Channel estimation equalization method and system

Abstract

The present disclosure relates to a channel estimation and equalization method and system. The channel estimation and equalization method of the present disclosure comprises: generating a local reference sequence in real time; using the local reference sequence to perform channel estimation on a data signal to obtain a channel estimation parameter; performing inverse processing on the channel estimation parameter to obtain an equalization parameter; and using the equalization parameter to process the data signal to obtain a channel equalization signal. The present disclosure generates a local reference sequence in real time to achieve channel estimation, which can well meet different subcarrier numbers in an OFDM system. Meanwhile, performing inverse division operation on the channel estimation can reduce the operation delay of equalization and reduce the total processing delay.

Description

Technical Field

[0001] This disclosure relates to the field of wireless communication, specifically to a channel estimation equalization method and system. Background Technology

[0002] In related technologies, some wireless communication systems, such as 4G and 5G, use wireless channels with multipath propagation, leading to inter-symbol interference (ISI) in digital communication that requires equalization. In single-carrier communication, as multipath propagation increases, the number of taps required for the transverse filter increases significantly, increasing the complexity of equalization. Orthogonal multicarrier (OFDM) communication simplifies equalization by incorporating a cyclic prefix of appropriate length. Regardless of whether it's single-carrier or multi-carrier communication, equalization requires channel estimation. Summary of the Invention

[0003] This disclosure aims to address at least one of the technical problems existing in the prior art by proposing a channel estimation equalization method and system.

[0004] To achieve the above objectives, this disclosure provides a channel estimation equalization method, comprising:

[0005] Generate local reference sequences in real time;

[0006] Channel estimation parameters are obtained by performing channel estimation on the data signal using a local reference sequence.

[0007] The equalization parameters are obtained by taking the reciprocal of the channel estimation parameters;

[0008] The data signal is processed using equalization parameters to obtain the channel equalization signal.

[0009] In some embodiments, channel estimation parameters are obtained by using a local reference sequence to perform channel estimation on the data signal, including:

[0010] The first channel frequency response is obtained by using the least squares estimation algorithm to estimate the channel.

[0011] The channel estimation parameters are obtained by denoising the frequency response of the first channel using the Fast Discrete Fourier Transform.

[0012] In some embodiments, the channel estimation parameters are obtained by denoising the frequency response of the first channel using the Fast Discrete Fourier Transform, including:

[0013] The intermediate channel time-domain response sequence is obtained by performing an inverse fast Fourier transform on the first channel frequency response sequence.

[0014] The intermediate channel time-domain response sequence is denoised, and the channel estimation parameters are obtained by performing a fast Fourier transform on the denoised intermediate channel time-domain response sequence.

[0015] In some embodiments, performing an inverse fast Fourier transform on the first channel frequency response sequence to obtain the intermediate channel time-domain response sequence includes:

[0016] The intermediate channel frequency response sequence is obtained by expanding the first channel frequency response sequence;

[0017] The intermediate channel time-domain response sequence is obtained by performing an inverse fast Fourier transform on the extended intermediate channel frequency response sequence.

[0018] In some embodiments, the first channel frequency response sequence is represented as [H LS (0),…,H LS (N-1)]

[0019] Among them, H LS (i) represents the (i+1)th sequence value in the first channel frequency response sequence, where i = 0, 1, ..., N-1, and N represents the length of the local reference sequence.

[0020] The extended intermediate channel frequency response sequence is represented as follows:

[0021]

[0022] Where, N FFT This indicates the number of sampling points used when performing the Fast Fourier Transform.

[0023] In some embodiments, the intermediate channel time-domain response sequence is denoised, including:

[0024] Retain the first N from the intermediate time-domain response sequence CP / 4 sequence values ​​and the last N CP / 4 sequence values, the rest of the sequence values ​​are considered noise to be filtered out, where N CP The length of the cyclic prefix.

[0025] In some embodiments, a local reference sequence is generated in real time, including:

[0026] Determine the length N of the local reference sequence;

[0027] The ZC sequence is calculated using the length N of the local reference sequence as the sequence period, and the calculated ZC sequence is the local reference sequence.

[0028] In some embodiments, processing the data signal using equalization parameters includes multiplying the equalization parameters by the data signal.

[0029] This disclosure also provides a channel estimation equalization system, including:

[0030] The channel estimation module is used to generate a local reference sequence in real time, use the local reference sequence to perform channel estimation on the data signal to obtain channel estimation parameters, and perform reciprocal processing on the channel estimation parameters to obtain equalization parameters.

[0031] The channel equalization module is used to process data signals using equalization parameters to obtain channel equalized signals.

[0032] In some embodiments, the channel estimation module includes:

[0033] The channel estimation submodule is used to perform channel estimation using the least squares estimation algorithm to obtain the first channel frequency response;

[0034] The noise reduction submodule is used to reduce the noise of the first channel frequency response using fast discrete Fourier transform to obtain channel estimation parameters, and to perform reciprocal processing on the channel estimation parameters to obtain equalization parameters. Attached Figure Description

[0035] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings:

[0036] Figure 1 This is a flowchart of downlink channel data propagation in some embodiments of this disclosure;

[0037] Figure 2 This is a flowchart of a downlink channel estimation equalization method in some embodiments of this disclosure;

[0038] Figure 3 This is a flowchart illustrating the real-time generation of local reference sequences in some embodiments of this disclosure. Detailed Implementation

[0039] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0040] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the described embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0041] It will be understood by those skilled in the art that all or some of the steps, modules / units in the methods, systems, and apparatuses disclosed herein can be implemented as software, firmware, hardware, and suitable combinations thereof. In hardware implementations, the division between modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0042] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0043] In related technologies, downlink user-side parameters in 4G or 5G communication systems change dynamically in real time, leading to dynamic changes in the length of the reference sequence. Therefore, traditional fixed-length reference sequences cannot meet the requirements. Furthermore, due to the large number of user connections, downlink reception inevitably involves multiple users coexisting, and the channels used by each user are dynamic. Thus, for a specific user, the reference signal used for reception, its sequence length, the time slot it occupies, and even the subcarrier position in OFDM are not fixed. In this situation, using a fixed local sequence would require storing multiple sequences of different lengths simultaneously, consuming significant hardware logic resources, making practical use inconvenient, and increasing costs.

[0044] Traditional channel estimation methods include time-domain and frequency-domain approaches, but frequency-domain channel estimation is more efficient in OFDM systems. In frequency-domain channel estimation, reference-symbol-based channel estimation algorithms typically include the Least Squares (LS) algorithm, the Minimum Mean Square Error (MMSE) algorithm, and simplified versions thereof. Among these, the LS estimation algorithm is simple, but its mean square error is relatively large; the MMSE algorithm offers the best estimation performance, but its complexity is very high.

[0045] For OFDM systems with constant-parameter channels, the traditional approach is to use only one OFDM symbol as a pilot, then use this pilot for channel estimation, and finally use this channel estimate to equalize all subsequent data symbols. This approach fails to consider that data symbols also contain channel information, which reduces the accuracy of the estimation to some extent, resulting in a relatively high error rate after equalization.

[0046] Therefore, in order to at least alleviate or avoid one of the aforementioned technical problems, this disclosure provides a channel estimation equalization method.

[0047] This disclosure provides a channel estimation equalization method, including the following steps:

[0048] Step S1: Generate a local reference sequence in real time.

[0049] Step S2: Use the local reference sequence to perform channel estimation on the data signal to obtain the channel estimation parameters.

[0050] Step S3: Obtain the equalization parameters by performing reciprocal processing on the channel estimation parameters.

[0051] Step S4: Process the data signal using equalization parameters to obtain the channel equalization signal.

[0052] The embodiments disclosed herein achieve channel estimation by generating local reference sequences in real time, which can well meet the different numbers of subcarriers in OFDM systems. At the same time, performing the reciprocal division operation at the channel estimation point can reduce the computational delay of equalization and reduce the total processing delay.

[0053] In some embodiments, step S1, generating a local reference sequence in real time, includes the following steps:

[0054] Step S11: Determine the length N of the local reference sequence.

[0055] Step S12: Use the length N of the local reference sequence as the sequence period to calculate the ZC sequence. The calculated ZC sequence is the local reference sequence.

[0056] In this embodiment of the disclosure, the reference sequence length N of the reference signal required by the local user terminal is determined, the ZC sequence is used as the reference sequence for channel estimation, and the reference sequence length N is used as the sequence period to calculate the value of each sequence, thereby generating N sequences, namely the local reference sequence.

[0057] In some embodiments, step S2, obtaining channel estimation parameters by performing channel estimation on the data signal using a local reference sequence, specifically includes the following steps:

[0058] Step S21: Use the least squares estimation algorithm to estimate the channel frequency response and obtain the first channel frequency response.

[0059] Step S22: Use Fast Discrete Fourier Transform to reduce the noise of the first channel frequency response to obtain the channel estimation parameters.

[0060] This disclosure presents a trade-off between the performance and complexity of the channel estimation algorithm. Specifically, it employs a channel estimation algorithm based on least squares estimation and fast discrete Fourier transform, i.e., using the LS-FFT channel estimation algorithm for channel estimation, which has lower complexity and better estimation performance.

[0061] In some embodiments, step S22, using Fast Discrete Fourier Transform to denoise the frequency response of the first channel to obtain channel estimation parameters, specifically includes the following steps:

[0062] Step S221: Perform an inverse fast Fourier transform on the first channel frequency response sequence to obtain the intermediate channel time domain response sequence.

[0063] Step S222: Denoise the intermediate channel time domain response sequence, and perform a fast Fourier transform on the denoised intermediate channel time domain response sequence to obtain the channel estimation parameters.

[0064] The embodiments of this disclosure use Fast Fourier Transform (FFT) for channel estimation, which can achieve real-time changes. Furthermore, FFT does not require more logic resources in its implementation. Therefore, the embodiments of this disclosure can work in conjunction with real-time generated local reference sequences to perform channel estimation, while also saving resources.

[0065] In some embodiments, step S221, performing an inverse fast Fourier transform on the first channel frequency response sequence to obtain an intermediate channel time-domain response sequence, specifically includes the following steps: expanding the first channel frequency response sequence to obtain an intermediate channel frequency response sequence, and performing an inverse fast Fourier transform on the expanded intermediate channel frequency response sequence to obtain an intermediate channel time-domain response sequence.

[0066] In this embodiment of the disclosure, the number of sequence values ​​of the first channel frequency response sequence may be different from the number of sampling points used in the fast Fourier transform process. In this case, the first channel frequency response sequence needs to be extended, and the number of sequence values ​​of the extended intermediate channel frequency response sequence can be the same as the number of sampling points used in the fast Fourier transform process, which is beneficial to realize the fast Fourier transform of the first channel frequency response and subsequent processing.

[0067] In some embodiments, the first channel frequency response sequence is represented as:

[0068] [H LS (0),…,H LS (N-1)]

[0069] Among them, H LS (i) represents the (i+1)th sequence value in the first channel frequency response sequence, i = 0, 1, ..., N-1, where N represents the length of the local reference sequence.

[0070] The extended intermediate channel frequency response sequence is represented as follows:

[0071]

[0072] Where, N FFT This indicates the number of sampling points used when performing the Fast Fourier Transform.

[0073] In this embodiment of the disclosure, the first channel frequency response sequence containing N data points is expanded to contain N... FFT A sequence of data, and specifically, symmetrically expanded (N) on both the left and right sides. FFT -N) / 2 data points. The number of sampling points used when performing the Fast Fourier Transform can be the same as the maximum number of OFDM points specified by the communication protocol. Therefore, the embodiments of this disclosure extend the frequency response sequence of the first channel to meet the maximum number of OFDM points specified by the communication protocol.

[0074] In some embodiments, step S222, which involves denoising the intermediate channel time-domain response sequence, specifically includes: retaining the first N values ​​of the intermediate time-domain response sequence. CP / 4 sequence values ​​and the last N CP / 4 sequence values, the rest of the sequence values ​​are considered noise to be filtered out, where N CP The length of the cyclic prefix.

[0075] In this embodiment of the disclosure, noise reduction processing of the intermediate channel time domain response sequence can improve the data signal quality and reduce the impact of noise.

[0076] In some embodiments, processing the data signal using equalization parameters includes multiplying the equalization parameters by the data signal.

[0077] In this embodiment of the disclosure, the channel estimation parameters are divided by the reciprocal during channel estimation. Therefore, it is not necessary to perform the reciprocal operation again during the channel equalization process. Thus, this embodiment of the disclosure can reduce the operation delay of equalization and reduce the total processing delay.

[0078] This disclosure also provides a channel estimation equalization system, including: a channel estimation module and a channel equalization module.

[0079] The channel estimation module is used to generate a local reference sequence in real time, use the local reference sequence to perform channel estimation on the data signal to obtain channel estimation parameters, and perform reciprocal processing on the channel estimation parameters to obtain equalization parameters.

[0080] The channel equalization module is used to process data signals using equalization parameters to obtain channel equalized signals.

[0081] The channel estimation equalization system in this embodiment is applied to the user end. It can perform corresponding channel estimation and equalization for different user ends, thereby realizing the transmission of downlink data signals from the base station to multiple user ends.

[0082] Meanwhile, in this embodiment of the disclosure, the channel estimation module performs a reciprocal division operation on the channel estimation parameters while estimating the channel. Therefore, it is not necessary to perform the reciprocal operation again during the channel equalization process of the channel equalization module. Thus, this embodiment of the disclosure can reduce the equalization operation delay in the channel equalization module and reduce the total system processing delay.

[0083] In some embodiments, the channel estimation module includes a channel estimation submodule and a noise reduction submodule.

[0084] The channel estimation submodule is used to estimate the first channel frequency response using the least squares estimation algorithm.

[0085] The noise reduction submodule is used to reduce the noise of the first channel frequency response using fast discrete Fourier transform to obtain channel estimation parameters, and to perform reciprocal processing on the channel estimation parameters to obtain equalization parameters.

[0086] Since the length of the reference sequence is dynamically changing, in this embodiment, the channel estimation submodule uses the least squares estimation algorithm for channel estimation, while the noise reduction submodule uses discrete fast Fourier transform for noise reduction. This enables real-time changes in channel estimation and meets the channel estimation requirements of different user terminals.

[0087] The channel estimation and equalization method of this disclosure will be specifically described below with reference to specific embodiments.

[0088] Figure 1 This is a flowchart of downlink channel data propagation in some embodiments of this disclosure.

[0089] like Figure 1 As shown, the present invention provides a downlink channel estimation equalization method, which includes the following steps in sequence:

[0090] The first step is to sample the baseband digital signal from the ADC, and then perform OFDM symbol de-CP after synchronization processing, frequency offset estimation and compensation.

[0091] The second step is to perform OFDM symbol decomposition, i.e., FFT operation.

[0092] The third step is to separate the reference signal from the data signal.

[0093] The fourth step is to use the LS-FFT algorithm to estimate the channel on the reference signal.

[0094] The fifth step is to equalize the channel estimation parameters and the data signal using a zero-forcing equalization algorithm. The equalization result can be used for subsequent demodulation and channel decoding.

[0095] This invention implements the LS-FFT channel estimation algorithm by generating local reference sequences in real time. This method can well meet the variable-length reference sequence requirements of OFDM systems with dynamic subcarrier number requirements. LS-FFT satisfies both good estimation performance and low complexity. At the same time, performing the reciprocal division operation at the channel estimation stage can reduce the computational delay of equalization and reduce the total system processing delay.

[0096] Specifically, in one embodiment, for downlink signals transmitted from a 4G or 5G base station to the user terminal, such as Figure 1As shown, after the downlink signal is processed by the RF link in the antenna, it needs to be sampled by an ADC to obtain a baseband digital signal. Then, the baseband digital signal is synchronized through digital domain processing to obtain a synchronization signal. The synchronization signal is then corrected for Doppler and frequency offset between the base station and the user terminal through frequency offset estimation and compensation. Based on the frequency offset signal, channel estimation and equalization are performed to resolve multipath and residual phase offset in the base station system, so as to facilitate subsequent signal demodulation and decoding.

[0097] Specifically, channel estimation and equalization include:

[0098] Step S01: The first step is to remove the CP data from the signal.

[0099] Step S02: Then, perform OFDM symbol decomposition on the data after CP removal, i.e., perform FFT operation.

[0100] Step S03: After obtaining the data after FFT operation, it is necessary to find the number of OFDM symbols used by the user terminal, the initial position of its subcarrier in OFDM, the total number of subcarriers, etc., based on the parameters of the base station and user terminal, and extract the user data.

[0101] Step S04: Extract channel reference data from the user data and cache it. The channel reference data is used for channel estimation. Simultaneously, the data signal is cached locally for subsequent equalization processing.

[0102] Figure 2 This is a flowchart of a downlink channel estimation equalization method in some embodiments of this disclosure.

[0103] The channel estimation and equalization method disclosed herein includes two processes: channel estimation and equalization processing in step S04.

[0104] like Figure 2 As shown, specifically in one example, the channel estimation process includes:

[0105] Step S041: Generate a local reference sequence in real time.

[0106] This embodiment uses the ZC sequence as the reference sequence for channel estimation. The generation method for each sequence value in the N reference sequences r(0),…,r(N-1) is as follows:

[0107] r(n) = x q (nmodN zc )

[0108]

[0109] Where, N ZC It is a sequence period, satisfying N ZCThe largest prime number ≤ N is, in practice, N. ZC =N is used to calculate the sequence, that is, when generating local sequences in real time, the value of N is the length of the sequence.

[0110] Figure 3 This is a flowchart illustrating the real-time generation of local reference sequences in some embodiments of this disclosure.

[0111] like Figure 3 As shown, the real-time generation process of the local reference sequence includes:

[0112] First, the actual reference sequence length N used by the user needs to be obtained based on the system parameters. Then, the calculation should end at a certain point based on the size of N. The value of m*(m+1) is calculated using a multiplier, and then this value is multiplied by the constant π (Pi). Considering the implementation complexity and accuracy requirements, π is quantized to the power of 2, meaning the value of the constant π is retained to the nearest 2. 13 Simultaneously, based on the sequence length N, the value of 1 / N is read from the ROM. Also, considering the sensitivity of complex exponentiation, the number of decimal places for 1 / N needs to be 2. 36 Therefore, in this embodiment of the disclosure, 1 / N stored in ROM is multiplied by 2. 36 The value after rounding.

[0113] In 4G or 5G communication standards, the selection of subcarriers must meet the requirement of being an integer multiple of 12, and the maximum number of OFDM points supported is 4096. In actual use, a maximum of 275*12 subcarriers are used to transmit useful data. Therefore, the ROM only needs a 36-bit width and a depth of 275 for data storage.

[0114] After determining the sequence length N used by the user, the corresponding ROM address can be found through N to obtain the value of 1 / N. Multiplying this value by the previously calculated sequence value in real time yields the required phase value for calculating the complex exponent. Since this process may result in a value greater than 2π, it is necessary to perform a 2π remainder operation. In implementing the calculation of the complex exponent, this embodiment uses a Cordic IP core. The remainder obtained after processing is used as input, and the Cordic IP core uses sin and cos modes. The output of its IP core is the local reference sequence we need.

[0115] Step S042: After obtaining the local reference sequence, channel estimation is performed using the LS estimation algorithm. The estimated value obtained using the LS estimation algorithm is calculated using the following formula:

[0116]

[0117] Where X is the transmitted reference symbol, Y is the corresponding received reference symbol, and H... LSThis represents the channel frequency response of the corresponding subcarrier.

[0118] In this embodiment of the disclosure, X is the locally generated reference sequence in real time. According to the characteristics of the ZC sequence, the inverse X-1 of X is the complex conjugate of X. Therefore, the channel estimate obtained by the LS estimation algorithm is actually obtained by performing complex multiplication on the complex conjugate of the corresponding points of the received reference sequence and the local reference sequence.

[0119] The estimated value obtained using the LS estimation algorithm is the frequency response of the first channel.

[0120] Step S042: Perform DFT noise reduction on the frequency response of the first channel.

[0121] The estimation algorithm based on Discrete Fourier Transform (DFT) was proposed by Van-De-Beek et al. It is a correction of the estimated value of the channel frequency response based on the LS estimation algorithm. This technique improves the accuracy of the estimation by reducing noise other than the maximum multipath delay.

[0122] The principle of the DFT channel estimation algorithm is to first perform a discrete Fourier transform on the channel frequency response obtained by LS estimation, then perform reasonable linear or nonlinear processing in the time domain to reduce the noise of the channel impulse response, and finally restore the noise-reduced impulse response to the frequency domain by performing a discrete Fourier transform again.

[0123] However, in practical use, since the length of the reference sequence is dynamically changing, but the DFT implementation usually cannot change in real time, and more logic resources are required in several variable DFT implementations, the embodiments of this disclosure adopt the Fast Discrete Fourier Transform (FFT) mode for implementation.

[0124] Specifically, FFT is used to denoise the frequency response of the first channel.

[0125] During downlink channel transmission, the signal length needs to be a power of 2. This cannot be directly satisfied by the reference sequence length N, because the protocol specifies that the reference sequence length must be an integer multiple of 12.

[0126] The following describes the specific implementation process of DFT noise reduction for the frequency response of the first channel.

[0127] The first step is to expand the generated local reference sequence, which means expanding the length of the LS-estimated sequence.

[0128] Since the protocol specifies a maximum OFDM point count of 4096, this embodiment of the disclosure uses 4096 points as an example for FFT noise reduction processing.

[0129] A symmetric expansion method is used when expanding the sequence. That is, in H... LSSymmetrical expansion on both the left and right sides (N) FFT -N) / 2 data points, that is, expanding from a sequence containing N data points to a sequence containing N data points. FFT A sequence of data points, denoted as Hls1.

[0130]

[0131] The expanded sequence Hls1 is the intermediate channel frequency response sequence.

[0132] Step 2: Perform an IFFT transform on the expanded data Hls1 to obtain the time-domain reference sequence hls1:

[0133]

[0134] Where h(n) represents the reference signal and w(n) represents the noise signal.

[0135] Step 3: Denoising the time-domain reference sequence hls1, i.e., retaining useful data. Specifically, retain the points with higher energy at both ends, and filter out the rest as noise. hls1 can be expressed by the following formula:

[0136]

[0137] After filtering out noise, hls2 is obtained and can be expressed by the following formula:

[0138]

[0139] Where, N CP N is the length of the cyclic prefix. In specific implementations, N... CP Based on the CP length used in OFDM symbols, select the first 1 to N after IFFT. CP / 4, and 4096-N CP Points from 4+1 to 4096 are the useful data, and all other points are assigned a value of 0 to obtain the data after noise reduction.

[0140] Step 4: Perform an FFT transform on the denoised data hls2, and transform it to the frequency domain to obtain the channel frequency response Hls2.

[0141]

[0142] For example, in this implementation with a 4096-point FFT, for the frequency response sequence Hls2 of the channel estimation obtained again, the data corresponding to the first step of symmetric extension position is removed to obtain the channel frequency response H' of the subcarrier corresponding to the reference symbol. This value is the frequency estimation response of the channel adopted by the user, i.e., the channel estimation parameter.

[0143] Step 5: The equalization parameters can be obtained by taking the reciprocal of the channel estimation sequence (i.e., the channel estimation parameters H') after noise reduction. Then calculate the value The value is stored in RAM. This is the calculated value used for channel equalization.

[0144] In this embodiment of the disclosure, the division of the channel estimate is performed to reduce system computational latency, as well as the computational cost of subsequent equalization and hardware power consumption. This is because the number of data paths during equalization is usually several times the length of the reference sequence. Therefore, by calculating the division in advance, the equalization module does not need to perform the division again, thereby reducing the total number of division operations.

[0145] Step S043: Equalize parameters Multiplying the signal by the data to be equalized will give the signal equalization result.

[0146] Equalization techniques are used to compensate for linear distortion caused by channel frequency selectivity, and it is a crucial component of digital communication systems. Equalization can be divided into time-domain equalization and frequency-domain equalization. Time-domain equalization considers the system's impulse response from a time-domain perspective, ensuring it meets the condition of no inter-symbol interference. Frequency-domain equalization designs equalization filters in the frequency domain, using the frequency characteristics of tunable filters to compensate for the frequency characteristics of the baseband system, ensuring that the overall characteristics of the baseband system, including the tunable filters, meet the requirements of distortion-free transmission. For OFDM systems, frequency-domain equalization is more suitable; therefore, the embodiments of this disclosure use frequency-domain equalization.

[0147] Specifically, the frequency domain equalization in this disclosure uses the Zero Forcing (ZF) equalization algorithm. Zero Forcing equalization is derived from the peak distortion criterion. Peak distortion is defined as the worst-case inter-symbol interference at the equalizer output. Peak distortion describes the worst-case inter-symbol interference after equalization, and the process of minimizing this parameter is called the peak distortion criterion.

[0148] In the continuous case, the transfer function or coefficients w(z) of the channel equalization module, or equalizer, can be derived as follows:

[0149]

[0150] Therefore, the transfer function w(z) of the equalizer based on the peak distortion criterion is the reciprocal of the channel filter transfer function H′(z). The equalized data is then obtained by multiplying the data by the transfer function.

[0151] This equalization algorithm is called zero-forcing equalization. The frequency domain feature H′(z) is the channel estimate value H' obtained from the previous channel estimation algorithm; then, by multiplying the data signal with the equalizer coefficient w(z), this is the frequency domain equalization process. That is, in the continuous case, the equalization parameters... This is the transfer function.

[0152] In the discrete case, the coefficients in the channel equalization module can be expressed as: k = 0, 1, ..., L. H′ k This is also the channel estimate value H' obtained from the previous channel estimation algorithm. Finally, for the data signal S... k Perform equalization processing, that is, perform equalization processing on the data signal S k With equilibrium parameter w k Multiplying them together yields the channel equalization signal Y. k Y k =S k *w k In this embodiment, the step of calculating the reciprocal has already been implemented during channel estimation. Here, the signal equalization result can be obtained simply by multiplying the estimated channel value by the data to be equalized. This equalized data, i.e., the channel-equalized signal, can be used for subsequent demodulation and decoding.

[0153] In summary, this invention implements the LS-FFT channel estimation algorithm by generating local reference sequences in real time. This method can effectively meet the requirements of different numbers of subcarriers in OFDM systems and the channel estimation of reference symbols at different positions within OFDM symbols. LS-FFT satisfies both good estimation performance and low complexity; furthermore, performing the reciprocal division operation at the channel estimation stage reduces the computational delay for equalization, thereby reducing the overall system processing delay.

[0154] It is understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of this disclosure, and this disclosure is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this disclosure, and these modifications and improvements are also considered to be within the scope of protection of this disclosure.

Claims

1. A channel estimation equalization method, characterized in that, include: Generate local reference sequences in real time; Channel estimation parameters are obtained by performing channel estimation on the data signal using a local reference sequence. The equalization parameters are obtained by taking the reciprocal of the channel estimation parameters; The data signal is processed using equalization parameters to obtain the channel equalization signal.

2. The channel estimation equalization method according to claim 1, characterized in that, Channel estimation parameters are obtained by performing channel estimation on the data signal using a local reference sequence, including: The first channel frequency response is obtained by using the least squares estimation algorithm to estimate the channel. The channel estimation parameters are obtained by denoising the frequency response of the first channel using the Fast Discrete Fourier Transform.

3. The channel estimation equalization method according to claim 2, characterized in that, The channel estimation parameters are obtained by denoising the frequency response of the first channel using the Fast Discrete Fourier Transform, including: The intermediate channel time-domain response sequence is obtained by performing an inverse fast Fourier transform on the first channel frequency response sequence. The intermediate channel time-domain response sequence is denoised, and the channel estimation parameters are obtained by performing a fast Fourier transform on the denoised intermediate channel time-domain response sequence.

4. The channel estimation equalization method according to claim 3, characterized in that, The process of obtaining the intermediate channel time-domain response sequence by performing an inverse fast Fourier transform on the first channel frequency response sequence includes: expanding the first channel frequency response sequence to obtain the intermediate channel frequency response sequence; and performing an inverse fast Fourier transform on the expanded intermediate channel frequency response sequence to obtain the intermediate channel time-domain response sequence.

5. The channel estimation equalization method according to claim 4, characterized in that, The first channel frequency response sequence is represented as follows: Among them, H LS (i) represents the (i+1)th sequence value in the first channel frequency response sequence, where i = 0, 1, ..., N-1, and N represents the length of the local reference sequence. The extended intermediate channel frequency response sequence is represented as follows: Where, N FFT This indicates the number of sampling points used when performing the Fast Fourier Transform.

6. The channel estimation equalization method according to claim 5, characterized in that, The intermediate channel time-domain response sequence is denoised, including: Retain the first N from the intermediate time-domain response sequence CP / 4 sequence values ​​and the last N CP / 4 sequence values, the rest of the sequence values ​​are considered noise to be filtered out, where N CP The length of the cyclic prefix.

7. The channel estimation equalization method according to any one of claims 1 to 6, characterized in that, Real-time generation of local reference sequences, including: Determine the length N of the local reference sequence; The ZC sequence is calculated using the length N of the local reference sequence as the sequence period, and the calculated ZC sequence is the local reference sequence.

8. The channel estimation equalization method according to any one of claims 1 to 6, characterized in that, Processing data signals using equalization parameters includes multiplying the equalization parameters by the data signal.

9. A channel estimation equalization system, characterized in that, include: The channel estimation module is used to generate a local reference sequence in real time, use the local reference sequence to perform channel estimation on the data signal to obtain channel estimation parameters, and perform reciprocal processing on the channel estimation parameters to obtain equalization parameters. The channel equalization module is used to process data signals using equalization parameters to obtain channel equalized signals.

10. The channel estimation equalization system according to claim 9, characterized in that, The channel estimation module includes: The channel estimation submodule is used to perform channel estimation using the least squares estimation algorithm to obtain the first channel frequency response; The noise reduction submodule is used to reduce the noise of the first channel frequency response using fast discrete Fourier transform to obtain channel estimation parameters, and to perform reciprocal processing on the channel estimation parameters to obtain equalization parameters.

Images