Signal detection method, signal detection device, and electronic device

Abstract

This disclosure relates to the field of communication transmission technology, specifically to a signal detection method, a signal detection device, and an electronic device. The signal detection method includes: acquiring a frequency domain signal to be processed based on wireless transmission; determining a first power delay spectrum of multiple traditional long training fields; determining a decision symbol sequence through hard-decision traditional signal fields; determining a second power delay spectrum of the decision symbol sequence; performing channel estimation on the extended long training fields based on the first and second power delay spectra to obtain the channel estimation result of the extended long training fields; and performing data detection based on the channel estimation result to obtain data information transmitted through the frequency domain signal to be processed. This method enables multi-dimensional joint characterization of the multipath characteristics of wireless channels, effectively suppressing noise and improving the accuracy of channel statistical characteristics, thereby effectively reducing transmission errors, improving data transmission reliability, and ensuring overall wireless communication performance.

Description

Technical Field

[0001] This disclosure relates to the field of communication transmission technology, specifically to a signal detection method, a signal detection device, and an electronic device. Background Technology

[0002] In wireless communication systems based on Orthogonal Frequency Division Multiplexing (OFDM) technology, such as Wi-Fi, LTE, and NR, pilot-assisted two-dimensional linear minimum mean square error (2D-LMMSE) estimation is a linearly optimal channel estimation method. However, due to its high computational complexity, practical systems typically simplify this by performing LMMSE channel estimation separately in the time and frequency domains.

[0003] In related technologies, when detecting wireless signals that are not in the 802.11a / g frame format, channel detection is mainly based on the traditional Long Training Field (L-LTF) and the Extended Long Training Field (X-LTF).

[0004] However, in actual detection, subsequent processing is mainly based on the channel estimation results of L-LTF signals in the time domain. The information dimension used is relatively simple, which affects the accuracy of X-LTF detection and thus the overall signal detection accuracy. Summary of the Invention

[0005] To overcome the problems existing in related technologies, an exemplary embodiment of this disclosure provides a signal detection method, the method comprising: acquiring a frequency domain signal to be processed based on wireless transmission, wherein the frequency domain signal to be processed includes a conventional long training field, a conventional signal field, and an extended long training field; determining a first power delay spectrum of multiple conventional long training fields; determining a decision symbol sequence by hard-decision conventional signal fields; determining a second power delay spectrum of the decision symbol sequence; performing channel estimation on the extended long training field based on the first power delay spectrum and the second power delay spectrum to obtain a channel estimation result of the extended long training field; and performing data detection based on the channel estimation result to obtain data information transmitted through the frequency domain signal to be processed.

[0006] In some embodiments, determining the first power delay spectrum of multiple traditional long training fields includes: performing channel estimation processing on each traditional long training field sequentially according to the reception timing of the multiple traditional long training fields to obtain the channel frequency domain response of each traditional long training field; performing time domain transformation processing on each channel frequency domain response to obtain multiple first time domain channel responses; and determining the first power delay spectrum based on the fusion result of the power delay spectra of the multiple first time domain channel responses.

[0007] In some embodiments, the number of traditional long training fields is two. The first power delay spectrum is determined based on the fusion result of the power delay spectra of multiple first time-domain channel responses, including: determining the power delay spectrum of each first time-domain channel response respectively; averaging the two power delay spectra and using the averaged result as the first power delay spectrum.

[0008] In some embodiments, channel estimation is performed on the extended-length training field based on the first power delay spectrum and the second power delay spectrum to obtain the channel estimation result of the extended-length training field, including: determining the target frequency domain correlation coefficient based on the fusion result of the first power delay spectrum and the second power delay spectrum; and performing channel estimation on the extended-length training field by frequency domain Wiener filtering based on the target frequency domain correlation coefficient to determine the channel estimation result of the extended-length training field.

[0009] In some embodiments, determining the target frequency domain correlation coefficient based on the fusion result of the first power delay spectrum and the second power delay spectrum includes: averaging the first power delay spectrum and the second power delay spectrum to obtain the target power delay spectrum; performing noise reduction processing on the target power delay spectrum to obtain the noise reduction result; and performing frequency domain transformation processing on the noise reduction result to obtain the target frequency domain correlation coefficient.

[0010] In some embodiments, channel estimation of the extended-length training field is performed based on the target frequency domain correlation coefficient using frequency domain Wiener filtering to determine the channel estimation result of the extended-length training field, including: determining the Wiener filter coefficient matrix based on the channel statistical characteristics of the target frequency domain correlation coefficient; and performing channel estimation of the extended-length training field using the Wiener filter coefficient matrix to obtain the channel estimation result of the extended-length training field.

[0011] In some embodiments, determining the Wiener filter coefficient matrix based on the channel statistical characteristics of the target frequency domain correlation coefficient includes: determining the frequency domain correlation matrix based on multiple subcarriers corresponding to the target frequency domain correlation coefficient; and determining the Wiener filter coefficient matrix based on the frequency domain correlation matrix.

[0012] In some embodiments, the frequency domain signal to be processed further includes an extended signal field, and the method further includes: performing signal detection on the traditional signal field and the extended signal field sequentially based on the traditional long training field to obtain a detection result.

[0013] Secondly, this disclosure also provides a signal detection device, comprising: an acquisition module for acquiring a frequency domain signal to be processed based on wireless transmission, wherein the frequency domain signal to be processed includes a conventional long training field, a conventional signal field, and an extended long training field; a first processing module for determining a first power delay spectrum of multiple conventional long training fields; determining a decision symbol sequence through hard-decision conventional signal fields; and determining a second power delay spectrum of the decision symbol sequence; a channel estimation module for performing channel estimation on the extended long training field based on the first power delay spectrum and the second power delay spectrum to obtain a channel estimation result for the extended long training field; and a first detection module for performing data detection based on the channel estimation result to obtain data information transmitted through the frequency domain signal to be processed.

[0014] Thirdly, this disclosure also provides an electronic device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the signal detection method provided in any of the above aspects by executing the computer instructions.

[0015] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.

[0016] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: According to the signal detection method provided by this disclosure, determining the first power delay spectrum through multiple traditional long training fields can improve the accuracy of the power delay spectrum, reduce errors caused by noise interference, and improve the reliability of channel characteristics. By performing hard decision on traditional signal fields to obtain a decision symbol sequence, and calculating the second power delay spectrum based on this sequence, the detection bias caused by relying solely on traditional long training fields for channel estimation can be avoided. Fusing the first power delay spectrum and the second power delay spectrum can achieve a multi-dimensional joint representation of the multipath characteristics of the wireless channel, effectively suppressing noise, improving the accuracy of channel statistical features, and thus providing a more accurate convergence direction for the precise estimation of extended long training fields, making the channel estimation results more accurate and reliable. Based on this, data detection can effectively reduce transmission errors, improve data transmission reliability, and thus ensure the overall wireless communication performance. Attached Figure Description

[0017] This disclosure can be better understood by describing exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, in which: Figure 1 This is a schematic diagram illustrating a signal detection process according to an exemplary embodiment of a public document; Figure 2 This is a flowchart illustrating a signal detection method according to an exemplary embodiment of a published document; Figure 3 This is a schematic diagram illustrating a process for determining a target frequency domain correlation coefficient according to an exemplary embodiment of a public document; Figure 4 This is a flowchart illustrating another signal detection method according to an exemplary embodiment of a published document; Figure 5 This is a schematic diagram of another signal detection process according to an exemplary embodiment of a public disclosure; Figure 6 This is a schematic diagram of the structure of a signal detection device according to an exemplary embodiment disclosed in a book. Detailed Implementation

[0018] The following describes specific embodiments of this disclosure. It should be noted that, in order to provide a concise description, this specification cannot exhaustively describe all features of the actual embodiments. It should be understood that, in the actual implementation of any embodiment, just as in any engineering or design project, various specific decisions are often made to achieve the developer's specific goals and to meet system-related or business-related constraints, and this can change from one embodiment to another. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this disclosure, changes in design, manufacturing, or production based on the technical content disclosed in this disclosure are merely conventional technical means and should not be construed as insufficient content of this disclosure.

[0019] Unless otherwise defined, the technical or scientific terms used in this disclosure shall have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms “first,” “second,” and similar terms used in this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. The terms “a” or “one,” etc., do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “comprising,” “including,” etc., mean that the element or object preceding “comprising” or “including” encompasses the elements or objects listed following “comprising” or “including” and their equivalents, and do not exclude other elements or objects. The terms “connected,” “linked,” etc., are not limited to physical or mechanical connections, nor are they limited to direct or indirect connections.

[0020] In related technologies, the signal detection process for performing LMMSE channel estimation on wireless signals in the frequency domain can be as follows: Figure 1As shown, radio frequency (RF) data is obtained by receiving airborne wireless signals through an antenna. The RF data is then converted into a time-domain baseband digital signal through digital front-end processing. A Fast Fourier Transform (FFT) is performed on the time-domain baseband digital signal to convert it into a frequency-domain signal. Coarse channel estimation (e.g., least squares (LS) channel estimation) is performed using the traditional long training field (L-LTF) in the frequency-domain received signal to obtain the channel's frequency response. Based on the L-LTF coarse channel estimation results, demodulation detection is performed on the traditional signal field (L-SIG) to complete frame synchronization and frame format identification, outputting the log-likelihood ratio (LLR). Then, demodulation detection is performed on the extended signal field (X-SIG) to obtain relevant configuration parameters of the wireless signal, such as modulation and coding scheme (MCS), spatial stream number, and OFDMA parameters. Based on the coarse channel estimation results of L-LTF and the detection information of L-SIG, frequency domain correlation coefficients are calculated. These coefficients are then used to perform fine channel estimation on the extended long training field (X-LTF), yielding the optimized channel frequency response. According to the fine channel estimation results of X-LTF, the data field (DATA) signal is equalized, demodulated, and demapped to obtain the encoded bitstream. Channel decoding is then performed on the encoded bitstream to correct transmission errors, ultimately restoring the original information bits or Physical Layer (PHY) Protocol Data Units (PPDUs) for processing by upper-layer protocols.

[0021] When detecting wireless signals in 802.11a / g frame format, since there is no Extended Long Training Field (X-LTF), there is no need to perform fine channel estimation processing for X-LTF.

[0022] When detecting wireless signals that are not in the 802.11a / g frame format, channel detection using the Extended Long Training Field (X-LTF) is mainly based on the channel estimation results of the L-LTF signal in the time domain. The information dimension used is relatively simple, which will affect the insufficient accuracy of the channel estimation of X-LTF, thus affecting the overall accuracy of signal detection.

[0023] Therefore, this disclosure provides a signal detection method. For example... Figure 2 As shown, the signal detection method may include the following steps: Step S210: Obtain the frequency domain signal to be processed based on wireless transmission.

[0024] The frequency domain signal to be processed can be understood as the frequency domain signal obtained after processing the wireless signal based on wireless transmission through analog-to-digital conversion, which is used as the signal basis for subsequent channel estimation. The frequency domain signal to be processed includes a traditional long training field, a traditional signal field, and an extended long training field.

[0025] The acquisition process may include: in response to receiving a wireless signal transmitted over the air, performing analog-to-digital conversion on the wireless signal to convert the analog radio frequency signal into digital radio frequency data; processing the digital radio frequency data through a digital front-end to convert the digital radio frequency data into a time-domain baseband digital signal; and performing a Fast Fourier Transform (FFT) on the time-domain baseband digital signal to convert the time-domain baseband digital signal into a frequency-domain signal, thereby obtaining the frequency-domain signal to be processed.

[0026] Step S220: Determine the first power delay spectrum of multiple traditional long training fields.

[0027] Since each frame of a wireless signal includes a traditional long training field, a traditional signal field, and an extended long training field at fixed locations within its frame structure, a first power delay profile (PDP) is determined for multiple traditional long training fields to ensure the accuracy of subsequent channel estimation using the extended long training field. This PDP reflects the multipath delay distribution of the wireless channel. Determining the first PDP using multiple traditional long training fields helps improve the accuracy of the power delay profile, reduces errors caused by noise interference, and thus enhances the reliability of channel characteristics.

[0028] Step S230: Determine the decision symbol sequence using the hard decision conventional signal field.

[0029] Since the channel information of the traditional signal field can also help improve the channel estimation accuracy of the subsequent extended long training field, in order to further enhance the reliability of channel estimation, the traditional signal field is demodulated and hard-determined to transform the originally unknown data symbols into a known decision symbol sequence, so that the channel estimation can be performed based on the decision symbol sequence, which can provide additional channel feature references for the processing of the extended long training field.

[0030] For example, the frequency domain signal to be processed is For example, the logic for hard-decision judgment on traditional signal fields can be as follows: ; in, This represents the frequency domain received signal on the k-th subcarrier. Indicates the first k Channel frequency response on each subcarrier Indicates the first kTraditional signal fields on each subcarrier Indicates the first k Additive white Gaussian noise on each subcarrier k This indicates the number of the data subcarrier in the traditional signal field of the frequency domain (Data Tone index).

[0031] Step S240: Determine the second power delay spectrum of the decision symbol sequence.

[0032] Since the traditional signal field and the traditional long training field experience the same wireless channel, the traditional signal field is converted into a known decision symbol sequence through hard decision, and the second power delay spectrum calculated based on the decision symbol sequence can be used as a supplementary reference to the first power delay spectrum to correct and optimize the multipath distribution results of the first power delay spectrum. This can effectively suppress the interference of noise on channel estimation, improve the accuracy and reliability of channel feature extraction, and thus provide more comprehensive channel feature support for the subsequent fine channel estimation of the extended long training field.

[0033] Step S250: Based on the first power delay spectrum and the second power delay spectrum, channel estimation is performed on the extended long training field to obtain the channel estimation result of the extended long training field.

[0034] By jointly characterizing the first and second power delay spectra, the multipath characteristics of the wireless channel can be comprehensively depicted from multiple dimensions, thereby effectively suppressing noise interference and reducing the risk of misidentification, ensuring the reliability and stability of the coarse estimation. Based on this, channel estimation using extended training fields can provide a more precise convergence direction for fine estimation, resulting in more accurate and reliable channel estimation results.

[0035] Step S260: Based on the channel estimation results, data detection is performed to obtain the data information transmitted through the frequency domain signal to be processed.

[0036] By improving the accuracy of precise estimation of the wireless channel, and then performing data detection based on the obtained channel estimation results, the final reconstructed data information can be more accurate, effectively reducing transmission errors and improving data transmission efficiency, thereby helping to ensure the overall performance of wireless communication.

[0037] According to the signal detection method provided in this disclosure, determining the first power delay spectrum through multiple traditional long training fields can improve the accuracy of the power delay spectrum, reduce errors caused by noise interference, and improve the reliability of channel features. By performing hard decision on traditional signal fields to obtain a decision symbol sequence, and calculating the second power delay spectrum based on this sequence, detection bias caused by relying solely on traditional long training fields for channel estimation can be avoided. Fusing the first and second power delay spectra enables a multi-dimensional joint representation of the multipath characteristics of the wireless channel, effectively suppressing noise and improving the accuracy of channel statistical features. This provides a more accurate convergence direction for the precise estimation of extended long training fields, making the channel estimation results more accurate and reliable. Based on this, data detection can effectively reduce transmission errors, improve data transmission reliability, and thus ensure overall wireless communication performance.

[0038] In some embodiments, step S220 may include: Step a1: According to the reception timing of multiple traditional long training fields, channel estimation processing is performed on each traditional long training field in sequence to obtain the channel frequency domain response of each traditional long training field.

[0039] Since multiple traditional long training fields are consecutive in transmission timing and generated in the same way, in order to avoid inaccurate channel estimation due to noise during the detection process, channel estimation is performed on each traditional long training field sequentially according to the reception timing of the multiple traditional long training fields to obtain the channel frequency domain response of each traditional long training field. This allows the multiple channel frequency domain responses to be cross-checked and compared, ensuring the stability and reliability of the subsequent determination of the first power delay spectrum.

[0040] In some examples, the channel frequency domain response can be obtained by least squares (LS) channel estimation on a conventional long training field.

[0041] Step a2: Perform time-domain transformation on each channel frequency domain response to obtain multiple first time-domain channel responses.

[0042] Converting the channel frequency domain response into a time domain form helps to extract channel multipath features and quickly determine the time domain channel features corresponding to each traditional long training field, thus providing a basis for the subsequent calculation of the first power delay spectrum.

[0043] The time-domain transformation of the channel frequency domain response can be achieved by performing an Inverse Fast Fourier Transform (IFFT) on the channel frequency domain response. For example, when the traditional long training field corresponds to 64 subcarriers, the time-domain transformation of the channel frequency domain response can be performed by performing an N=64-point IFFT to obtain the corresponding first time-domain channel response.

[0044] Step a3: Determine the first power delay spectrum based on the fusion result of the power delay spectra of multiple first time-domain channel responses.

[0045] The power delay spectrum of each first time-domain channel response is determined and fused to reduce errors caused by noise interference, thereby obtaining a reliable first power delay spectrum. The fusion process may include, but is not limited to, weighted averaging or direct averaging of the power delay spectra of multiple first time-domain channel responses.

[0046] In some examples, the number of traditional long training fields is two. Therefore, the process of determining the first power delay spectrum can include: determining the power delay spectrum of each first time-domain channel response separately; and averaging the two power delay spectra, using the averaged result as the first power delay spectrum. That is, for the first time-domain channel responses of two consecutive traditional long training fields, the corresponding power delay spectra are calculated separately, and then the two power delay spectra are averaged to suppress noise errors. The averaged result is then directly used as the first power delay spectrum, which helps simplify the processing and improve the efficiency of determining the first power delay spectrum.

[0047] In other examples, a weighted average of the two power delay spectra can be performed, and the result can be used as the first power delay spectrum. Since the channel changes slowly as the signal propagates through the wireless channel, and the later symbol is closer to the current processing time, for two traditional long training fields, the weight of the power delay spectrum corresponding to the later transmitted traditional long training field can be configured to be greater than the weight of the earlier transmitted power delay spectrum. For example, the weight of the power delay spectrum corresponding to the later transmitted traditional long training field is 0.6, and the weight of the earlier transmitted field is 0.4. Obtaining the first power delay spectrum by weighted summation of the power delay spectra corresponding to the two traditional long training fields helps improve the accuracy of determining the first power delay spectrum. Alternatively, weights can be assigned based on the energy of the two first time-domain channel responses, configuring the weight corresponding to the first time-domain channel response with larger energy to be greater than that with smaller energy.

[0048] In some embodiments, step S240 may include: Step b1: Determine the target frequency domain correlation coefficient based on the fusion result of the first power delay spectrum and the second power delay spectrum.

[0049] Since both the first and second power delay spectra belong to the time-domain channel characteristics, while the extended training field is a frequency-domain signal, the first and second power delay spectra are fused to ensure the effectiveness of subsequent channel estimation. This results in a fusion result that fully characterizes the multipath characteristics of the current wireless channel in the time domain. A Fast Fourier Transform (FFT) is then performed on this fusion result to transform it to the frequency domain, yielding the target frequency domain correlation coefficient.

[0050] By fusing the dual-power delay spectrum, noise interference can be effectively suppressed, the reliability of channel characteristics can be improved, and the target frequency domain correlation coefficient can be made to better fit the real channel state, thereby providing accurate prior support for subsequent fine channel estimation of extended long training fields.

[0051] In some examples, step b1 above may include: Step b11: Average the first power delay spectrum and the second power delay spectrum to obtain the target power delay spectrum; Step b12: Denoise the target power time delay spectrum to obtain the denoising result; Step b13: Perform frequency domain transformation on the noise reduction result to obtain the target frequency domain correlation coefficient.

[0052] Specifically, since the first power delay spectrum and the second power delay spectrum reflect the multipath delay distribution of the same wireless channel, but since the noise contained in the two is independent of each other, in order to obtain a target power delay spectrum that can characterize the wireless channel, the first power delay spectrum and the second power delay spectrum are averaged to enhance the real multipath components by mutually suppressing random noise, thereby ensuring the accuracy and stability of the target power delay spectrum.

[0053] Denoising the target power time delay spectrum can further reduce noise interference and preserve as many true and effective multipath components as possible. Performing frequency domain transformation on the denoising result can effectively improve transformation accuracy and ensure the reliability of the target frequency domain correlation coefficient. For example, when denoising the target power time delay spectrum, a noise power threshold can be estimated first, and then denoising can be performed based on the threshold. The noise threshold can be calculated using the following formula: ; in, The noise power threshold is represented by `noiseStart`, the starting index for noise estimation, and the ending index for noise estimation. Characterize the target power delay spectrum. With the target power delay spectrum corresponding to the 64 subcarriers in the channel, the values ​​from noiseStart to noiseEnd range from 0 to 63.

[0054] This formula is used to average the noise-only intervals (i.e., intervals without true multipath components) in the target power time delay spectrum to obtain an accurate noise floor. Taking the 64-point time-domain response corresponding to the target power time delay spectrum as an example, noiseStart and noiseEnd can be selected from 0 to 63 as the noise estimation interval, used to average the noise portion and avoid interference from true multipath components in the threshold calculation. After obtaining the noise threshold noiseThr, the components in the target power time delay spectrum with power lower than noiseThr are set to zero, and only the true multipath components with power higher than the threshold are retained, thus completing the noise reduction process and obtaining the noise reduction result. Noise Reduction Result The expression can be as follows: .

[0055] In some examples, the target power delay spectrum can be filtered before frequency domain transformation, for example, by using a finite impulse response (FIR) filter to remove noise spikes and improve the reliability and effectiveness of noise reduction.

[0056] In some optional application scenarios, taking two traditional long training fields (L-LFT), namely L-LFT sym#0 and L-LFT sym#1, as an example, the process of determining the target frequency domain correlation coefficient can be as follows: Figure 3 As shown. The transmission timing of L-LFTsym#0 is earlier than that of L-LFTsym#1, and consequently, the reception timing of L-LFTsym#0 is also earlier than that of L-LFTsym#1.

[0057] During the process of receiving the frequency domain signal to be processed, in response to the currently received L-LFT sym#0, channel estimation processing is performed on L-LFT sym#0 through LS to obtain the channel frequency domain response of L-LFT sym#0. Then, IFFT is performed on the channel frequency domain response of L-LFT sym#0 to obtain the power delay spectrum of L-LFT sym#0.

[0058] In response to the current received L-LFT sym#1, channel estimation processing is performed on L-LFT sym#1 through LS to obtain the channel frequency domain response of L-LFT sym#1. Then, IFFT is performed on the channel frequency domain response of L-LFT sym#1 to obtain the power delay spectrum of L-LFT sym#1.

[0059] Since there are two L-LFTs, after obtaining the power delay spectrum of L-LFT sym#1, it is averaged with the power delay spectrum of L-LFT sym#0 to obtain the first power delay spectrum corresponding to L-LFT.

[0060] In response to the currently received traditional signal field (L-SIG), a hard decision is made to determine the decision symbol sequence. Channel estimation is performed on the decision symbol sequence using LS, and then IFFT is performed on the processed result to obtain the second power delay spectrum.

[0061] The target power delay spectrum is obtained by averaging the first power delay spectrum and the second power delay spectrum.

[0062] The target power delay spectrum is denoised, and the denoised result is subjected to FFT to obtain the target frequency domain correlation coefficient.

[0063] Step b2: Based on the target frequency domain correlation coefficient, channel estimation is performed on the extended long training field using frequency domain Wiener filtering to determine the channel estimation result of the extended long training field.

[0064] Since the target frequency domain correlation coefficient is generated by frequency domain transformation of the target power delay spectrum obtained by fusing the dual power delay spectra, its channel characteristic accuracy and stability are greatly improved. Therefore, channel estimation of the extended long training field based on the target frequency domain correlation coefficient can integrate multi-dimensional channel information to achieve comprehensive estimation, making the channel estimation results of the extended long training field more accurate and reliable.

[0065] In some examples, step b2 above may include: Step b21: Determine the Wiener filter coefficient matrix based on the channel statistical characteristics of the target frequency domain correlation coefficient; Step b22: Channel estimation is performed on the extended long training field using the Wiener filter coefficient matrix to obtain the channel estimation result of the extended long training field.

[0066] Specifically, Wiener filtering is an optimal linear filtering method based on the minimum mean square error criterion, and the accuracy of its filter coefficients directly determines the accuracy of channel estimation.

[0067] By determining the channel statistical characteristics (such as the channel autocorrelation matrix, noise power, etc.) of the target frequency domain correlation coefficient, a more realistic channel state can be reflected through the channel statistical characteristics, thereby enabling the determined Wiener filter coefficient matrix to achieve the optimal noise reduction effect.

[0068] Based on the Wiener filter coefficient matrix, the initial channel estimation results of the extended long training field are filtered, which can effectively suppress noise interference. While preserving the true multipath characteristics of the channel, the accuracy and stability of the channel estimation are greatly improved, and finally, high-precision channel estimation results of the extended long training field are obtained, providing reliable support for signal detection in subsequent data fields.

[0069] In other examples, the frequency domain correlation matrix can be determined based on multiple subcarriers corresponding to the target frequency domain correlation coefficient, and then the Wiener filter coefficient matrix can be determined based on the frequency domain correlation matrix.

[0070] Since the extended-length training field is a known sequence, it can first be subjected to least squares (LS) channel estimation to obtain an initial estimation result. Then, based on the channel statistical characteristics of the target frequency domain correlation coefficient, the cross-correlation matrix between the real channel and the LS estimation result of the extended-length training field, the autocorrelation matrix of the LS estimation result of the extended-length training field, and the signal-to-noise ratio of the received signal at the receiver are calculated. Among them, the frequency domain correlation matrix includes the cross-correlation matrix between the real channel (target frequency domain correlation coefficient) and the LS estimation result of the extended-length training field, as well as the autocorrelation matrix of the LS estimation result of the extended-length training field.

[0071] The Wiener filter coefficient matrix W is obtained using the following formula: ; in, This represents the cross-correlation matrix between the real channel and the LS estimation results of the extended-length training field. This represents the autocorrelation matrix of the LS estimation results for extended long training fields. This indicates the signal-to-noise ratio of the signal received at the receiving end. This represents the identity matrix. The formula is derived based on the Linear Minimum Mean Square Error (LMMSE) criterion. By introducing high-precision prior channel statistics provided by the target frequency domain correlation coefficient, the optimal Wiener filter coefficient matrix is ​​obtained, thereby achieving optimal noise reduction for channel estimation of extended long training fields.

[0072] In some examples, the frequency domain signal to be processed also includes an extended signal field, such as Figure 4 As shown, the signal detection method may further include: Step S270: Based on the traditional long training field, perform signal detection on the traditional signal field and the extended signal field in sequence to obtain the detection result.

[0073] For traditional long training fields, channel estimation and detection can be performed directly (e.g., using least squares (LS) channel estimation) to obtain the channel's frequency response. Based on this channel's frequency response, demodulation and detection are performed on the traditional signal field (L-SIG) to complete frame synchronization and frame format identification, outputting the log-likelihood ratio (LLR). Then, demodulation and detection are performed on the extended signal field (X-SIG) to obtain the relevant configuration parameters of the frequency domain signal to be processed, such as modulation and coding scheme (MCS), spatial stream number, OFDMA parameters, etc.

[0074] By determining the test results, we can provide clear parameter basis for the subsequent processing of data fields, ensure the correctness and effectiveness of data reception, and thus improve the data reception rate.

[0075] In some optional application scenarios, such as Figure 5 As shown, the process of frequency domain detection for WiFi signals can be as follows: Radio frequency (RF) data is obtained by receiving wireless signals from the air through an antenna. This RF data is then converted into a time-domain baseband digital signal through digital front-end processing. Finally, a Fast Fourier Transform (FFT) is performed on the time-domain baseband digital signal to convert it into a frequency-domain signal to be processed.

[0076] LS channel estimation is performed using the traditional long training field (L-LTF) in the received signal in the frequency domain to obtain the channel's frequency response. Based on the coarse channel estimation results of the traditional long training field, demodulation detection is performed on the traditional signal field (L-SIG) to complete frame synchronization and frame format identification, and the log-likelihood ratio (LLR) is output. Then, demodulation detection is performed on the extended signal field (X-SIG) to obtain the relevant configuration parameters of the wireless signal, providing clear parameter basis for subsequent data field processing.

[0077] In the process of receiving traditional long training fields, channel estimation is performed on two consecutive traditional long training fields to obtain the channel frequency domain response of each traditional long training field. Then, time-domain transformation is performed on each channel frequency domain response to obtain two first time-domain channel responses. The two first time-domain channel responses are averaged to obtain the first power delay spectrum. Hard decision is performed on the traditional signal field (L-SIG) to determine the decision symbol sequence. Channel estimation is performed on the decision symbol sequence using LS, and then IFFT is performed on the processed result to obtain the second power delay spectrum. The first and second power delay spectra are averaged to obtain the target power delay spectrum. Noise reduction is performed on the target power delay spectrum, and FFT is performed on the output noise reduction result to obtain the target frequency domain correlation coefficient. Based on the target frequency domain correlation coefficient, channel estimation is performed on the extended long training field using frequency domain Wiener filtering to determine the channel estimation result of the extended long training field.

[0078] Based on the precise channel estimation results of X-LTF, the data field (DATA) is equalized, demodulated, and demapped to obtain the encoded bit stream. Channel decoding is then performed on the encoded bit stream to correct transmission errors, and finally the original information bits or Physical Layer (PHY) Protocol Data Unit (PPDU) are restored and delivered to the upper layer protocol for processing.

[0079] Based on the same inventive concept, this disclosure also provides a signal detection device. For example... Figure 6 As shown, the signal detection device 300 may include: The acquisition module 310 is used to acquire the frequency domain signal to be processed based on wireless transmission, wherein the frequency domain signal to be processed includes a traditional long training field, a traditional signal field, and an extended long training field; The first processing module 320 is used to determine the first power delay spectrum of multiple traditional long training fields; determine the decision symbol sequence through hard-decision traditional signal fields; and determine the second power delay spectrum of the decision symbol sequence. The channel estimation module 330 is used to perform channel estimation on the extended long training field based on the first power delay spectrum and the second power delay spectrum, so as to obtain the channel estimation result of the extended long training field. The first detection module 340 is used to perform data detection based on the channel estimation results to obtain data information transmitted through the frequency domain signal to be processed.

[0080] In some embodiments, the first processing module 320 includes: a first processing unit, configured to perform channel estimation processing on each traditional long training field sequentially according to the reception timing of multiple traditional long training fields, to obtain the channel frequency domain response of each traditional long training field; a conversion unit, configured to perform time domain conversion processing on each channel frequency domain response respectively, to obtain multiple first time domain channel responses; and a first determining unit, configured to determine a first power delay spectrum based on the fusion result of the power delay spectrum of the multiple first time domain channel responses.

[0081] In some embodiments, the number of conventional long training fields is two: a first determining unit, used to determine the power delay spectrum of each first time-domain channel response; and to perform averaging on the two power delay spectra, using the resulting average as the first power delay spectrum.

[0082] In some embodiments, the channel estimation module 330 includes: a fusion unit, configured to determine a target frequency domain correlation coefficient based on the fusion result of a first power delay spectrum and a second power delay spectrum; and an estimation unit, configured to perform channel estimation on an extended long training field using frequency domain Wiener filtering based on the target frequency domain correlation coefficient, and determine the channel estimation result of the extended long training field.

[0083] In some embodiments, the fusion unit includes: an averaging unit for averaging the first power delay spectrum and the second power delay spectrum to obtain a target power delay spectrum; a noise reduction unit for performing noise reduction processing on the target power delay spectrum to obtain a noise reduction result; and a conversion processing unit for performing frequency domain conversion processing on the noise reduction result to obtain a target frequency domain correlation coefficient.

[0084] In some embodiments, the estimation unit includes: a first execution unit, configured to determine the Wiener filter coefficient matrix based on the channel statistical characteristics of the target frequency domain correlation coefficient; and a second execution unit, configured to perform channel estimation on the extended-length training field using the Wiener filter coefficient matrix to obtain the channel estimation result of the extended-length training field.

[0085] In some embodiments, the first execution unit is configured to determine a frequency domain correlation matrix based on multiple subcarriers corresponding to the target frequency domain correlation coefficient; and to determine a Wiener filter coefficient matrix based on the frequency domain correlation matrix.

[0086] In some embodiments, the frequency domain signal to be processed further includes an extended signal field, and the method further includes: a second processing module, used to perform signal detection on the traditional signal field and the extended signal field sequentially based on the traditional long training field, to obtain a detection result.

[0087] Regarding the signal detection device in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0088] Based on the same inventive concept, this disclosure also provides an electronic device, including: a memory and a processor, which are interconnected and communicate with each other. The memory stores computer instructions, and the processor executes any of the signal detection methods provided in this disclosure by executing the computer instructions.

[0089] This disclosure uses specific terms to describe embodiments of the present disclosure. Terms such as "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of the present disclosure. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Furthermore, certain features, structures, or characteristics in one or more embodiments of the present disclosure can be appropriately combined.

[0090] In the context of this disclosure, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0091] Similarly, it should be noted that, in order to simplify the description of this disclosure and thus aid in the understanding of one or more embodiments, the foregoing description of embodiments of this disclosure may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of this disclosure requires more features than the features claimed. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.

[0092] The basic concepts have been described above. It is obvious that the above disclosure is merely illustrative and does not constitute a limitation of this disclosure. Although not explicitly stated herein, various modifications, improvements, and corrections may be made to this disclosure by those skilled in the art. Such modifications, improvements, and corrections are suggested in this disclosure and therefore remain within the spirit and scope of the embodiments of this disclosure.

Claims

1. A signal detection method, characterized in that, The method includes: Acquire a frequency domain signal to be processed based on wireless transmission, wherein the frequency domain signal to be processed includes a traditional long training field, a traditional signal field, and an extended long training field; Determine the first power delay spectrum of multiple traditional long training fields; The decision symbol sequence is determined by using the traditional signal field described in the hard decision method; Determine the second power delay spectrum of the decision symbol sequence; Based on the first power delay spectrum and the second power delay spectrum, channel estimation is performed on the extended-length training field to obtain the channel estimation result of the extended-length training field; Data detection is performed based on the channel estimation results to obtain data information transmitted through the frequency domain signal to be processed.

2. The signal detection method according to claim 1, characterized in that, Determining the first power delay spectrum of the plurality of traditional long training fields includes: According to the reception timing of the multiple traditional long training fields, channel estimation processing is performed on each traditional long training field in sequence to obtain the channel frequency domain response of each traditional long training field. Each of the channel frequency domain responses is subjected to time domain transformation to obtain multiple first time domain channel responses; The first power delay spectrum is determined based on the fusion result of the power delay spectra of multiple first time-domain channel responses.

3. The signal detection method according to claim 2, characterized in that, The traditional long training field has two elements. The determination of the first power delay spectrum based on the fusion result of multiple power delay spectra of the first time-domain channel responses includes: Determine the power delay spectrum for each of the first time-domain channel responses; The two power delay spectra are averaged, and the averaged result is used as the first power delay spectrum.

4. The signal detection method according to claim 1, characterized in that, The process of performing channel estimation on the extended-length training field based on the first power delay spectrum and the second power delay spectrum to obtain the channel estimation result of the extended-length training field includes: Based on the fusion result of the first power delay spectrum and the second power delay spectrum, the target frequency domain correlation coefficient is determined; Based on the target frequency domain correlation coefficient, channel estimation is performed on the extended long training field using frequency domain Wiener filtering to determine the channel estimation result of the extended long training field.

5. The signal detection method according to claim 4, characterized in that, The determination of the target frequency domain correlation coefficient based on the fusion result of the first power delay spectrum and the second power delay spectrum includes: The first power delay spectrum and the second power delay spectrum are averaged to obtain the target power delay spectrum; The target power time delay spectrum is subjected to noise reduction processing to obtain the noise reduction result; The noise reduction result is subjected to frequency domain transformation to obtain the target frequency domain correlation coefficient.

6. The signal detection method according to claim 4 or 5, characterized in that, The step of performing channel estimation on the extended-length training field based on the target frequency domain correlation coefficient and using frequency domain Wiener filtering to determine the channel estimation result of the extended-length training field includes: Based on the channel statistical characteristics of the target frequency domain correlation coefficient, the Wiener filter coefficient matrix is ​​determined; Channel estimation is performed on the extended-length training field using the Wiener filter coefficient matrix to obtain the channel estimation result of the extended-length training field.

7. The signal detection method according to claim 6, characterized in that, The determination of the Wiener filter coefficient matrix based on the channel statistical characteristics of the target frequency domain correlation coefficient includes: Based on the multiple subcarriers corresponding to the target frequency domain correlation coefficient, determine the frequency domain correlation matrix; Based on the frequency domain correlation matrix, the Wiener filter coefficient matrix is ​​determined.

8. The signal detection method according to claim 1, characterized in that, The frequency domain signal to be processed further includes an extended signal field, and the method further includes: Based on the traditional long training field, signal detection is performed sequentially on the traditional signal field and the extended signal field to obtain the detection result.

9. A signal detection device, characterized in that, The device includes: The acquisition module is used to acquire a frequency domain signal to be processed based on wireless transmission, wherein the frequency domain signal to be processed includes a traditional long training field, a traditional signal field, and an extended long training field; A first processing module is configured to determine a first power delay spectrum of the plurality of conventional long training fields; determine a decision symbol sequence by hard-deciding the conventional signal fields; and determine a second power delay spectrum of the decision symbol sequence. The channel estimation module is used to perform channel estimation on the extended-length training field based on the first power delay spectrum and the second power delay spectrum, and obtain the channel estimation result of the extended-length training field. The first detection module is used to perform data detection based on the channel estimation results to obtain data information transmitted through the frequency domain signal to be processed.

10. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the signal detection method of any one of claims 1-8 by executing the computer instructions.

Images