9 is a schematic diagram of MSE performance of the frequency domain filtering method and other channel estimation methods used in the present invention on full frequency sub-carriers for EVA channels with doppler shift of 70 Hz.

Fig. 10 is a schematic diagram of MSE performance of the frequency domain filtering method and other channel estimation methods used in the present invention on full frequency domain subcarriers for an ETU channel with a doppler shift of 300 Hz.

FIG. 11 is a schematic diagram of the system architecture of the present invention;

fig. 12 is a prior art schematic of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

As shown in fig. 1, a channel estimation method with noise cancellation includes the following steps:

receiving a signal to be processed, and performing LS estimation on the signal to be processed to obtain an estimation result;

performing low-pass filtering on the estimation result to remove noise and obtain a denoising result;

The LS estimation calculation process of the signal to be processed is as follows:

g _LS ＝(X) ^H y

The interpolation algorithm determines by the requirement for channel estimation accuracy.

The signal to be processed includes a frequency domain received signal and an ideal reference signal.

The process of low-pass filtering the estimation result and removing noise to obtain a denoising result comprises the following steps:

an appropriate window function is selected for filtering and filter parameters, including order and cut-off frequency, are determined. Wherein the filter order affects the filter performance and computational complexity; the cut-off frequency determines the transition bandwidth of the filter in the frequency domain, affected by the maximum multipath delay. This step aims at designing a low pass filter suitable for a particular application to balance the smoothness and accuracy of the channel estimation. The window function and parameters are selected depending on the requirements and performance goals of the application.

And (3) performing value supplementing operation on the original sequence, intercepting the number of M/2 points at two ends of the sequence, and supplementing the number of M/2 points to two ends of the original sequence through a mirror image continuation method. This operation helps to avoid boundary effects caused during filtering and compensates for the delay introduced by the filter. The complement operation is specifically as follows:

wherein,for the channel estimation at the reference signal, N is the input sequence length, M is the filter order,/->Representation pair->Before->Dot inversion, +_>Indicate->The dots are inverted.

And filtering the N+M channel estimated values after the complementary values by using the generated filter coefficients. The filtering operation will apply a selected filter to remove noise and unwanted high frequency components to obtain smoother channel estimation results. The filtered output is as follows:

wherein the method comprises the steps ofIs->The output after passing through the filter; />Estimating a value sequence for the LS algorithm after the complement, wherein the value range of N is 0-N+M-1; f (F) _cof Is a filter coefficient.

And extracting effective channel estimation values from the filtered output sequence. Since the original sequence is subjected to the value compensation operation before, the transition part needs to be removed from the filtering output sequence so as to ensure that the channel estimation value after denoising is accurate and is suitable for the subsequent interpolation processing operation. The effective output value at this time isStarting from the Mth value at a total of N points, i.e

The process for carrying out interpolation algorithm on the denoising result comprises the following steps:

and selecting a proper interpolation algorithm, and comprehensively considering the influence of precision and complexity. This choice is affected by factors such as the accuracy of the desired channel estimate, computational resources, and the actual application scenario.

And carrying out interpolation processing on the data subcarriers among the reference signals by using the selected interpolation algorithm so as to obtain channel estimation values of the positions of the data subcarriers. This step aims at filling up the channel estimation values of the data subcarrier locations to obtain complete frequency domain information.

And obtaining channel estimation values of all available subcarriers before the first reference signal subcarrier and all available subcarriers after the last reference signal subcarrier by using an extrapolation method. The range of channel estimates is intended to be extended to cover the entire frequency domain.

In another aspect, in order to achieve the above object, as shown in fig. 11, an embodiment of the present invention discloses a channel estimation system with noise cancellation, including:

The key point of the invention is that the channel estimation process adopts a frequency domain direct filtering method, rather than converting the frequency domain estimation result into the time domain for processing through inverse discrete Fourier transform (IFFT). This innovation brings the following significant advantages:

improving the accuracy of the channel estimation result: the direct filtering method allows the estimation result to be filtered in the frequency domain, so that the influence of noise on channel estimation can be effectively reduced, and the accuracy of the estimation result is improved.

Reducing the overall processing delay of channel estimation: by avoiding frequency domain to time domain to frequency domain conversion, the present invention significantly reduces the processing delay of channel estimation. In addition, by means of special hardware structures such as a DSP (digital signal processor), an FPGA (field programmable gate array) and the like, the invention can realize low-complexity operation and further reduce the processing time delay of channel estimation. It should be further explained that, in this embodiment:

simulation conditions

The simulation channel selects three tap delay models provided by the LTE system, wherein the EPA channel is used for simulating indoor, indoor to outdoor and pavement tests, the EVA channel is mainly used for simulating vehicle-mounted tests of delay in simulation, and the ETU channel is mainly used for simulating tests of typical urban areas and the like with high delay. Specific parameters of the three models are detailed in the following table.

TABLE 1 typical channel model delay power spectrum

The following simulation process involves configuring a communication system with 20MHz bandwidth (1200 effective subcarriers) and DQPSK modulation, configuring multiple antennas to 2 x 2, simulating under different channel conditions (EPA, EVA, ETU), selecting channel estimation values of 200 noisy reference signal subcarriers at equal intervals, performing different denoising methods on the use of the channel estimation values, and then obtaining 1200 point subcarrier channel estimation values through a first order linear difference algorithm. The simulation process evaluates the performance of different channel estimation methods (LS, LMMSE, DFT transform domain, hamming window low-pass filtering) through signal generation, multi-antenna processing, channel simulation, signal processing and performance evaluation, and compares the performance of the denoising processing algorithm by calculating the estimation errors separately.

Under the typical 20MHz bandwidth and EPA and EVA two channel environments, the channel frequency response amplitude obtained by adopting different channel estimation methods is shown in figures 3 and 4. It can be seen that noise is added at each position in the frequency domain, and the amplitude value transmits larger fluctuation; and compared with EPA channels, EVA channels introduce stronger frequency selectivity due to the larger diameter number.

Fig. 5 and 6 show time domain delay power spectra in EPA, EVA channel environments, where significant multi-channel can be seen and energy is concentrated at the beginning, i.e. in the delay spread interval, with noise interference in the latter part.

Simulation results

Fig. 3 and fig. 4 show frequency domain estimation results obtained by different channel estimation methods in two channel environments, and it can be seen that, compared with the method without any denoising, the different methods achieve a certain smoothing effect, and the channel estimation values based on filtering tend to be smooth and approach to ideal values. The LMMSE algorithm of the exponential distribution model has a better smoothing effect, but is often difficult to apply in occasions with higher requirements on processing delay in view of higher algorithm complexity.

Fig. 7 shows MSE performance curves for the LS algorithm, LMMSE algorithm, simplified LMMSE algorithm, DFT-based transform domain algorithm, and frequency-based filtering algorithm for the EVA and ETU channels shown in fig. 8. It can be seen that several filtering methods can achieve a certain noise reduction effect. The frequency domain filtering-based algorithm is slightly inferior to the LMMSE algorithm of the exponential distribution model before the 0dB signal-to-noise ratio, and is superior to all algorithms except the LMMSE algorithm as a whole. Because the transform domain filtering needs to estimate the noise energy and performs filtering according to the noise energy value, the transform domain filtering is susceptible to noise, and the estimation result is inferior to the frequency domain filtering.

Fig. 9 shows MSE performance curves for the LS algorithm, LMMSE algorithm, simplified LMMSE algorithm, DFT-transform-domain-based algorithm, and frequency-domain-filtering-based algorithm for the EVA and ETU channels given in fig. 10. MSE performance on full-band subcarriers under 20M system bandwidth, with a signal-to-noise ratio of-1 dB, is aimed at analyzing performance of various estimation algorithms under boundary conditions. From the results, it can be seen that the result of LS channel estimation is substantially consistent with the noise variance for the full-band subcarrier MSE, since no special processing is performed. The LMMSE algorithm performs best, but its complexity is too high to be applied in practical engineering. The improved algorithm based on the frequency domain filtering is ideal in overall estimation performance and is only inferior to the LMMSE algorithm. Whereas estimation algorithms based on DFT transforms are superior to LS channel estimation in overall noise suppression, at the boundaries, estimation performance fluctuates significantly due to the gibbs phenomenon caused by transform domain windowing.

From the analysis of the simulation results, it can be seen. The frequency domain digital filtering method has better performance than the transform domain filtering method, and compared with the transform domain filtering method, the complex operations such as IFFT, FFT and the like are reduced. The frequency domain can thus be considered to directly use the digital filter design for the filtering operation.

Based on the same inventive concept, the present invention also provides a computer apparatus comprising: one or more processors, and memory for storing one or more computer programs; the program includes program instructions and the processor is configured to execute the program instructions stored in the memory. The processor may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application SpecificIntegrated Circuit, ASIC), field-Programmable gate arrays (FPGAs) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which are the computational core and control core of the terminal for implementing one or more instructions, in particular for loading and executing one or more instructions within a computer storage medium to implement the methods described above.

It should be further noted that, based on the same inventive concept, the present invention also provides a computer storage medium having a computer program stored thereon, which when executed by a processor performs the above method. The storage media may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

In the description of the present specification, the descriptions of the terms "one embodiment," "example," "specific example," and the like, mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

The foregoing has shown and described the basic principles, principal features, and advantages of the present disclosure. It will be understood by those skilled in the art that the present disclosure is not limited to the embodiments described above, which have been described in the foregoing and description merely illustrates the principles of the disclosure, and that various changes and modifications may be made therein without departing from the spirit and scope of the disclosure, which is defined in the appended claims.

Claims

1. A method of channel estimation with noise cancellation, the method comprising the steps of:

2. The channel estimation method with noise cancellation as claimed in claim 1, wherein the LS estimation calculation process for the signal to be processed is as follows:

g _LS ＝(X) ^H y

3. The channel estimation method with noise cancellation of claim 1 wherein said signal to be processed comprises a frequency domain received signal and an ideal reference signal.

4. A channel estimation method with noise cancellation as claimed in claim 1, wherein the interpolation algorithm determines the requirement for channel estimation accuracy.

5. The channel estimation method with noise cancellation as claimed in claim 1, wherein the process of performing low-pass filtering on the estimation result to remove noise and obtain the denoising result comprises the following steps:

selecting a proper window function for filtering, and determining filter parameters including an order and a cut-off frequency, wherein the filter order influences the filter performance and the computational complexity; the cut-off frequency determines the transition bandwidth of the filter in the frequency domain and is influenced by the maximum multipath time delay;

wherein,for the channel estimation at the reference signal, N is the input sequence length, M is the filter order,/->Representation pairBefore->Dot inversion, +_>Indicate->Inverting the dots;

6. The channel estimation method with noise cancellation as claimed in claim 1, wherein the process of performing the interpolation algorithm on the denoising result comprises the steps of:

7. A channel estimation system with noise cancellation, comprising:

8. An apparatus, comprising:

one or more processors;

a memory for storing one or more programs;

when executed by one or more of the processors, causes the one or more processors to implement a channel estimation method with noise cancellation as claimed in any one of claims 1-6.

9. A storage medium containing computer executable instructions which, when executed by a computer processor, are for performing a channel estimation method with noise cancellation as claimed in any one of claims 1 to 6.