Channel equalization method and device based on channel updating

Abstract

The invention relates to a channel equalization method and device based on channel updating, and relates to the technical field of signal processing, and the method comprises the steps: carrying out the channel equalization of a data unit of a current level based on a channel updating result of a previous level, and obtaining a first channel equalization result of the current level; determining a reference signal of the current level based on the first channel equalization result; performing channel estimation based on the reference signal of the current level to obtain a channel estimation result of the current level; performing channel updating based on the channel updating result of the previous level and the channel estimation result of the current level to obtain a channel updating result of the current level; and performing channel equalization on the data unit of the current level based on the channel updating result of the current level to obtain a second channel equalization result of the current level, or taking the next level as the updated current level, and performing channel equalization on the data unit of the current level based on the channel updating result of the previous level. According to the embodiment of the invention, the channel equalization result quality can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of signal processing, and more specifically, to a channel equalization method and apparatus based on channel update. Background Art

[0002] Channel equalization is an anti-fading measure taken to improve the transmission performance of fading channels in a communication system, and its core lies in eliminating or weakening the inter-symbol interference problem caused by multi-path time delay in broadband communication.

[0003] The prior art usually performs channel estimation based on long training fields in received signals and uses the channel estimation results for all data units of the received signals. Herein, a data unit is a basic unit for channel equalization processing.

[0004] However, the channel may change due to factors such as relative movement between the transceiver devices and temperature change of components, and the channel estimation results are not applicable to the channel equalization of data units with a relatively later reception time. Therefore, the quality of the channel equalization results of the prior art needs to be improved. Summary of the Invention

[0005] The present application is provided to solve the above problems existing in the prior art. A channel equalization method and apparatus based on channel update according to an embodiment of the present application can improve the quality of channel equalization results.

[0006] In a first aspect, an embodiment of the present application provides a channel equalization method based on channel update, including:

[0007] Performing channel equalization on data units of the current level based on channel update results of the upper level to obtain a first channel equalization result of the current level;

[0008] Determining a reference signal of the current level based on the first channel equalization result;

[0009] Performing channel estimation based on the reference signal of the current level to obtain a channel estimation result of the current level;

[0010] Performing channel update based on the channel update results of the upper level and the channel estimation result of the current level to obtain a channel update result of the current level;

[0011] Performing channel equalization on data units of the current level based on the channel update result of the current level to obtain a second channel equalization result of the current level, or using the next level as the updated current level and performing channel equalization on data units of the current level based on channel update results of the upper level;

[0012] Among them, the received signal includes data units of multiple levels. If the current level is the first level, the channel update result of its upper level is the channel estimation result based on the long training field in the received signal; if the channel update result of the current level is used for the channel update of the current level, the channel equalization result of the data unit of the current level is the second channel equalization result, otherwise, the channel equalization result of the data unit of the current level is the first channel equalization result.

[0013] In a second aspect, an embodiment of the present application provides a channel equalization device based on channel update, including:

[0014] A channel equalization module, configured to perform channel equalization on the data unit of the current level based on the channel update result of the upper level to obtain the first channel equalization result of the current level;

[0015] A reference signal update module, configured to determine the reference signal of the current level based on the first channel equalization result;

[0016] A channel estimation module, configured to perform channel estimation based on the reference signal of the current level to obtain the channel estimation result of the current level;

[0017] A channel update module, configured to perform channel update based on the channel update result of the upper level and the channel estimation result of the current level to obtain the channel update result of the current level;

[0018] The channel equalization module is further configured to perform channel equalization on the data unit of the current level based on the channel update result of the current level to obtain the second channel equalization result of the current level, or, the channel update module is further configured to use the next level as the updated current level and trigger the channel equalization module to perform channel equalization on the data unit of the current level based on the channel update result of the upper level;

[0019] Among them, the received signal includes data units of multiple levels. If the current level is the first level, the channel update result of its upper level is the channel estimation result based on the long training field in the received signal; if the channel update result of the current level is used for the channel update of the current level, the channel equalization result of the data unit of the current level is the second channel equalization result, otherwise, the channel equalization result of the data unit of the current level is the first channel equalization result.

[0020] In a third aspect, an embodiment of the present application provides a computer program product, including computer programs / instructions, and when the computer programs / instructions are executed by a processor, the method described in any of the above embodiments is implemented.

[0021] The beneficial effects of the embodiments of the present application are as follows: Compared with using the channel estimation results obtained from long training fields for channel equalization, the present application takes into account the influence of time variation on the channel. Based on the channel equalization results, the reference signal used for channel estimation is re-determined, and more accurate channel estimation results can be obtained. The channel equalization results obtained based on this channel estimation result have higher accuracy and reliability. Description of the Drawings

[0022] In the drawings, which are not necessarily to scale, the same reference numerals may describe similar components in different views. The same reference numerals with letter suffixes or different letter suffixes may represent different instances of similar components. The drawings generally illustrate various embodiments by way of example and not limitation, and are used together with the description and the claims to explain the disclosed embodiments. Where appropriate, the same reference numerals are used throughout the drawings to refer to the same or similar parts. Such embodiments are illustrative and are not intended to be exhaustive or exclusive embodiments of the apparatus or method.

[0023] Figure 1 is a schematic diagram of channel equalization based on long training fields provided by an embodiment of the present application;

[0024] Figure 2 is a flowchart of a channel equalization method based on channel update provided by an embodiment of the present application;

[0025] Figure 3 is a schematic diagram of a feedback channel update scheme provided by an embodiment of the present application;

[0026] Figure 4 is a schematic diagram of a forward channel update scheme provided by an embodiment of the present application;

[0027] Figure 5 is a schematic diagram of the variation of the normalized root mean square error with the signal-to-noise ratio provided by an embodiment of the present application;

[0028] Figure 6 is a graph of the error vector magnitude varying with the signal duration provided by an embodiment of the present application;

[0029] Figure 7 is a curve of the packet error rate varying with the signal-to-noise ratio provided by an embodiment of the present application;

[0030] Figure 8 is a schematic diagram of a channel equalization device based on channel update provided by an embodiment of the present application. Detailed Embodiments

[0031] To enable those skilled in the art to better understand the technical solution of this application, the following provides a detailed description of this application in conjunction with the accompanying drawings and specific embodiments. The following further describes the embodiments of this application in detail in conjunction with the accompanying drawings and specific examples, but it is not a limitation to this application. The terms "first", "second", and "third" used in this application are only intended to distinguish the corresponding features, and do not represent the need for such sorting, nor necessarily indicate the singular form.

[0032] As Figure 1 shown, the existing solution performs channel estimation based on long training fields, equalizes all data units based on the channel estimation result, and then restores the data stream based on the equalization result of the data units. Taking the received signal as an OFDM (Orthogonal Frequency Division Multiplexing) signal as an example, the existing technology performs channel equalization on all OFDM symbols based on the channel estimation result, and then restores the data stream based on the equalization result of each OFDM symbol.

[0033] However, this method only performs channel estimation once using long training fields and cannot update the channel over time. When the channel changes during transmission, the channel estimated using long training fields will no longer be applicable for unpacking the data field, resulting in the deterioration of the channel equalization effect over time. Especially for long data packets, such as aggregated A-MPDU (Aggregated MAC Protocol Data Unit) and A-MSDU (Aggregated MAC Service Data Unit) packets, channel aging will cause the decoding failure rate of long packets to continuously increase over time. Moreover, the channel estimation error of this method is relatively large. Especially for low-order modulation signals, the received SNR (Signal-to-Noise Ratio) threshold is relatively low, thus affecting the channel equalization and the final decoding and packet reception effects.

[0034] In view of this, as Figure 2 shown, the embodiments of this application provide a channel equalization method based on channel update, including:

[0035] Step 201: Perform channel equalization on the data units of the current level based on the channel update result of the upper level to obtain the first channel equalization result of the current level.

[0036] Among them, this method is used for channel equalization of received signals. The received signals include data units of multiple levels. Channel equalization is performed on each level such as the first level, the second level, etc. in chronological order. The data unit is the basic unit for channel equalization processing. Taking an OFDM signal as an example, the data unit can be a coded block or a certain number of OFDM symbols, etc. The channel update result is the channel estimation value obtained through channel update.

[0037] Step 202: Determine the reference signal of the current level based on the first channel equalization result.

[0038] Step 203: Perform channel estimation based on the reference signal of the current level to obtain the channel estimation result of the current level.

[0039] Step 204: Perform channel update based on the channel update result of the previous level and the channel estimation result of the current level to obtain the channel update result of the current level.

[0040] Step 205: Perform channel equalization on the data unit of the current level based on the channel update result of the current level to obtain the second channel equalization result of the current level, or use the next level as the updated current level and perform channel equalization on the data unit of the current level based on the channel update result of the previous level.

[0041] Among them, if the current level is the first level, the channel update result of its previous level is the channel estimation result based on the long training field in the received signal; if the channel update result of the current level is used for the channel update of the current level, the channel equalization result of the data unit of the current level is the second channel equalization result, otherwise, the channel equalization result of the data unit of the current level is the first channel equalization result.

[0042] The embodiments of this application provide two channel update schemes, one is a feedback type and the other is a forward type.

[0043] Feedback channel update scheme: As Figure 3 shown, the channel update result of the current level is used for channel equalization of the data unit of the next level, rather than for channel equalization of the data unit of the current level. Corresponding to "use the next level as the updated current level and perform channel equalization on the data unit of the current level based on the channel update result of the previous level" in step 205. This scheme has no delay and is easy to implement.

[0044] Forward channel update scheme: As Figure 4 shown, the channel update result of the current level is used for channel equalization of the data unit of the current level. Corresponding to "perform channel equalization on the data unit of the current level based on the channel update result of the current level to obtain the second channel equalization result of the current level" in step 205. The channel equalization result of this scheme has a higher quality.

[0045] Compared with using the channel estimation results obtained from long training fields for channel equalization, the present application takes into account the impact of time variation on the channel. Based on the channel equalization results, the reference signal used for channel estimation is re-determined, and more accurate channel estimation results can be obtained. The channel equalization results obtained based on this channel estimation result have higher accuracy and reliability.

[0046] In practical application scenarios, there are at least the following two ways of channel update. One is symbol-level channel update, and the other is code-block-level channel update.

[0047] Among them, for symbol-level channel update: channel update is performed once every preset number M of OFDM symbols, where M is the number of symbols required for single-channel estimation. After receiving every M OFDM symbols, channel estimation and channel update are performed once.

[0048] For code-block-level channel update: channel update is performed in units of coding blocks. For example, packet blocks encoded by LDPC (Low-Density Parity-Check Code) or interleaved blocks encoded by BCC (Binary Convolutional Code). Each coding block may contain multiple OFDM symbols, and channel update is performed only after receiving a complete coding block. Since the number of symbols required for single-channel estimation is M, the coding block corresponds to multiple channel estimations.

[0049] Symbol-level channel update is simpler to implement and has a fast channel update frequency. Code-block-level channel update can significantly improve the accuracy of the determined reference signal.

[0050] Table 1 shows the explanations of some symbols involved in the transceiver system.

[0051] Table 1 Symbol Explanation

[0052]

[0053] Number of transmit antennas N T , Number of receive antennas N R , Number of data spatial streams N SS ≤min{N T , N R}. To simplify the subsequent description, the embodiments of the present application ignore the impact of precoding.

[0054] Transmitted signal on the k-th subcarrier of the first OFDM symbol: is an N T dimensional vector space;

[0055] Corresponding channel matrix:

[0056] Received signal:

[0057] Gaussian noise: σ 2 is the noise power, and I is used to represent the identity matrix.

[0058] The embodiments of this application can be used in scenarios such as single-input single-output (N T = N R = 1), single-input multiple-output (N T = 1, N R > 1), multiple-input single-output (N T > 1, N R = 1), multiple-input multiple-output (N T > 1, N R > 1), etc.

[0059] In the WLAN scenario, the channel matrix is slowly varying compared to the signal transmission time. Therefore, assume: H k = H 1，k = … = H l，k = … = H L，k .

[0060] The channel estimation algorithm used in this application will be outlined below.

[0061] Transmit M groups of reference signals in the time domain M ≥ N T , so as to be able to recover the channel information.

[0062] Represent it in matrix form:

[0063] Correspondingly, there is:

[0064] Among them,

[0065] Estimate the channel of the k-th subcarrier according to the least squares criterion:

[0066]

[0067] Among them, represents the conjugate transpose of matrix X k .

[0068] For the OFDM scenario, the estimation error can be reduced by weighted averaging the channel estimation results of different subcarriers. Denote SF{·} as the smoothing operator.

[0069]

[0070] k represents the k-th subcarrier in the OFDM symbol, and the window length of the smoothing window is 2w + 1. is a set of smoothing coefficients, which can be determined by Wiener filtering.

[0071] It should be noted that the above is an overview of the channel estimation algorithm. This application involves channel estimation based on long training fields and channel estimation based on reference signals at the current level. The specific implementation of this algorithm in the two channel estimation processes will be further described later.

[0072] Channel equalization based on long training fields includes: channel estimation based on long training fields, demodulation of signaling, phase tracking, and channel equalization. Each process will be described separately below.

[0073] (1) Channel estimation based on long training fields:

[0074] Matrix X composed of M groups of reference signals LTF,k , received signal matrix Y LTF,k , perform channel estimation and subcarrier smoothing:

[0075]

[0076] H LTF,k = SF{H LTF,k}

[0077] Take the channel estimation result based on the long training field as the initialization of channel update, that is, H 0,k = H LrF,k , which is used for channel equalization of the first-level data unit.

[0078] (2) Phase tracking:

[0079] Due to residual frequency offset and sampling clock deviation, the received signal will continuously accumulate phase error:

[0080]

[0081] CPE l = l∈

[0082] STO l = lμ

[0083] Among them, the OFDM symbol number l = 1,..., L, the total number of subcarriers K, the subcarrier number k, the common phase error (CPE) increases with time in steps of frequency offset ∈, and the symbol time offset (STO) increases with time in steps of clock deviation μ.

[0084] The CPE and STO parameter estimation utilizes the known pilot signals on each OFDM symbol. is the subcarrier index corresponding to the pilot signal, and the pilot position set is determined by the communication protocol.

[0085] The corresponding received signal is The phase error estimation value can be obtained as:

[0086]

[0087] Utilize the phase difference estimation values on all pilot subcarriers of the l-th OFDM symbol A system of equations can be established, and the estimated values of CPE and STO, CPE l and STO l ;

[0088] It is also possible to perform time-domain filtering on the sequence of estimated values accumulated historically and to reduce the error of the estimation result.

[0089]

[0090] Denote TF{·} as the time-domain filtering operator.

[0091] Up to this point, the phase error estimation value on any subcarrier can be obtained:

[0092]

[0093] Furthermore, perform phase compensation on the received signal:

[0094]

[0095] (3) Channel equalization:

[0096] The compensated signal: Channel equalization can adopt zero-forcing equalization or minimum mean square error equalization.

[0097] For zero-forcing equalization, if N T ≥ N R (H l,k is a row full-rank matrix), then If N r ≤ N R (H l,k is a column full-rank matrix), then

[0098] For minimum mean square error equalization, if N T ≥ N R (H l，kIf it is a row full-rank matrix, then If N T ≤N R (H l,k is a column full-rank matrix), then

[0099] where σ 2 is the noise power.

[0100] The following will separately elaborate on the two channel update methods in detail.

[0101] Method 1: Symbol-level channel update.

[0102] In an embodiment of the present application, where the data unit is a preset number of OFDM symbols;

[0103] Based on the first channel equalization result, determining the reference signal of the current layer, including:

[0104] Demapping the first channel equalization result;

[0105] Performing soft decision on the demapping result;

[0106] Performing hard decision on the soft decision result;

[0107] Performing modulation mapping on the hard decision result to obtain the reference signal of the current layer.

[0108] The received signal includes L consecutive OFDM symbols. Channel estimation can be performed every M received OFDM symbols, M≥N T , generally taking M = N T is sufficient. The number of channel updates is:

[0109]

[0110] Initializing the channel as the long training field estimation result: H 0,k = H LTF,k

[0111] At the t-th channel update, the corresponding M received OFDM symbols k represents the subcarrier number, corresponding to the received matrix:

[0112] Y t，k = [y M(t-1)+1，k , …, y k,Mt

[0113] The equalized symbol set:

[0114]

[0115] ​For a single constellation symbol Let each symbol carry Q bits of information.

[0116] After equalization Map it to a Q-dimensional data sequence:

[0117]

[0118] The demapping process is the inverse process of constellation modulation

[0119] b s is the soft bit obtained after soft decision, calculated through the log-likelihood ratio

[0120]

[0121] respectively represent the probabilities that the q-th bit is 1 and 0.

[0122] Infer the transmitted bit information through hard decision:

[0123]

[0124] When the soft bit is 0, the decision is ambiguous. Therefore, two groups of sequences b h，0 and b h，1 , b h，0 sequence determines all ambiguous bits as 0, b h，1 determines all ambiguous bits as 1.

[0125] Remap the bits obtained by hard decision to constellation symbols:

[0126] x r,0 ←b h，0

[0127] x r，1 ←b h，1

[0128] Take the average of the two to obtain the updated reference signal:

[0129]

[0130] It should be noted that in the case of low-order modulation BPSK (Binary Phase Shift Keying) (MCS = 0, 1), one bit corresponds to a modulation symbol on a subcarrier. If the reference signal is updated in the above manner, a zero-valued reference signal will be generated at the ambiguous decision bit, which may cause the matrix to be singular and thus lead to the failure of channel estimation. Therefore, for low-order modulation, when encountering an ambiguous decision soft bit of 0, hard decision cannot provide effective information, and the channel update at the corresponding subcarrier position is directly cancelled.

[0131] In one embodiment of the present application, channel estimation is performed based on the reference signal of the current layer to obtain the channel estimation result of the current layer, including:

[0132] Performing primary channel estimation on the reference signal of the current layer based on the least squares criterion to obtain the channel estimation result of the current layer; wherein, the number of reference signals of the current layer is a preset number.

[0133] In summary, the updated reference signal in N T dimensions can be obtained:

[0134]

[0135] Accumulating M reference signals, the corresponding matrix:

[0136] X t，k =[x k，M(t-1)+1 , …, x k

[0137] According to the channel estimation formula, the real-time channel estimation result can be obtained:

[0138] ΔH t,k =Y t,k (X t,k ) H [X t，k (X t,k ) H -1

[0139] Since there may be errors in the hard decision result, when performing symbol-level real-time channel estimation, the subcarrier smoothing function is turned off to avoid error spread.

[0140] In one embodiment of the present application, channel update is performed based on the channel update result of the previous layer and the channel estimation result of the current layer to obtain the channel update result of the current layer, including:

[0141] Calculating the channel update result of the current layer based on the preset update step size, the channel update result of the previous layer, and the channel estimation result of the current layer;

[0142] wherein, the update step size is determined by the soft decision result or the frequency domain channel two-norm.

[0143] Updating the previous channel estimation value using the real-time channel estimation result:

[0144] H t，k =α t，k H t-1，k +β t，k ΔH t，k ​​

[0145] Among them, β t，k is the update step size, and its value range is 0 to 1. α t，k = 1 - β t，k .

[0146] The update step size can be a constant independent of both t and k, and can also be set based on the soft decision result or the frequency domain channel two-norm. Specifically, one is to set the update step size by synthesizing soft bit information, that is, the larger the absolute value of the soft bit, the greater the confidence in the hard decision, and a larger update step size (always less than 1) is set; the other is to adjust the update step size according to the frequency domain channel two-norm. The larger the channel two-norm, the stronger the signal reception strength at that frequency point, and a larger update step size can be set.

[0147] In summary, the present invention provides a feasible step size calculation formula:

[0148] β t，k = β0γ t,k ρ t,k ,

[0149]

[0150] Among them, β0 < 1 is a constant coefficient, γ t,k is determined by the current decoded soft bit, and ρ t,k is determined according to the two-norm of the previous channel update result, and ||·|| F represents the two-norm operator.

[0151] Method 2: Channel update at the code block level.

[0152] The channel update at the code block level can be used alone or in combination with the symbol-level signal update, such as taking the mean of the update results obtained from both as the channel update result.

[0153] The channel is updated in units of the coding block (referred to as the code block). After a code block is decoded, the reference signal is determined by re-encoding.

[0154] This method further includes: determining the bit number threshold based on the number of data subcarriers carried by each OFDM symbol, the number of information bits carried by each modulation symbol, and the number of reference signals required for a single channel estimation.

[0155] If the number of information bits B of the decoded code block is less than the bit number threshold, the code block information needs to be accumulated, and after the number of bits reaches the bit number threshold, the channel update is performed.

[0156] If the number of information bits B of the decoded code block is not less than the bit number threshold, the number of updates is

[0157] That is, after receiving a complete coding block (the number of information bits in the code block is the bit number threshold), an overall channel update operation is performed once, and each coding block can cover at least one channel estimation operation.

[0158] Since B is not necessarily an integer multiple of MK, when updating the channel for each coding block, the OFDM symbols involved in the current code block may contain information of the next coding block. Define R D as the number of bits belonging to the current code block among the OFDM symbols involved in each channel update, and D t as the number of bits belonging to the next code block. t Whenever it exceeds the number of bits B of the next coding block

[0159] t+1 it is determined that a code block is accumulated and a channel update can be performed. U t is the number of channel estimations that need to be performed during the current t-th channel update. Among them, l heaa,t is the starting OFDM symbol number of the t-th channel update, and l tail,t is the ending OFDM symbol number of the t-th channel update.

[0160] In an embodiment of the present application, the data unit is a coding block;

[0161] Based on the first channel equalization result, determining the reference signal of the current layer includes:

[0162] Demapping the first channel equalization result;

[0163] Soft-decision making on the demapping result;

[0164] Decoding the soft-decision result;

[0165] When the number of information bits in the code block obtained by decoding is not less than the bit number threshold, encoding is performed in units of the code block information of the bit number threshold; where the bit number threshold is determined by the number of data subcarriers carried by each OFDM symbol, the number of information bits carried by each modulation symbol, and the number of reference signals required for a single channel estimation;

[0166] Modulation mapping is performed on the encoded code block vector to obtain the reference signal of the current layer.

[0167] The demapping and soft-decision processes are the same as those in the foregoing symbol-level channel update and will not be elaborated here.

[0168] Decoding is used to recover the originally transmitted code block information.

[0169] For BCC coding, the commonly used Viterbi decoding method;

[0170] ​For LDPC coding, methods such as the sum-product decoding algorithm, the min-sum algorithm, and the layered decoding algorithm can be used.

[0171] The original bit information obtained by recovery is re-encoded to obtain the code block B of dimension t .

[0172] For LDPC coding, the parity check bits of LDPC coding can be used to check the decoding result, and the subsequent process is only executed when the check passes.

[0173] Through modulation mapping, symbols of the QAM modulation constellation are obtained as the reference signal.

[0174] In an embodiment of the present application, decoding the soft decision result includes:

[0175] Deinterleaving the soft decision result;

[0176] Converting the coded block matrix obtained by deinterleaving into a column vector;

[0177] Decoding the column vector;

[0178] Modulation mapping the coded block vector obtained by coding to obtain the reference signal of the current layer, including:

[0179] Mapping the coded block vector obtained by coding onto the code block matrix;

[0180] Interleaving the code block matrix;

[0181] Performing modulation mapping on the interleaving result to obtain the reference signal of the current layer.

[0182] For the BCC coding scenario, the method further includes interleaving and deinterleaving steps. When the number of data spatial streams is greater than 1, the method further includes parsing and inverse parsing steps. Parsing refers to mapping the coded block vector obtained by coding onto the code block matrix, and inverse parsing refers to converting the coded block matrix obtained by deinterleaving into a column vector.

[0183] The following will elaborate on each processing process in detail.

[0184] Deinterleaving:

[0185] Code block BL t Composition:

[0186] BL t,1 Contains the last D head,tt ~l head,t +M OFDM symbols corresponding bits, which can be an empty set; t-1

[0187] BL t,2 Contains the lhead,t ~l tail，t OFDM symbols corresponding to the first D t-1 +R t-1 +1 bits up to the last bit;

[0188] BL t，3 including the first l tail，t ~l tail,t +M OFDM symbols corresponding to the first R t bits, which can be an empty set;

[0189] code block BL t can be composed of N SS column vectors:

[0190] BL t =[BL t [1], …, BL t [n t , … BL t，3 [N SS

[0191] Each column vector corresponds to the soft bit information of a data stream, which is the operation unit of deinterleaving. Deinterleaving is the inverse process of interleaving, that is, according to a predefined rule, the elements in the column vector BL t [n t are rearranged, and the rule is determined by the communication protocol.

[0192] Inverse parsing:

[0193] Inverse parsing is the inverse process of stream parsing. The coded block matrix BL t is transformed into a column vector of dimension B t according to a predefined rule, and the rule is determined according to the communication protocol.

[0194]

[0195] Stream parsing:

[0196] Code blocks (code block vectors) of dimension B t are mapped to N SS data space streams (coded block matrices) according to a predefined rule.

[0197] Interleaving:

[0198] The data of each data space stream is rearranged according to a predefined rule.

[0199] In an embodiment of the present application, channel estimation is performed based on the reference signal of the current layer to obtain the channel estimation result of the current layer, including:

[0200] ​Perform multiple channel estimations on the reference signal of the current level based on the least squares criterion; wherein, each channel estimation is performed on a preset number of OFDM symbols.

[0201] Perform weighted averaging on the results of each channel estimation to obtain the channel estimation result of the current level.

[0202] Specifically, in the t-th channel update, a total of Ut real-time channel estimations are performed.

[0203] For the ut-th channel estimation, based on l head,t +(u t -1)M + 1 to l head，t + u t M + 1 of the OFDM symbols, perform real-time channel estimation:

[0204]

[0205] "Decoding + encoding" can significantly improve the accuracy of the reference signal because subcarrier smoothing can enhance the channel estimation performance.

[0206]

[0207] Take the average of the real-time channel estimation results accumulated for Ut times:

[0208]

[0209] In an embodiment of the present application, perform channel update based on the channel update result of the previous level and the channel estimation result of the current level to obtain the channel update result of the current level, including:

[0210] Based on a preset update step size, the channel update result of the previous level, and the channel estimation result of the current level, calculate the channel update result of the current level;

[0211] Wherein, the update step size is a positive number less than 1.

[0212] Use the channel estimation result to update the previous channel estimation value. The method can be the same as the symbol-level channel update. However, different from the symbol-level channel update, the code block-level channel update ensures the accuracy of the reference signal. Therefore, the requirement for the accuracy of the update step size can be not as high as that of the symbol-level channel update. It can be directly set as a positive number less than 1. If higher channel equalization performance is required, it can also be determined based on the soft decision result or the frequency domain channel two-norm.

[0213] As Figure 5 shown, it is a schematic diagram of the change of the normalized root mean square error with the signal-to-noise ratio.

[0214] baseline is used to represent the baseline, the normalized root mean square error in the case of no channel update;

[0215] symbol-based is used to represent the normalized root mean square error in the case of symbol-level channel update;

[0216] code-based is used to represent the normalized root mean square error in the case of code-block-level channel update;

[0217] symbol / code combined is used to represent the normalized root mean square error in the case where symbol-level channel update and coding-level channel update coexist;

[0218] The horizontal axis is used to represent the signal-to-noise ratio SNR (unit: dB)

[0219] The vertical axis NRMSE is used to represent the normalized root mean square error (unit: %)

[0220] The normalized root mean square error is used to evaluate the magnitude of the error in the channel estimation result.

[0221] Simulation conditions: VHT (Very High Throughput), 20M, MCS (Modulation and Coding Scheme) 7, 33000-byte-long A-MPDU aggregation packet, Gaussian channel.

[0222] Through Figure 5 It can be seen that the channel update method proposed in this application can effectively reduce the error of the channel estimation result. Under the simulation conditions, near an SNR of 20dB, the normalized root mean square error can be reduced by approximately 0.8%, 1.1%, and 1.3% respectively.

[0223] As Figure 6 shown, it is a graph of the error vector magnitude varying with the signal duration. The horizontal axis is the signal duration, and the vertical axis is the error vector magnitude.

[0224] Experimental conditions: VHT, 20M, MCS7, A-MPDU aggregation packet with a payload of 33000 bytes, slow-varying channel. A slow-varying channel refers to a channel whose characteristics change very slowly on a relatively long time scale.

[0225] From Figure 6It can be seen that the magnitude of the baseline error vector deteriorates continuously over time. The channel update method proposed in this application can reduce the magnitude of the error vector in the unpacking result of the received signal, and a gain of about 1 dB is measured under experimental conditions. From the perspective of the error vector magnitude, the channel update performance at the code block level is better than that at the symbol level. Compared with only using the channel update at the code block level, the additional gain in the error vector magnitude after the two are superimposed is small. Due to the channel update, the error vector magnitude decreases continuously over time and converges, indicating that within a certain range, the longer the packet, the better the improvement effect of the error vector magnitude.

[0226] As Figure 7 shown, it is the curve of the packet error rate varying with the signal-to-noise ratio. Simulation conditions: VHT, 20M, MCS7, long A-MPDU aggregated packet with a payload of 33000 bytes, slow-varying channel.

[0227] From Figure 7 it can be seen that the channel update method proposed in this application can effectively reduce the received packet error rate, thereby reducing retransmissions and improving the throughput. The performance is best when the two update schemes are combined. Under the simulation conditions, if the signal-to-noise ratio corresponding to the standard of packet error rate = 10% is used, it can be reduced from 20.7 dB to 20.0 dB, 19.6 dB, and 19.6 dB respectively.

[0228] The method provided in this application can also reduce the signal-to-noise ratio threshold for the receiver to work and reduce the working power consumption.

[0229] The signal-to-noise ratio threshold is the signal-to-noise ratio when the false alarm rate = 10%, and it is commonly used as an index to evaluate the packet reception performance and the power consumption of the receiver. The lower the signal-to-noise ratio threshold, the better the packet reception performance or the lower the requirements for the receiver, and the lower the power consumption can be achieved.

[0230] Figure 2 It is the signal-to-noise ratio threshold under different scenarios obtained through simulation.

[0231] Table 2 Signal-to-noise ratio thresholds under different scenarios

[0232]

[0233] The simulation settings corresponding to Table 2 are: HT _ MF signal, 20M, non-aggregated packet, payload 2048 bytes.

[0234] As Figure 8 shown, the embodiment of this application provides a channel equalization device based on channel update, including:

[0235] A channel equalization module 801, configured to perform channel equalization on the data unit at the current level based on the channel update result of the upper level to obtain the first channel equalization result at the current level;

[0236] A reference signal update module 802, configured to determine a reference signal of a current level based on a first channel equalization result;

[0237] A channel estimation module 803, configured to perform channel estimation based on the reference signal of the current level to obtain a channel estimation result of the current level;

[0238] A channel update module 804, configured to perform channel update based on a channel update result of a previous level and the channel estimation result of the current level to obtain a channel update result of the current level;

[0239] A channel equalization module 801, further configured to perform channel equalization on a data unit of the current level based on the channel update result of the current level to obtain a second channel equalization result of the current level, or, the channel update module is further configured to use a next level as the updated current level, and trigger the channel equalization module to perform channel equalization on the data unit of the current level based on the channel update result of the previous level;

[0240] Wherein, the received signal includes data units of multiple levels. If the current level is the first level, the channel update result of its previous level is a channel estimation result based on a long training field in the received signal; if the channel update result of the current level is used for channel update of the current level, the channel equalization result of the data unit of the current level is the second channel equalization result, otherwise, the channel equalization result of the data unit of the current level is the first channel equalization result.

[0241] In an embodiment of the present application, wherein the data unit is a preset number of orthogonal frequency division multiplexing (OFDM) symbols;

[0242] The reference signal update module 802 is configured to demap the first channel equalization result; perform soft decision on the demapped result; perform hard decision on the soft decision result; and perform modulation mapping on the hard decision result to obtain a reference signal of the current level.

[0243] In an embodiment of the present application, the channel estimation module 803 is configured to perform primary channel estimation on the reference signal of the current level based on the least squares criterion to obtain a channel estimation result of the current level; wherein, the number of the reference signals of the current level is a preset number.

[0244] In an embodiment of the present application, the channel update module 804 is configured to calculate a channel update result of the current level based on a preset update step size, the channel update result of the previous level, and the channel estimation result of the current level; wherein, the update step size is determined by a soft decision result or a frequency domain channel two-norm.

[0245] In an embodiment of the present application, wherein the data unit is a coding block;

[0246] The reference signal updating module 802 is configured to demap the first channel equalization result; perform soft decision on the demapped result; perform decoding on the soft decision result; when the number of information bits of the code block obtained by decoding is not less than the bit number threshold, perform encoding in units of the code block information of the bit number threshold; wherein, the bit number threshold is determined by the number of data subcarriers carried by each OFDM symbol, the number of information bits carried by each modulation symbol, and the number of reference signals required for single-channel estimation; perform modulation mapping on the encoded code block vector to obtain the reference signal of the current layer.

[0247] In an embodiment of the present application, the reference signal updating module 802 is configured to deinterleave the soft decision result; convert the encoded block matrix obtained by deinterleaving into a column vector; perform decoding on the column vector; map the encoded code block vector onto the code block matrix; perform interleaving on the code block matrix; perform modulation mapping on the interleaved result to obtain the reference signal of the current layer.

[0248] In an embodiment of the present application, the channel estimation module 803 is configured to perform multiple channel estimations on the reference signal of the current layer based on the least squares criterion; wherein, each channel estimation is performed on a preset number of OFDM symbols; perform weighted averaging on the results of each channel estimation to obtain the channel estimation result of the current layer.

[0249] In an embodiment of the present application, the channel updating module 804 is configured to calculate the channel updating result of the current layer based on a preset update step size, the channel updating result of the previous layer, and the channel estimation result of the current layer; wherein, the update step size is a positive number less than 1.

[0250] An embodiment of the present application provides a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the method described in any of the above embodiments is implemented.

[0251] In addition, although exemplary embodiments have been described herein, their scope includes any and all embodiments based on the present application having equivalent elements, modifications, omissions, combinations (e.g., solutions that cross various embodiments), adaptations, or changes. Although several embodiments have been described separately for wireless communication methods and wireless communication components, it should be noted that the method details involved in the description of wireless communication components can also be incorporated into the various embodiments of the wireless communication method, and vice versa.

[0252] The elements in the claims will be broadly interpreted based on the language used in the claims and are not limited to the examples described in this specification or during the implementation of the present application, and the examples will be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered only as examples, and the true scope and spirit are indicated by the full scope of the claims and their equivalents.

[0253] The sequence of each step in this application is merely exemplary and not restrictive. Without affecting the implementation of this application (without destroying the logical relationship between the required steps), the execution sequence of the steps can be adjusted, and the various embodiments obtained after the adjustment still fall within the scope of this application.

[0254] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more of their aspects) can be used in combination with each other. For example, those of ordinary skill in the art can use other embodiments when reading the above description. Additionally, in the above detailed description, various features can be grouped together to simplify this application. This should not be construed as an intention that the disclosed features not claimed are necessary for any claim. On the contrary, the subject matter of the present invention may be less than all the features of a particular disclosed embodiment. Thus, the claims are hereby incorporated by way of example or embodiment into the detailed description, where each claim stands on its own as a separate embodiment, and it is contemplated that these embodiments can be combined with each other in various combinations or permutations. The scope of the present invention should be determined with reference to the appended claims and the full scope of equivalents to which those claims are entitled.

Claims

1. A channel equalization method based on channel update, characterized in that, Including: Performing channel equalization on the data unit of the current level based on the channel update result of the upper level to obtain the first channel equalization result of the current level; Determining the reference signal of the current level based on the first channel equalization result; Performing channel estimation based on the reference signal of the current level to obtain the channel estimation result of the current level; Performing channel update based on the channel update result of the upper level and the channel estimation result of the current level to obtain the channel update result of the current level; Performing channel equalization on the data unit of the current level based on the channel update result of the current level to obtain the second channel equalization result of the current level, or taking the next level as the updated current level and performing channel equalization on the data unit of the current level based on the channel update result of the upper level; Wherein, the received signal includes data units of multiple levels. If the current level is the first level, the channel update result of its upper level is the channel estimation result based on the long training field in the received signal; If the channel update result of the current level is used for the channel update of the current level, the channel equalization result of the data unit of the current level is the second channel equalization result; otherwise, the channel equalization result of the data unit of the current level is the first channel equalization result.

2. The method according to claim 1, Characterized in that Wherein, the data unit is a preset number of orthogonal frequency division multiplexing (OFDM) symbols; Determining the reference signal of the current level based on the first channel equalization result includes: Demapping the first channel equalization result; Performing soft decision on the demapping result; Performing hard decision on the soft decision result; Performing modulation mapping on the hard decision result to obtain the reference signal of the current level.

3. The method according to claim 2, characterized in that Performing channel estimation based on the reference signal of the current level to obtain the channel estimation result of the current level includes: Performing primary channel estimation on the reference signal of the current level based on the least squares criterion to obtain the channel estimation result of the current level; wherein, the number of the reference signals of the current level is the preset number.

4. The method according to claim 3, characterized in that Performing channel update based on the channel update result of the upper level and the channel estimation result of the current level to obtain the channel update result of the current level includes: Calculating the channel update result of the current level based on a preset update step size, the channel update result of the upper level, and the channel estimation result of the current level; Wherein, the update step size is determined by the soft decision result or the frequency domain channel two-norm.

5. The method according to claim 1, Characterized in that Wherein, the data unit is a coding block; Determining the reference signal of the current level based on the first channel equalization result includes: Demapping the first channel equalization result; Performing soft decision on the demapping result; Performing decoding on the soft decision result; When the number of information bits of the decoded code block is not less than the bit number threshold, encoding is performed in units of the code block information with the bit number threshold; wherein, the bit number threshold is determined by the number of data subcarriers carried by each OFDM symbol, the number of information bits carried by each modulation symbol, and the number of reference signals required for single-channel estimation. Perform modulation mapping on the encoded code block vector to obtain the reference signal of the current layer.

6. The method according to claim 5, wherein Decoding the soft decision result includes: Deinterleaving the soft decision result; Converting the encoded block matrix obtained by deinterleaving into a column vector; Decoding the column vector; Performing modulation mapping on the encoded code block vector to obtain the reference signal of the current layer, including: Mapping the encoded code block vector onto a code block matrix; Interleaving the code block matrix; Performing modulation mapping on the interleaving result to obtain the reference signal of the current layer.

7. The method according to claim 5, wherein Performing channel estimation based on the reference signal of the current layer to obtain the channel estimation result of the current layer, including: Performing multiple channel estimations on the reference signal of the current layer based on the least squares criterion; wherein, each channel estimation is for a preset number of OFDM symbols; Performing weighted averaging on the channel estimation results of each time to obtain the channel estimation result of the current layer.

8. The method according to claim 7, wherein Performing channel update based on the channel update result of the previous layer and the channel estimation result of the current layer to obtain the channel update result of the current layer, including: Calculating the channel update result of the current layer based on a preset update step size, the channel update result of the previous layer, and the channel estimation result of the current layer; wherein, the update step size is a positive number less than 1.

9. A channel equalization device based on channel update, characterized in that including: A channel equalization module configured to perform channel equalization on the data unit of the current layer based on the channel update result of the previous layer to obtain the first channel equalization result of the current layer; A reference signal update module configured to determine the reference signal of the current layer based on the first channel equalization result; A channel estimation module configured to perform channel estimation based on the reference signal of the current layer to obtain the channel estimation result of the current layer; A channel update module configured to perform channel update based on the channel update result of the previous layer and the channel estimation result of the current layer to obtain the channel update result of the current layer; The channel equalization module is further configured to perform channel equalization on the data unit of the current layer based on the channel update result of the current layer to obtain the second channel equalization result of the current layer, or, the channel update module is further configured to use the next layer as the updated current layer and trigger the channel equalization module to perform channel equalization on the data unit of the current layer based on the channel update result of the previous layer; Among them, the received signal includes multiple levels of data units. If the current level is the first level, the channel update result of its upper level is the channel estimation result based on the long training field in the received signal; If the channel update result of the current level is used for the channel update of the current level, the channel equalization result of the data unit of the current level is the second channel equalization result; otherwise, the channel equalization result of the data unit of the current level is the first channel equalization result.

10. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, the method according to any one of claims 1 to 8 is implemented.

Images