Time delay-Doppler domain channel measurement method and device
By designing channel detection signals in the delay-Doppler domain and using orthogonal time-frequency air conditioning technology to extract the channel expansion function, the problem of insufficient channel delay and Doppler shift measurement accuracy in the existing technology is solved, and the implementation of the following method High-precision channel measurement.
Patent Information
- Application Number
- CN202410871198.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-01
- Publication Date
- 2024-11-26
AI Technical Summary
The existing delay-Doppler domain channel measurement methods cannot accurately measure the channel's delay, Doppler shift and channel gain coefficient, and are greatly affected by fractional delay-Doppler, resulting in insufficient measurement accuracy.
Design a time-delay-Doppler domain channel measurement method, and extract channel expansion function (CSF) by designing channel detection signals in measurement scenarios, using orthogonal time-frequency air conditioning technology to perform OTFS modulation and demodulation of channel detection signals. ), and perform fractional delay-Doppler joint estimation to obtain channel delay, Doppler shift and gain coefficients.
Accurate measurement of channels, especially multipath fractional delay and Doppler shift measurement, significantly improving channel measurement accuracy.
Smart Images

Figure CN119030641A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of wireless communication technology, and in particular to a delay-Doppler domain channel measurement method and device. Background Art
[0002] Future Beyond 5G (B5G) and 6G communication systems will use higher frequency bands (millimeter wave, terahertz and even visible light) and wider bandwidths. In order to promote the design and network deployment of next-generation wireless communication systems, it is necessary to fully understand the characteristics of wireless channels and perform accurate measurements.
[0003] Among them, based on the Orthogonal Time Frequency Space (OTFS) modulation technology, the channel measurement in the delay-Doppler domain can be realized, and the CSF can be directly measured. However, the existing delay-Doppler domain channel measurement method is affected by the fractional delay-Doppler and cannot accurately measure the channel delay, Doppler frequency shift and channel gain coefficient.
[0004] It can be seen that how to design a delay-Doppler domain channel measurement method and improve the measurement accuracy is a problem that needs to be solved urgently by those skilled in the art. Summary of the invention
[0005] The present invention aims to provide a channel measurement method in the delay-Doppler domain to solve the above problems.
[0006] The technical solution of the present invention is: a channel measurement method in the delay-Doppler domain, comprising:
[0007] include:
[0008] S1, design the channel sounding signal in the delay-Doppler domain according to the measurement scenario;
[0009] S2, the detection signal is modulated by orthogonal time frequency space (OTFS) to become a time domain channel detection signal and sent to the wireless channel;
[0010] S3, the receiving end obtains the channel detection signal, performs synchronization and OTFS demodulation, and obtains the delay-Doppler domain receiving signal;
[0011] S4, extracting the channel spreading function (CSF) according to the input-output relationship in the delay-Doppler domain;
[0012] S5, perform fractional delay-Doppler joint estimation on CSF to obtain accurate channel delay, Doppler frequency shift and gain coefficient.
[0013] Preferably, S1 includes:
[0014] S11, based on the measurement scenario characteristics including channel stationary interval (SI), maximum excess delay (MED), and maximum Doppler shift;
[0015] S12, the delay-Doppler domain channel detection signal adopts an embedded pilot; specifically:
[0016] S121, selecting an appropriate number of delay dimension grids M, number of Doppler dimension grids N, and bandwidth B according to scene characteristics;
[0017] S122, insert a pilot at the center of the delay-Doppler domain resource grid, the pilot position is (k p ,l p ), where k p and l p They respectively represent the Doppler grid and the delay grid corresponding to the pilot position in the delay-Doppler domain resource grid;
[0018] S123, insert guard bands on both sides of the pilot, and the number of grids of the guard bands in the delay dimension and Doppler dimension are l respectively. τ and N, where l τ is the maximum measurable delay τ of the detection waveform max The corresponding grid number;
[0019] S124, inserting a pseudo-noise (PN) sequence into all positions except the center and the guard band in the delay-Doppler domain resource grid to synchronize and reduce the peak to average power ratio (PAPR) of the waveform;
[0020] S125, let x k,l is the detection signal at the kth Doppler grid and the lth delay grid in the OTFS detection signal resource grid, which can be expressed as:
[0021]
[0022] Among them, x p is the amplitude of the pilot signal; A PN is the amplitude of the inserted PN sequence; p ≤l≤l p +l τ , 0≤k≤N-1 is the delay-Doppler domain grid range used to extract CSF.
[0023] Preferably, in S121, the selected scene characteristics include:
[0024] The designed channel detection signal frame length should be less than or equal to the channel stable interval;
[0025] The maximum measurable delay τ of the designed detection signal max It must be greater than the maximum additional delay of the channel;
[0026] The maximum measurable Doppler shift of the designed detection signal must be greater than the maximum Doppler shift of the channel.
[0027] Preferably, in S2,
[0028] The designed delay-Doppler domain channel detection signal is transformed into a time domain channel detection signal through inverse symplectic finite Fourier transform (ISFFT) and Heisenberg transform and sent to the wireless channel.
[0029] Preferably, in S3,
[0030] The received time domain channel detection signal is synchronized with the local synchronization sequence in the time domain, and the synchronized received signal is subjected to Wigner Transform and Symplectic Finite Fourier Transform (SFFT) to obtain a delay-Doppler domain received signal;
[0031] The time domain synchronization of the received time domain channel sounding signal and the local synchronization sequence is performed by finding the starting position of the received channel sounding signal by means of sliding correlation; specifically:
[0032] The local synchronization sequence is the first L lengths of the time domain channel sounding signal;
[0033] The length L of the local synchronization sequence is limited by the signal to noise ratio (SNR) and the computational complexity. When the SNR is high, the value of L is reduced to reduce the computational complexity while ensuring the synchronization performance. When the SNR is low, the value of L is increased to improve the synchronization performance.
[0034] Preferably, in S4,
[0035] Extract the estimated value of CSF based on the delay-Doppler domain input-output relationship within the delay-Doppler domain grid used to extract CSF
[0036]
[0037] Among them, k p and lp They represent the Doppler grid and delay grid corresponding to the pilot position in the delay-Doppler domain resource grid; y[k,l] is the delay-Doppler domain received signal; x p is the amplitude of the pilot signal; P is the total number of multipath channels; h i represents the channel gain coefficient of the i-th path; k i and l i Respectively represent the Doppler grid and delay grid corresponding to the i-th path in the delay-Doppler domain grid; -0.5≤α i ,β i ≤0.5 indicates the fractional offset of the i-th path from the integer grid; M and N represent the number of grids in the delay dimension and the number of grids in the Doppler dimension, respectively; n[k,l] is the additive white Gaussian noise in the delay-Doppler domain.
[0038] Preferably, in S5,
[0039] Perform fractional delay-Doppler joint estimation on CSF, including:
[0040] S51, determine the noise floor according to the CSF and select a suitable effective multipath extraction threshold;
[0041] S52, determine the integer delay grid position corresponding to the maximum path (pth path) of the channel expansion function power and integer Doppler shift grid positions
[0042] S53, establish the equivalent channel function h of the pth path eq,p [k,l]:
[0043]
[0044] Among them, h p represents the p-th path complex gain; represents the phase of the pth path; k p and l p They represent the discrete forms of the delay and Doppler frequency shift of the pth path respectively; M and N represent the number of grids in the delay dimension and the number of grids in the Doppler dimension respectively;
[0045] S54, respectively set the fractional delay estimation accuracy Δτ step and fractional Doppler shift estimation accuracy Δυ step ;
[0046] S55, calculating the two-dimensional correlation function of the equivalent channel model of the p-th path and the channel extension function with the set accuracy
[0047]
[0048] S56, obtaining the estimated delay values of the p-th path through the calculation results of the two-dimensional correlation function Doppler shift estimate And the channel gain coefficient
[0049]
[0050] in, is the two-dimensional correlation function of the equivalent channel model of the p-th path and the channel spread function; and are the integer delay grid position and integer Doppler frequency shift grid position corresponding to the pth path respectively; Δτ is the delay resolution; Δυ is the Doppler resolution;
[0051] S57, determine the channel gain coefficient Is it greater than the effective multipath extraction threshold? If it is less than the threshold, the algorithm ends; otherwise, the influence of the pth path is eliminated from the channel extension function, and then S52 to S57 are repeated.
[0052] A delay-Doppler domain channel measurement device, characterized in that the transmitting end comprises:
[0053] The detection signal design module designs the channel detection signal in the delay-Doppler domain according to the characteristics of the measurement scenario, including the channel stationary interval, the maximum additional delay, and the maximum Doppler shift;
[0054] The OTFS modulation module performs ISFFT and Heisenberg transform on the designed delay-Doppler domain channel detection signal to convert it into a time domain channel detection signal;
[0055] The sending module up-converts the time domain channel detection signal to obtain a radio frequency signal, and adjusts the transmission power to send it to the wireless channel. The equipment used in the sending module includes a radio frequency signal source, an up-conversion module, a power amplifier, and a transmitting antenna.
[0056] The receiving end includes:
[0057] The receiving module is used to receive the channel detection signal; specifically, the radio frequency signal is converted into an intermediate frequency signal through a low noise amplifier (LNA), a band-pass filter (BPF) and a down converter; the band-pass sampling of the intermediate frequency signal is completed through an analog to digital converter (ADC), and the intermediate frequency digital signal is stored in a PC solid state drive; the intermediate frequency digital signal is processed offline; including obtaining a baseband channel detection signal through a digital down converter (DDC) and a low-pass filter (LPF);
[0058] A synchronization module synchronizes the baseband channel detection signal with the local synchronization sequence in the time domain;
[0059] The OTFS demodulation module performs Wigner transform and SFFT on the synchronized detection signal to obtain the delay-Doppler domain received signal;
[0060] A CSF extraction module extracts CSF according to a delay-Doppler domain input-output relationship within a delay-Doppler domain grid range used for extracting CSF;
[0061] The fractional delay-Doppler joint estimation module establishes an equivalent channel function based on the matched filtering algorithm, performs fractional delay-Doppler joint estimation on the extracted CSF, and obtains accurate channel delay, Doppler frequency shift and gain coefficient.
[0062] The beneficial effects of the present invention are:
[0063] The method of the present invention can directly measure CSF, and can measure the fractional delay and Doppler frequency shift of multipath, thereby effectively improving the channel measurement accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0064] Figure 1 A schematic diagram of the flow chart of the delay-Doppler domain channel measurement method provided by the present invention;
[0065] Figure 2 A schematic diagram of the delay-Doppler domain detection signal design provided by the present invention;
[0066] Figure 3 This is a schematic diagram of the delay-Doppler domain channel measurement device provided by the present invention;
[0067] Figure 4 It is a flow chart of the channel spread function fractional delay-Doppler joint estimation;
[0068] Figure 5A diagram showing the synchronization performance analysis results of the delay-Doppler domain channel measurement method provided by the present invention;
[0069] Figure 6 This is a schematic diagram of the verification results of the delay-Doppler domain channel measurement method provided by the present invention. DETAILED DESCRIPTION
[0070] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. The implementation methods of the present invention are not limited thereto.
[0071] Example 1
[0072] Figure 1 The figure is a flow chart of a delay-Doppler domain channel measurement method, including:
[0073] Step S1: Design a channel sounding signal in the delay-Doppler domain according to the measurement scenario;
[0074] For different measurement scenarios, it is necessary to flexibly design the delay-Doppler domain channel detection signal. The following factors need to be considered during the design process: delay grid number M, Doppler grid number N, bandwidth B, signal frame length, delay resolution, Doppler resolution, maximum measurable delay, maximum measurable Doppler frequency shift, signal PAPR, synchronization method, etc.
[0075] Figure 2 To detect the signal in the delay-Doppler domain grid resource allocation diagram, a pilot is inserted at the center of the grid resource. The pilot position is given by (l p ,k p ) represents, where k p and l p They represent the Doppler grid and delay grid corresponding to the pilot position in the delay-Doppler domain resource grid. Guard bands are inserted on both sides of the pilot, and the number of grids of the guard band in the delay dimension and Doppler dimension are l τ and N, where l τ is the maximum measurable delay τ of the detection waveform max The corresponding grid number. In the delay-Doppler domain resource grid, except for the center and the guard band, pseudo-random (PN) sequences are inserted in other positions. The purpose is to synchronize and reduce the peak to average power ratio (PAPR) of the waveform. Let x k,l is the detection signal at the kth Doppler tap and the lth delay tap in the OTFS detection signal resource grid, which can be expressed as:
[0076]
[0077] Among them, x p is the amplitude of the pilot signal, A PN is the amplitude of the inserted PN sequence; p ≤l≤l p +l τ , 0≤k≤N-1 is the delay-Doppler domain grid range used to extract CSF. The above formula has three cases, which correspond to the pilot, guard band and PN sequence in the delay-Doppler domain resource grid. Due to the unidirectional extension characteristics of the delay, the delay-Doppler domain grid range is used to extract CSF. This grid range is the guard band to the right of the pilot, not the entire guard band range.
[0078] Assume that bandwidth is represented by B and delay resolution is represented by Δτ. Then there is the following relationship:
[0079] B=1 / Δτ
[0080] That is, the larger the bandwidth, the smaller the interval of the delay domain, the higher the delay resolution, and the stronger the ability to distinguish multipath. The detection signal frame length is expressed as T OTFS denoted by , and Doppler resolution is denoted by Δυ. Then there is the following relationship:
[0081] T OTFS =1 / Δυ
[0082] That is, the longer the duration of the signal, the smaller the interval in the Doppler domain, the higher the Doppler resolution, and the more detailed the changes in the Doppler domain can be observed. The frame length of a detection signal is T OTFS =M·N / B, and the corresponding Doppler resolution is Δυ=B / (M·N). The frame length of the detection signal is limited by the stationary interval (SI) of the wireless channel. The SI in different scenarios will be different. For different scenarios, the selection of waveform parameters must meet the following requirements:
[0083] T OTFS =M·N / B≤T SI
[0084] For example, in the high-speed rail scenario, when the SI is 20 ms and the bandwidth of the detection signal is 100 MHz, the delay-Doppler domain grid resource size M·N≤2×10 6 .
[0085] The maximum measurable time delay and the maximum measurable Doppler frequency shift are τ max and max The maximum measurable time delay and the maximum measurable Doppler frequency shift are:
[0086] τ max = l τ Δτ
[0087] υmax = ±N·Δυ / 2
[0088] Among them, τ max The value of M must be greater than the maximum delay difference that can be produced in the actual propagation scenario, that is, the maximum excess delay (MED). Usually, when the maximum measurable delay is greater than 2μs, it can meet the channel measurement requirements in most high-speed mobile scenarios. When the bandwidth is fixed at 100MHz, the value of the delay grid number M must be greater than 800.
[0089] Therefore, the selected scenario characteristics are: the frame length of the channel detection signal must be less than or equal to the channel stable interval; the maximum measurable delay of the detection signal must be greater than the maximum additional delay of the channel; the maximum measurable Doppler shift of the detection signal must be greater than the maximum Doppler shift of the channel.
[0090] Step S2: The detection signal is modulated by orthogonal time frequency space (OTFS) to become a time domain channel detection signal and sent to the wireless channel;
[0091] OTFS modulation consists of two steps: the delay-Doppler domain channel detection signal x is obtained by ISFFT. k,l Get the time-frequency domain channel detection signal X[n,m]:
[0092]
[0093] Through the Heisenberg Transform, the time-domain channel detection signal x(t) is obtained from the time-frequency domain channel detection signal X[n,m]:
[0094]
[0095] Among them, g tx (t) is the transmitting pulse signal.
[0096] Step S3: The receiving end obtains the channel detection signal, performs synchronization and OTFS demodulation, and obtains the delay-Doppler domain receiving signal;
[0097] The channel detection signal y(t) obtained by the receiving end can be expressed as:
[0098] y(t)=∫∫h(τ,υ)x(t-τ)e j2πυ(t-τ) dτdυ+n(t)
[0099] Among them, n(t) is the additive white Gaussian noise in the time domain; h(τ,υ) is the channel spread function, which can be sparsely represented in the delay-Doppler domain:
[0100]
[0101] Where P is the total number of multipath channels; h i represents the channel gain coefficient of the i-th path; τ i and i are the time delay and Doppler shift of the ith path respectively; δ(·) is the Dirac function.
[0102] The received time domain channel sounding signal is synchronized with the local synchronization sequence in the time domain, and the starting position of the received channel sounding signal is found by sliding correlation:
[0103]
[0104] Wherein, L is the length of the local synchronization sequence; x[i] and y[i] are the local synchronization sequence and the received channel detection signal, respectively.
[0105] The synchronized channel detection signal is demodulated by OTFS, and the time-frequency domain channel detection signal Y[n,m] is obtained by Wigner Transform:
[0106]
[0107] Among them, g rx (t) is the received pulse signal.
[0108] Through SFFT, the time-domain channel detection signal y[k,l] is obtained from the time-frequency domain channel detection signal Y[n,m]:
[0109]
[0110] The length L of the local synchronization sequence is limited by the signal to noise ratio (SNR) and the computational complexity. When the SNR is high, the value of L is reduced to reduce the computational complexity while ensuring the synchronization performance. When the SNR is low, the value of L is increased to improve the synchronization performance.
[0111] For example, the length of the time domain channel sounding signal is 2048×256. When the SNR is high, the value of L may be 2048; when the SNR is low, the value of L may be 2048×256. In this case, the local synchronization sequence is the time domain channel sounding signal.
[0112] Step S4: extracting the channel spreading function (CSF) according to the input-output relationship in the delay-Doppler domain;
[0113] Since the delay-Doppler domain received signal is the two-dimensional convolution result of the channel spread function and the delay-Doppler domain detection signal of the transmitter, it is necessary to extract the channel spread function from the received signal according to the delay-Doppler domain input-output relationship and the designed channel detection signal.
[0114] Extract the estimated value of CSF based on the delay-Doppler domain input-output relationship within the delay-Doppler domain grid used to extract CSF
[0115]
[0116] In the above formula, the first equal sign means that the estimated value of CSF can be obtained by receiving the signal in the delay-Doppler domain. The second equal sign means that the estimated value of CSF can be expressed by formula. The detailed description is as follows:
[0117] Because the wireless channel can be sparsely represented in the delay-Doppler domain by the channel spread function h(τ,υ) as follows:
[0118]
[0119] Where P is the total number of multipath channels; h i represents the channel gain coefficient of the i-th path; τ i and i are the delay and Doppler shift of the i-th path respectively; δ(·) is the Dirac function. According to the input-output relationship in the delay-Doppler domain, the CSF estimate can be expressed as shown in the second equal sign.
[0120] The meaning of the parameters in the above formula is: k p and l p They represent the Doppler grid and delay grid corresponding to the pilot position in the delay-Doppler domain resource grid; y[k,l] is the delay-Doppler domain received signal; x p is the amplitude of the pilot signal; P is the total number of multipath channels; h i represents the channel gain coefficient of the i-th path; k i and l i Respectively represent the Doppler grid and delay grid corresponding to the i-th path in the delay-Doppler domain grid; -0.5≤α i ,β i ≤0.5 indicates the fractional offset of the i-th path from the integer grid; M and N represent the number of grids in the delay dimension and the number of grids in the Doppler dimension, respectively; n[k,l] is the additive white Gaussian noise in the delay-Doppler domain;
[0121] Among them, -0.5≤α i ,β i≤0.5 indicates that the i-th path is offset by a fraction of the integer grid, satisfying the following relationship:
[0122]
[0123] Wherein, Δf and T are the subcarrier spacing and time interval respectively.
[0124] Step S5: Perform fractional delay-Doppler joint estimation on the CSF to obtain more accurate channel delay, Doppler frequency shift and gain coefficient.
[0125] Based on the matched filtering algorithm, an equivalent channel function is established to perform fractional delay-Doppler joint estimation on CSF. The specific estimation is divided into the following steps:
[0126] S51: Determine the noise floor according to the CSF and select a suitable effective multipath extraction threshold;
[0127] S52: Determine the integer delay grid position corresponding to the maximum path (pth path) of the channel expansion function power and integer Doppler shift grid positions
[0128] S53: Establishing the equivalent channel function h of the pth path eq,p [k,l]:
[0129]
[0130] Among them, h p represents the p-th path complex gain; represents the phase of the pth path; k p and l p They represent the discrete forms of the delay and Doppler frequency shift of the pth path respectively; M and N represent the number of grids in the delay dimension and the number of grids in the Doppler dimension respectively;
[0131] S54: respectively set the fractional delay estimation accuracy Δτ step and fractional Doppler shift estimation accuracy Δυ step ;
[0132] S55: Calculate the equivalent channel model of the p-th path and the two-dimensional correlation function of the channel extension function with the set accuracy
[0133]
[0134] S56: Obtain the estimated delay value of the p-th path through the calculation result of the two-dimensional correlation function Doppler shift estimate And the channel gain coefficient
[0135]
[0136] in, is the two-dimensional correlation function of the equivalent channel model of the p-th path and the channel spread function; and are the integer time delay grid position and integer Doppler frequency shift grid position corresponding to the p-th path respectively; Δτ is the time delay resolution; Δυ is the Doppler resolution.
[0137] S57: Determine channel gain coefficient Is it greater than the effective multipath extraction threshold? If it is less than the threshold, the algorithm ends; otherwise, the influence of the pth path is eliminated from the channel extension function, and then S52 to S57 are repeated.
[0138] Figure 4 The figure shows the principle diagram of the delay-Doppler domain channel measurement device provided by the present invention. The transmitting end of the device includes: a detection signal design module, an OTFS modulation module and a sending module. The receiving end includes: a receiving module, a synchronization module, an OTFS demodulation module, a CSF extraction module and a fractional delay-Doppler joint estimation module.
[0139] The detection signal design module designs the delay-Doppler domain channel detection signal according to the channel's stable interval, maximum additional delay and maximum Doppler frequency shift, and obtains the time domain channel detection signal through the OTFS modulation module. The transmission module up-converts the time domain detection signal to obtain the RF signal, and adjusts the transmission power to send it to the wireless channel.
[0140] The receiving module converts the RF signal into an IF signal through LNA, BPF and downconverter, completes the bandpass sampling of the IF signal through ADC, and stores it on the PC hard disk. The sampling rate is calculated by the bandpass sampling theorem. According to the bandpass sampling theorem, if the sampling rate is f s , then it satisfies:
[0141]
[0142] Where N is an integer and its maximum value does not exceed f H / (f H -f L ), f H With f L are the lowest and highest frequencies of the signal respectively, then the sampling frequency f that satisfies the bandpass sampling theorem is used s After sampling at equal intervals, the original signal can be restored without distortion through spectrum shifting. For an intermediate frequency signal with a center frequency of f0 = 180 MHz and a bandwidth of B = 100 MHz, the sampling rate can be f s=500MHz. Offline processing of the intermediate frequency digital signal: obtaining the baseband signal through DDC and LPF. Correlation of the locally stored time domain channel detection signal with the received signal to achieve symbol synchronization.
[0143] Obtain the delay-Doppler domain signal from the synchronized detection signal through the OTFS demodulation module;
[0144] Extracting CSF from the acquired delay-Doppler domain signal through the CSF extraction module;
[0145] The fractional delay-Doppler joint estimation module is used to perform fractional delay-Doppler joint estimation on the CSF to obtain more accurate channel delay, Doppler frequency shift and gain coefficient.
[0146] In order to verify the synchronization performance of the delay-Doppler domain channel measurement method proposed in the present invention, the detection signal passes through the additive white Gaussian noise (AWGN) channel, and the synchronization gain changes with the signal-to-noise ratio when the local synchronization sequence length is different. The results are as follows: Figure 5 (a). The longer the local synchronization sequence is, the higher the synchronization gain is. As the channel deteriorates, the synchronization gain will decrease, but it can still maintain good synchronization performance at a low signal-to-noise ratio. Therefore, the length of the local synchronization sequence can be adjusted according to the signal-to-noise ratio and computational complexity. Figure 5 (b) shows the normalized correlation peak value obtained when the detection signal passes through the AWGN channel. The simulation signal-to-noise ratio is set to 20 dB, the length of the local synchronization sequence L is M×N, and the synchronization gain is about 44 dB.
[0147] In order to verify the delay-Doppler domain channel measurement method proposed in the present invention, a pure Doppler shift channel is measured. A pure Doppler shift channel refers to a channel in which the Doppler shifts of all multipaths are independent of each other, and the delay and power of the multipaths can be set separately. Among them, a delay-Doppler domain grid with a delay tap number M of 2048 and a Doppler tap number N of 256 is used to design the channel detection signal. The detection signal bandwidth is selected as 10MHz, and the corresponding waveform parameter configuration is shown in the following table.
[0148]
[0149] The simulation channel parameter configuration used for verification is shown in the following table. The delays of the three paths are [0, 2, 3.02] μs, and the corresponding delay resolution multiples are [0, 20, 30.2], and the Doppler shifts of the three paths are [572.2, -381.47, 387.19] Hz, and the corresponding Doppler resolution multiples are [30, -20, 20.3].
[0150]
[0151] The synchronized received signal is demodulated by OTFS. According to the input-output relationship in the delay-Doppler domain, the channel expansion function is obtained as follows: Figure 6 As shown in (a), the integer delays and Doppler shifts of the three multipaths are all at the correct position. Due to the fractional delay-Doppler effect of the third path, the power is leaked in the delay dimension and the Doppler dimension. The peak power is -11.84dB, which is a large error from the set -10dB. Figure 6 As shown in (b), after performing fractional delay-Doppler joint estimation on CSF, the delays and Doppler shifts corresponding to the three effective multipaths can be accurately extracted. At the same time, the power of the third path is also compensated. The compensated power is -10.11 dB, which is closer to the preset value of -10 dB.
[0152] The above experiments verify that the delay-Doppler domain channel measurement method proposed in the present invention can directly measure CSF, and can measure the fractional delay and Doppler shift of multipath, effectively improving the channel measurement accuracy. Compared with the traditional time domain or frequency domain channel measurement method, the present invention can directly obtain the Doppler dimension information of the channel without further processing from the time domain or frequency domain channel. In addition, by establishing an equivalent channel, the fractional delay-Doppler joint estimation of CSF is performed, which effectively improves the channel measurement accuracy, and has better measurement performance than other delay-Doppler domain channel measurement methods. This method has convenience, accuracy, high efficiency, etc., and provides a reference for channel measurement and channel modeling of wireless communications.
[0153] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A channel measurement method in delay-Doppler domain, characterized in that: include: S1, design the channel sounding signal in the delay-Doppler domain according to the measurement scenario; S2, the detection signal is modulated by orthogonal time frequency space (OTFS) to become a time domain channel detection signal and sent to the wireless channel; S3, the receiving end obtains the channel detection signal, performs synchronization and OTFS demodulation, and obtains the delay-Doppler domain receiving signal; S4, extracting the channel spreading function (Channel Spreading Function, CSF) according to the input-output relationship in the delay-Doppler domain; S5, perform fractional delay-Doppler joint estimation on CSF to obtain accurate channel delay, Doppler frequency shift and gain coefficient.
2. The delay-Doppler domain channel measurement method according to claim 1, characterized in that: S1 includes: S11, based on the measurement scenario characteristics including channel stationary interval (SI), maximum excess delay (MED), maximum Doppler shift; S12, the delay-Doppler domain channel detection signal adopts an embedded pilot; specifically: S121, selecting an appropriate number of delay dimension grids M, number of Doppler dimension grids N, and bandwidth B according to scene characteristics; S122, insert a pilot at the center of the delay-Doppler domain resource grid, the pilot position is (kp, lp), where k p and l p They respectively represent the Doppler grid and the delay grid corresponding to the pilot position in the delay-Doppler domain resource grid; S123, insert guard bands on both sides of the pilot, and the number of grids of the guard bands in the delay dimension and Doppler dimension are l respectively. τ and N, where l τ is the maximum measurable delay τ of the detection waveform max The corresponding grid number; S124, inserting a pseudo-random (Pseudo-Noise, PN) sequence in all positions except the center and the guard band in the delay-Doppler domain resource grid to synchronize and reduce the peak to average power ratio (Peak to Average Power Ratio, PAPR) of the waveform; S125, let x k,l is the detection signal at the kth Doppler grid and the lth delay grid in the OTFS detection signal resource grid, which can be expressed as: Among them, x p is the amplitude of the pilot signal; A PN is the amplitude of the inserted PN sequence; p ≤l≤l p +l τ , 0≤k≤N-1 is the delay-Doppler domain grid range used to extract CSF.
3. The delay-Doppler domain channel measurement method according to claim 1, characterized in that: In S121, the selected scene characteristics include: The designed channel detection signal frame length should be less than or equal to the channel stable interval; The maximum measurable delay τ of the designed detection signal max It must be greater than the maximum additional delay of the channel; The maximum measurable Doppler shift of the designed detection signal must be greater than the maximum Doppler shift of the channel.
4. The delay-Doppler domain channel measurement method according to claim 1, characterized in that: In S2, The designed delay-Doppler domain channel detection signal is transformed into a time domain channel detection signal through inverse symplectic finite Fourier transform (ISFFT) and Heisenberg transform and sent to the wireless channel.
5. The delay-Doppler domain channel measurement method according to claim 1, characterized in that: In S3, The received time domain channel detection signal is synchronized with the local synchronization sequence in the time domain, and the synchronized received signal is subjected to Wigner Transform and Symplectic Finite Fourier Transform (SFFT) to obtain a delay-Doppler domain received signal; The time domain synchronization of the received time domain channel sounding signal and the local synchronization sequence is performed by finding the starting position of the received channel sounding signal by means of sliding correlation; specifically: The local synchronization sequence is the first L lengths of the time domain channel sounding signal; The length L of the local synchronization sequence is limited by the signal to noise ratio (SNR) and the computational complexity. When the SNR is high, the value of L is reduced to reduce the computational complexity while ensuring the synchronization performance. When the SNR is low, the value of L is increased to improve the synchronization performance.
6. The delay-Doppler domain channel measurement method according to claim 1, characterized in that: In S4, Extract the estimated value of CSF based on the delay-Doppler domain input-output relationship within the delay-Doppler domain grid used to extract CSF Among them, k p and l p They represent the Doppler grid and delay grid corresponding to the pilot position in the delay-Doppler domain resource grid; y[k,l] is the delay-Doppler domain received signal; x p is the amplitude of the pilot signal; P is the total number of multipath channels; h i represents the channel gain coefficient of the i-th path; k i and l i Respectively represent the Doppler grid and delay grid corresponding to the i-th path in the delay-Doppler domain grid; -0.5≤α i ,β i ≤0.5 indicates the fractional offset of the i-th path from the integer grid; M and N represent the number of grids in the delay dimension and the number of grids in the Doppler dimension, respectively; n[k,l] is the additive white Gaussian noise in the delay-Doppler domain.
7. The delay-Doppler domain channel measurement method according to claim 1, characterized in that: In S5, Perform fractional delay-Doppler joint estimation on CSF, including: S51, determine the noise floor according to the CSF and select a suitable effective multipath extraction threshold; S52, determine the integer delay grid position corresponding to the maximum path (pth path) of the channel expansion function power and integer Doppler shift grid positions S53, establish the equivalent channel function h of the pth path eq,p [k,l]: Among them, h p represents the p-th path complex gain; represents the phase of the pth path; k p and l p They represent the discrete forms of the delay and Doppler frequency shift of the pth path respectively; M and N represent the number of grids in the delay dimension and the number of grids in the Doppler dimension respectively; S54, respectively set the fractional delay estimation accuracy Δτ step and fractional Doppler shift estimation accuracy Δυ step ; S55, calculating the two-dimensional correlation function of the equivalent channel model of the p-th path and the channel extension function with the set accuracy S56, obtaining the estimated delay values of the p-th path through the calculation results of the two-dimensional correlation function Doppler shift estimate And the channel gain coefficient in, is the two-dimensional correlation function of the equivalent channel model of the p-th path and the channel spread function; and are the integer delay grid position and integer Doppler frequency shift grid position corresponding to the pth path respectively; Δτ is the delay resolution; Δυ is the Doppler resolution; S57, determine the channel gain coefficient Is it greater than the effective multipath extraction threshold? If it is less than the threshold, the algorithm ends; otherwise, the influence of the pth path is eliminated from the channel extension function, and then S52 to S57 are repeated.
8. A delay-Doppler domain channel measurement device, characterized in that the transmitting end include: The detection signal design module designs the channel detection signal in the delay-Doppler domain according to the characteristics of the measurement scenario, including the channel stationary interval, the maximum additional delay, and the maximum Doppler shift; The OTFS modulation module performs ISFFT and Heisenberg transform on the designed delay-Doppler domain channel detection signal to convert it into a time domain channel detection signal; The sending module up-converts the time domain channel detection signal to obtain a radio frequency signal, and adjusts the transmission power to send it to the wireless channel. The equipment used in the sending module includes a radio frequency signal source, an up-conversion module, a power amplifier, and a transmitting antenna.
9. A delay-Doppler domain channel measurement device, characterized in that: The receiving end includes: The receiving module is used to receive the channel detection signal; specifically, the radio frequency signal is converted into an intermediate frequency signal through a low noise amplifier (LNA), a band-pass filter (BPF) and a down converter; the band-pass sampling of the intermediate frequency signal is completed through an analog to digital converter (ADC), and the intermediate frequency digital signal is stored in a PC solid state drive; the intermediate frequency digital signal is processed offline; including obtaining a baseband channel detection signal through a digital down converter (DDC) and a low-pass filter (LPF); A synchronization module synchronizes the baseband channel detection signal with the local synchronization sequence in the time domain; The OTFS demodulation module performs Wigner transform and SFFT on the synchronized detection signal to obtain the delay-Doppler domain received signal; A CSF extraction module extracts CSF according to a delay-Doppler domain input-output relationship within a delay-Doppler domain grid range used for extracting CSF; The fractional delay-Doppler joint estimation module establishes an equivalent channel function based on the matched filtering algorithm, performs fractional delay-Doppler joint estimation on the extracted CSF, and obtains accurate channel delay, Doppler frequency shift and gain coefficient.
Citation Information
Cited By
Off-network block sparse orthogonal matching pursuit underwater acoustic time delay Doppler domain channel estimation method
CN119652701A
An Off-Network Block Sparse Orthogonal Matching Pursuit Method for Underwater Acoustic Delay Doppler Domain Channel Estimation
CN119652701B
Efficient channel measurement method based on sparse non-uniform frequency sampling and likelihood correction SAGE algorithm
CN121966749A
Efficient channel measurement method based on sparse non-uniform frequency sampling and likelihood correction SAGE algorithm
CN121966749B
