Chirp signal phase based picosecond level wireless time of arrival estimation method and system

Abstract

The application belongs to the technical field of communication, and discloses a picosecond-level wireless time of arrival estimation method and system based on chirp signal phase. Synchronization sequence capture is performed through sliding dechirp FFT search, integer sampling point level capture timing estimation is obtained, phase difference between continuous upper chirp symbols is used to estimate sampling frequency offset and construct a compensation factor, residual quadratic phase and inter-symbol cumulative timing drift are eliminated, carrier frequency offset and timing deviation are jointly solved based on the spectrum peak positions of the compensated upper and lower chirps, sub-sampling point level time of arrival coarse estimation is obtained, the dechirp signal is shifted to zero frequency to extract the phase, the coarse estimation result is used to eliminate the phase cycle ambiguity, then the residual timing correction amount is calculated and the final estimation result is obtained. The frequency domain coarse estimation capability of the chirp signal is combined with the high sensitivity of the signal phase to slight time changes, so that the time of arrival estimation is improved from the sub-sampling point level to the carrier level, thereby providing a technical foundation for realizing picosecond-level wireless time of arrival estimation.

Description

Technical Field

[0001] This invention belongs to the field of communication technology, specifically a picosecond-level wireless time-of-arrival estimation method and system based on the phase of a chirped signal. Background Technology

[0002] In systems such as wireless clock synchronization, wireless ranging, distributed cooperative sensing, and wireless positioning, the receiving end needs to accurately determine the physical time when synchronization messages or measurement signals arrive at the receiver. Physical layer arrival time is a crucial fundamental observation for round-trip time measurement, node clock offset calculation, propagation delay estimation, and distance inversion; its estimation accuracy directly affects the accuracy of wireless clock synchronization, ranging, and positioning results.

[0003] Chirped spread spectrum signals have a continuously changing instantaneous frequency, exhibiting a large time-bandwidth product and good noise immunity, narrowband interference immunity, and multipath immunity. They can still produce significant correlation peaks or spectral peaks even under low signal-to-noise ratio conditions, thus being widely used in wireless communication, wireless ranging, and time synchronization. Existing time-of-arrival estimation methods based on chirped signals typically perform dechirping by multiplying the received chirped signal with a local reference chirped signal, converting the wideband chirped signal into a complex single-tone signal with an approximately fixed frequency. Subsequently, a discrete Fourier transform is used to search for spectral peaks, and the time offset of the received signal relative to the local reference signal is estimated based on the position of these peaks. Some methods further combine the spectral peak positions of the upper and lower chirped signals to separate the carrier frequency deviation from the time-of-arrival deviation.

[0004] Several existing patented technologies have explored the application of chirped signals in time-of-arrival estimation. For example, Chinese Patent CN114337982A discloses a method for generating and estimating time-frequency chirped pilot signals, which presents a pilot structure composed of an upper linear chirped signal and a lower linear chirped signal in series. The receiver performs forward and reverse frequency domain correlation on two consecutive chirped symbols, respectively, and jointly calculates the time delay and frequency offset of the received signal based on the positions of the two frequency domain correlation peaks. This method can maintain the constant envelope characteristics of the chirped pilot signal and achieve joint estimation of time delay and frequency offset. However, its timing information mainly comes from the position of the frequency domain correlation peaks, and the estimation accuracy is limited by factors such as sampling frequency, Fourier transform frequency resolution, spectral leakage, and peak interpolation error. Another example is Chinese Patent CN114338297A, which discloses a joint timing synchronization and frequency offset estimation method for non-coherent LoRa systems, providing a method for joint timing synchronization and frequency offset estimation for satellite IoT communication. This method sequentially performs demodulation, dechirping, discrete Fourier transform, modulus extraction, peak search, and phase angle calculation on the received pilot signal to obtain a joint estimate of transmission delay and Doppler shift. This scheme focuses on reliable synchronization under conditions of extremely low signal-to-noise ratio, large transmission delay, and significant Doppler shift. Its main objective is to expand the acquisition range of timing and frequency offset, without further utilizing the chirped signal phase for high-resolution estimation of small timing deviations.

[0005] Furthermore, existing methods for estimating the arrival time of chirped signals typically use correlation peaks, matched filter peaks, or denoised spectral peaks as the primary timing observations. Zhang et al., in their paper "TOA Estimation of Chirp Signal in Dense Multipath Environment for Low-Cost Acoustic Ranging," addressed the problem of estimating the arrival time of chirped signals in dense multipath environments by employing fractional Fourier domain iterative processing to extract the first path signal and determining the signal arrival time by outputting the peak value through matched filtering. Bernier et al., in their paper "Low Complexity LoRa Frame Synchronization for Ultra-Low Power Software-Defined Radios," proposed a low-complexity frame synchronization method for LoRa chirped spread spectrum signals. This method achieves frame detection and timing / frequency synchronization by denoising and frequency domain processing of the synchronization preamble. These methods primarily utilize correlation domain or frequency domain peak positions to extract timing information, enabling arrival time estimation at the integer sampling point level or sub-sampling point level. While increasing the number of Fourier transform points or using spectral peak interpolation can improve resolution to some extent, it cannot fundamentally overcome the constraint of finite sampling rates on timing accuracy. Signal phase is highly sensitive to small time deviations and can be used to further extract residual timing deviations much smaller than one sampling period. However, the signal phase itself has integer period ambiguity, requiring reliable coarse arrival time estimation results to determine its integer phase period.

[0006] Most existing technologies use correlation peaks or spectral peaks to estimate the arrival time of chirped signals. The estimation accuracy is easily limited by sampling period, spectral resolution, peak interpolation errors, and noise. While some joint chirp estimation methods can separate carrier frequency offset from timing deviation, they do not fully consider residual secondary phase and inter-symbol timing drift caused by inconsistencies in the transceiver sampling clocks, making it difficult to guarantee estimation stability under long synchronization sequence conditions. Furthermore, existing methods do not combine sub-sampling point-level coarse estimation with fine signal phase estimation, nor do they form a complete processing flow that integrates sampling offset estimation and compensation, joint chirp coarse estimation, integer fuzzy determination of signal phase, and residual timing correction. Therefore, they cannot meet the high-precision physical layer arrival time estimation requirements of picosecond-level wireless clock synchronization systems. Summary of the Invention

[0007] This invention provides a picosecond-level wireless time-of-arrival estimation method and system based on the phase of a chirped signal, which solves the problems in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: Picosecond-level wireless time-of-arrival estimation methods based on chirped signal phase include: The receiving end performs chirped synchronization sequence acquisition on the received digital baseband signal and uses the sliding dechirped Fourier transform search method to detect spectral peaks. When stable and closely spaced significant spectral peaks appear in multiple consecutive processing windows or multiple consecutive chirped symbols, it is determined that the synchronization sequence has been acquired, and an integer sampling point-level acquisition timing estimate is obtained. A sampling offset observation is constructed using the phase difference between multiple consecutive chirped symbols. The observations of multiple symbols are coherently accumulated or averaged to obtain a sampling frequency offset estimate. Based on the sampling frequency offset estimate, a phase compensation factor is constructed for each sampling point and each symbol to compensate for the residual secondary phase term and the cumulative timing drift between symbols in the dechirped signal. Fourier transforms are performed on the compensated upper-chirped and lower-chirped dechirped signals respectively to obtain the integer positions of the upper-chirped and lower-chirped spectral peaks. Peak frequency points and their adjacent frequency points are selected near the integer spectral peaks. The fractional spectral peak corrections for upper-chirped and lower-chirped signals are obtained using the three-point spectral peak interpolation method. Based on the characteristics that the carrier frequency offset causes the upper-chirped and lower-chirped spectral peaks to shift in the same direction and the timing offset causes the upper-chirped and lower-chirped spectral peaks to shift in the opposite direction, a joint solution is performed to separate the carrier frequency offset and the arrival time offset, and a coarse estimate of the arrival time at the subsampling point level is obtained. The dechirped signal used for fine estimation is moved to the vicinity of zero frequency, coherently accumulated, and the phase observation value is extracted. The period integer corresponding to the phase is determined using the coarse arrival time estimate to eliminate the period ambiguity of the phase. The residual timing correction is calculated based on the mapping relationship between phase change and timing change. The residual timing correction is fused with the coarse arrival time estimate to obtain the final arrival time estimate.

[0009] According to one embodiment of the present invention, the chirping synchronization sequence includes a plurality of consecutive up-chirped symbols and at least one down-chirped symbol, wherein the instantaneous frequency of the up-chirped symbols increases with time, and the instantaneous frequency of the down-chirped symbols decreases with time.

[0010] According to one embodiment of the present invention, in the chirped synchronization sequence capture, for each candidate window, the received signal within the window is multiplied by the local lower chirped reference signal to complete the dechirping process of the received upper chirped signal, and a dechirped symbol is obtained. The amplitude spectrum is obtained by performing a Fourier transform on the dechirped symbol. The spectral peak positions of multiple consecutive windows or multiple consecutive upper chirped symbols are detected. When multiple consecutive spectral peaks exceed a set threshold and the peak position changes are within the allowable range, it is determined that a synchronization sequence has been captured.

[0011] According to one embodiment of the present invention, the sampling offset observation is constructed by performing conjugate multiplication and coherent accumulation on the dechirped results of two adjacent chirped symbols in the synchronization sequence, extracting the phase of the conjugate multiplication result, and calibrating it in combination with the chirped symbol length, the acquisition timing estimation, and the frequency doubling relationship between the system carrier and the sampling frequency to obtain the estimated sampling frequency offset value.

[0012] According to one embodiment of the present invention, the phase compensation factor includes an upper chirp compensation factor and a lower chirp compensation factor, which are respectively used to compensate for the residual quadratic phase term generated by time scale scaling and the cumulative timing drift term formed by sampling offset between different chirp symbols. After performing phase rotation compensation on the dechirped signal, the upper and lower chirped dechirped signals are transformed from signals containing residual quadratic phase into complex single-tone signals.

[0013] According to one embodiment of the present invention, in the three-point peak interpolation method, the peak frequency point and its left and right adjacent frequency points are selected, and the fractional peak correction amount is calculated based on the complex spectrum values ​​of the three frequency points. The corrected upper chirped peak position is the integer peak position minus the fractional peak correction amount, and the corrected lower chirped peak position is the integer peak position minus the fractional peak correction amount. The subsampling point level arrival time coarse estimation is calculated based on the difference between the corrected upper chirped peak position and the corrected lower chirped peak position.

[0014] According to one embodiment of the present invention, in the signal phase fine estimation, a chirped symbol is selected from the synchronization sequence as the signal phase fine estimation symbol. Based on the spectral peak position corresponding to the coarse arrival time estimation, the dechirped signal used for fine estimation is shifted to the vicinity of zero frequency. The near-zero frequency signal is coherently accumulated and the phase is extracted. The corresponding predicted phase is calculated using the coarse arrival time estimation. The predicted phase is compared with the observed phase to determine the period integer corresponding to the observed phase. After completing the period fuzzy determination, the continuous residual phase between the observed phase and the coarsely estimated predicted phase is calculated. The residual timing deviation is estimated based on the residual phase to obtain the residual timing correction amount.

[0015] This invention provides a wireless time-of-arrival estimation system based on the phase of a chirped signal, comprising: A digital signal processor (DSP) is used to generate a digital baseband synchronization sequence and perform digital signal processing for time-of-arrival estimation. A transmit subsystem is used to convert the digital baseband synchronization sequence into an analog signal and transmit it via an antenna. The transmit subsystem includes an interpolation filter, a digital up-converter, a digital-to-analog converter (DAC), a power amplifier, and an output filter. A receive subsystem is used to convert the wireless signal received by the antenna into a digital baseband sampling sequence and input it into the DSP. The receive subsystem includes a receive filter, a low-noise amplifier, an DAC, a digital down-converter, and a decimation filter. A reference clock and phase-locked loop (PLL) module includes a system PLL, a transmit PLL, and a receive PLL. A reference crystal drives the transmit PLL and the receive PLL through the system PLL, providing a clock or local oscillator reference for the DAC, DAC, digital mixer, and DSP.

[0016] According to one embodiment of the present invention, the digital signal processor operates a chirped spread spectrum baseband module; the synchronization sequence is processed sequentially by an interpolation filter and a digital up-converter, then converted into an analog signal by a digital-to-analog converter, and then transmitted by an antenna after passing through a power amplifier and an output filter.

[0017] According to one embodiment of the present invention, in the receiving subsystem, the wireless signal received by the antenna is sequentially passed through a receiving filter, a low-noise amplifier, and an analog-to-digital converter to form a digital sampling signal; the digital sampling signal is passed through a digital down-converter and a decimation filter to obtain a digital baseband sampling sequence, which is then input to the digital signal processor.

[0018] Compared with existing technologies, this invention has the following advantages: This invention provides a picosecond-level wireless time-of-arrival estimation method based on chirped signal phase. It acquires a synchronization sequence through a sliding dechirped FFT search, obtaining an integer-sampling-point-level acquisition timing estimate. It then uses the phase difference between consecutive chirped symbols to estimate the sampling frequency offset and constructs a compensation factor to eliminate residual secondary phase and cumulative timing drift between symbols. Next, based on the peak positions of the compensated chirped and unchirped spectral frequencies, it jointly calculates the carrier frequency offset and timing deviation to obtain a sub-sampling-point-level coarse estimate of the time of arrival. The dechirped signal is shifted to zero frequency to extract the phase. After eliminating phase period ambiguity using the coarse estimation result, the residual timing correction is calculated and fused to obtain the final estimation result. This method combines the frequency domain coarse estimation capability of chirped signals with the high sensitivity of signal phase to minute time changes, improving the time-of-arrival estimation from the sub-sampling-point level to the carrier level. It breaks through the resolution limitation formed by solely relying on the digital baseband sampling period, providing a technical foundation for achieving picosecond-level wireless time-of-arrival estimation. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a picosecond-level wireless time-of-arrival estimation method based on the phase of a chirped signal, according to an embodiment of the present invention. Figure 2 This is a hardware block diagram of a picosecond-level wireless time-of-arrival estimation system based on the phase of a chirped signal, according to an embodiment of the present invention.

[0021] Figure 3 This is a time-frequency structure diagram of the chirped synchronization sequence used in the embodiments of the present invention.

[0022] Figure 4 This is a flowchart illustrating the picosecond-level wireless time-of-arrival estimation based on the phase of a chirped signal, according to an embodiment of the present invention.

[0023] Figure 5 This is a simulation diagram of the sampling offset estimation accuracy in stage two of the embodiments of the present invention.

[0024] Figure 6 This is a diagram showing the effect of sampling offset compensation on the interpolation estimation of three-point spectral peaks in stage two of the present invention.

[0025] Figure 7 This is a diagram illustrating the effect of sampling offset compensation on arrival time estimation based on combined upper and lower chirps in stage two of this embodiment of the invention.

[0026] Figure 8 This is a comparison chart of the accuracy of signal phase estimation and the accuracy of joint estimation based on upper and lower chirps in an embodiment of the present invention.

[0027] Figure 9 The figure shows the overall performance simulation results of the picosecond-level wireless time-of-arrival estimation method based on the phase of the chirped signal in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0030] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0031] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0032] See Figure 1 In one embodiment of the present invention, a picosecond-level wireless time of arrival estimation method based on the phase of a chirped signal is provided, which solves the technical problem that existing chirped signal time of arrival estimation methods are limited by sampling frequency, spectral resolution and noise, and are difficult to meet the requirements of picosecond-level wireless clock synchronization systems.

[0033] Detailed steps include: S1: Chirped synchronization sequence capture The receiver performs chirped synchronization sequence acquisition on the received digital baseband signal and uses a sliding dechirped Fourier transform search method to detect spectral peaks. When a stable and closely spaced significant spectral peak appears in multiple consecutive processing windows or multiple consecutive chirped symbols, it is determined that a synchronization sequence has been acquired, and an integer sampling point-level acquisition timing estimate is obtained.

[0034] Specifically, the signal reaches the receiver via a wireless channel. The received signal undergoes analog-to-digital conversion, digital down-conversion, and decimation to obtain a discrete sampling sequence. Initially, the receiver cannot predict the exact arrival location of the synchronization sequence, so it uses a sliding window of length Ncs to search for the received digital baseband sequence. For each candidate window, the received signal within the window is multiplied by the local down-chirped reference signal to perform dechirping of the up-chirped signal, obtaining the dechirped symbol. An N-point FFT is performed on the dechirped symbol to obtain the corresponding amplitude spectrum. If the candidate window does not contain a complete synchronization symbol, a stable and significant single spectral peak will not form in the spectrum; when the candidate window contains a complete up-chirped symbol, a significant spectral peak with concentrated energy appears in the spectrum. The spectral peak positions of multiple consecutive windows or multiple consecutive up-chirped symbols are detected. When multiple consecutive spectral peaks exceed a set threshold and the changes in peak positions are within an allowable range, it is determined that a chirped synchronization sequence has been captured. Record the capture position of the synchronization sequence and obtain an integer sample point-level capture timing estimate, which is used to determine the processing window for subsequent up and down chirped symbols.

[0035] Explanatory: Dechirped signals can be converted from broadband signals to narrowband single-tone signals through dechirping, establishing a correspondence between spectral peak positions and signal arrival times. Sliding window search enables initial acquisition even with unknown synchronization sequence arrival times. Joint detection of multiple consecutive spectral peaks improves acquisition reliability and avoids false alarms or missed alarms from single detections. The chirped synchronization sequence consists of multiple consecutive up-chirped symbols and at least one down-chirped symbol. The instantaneous frequency of the up-chirped symbols increases with time, while the instantaneous frequency of the down-chirped symbols decreases with time. The up-chirped and down-chirped symbols have different slopes in the time-frequency plane, which is beneficial for distinguishing between carrier frequency offset and timing deviation during subsequent joint estimation.

[0036] S2: Sampling Offset Estimation and Compensation A sampling offset observation is constructed using the phase difference between multiple consecutive chirped symbols. The observations of multiple symbols are coherently accumulated or averaged to obtain a sampling frequency offset estimate. Based on the sampling frequency offset estimate, a phase compensation factor is constructed for each sampling point and each symbol to compensate for the residual secondary phase term and the cumulative timing drift between symbols in the dechirped signal.

[0037] Specifically, the dechirped results of two adjacent chirped symbols in the synchronization sequence are conjugately multiplied and coherently accumulated to construct a sampling offset observation. The phase of this observation is extracted and calibrated using the chirped symbol length, the acquisition timing estimate, and the frequency harmonic relationship between the system carrier and the sampling frequency to obtain a sampling frequency offset estimate. Based on the sampling offset estimate, chirped and lower chirped compensation factors are constructed respectively. These compensation factors compensate for the residual quadratic phase term generated by time-scale scaling, as well as the cumulative timing drift term formed between different chirped symbols. After performing phase rotation compensation on the dechirped signal, the chirped and lower chirped signals are transformed from non-ideal signals containing significant residual quadratic phase into complex single-tone signals approaching a fixed frequency. Their FFT peak width is reduced, local spectral symmetry and energy concentration are improved, and the timing drift between symbols is effectively compensated.

[0038] Explanatory: Inconsistency in the sampling clocks at the transmitting and receiving ends causes the received signal to stretch relative to the transmitted signal on the time scale. This stretching effect manifests as residual secondary phase and cumulative timing drift between symbols after dechirping. The phase difference method eliminates the common initial phase and carrier phase terms, ensuring that the phase difference between adjacent chirped symbols primarily reflects the influence of the sampling offset, thus achieving separate estimation of the sampling frequency offset. Coherent accumulation improves the accuracy of observations at low signal-to-noise ratios. After compensation, the dechirped signal is closer to the ideal single-tone signal, providing accurate basic observations for subsequent spectral peak interpolation and joint estimation.

[0039] S3: Rough estimation of combined upper and lower chirps Fourier transforms are performed on the compensated upper-chirped and lower-chirped dechirped signals respectively to obtain the integer positions of the upper-chirped and lower-chirped spectral peaks. Peak frequency points and their adjacent frequency points are selected near the integer spectral peaks, and the fractional spectral peak corrections for upper-chirped and lower-chirped signals are obtained using the three-point spectral peak interpolation method. Based on the characteristics that the carrier frequency offset causes the upper-chirped and lower-chirped spectral peaks to shift in the same direction and the timing offset causes the upper-chirped and lower-chirped spectral peaks to shift in the opposite direction, a joint solution is performed to separate the carrier frequency offset and the arrival time offset, and a coarse estimate of the arrival time at the subsampling point level is obtained.

[0040] Specifically, the compensated dechirped signal is processed using FFT. The maximum value of the upper-chirped FFT amplitude spectrum is searched, and the integer positions of the upper-chirped and lower-chirped FFT peaks are recorded. Peak frequencies and their adjacent frequencies are selected near the integer peaks, and the fractional peak correction is calculated based on the complex spectrum values ​​of these three frequencies. The corrected upper-chirped and lower-chirped peak positions are obtained. Carrier frequency offset will cause the upper-chirped and lower-chirped peaks to shift in the same direction; arrival time deviation will cause them to shift in opposite directions. A joint calculation is performed based on the compensated upper-chirped and lower-chirped peak positions to obtain a coarse timing estimate in the sub-sampling level digital domain.

[0041] Explanatory: The time-frequency slopes of the upper and lower chirps are opposite, therefore the carrier frequency offset produces spectral peak shifts in the same direction for both, while the timing offset produces spectral peak shifts in opposite directions. Utilizing this property, the two physical quantities can be separated simply by jointly calculating the compensated upper and lower chirp spectral peak positions. Three-point spectral peak interpolation uses the complex spectrum values ​​of the peak frequency and its adjacent frequencies, enabling high-precision estimation of the decimal position of the peak, thus breaking through the limitations of FFT frequency resolution in peak position estimation.

[0042] S4: Precise signal phase estimation The dechirped signal used for fine estimation is moved to the vicinity of zero frequency, coherently accumulated, and the phase observation value is extracted. The period integer corresponding to the phase is determined using the coarse arrival time estimate to eliminate the period ambiguity of the phase. The residual timing correction is calculated based on the mapping relationship between phase change and timing change. The residual timing correction is fused with the coarse arrival time estimate to obtain the final arrival time estimate.

[0043] Specifically, the lower chirped symbol is selected from the synchronization sequence as the signal phase fine estimation symbol. Based on the spectral peak position obtained by joint estimation of the upper and lower chirps, the dechirped signal used for fine estimation is shifted to near zero frequency to obtain a near-zero frequency complex baseband signal. After frequency shifting, the main linear phase rotation in the signal is removed, and the remaining phase mainly includes the carrier propagation phase, the phase corresponding to the residual timing error, and a fixed phase term that can be calculated from system parameters. The near-zero frequency signal is coherently accumulated, and the phase of the coherent accumulation result is extracted. Due to the 2π periodic ambiguity in the phase observation, the corresponding predicted phase is calculated using the coarse estimation of arrival time. The predicted phase is compared with the observed phase to determine the period integer corresponding to the observed phase. After the periodic ambiguity is determined, the continuous residual phase between the observed phase and the coarsely estimated predicted phase is calculated. The residual timing error is estimated based on the residual phase to obtain the residual timing correction. Therefore, the timing error fine estimation result is the coarse estimate plus the residual correction. The estimated final arrival time is combined with the start time of the capture window to generate the physical layer receive timestamp of the synchronization message, and then output to the RTT clock synchronization, wireless ranging, TOA positioning, or TDOA positioning module.

[0044] Explanatory: Signal phase is extremely sensitive to minute time deviations. The time corresponding to one carrier cycle is much shorter than one digital baseband sampling period; therefore, signal phase can provide timing resolution far superior to the sampling period. However, due to the inherent 2π integer multiple ambiguity in the phase, it is necessary to determine this integer multiple using a coarse estimation at the sub-sampling point level obtained from joint chirp estimation, thus correctly establishing the mapping relationship between the observed phase and the timing deviation. The final estimation result is composed of a progressively advancing process: "integer sampling point level acquisition + sub-sampling point level coarse estimation + signal phase level fine estimation," with each level providing prior information for the next, gradually approximating the true arrival time of the signal.

[0045] In one possible implementation, the chirped synchronization sequence includes multiple consecutive up-chirped symbols and at least one down-chirped symbol. This embodiment uses a synchronization sequence consisting of eight consecutive up-chirped symbols and two consecutive down-chirped symbols as an example. The instantaneous frequency of the up-chirped symbols increases with time, while the instantaneous frequency of the down-chirped symbols decreases with time.

[0046] In one possible implementation, during the chirped synchronization sequence capture, for each candidate window, the received signal within the window is multiplied by the local lower-chirped reference signal to complete the dechirping process of the received upper-chirped signal, resulting in a dechirped symbol. A Fourier transform is performed on the dechirped symbol to obtain the amplitude spectrum. The spectral peak positions of multiple consecutive windows or multiple consecutive upper-chirped symbols are detected. When multiple consecutive spectral peaks exceed a set threshold and the peak position changes are within an allowable range, it is determined that a synchronization sequence has been captured. For example, when eight consecutive upper-chirped symbols are detected to have stable spectral peaks and the peak position deviation is less than a preset threshold, the synchronization sequence capture is confirmed to be successful.

[0047] In one possible implementation, the sampling offset observation is constructed by performing conjugate multiplication and coherent accumulation on the dechirped results of two adjacent chirped symbols in the synchronization sequence; the phase of the conjugate multiplication result is extracted and calibrated by combining the chirped symbol length, the acquisition timing estimate, and the frequency harmonic relationship between the system carrier and the sampling frequency to obtain a sampling frequency offset estimate. In this embodiment, the sampling offset observation is constructed using the differential phase between two consecutive chirped symbols, and a high-precision sampling offset estimation is achieved by utilizing the frequency harmonic relationship between the sampling frequency and the carrier frequency.

[0048] In one possible implementation, the phase compensation factor includes an upper chirp compensation factor and a lower chirp compensation factor, which are used to compensate for the residual quadratic phase term generated by time-scale scaling and the cumulative timing drift term formed by sampling offset between different chirp symbols, respectively. After performing phase rotation compensation on the dechirped signal, the upper-chirped and lower-chirped dechirped signals are transformed from signals containing residual quadratic phase into complex single-tone signals. The compensated dechirped signal is closer to the ideal single-tone signal, the FFT main peak width is reduced, and the local spectral symmetry and energy concentration are improved.

[0049] In one possible implementation, the three-point peak interpolation method selects the peak frequency and its left and right adjacent frequencies, calculates the fractional peak correction amount based on the complex spectrum values ​​of the three frequencies, and the corrected upper chirped peak position is the integer peak position minus the fractional peak correction amount, and the corrected lower chirped peak position is the integer peak position minus the fractional peak correction amount; the subsampling point level arrival time coarse estimation is calculated based on the difference between the corrected upper chirped peak position and the corrected lower chirped peak position.

[0050] In one possible implementation, in the fine estimation of the signal phase, the lower chirped symbol is selected from the synchronization sequence as the fine estimation symbol. Based on the spectral peak position corresponding to the coarse estimation of arrival time, the dechirped signal used for fine estimation is shifted to near zero frequency. The near-zero frequency signal is coherently accumulated and its phase is extracted. The corresponding predicted phase is calculated using the coarse estimation of arrival time. The predicted phase is compared with the observed phase to determine the integer period corresponding to the observed phase. After completing the period fuzzy determination, the continuous residual phase between the observed phase and the coarsely estimated predicted phase is calculated. The residual timing deviation is estimated based on the residual phase to obtain the residual timing correction. This step utilizes the high sensitivity of the signal phase to small time changes to further correct the residual timing error in the joint coarse estimation result of upper and lower chirps.

[0051] like Figure 2 As shown, one embodiment of the present invention provides a wireless time-of-arrival estimation system based on the phase of a chirped signal, comprising: A digital signal processor for generating digital baseband synchronization sequences and performing digital signal processing for time-of-arrival estimation; A transmitting subsystem is used to convert the digital baseband synchronization sequence into an analog signal and transmit it via an antenna; the transmitting subsystem includes an interpolation filter, a digital up-converter, a digital-to-analog converter, a power amplifier, and an output filter; A receiving subsystem is used to convert the wireless signal received by the antenna into a digital baseband sampling sequence and input it into the digital signal processor; the receiving subsystem includes a receiving filter, a low-noise amplifier, an analog-to-digital converter, a digital down-converter, and a decimation filter; The reference clock and phase-locked loop module includes a system phase-locked loop, a transmit phase-locked loop, and a receive phase-locked loop. The reference crystal drives the transmit phase-locked loop and the receive phase-locked loop through the system phase-locked loop, respectively, to provide a clock or local oscillator reference for the digital-to-analog converter, analog-to-digital converter, and digital signal processor.

[0052] The following section uses a synchronization sequence consisting of 8 consecutive up-chirped symbols and 2 consecutive down-chirped symbols as an example to illustrate the wireless time-of-arrival estimation method based on chirped signal phase proposed in this invention.

[0053] like Figure 2As shown, the hardware system of this embodiment includes a digital signal processor, a transmitting subsystem, a receiving subsystem, a reference clock, a system phase-locked loop (PLL), a transmitting PLL, and a receiving PLL. The digital signal processor generates a digital baseband synchronization sequence in the transmitting direction. This synchronization sequence is processed sequentially by an interpolation filter, a digital mixer, and a digital up-converter, then converted into an analog signal by a digital-to-analog converter (DAC), and finally transmitted by the antenna after passing through a power amplifier and an output filter. In the receiving direction, the wireless signal received by the antenna sequentially passes through a receiving filter, a low-noise amplifier, and an analog-to-digital converter (ADC) to form a digital sampling signal. This signal is then processed by a digital down-converter and a decimation filter to obtain a digital baseband sampling sequence, which is then input to the digital signal processor. The reference crystal oscillator drives the transmitting PLL and the receiving PLL through the system PLL, providing a clock or local oscillator reference for the DAC, ADC, digital mixer, and digital signal processor. The reference crystal oscillators of different wireless nodes are independent of each other, thereby creating sampling frequency offset and carrier frequency offset between the transmitting and receiving nodes.

[0054] Figure 3 The time-frequency structure diagram for generating a synchronization preamble sequence consisting of 8 upper-chirped symbols and 2 lower-chirped symbols for the transmitter is shown. Its digital baseband expression is: (1) in, and These represent the upper and lower chirp symbols, respectively; Indicates the chirp symbol index in the synchronization sequence; This indicates the number of sampling points for one chirp symbol. The signal is transmitted via an antenna after interpolation, digital up-conversion, and digital-to-analog conversion.

[0055] Figure 4 The flowchart of the picosecond-level wireless time-of-arrival estimation method based on chirped signal phase of this invention is shown below. It mainly consists of four stages: chirped synchronization sequence acquisition stage, sampling offset estimation and compensation stage, joint coarse estimation stage of upper and lower chirps, and fine signal phase estimation stage. The specific operations are as follows: Chirped synchronization sequence capture phase (phase one): The signal reaches the receiver via a wireless channel. The received signal undergoes analog-to-digital conversion, digital down-conversion, and decimation to obtain a discrete sampling sequence, which can be represented as: (2) in, and These represent the up and down chirp symbols after discrete sampling at the receiving end, respectively. This indicates a sampling offset caused by inconsistency between the sampling clocks of the transmitting and receiving ends; Represents timing deviation in the digital field. This indicates the period of discrete sampling by the transmitting DSP; This indicates the carrier frequency offset of the transceiver; This indicates the carrier frequency of the transmitter.

[0056] Since the receiver cannot predict the exact arrival location of the synchronization sequence in the initial state, a sliding window of length Ncs is used to search for the received digital baseband sequence. For each candidate window, the received signal within the window is multiplied by the local lower chirp reference signal to complete the dechirping of the upper chirp signal, obtaining the dechirped symbol of the upper chirp.

[0057] (3) Perform dechirping on the symbol Point FFT is used to obtain the corresponding amplitude spectrum. If the candidate window does not contain a complete chirped symbol, a stable and significant single spectral peak will not form in the spectrum; when the candidate window contains a complete chirped symbol, a significant spectral peak with concentrated energy appears in the spectrum.

[0058] The system detects the spectral peak positions of multiple consecutive windows or multiple consecutive chirped symbols. When multiple consecutive spectral peaks exceed a set threshold, and the changes in peak positions are within the allowable range, a chirped synchronization sequence is determined to have been captured.

[0059] Record the capture position of the synchronization sequence and obtain an integer sample point-level capture timing estimate. The capture timing estimate is used to determine the processing window for subsequent upper and lower chirp symbols, but is not directly used as the final arrival time estimate.

[0060] Sampling offset estimation and compensation stage (stage two): Conjugate multiplication and coherent summation are performed on the dechirped results of two adjacent chirped symbols in the synchronization sequence to construct the sampling offset observation.

[0061] (4) Among them, symbols Indicates complex conjugation.

[0062] extract The phase, combined with the known chirp symbol length. Capture timing estimation The sampling offset estimate is obtained by calibrating the relationship between the system carrier and the sampling frequency multiples. .

[0063] Based on the sampling offset estimate Construct chirp compensation factors respectively And chirping compensation factor .

[0064] (5) The compensation factor is used to compensate for the residual secondary phase term per sampling point caused by time scale scaling, as well as the cumulative timing drift term formed between different chirp symbols due to sampling offset.

[0065] Perform phase rotation compensation on the dechirped signal to obtain the compensated dechirped symbol with upchirped input. The dechirp symbol for chirping.

[0066] (6) After compensation, the chirped signals of the upper and lower chirps are transformed from non-ideal signals with obvious residual secondary phase into complex single-tone signals with a near-fixed frequency. The width of the FFT main peak is reduced, and the local spectral symmetry and energy concentration are improved. Moreover, the timing drift between symbols is effectively compensated.

[0067] Joint rough estimation stage based on chirping from above and below (Stage 3): Perform FFT processing on the compensated dechirped signal. Search for the maximum value of the chirped FFT amplitude spectrum and record the integer position of the chirped spectrum peak. The integer position of the peak of the chirped FFT is .

[0068] Select the peak frequency point and its left and right adjacent frequencies near the integer spectral peak, and calculate the fractional spectral peak correction based on the complex spectrum values ​​of the three frequency points.

[0069] (7) in, This represents the complex spectrum value at the peak frequency. , These represent the complex spectral values ​​at the left and right adjacent frequencies of the peak, respectively. The corrected upper chirped peak position is obtained as follows: The corrected position of the lower chirp peak is .

[0070] Carrier frequency offset causes the peak values ​​of the up-chirped and down-chirped spectra to shift in the same direction; arrival time deviation causes them to shift in opposite directions. A joint calculation based on the compensated up-chirped and down-chirped peak positions yields a coarse timing estimate in the sub-sampling-point digital domain. (8) Phase-level fine estimation stage (stage four): The lower chirped symbol is selected from the synchronization sequence as the symbol for fine phase estimation of the signal. Based on the spectral peak position obtained by joint estimation of the upper and lower chirps, the dechirped signal used for fine estimation is shifted to near zero frequency to obtain the near-zero frequency complex baseband signal.

[0071] (9) After frequency shift, the main linear phase rotation in the signal is removed, and the remaining phase mainly includes the carrier propagation phase, the phase corresponding to the residual timing deviation, and the fixed phase term that can be calculated from the system parameters.

[0072] Coherent accumulation is performed on the near-zero frequency signal, and the phase of the coherent accumulation result is extracted.

[0073] (10) Due to the existence of phase observation The periodic ambiguity, the actual phase obtained for (11) in, This indicates phase unwinding.

[0074] Define the phase function as (12) in, Indicates the first Timing deviation of each symbol. This indicates the frequency harmonic relationship between the carrier frequency and the sampling frequency.

[0075] Because phase has Due to its periodicity, the full-range timing deviation cannot be directly deduced from the observed phase. A rough estimate is made based on the arrival time. Calculate the corresponding predicted phase. Compare the predicted phase with the observed phase to determine the integer period corresponding to the observed phase. After completing the period fuzzy determination, calculate the continuous residual phase between the observed phase and the coarsely estimated predicted phase. : (13) The residual timing error is estimated based on the residual phase to obtain the residual timing correction. Therefore, the first The precise estimation results of the timing deviation at each symbol are as follows: .

[0076] The estimated final arrival time is combined with the start time of the capture window to generate the physical layer receive timestamp of the synchronization message, and then output to the RTT clock synchronization, wireless ranging, TOA positioning, or TDOA positioning module.

[0077] In summary, this invention obtains the initial position at the integer sampling point level by capturing the chirped synchronization sequence, recovers the ideal single-tone characteristics of the dechirped signal by sampling offset estimation and compensation, obtains the sub-sampling point level coarse timing result by joint calculation of upper and lower chirps, and then completes the residual timing fine correction by using the signal phase, thereby forming a wireless time of arrival estimation link that is progressively advanced by "sampling point level capture, sub-sampling point level coarse estimation, and signal phase fine estimation".

[0078] The simulation of this invention verifies the performance of sampling offset estimation, sampling offset compensation, joint coarse estimation of upper and lower chirps, and fine estimation of signal phase by changing the signal-to-noise ratio (SNR) of the received signal and the crystal oscillator frequency offset between the transmitter and receiver. The SNR variation range is set to -5dB to 25dB, and the crystal oscillator frequency offset variation range is set to 0ppm to 2ppm. The system sampling rate is 20Msps, and the carrier frequency is 1GHz. The number of sampling points for the chirped symbol is 8192. The root mean square error (RMSE) of the estimation results is used as the performance evaluation index.

[0079] Figure 5 This is a simulation result graph of the sampling offset estimation accuracy in Phase Two of this invention, where the horizontal axis represents the signal-to-noise ratio and crystal oscillator frequency offset, and the vertical axis represents the root mean square error of the sampling offset estimation result. Figure 5 It is evident that, under different crystal oscillator frequency offset conditions, the sampling offset estimation error continuously decreases with increasing signal-to-noise ratio (SNR). When the SNR is low, noise significantly disturbs the differential phase between adjacent chirped symbols, resulting in a relatively large sampling offset estimation error. As the SNR gradually increases, the phase difference observation between adjacent chirped symbols tends to stabilize, and the estimation error decreases rapidly. Under the same SNR conditions, when the crystal oscillator frequency offset increases from 0 ppm to 2 ppm, the sampling offset estimation error does not show a significant increase, and the surface along the crystal oscillator frequency offset direction remains relatively flat. This indicates that the phase difference method between consecutive chirped symbols used in this invention can stably estimate the sampling frequency offset within a certain crystal oscillator offset range. The estimation accuracy is mainly affected by the received signal-to-noise ratio (SNR), while exhibiting good adaptability to the magnitude of the crystal oscillator offset.

[0080] Figure 6 This paper compares the interpolation errors of the three-point spectral peaks before and after sampling offset compensation at a crystal oscillator frequency offset of 2 ppm. The horizontal axis represents the signal-to-noise ratio, and the vertical axis represents the root mean square error of the three-point spectral peak interpolation in units of FFT frequency intervals. Figure 6As can be seen, without sampling offset compensation, the three-point peak interpolation error only decreases slightly with increasing signal-to-noise ratio (SNR) and eventually stabilizes around 0.016 FFT frequency points. This is because the sampling offset introduces residual quadratic phase into the dechirped signal, causing broadening of the FFT main peak and local spectral skew. Even with increased SNR, this deterministic error cannot be eliminated. After sampling offset compensation, the three-point peak interpolation error decreases significantly with increasing SNR, gradually approaching zero under medium-to-high SNR conditions. This result demonstrates that the sampling offset compensation factor constructed in this invention can effectively eliminate the residual quadratic phase caused by time-scale scaling, bringing the dechirped signal closer to an ideal complex single-tone signal, improving the energy concentration and local symmetry of the spectral main peak, and thus significantly improving the estimation accuracy of the fractional position of the spectral peak.

[0081] Figure 7 The graph shows the impact of sampling offset compensation on the joint arrival time estimation results of upper and lower chirps, where the horizontal axis represents the signal-to-noise ratio and the vertical axis represents the root mean square error of the arrival time estimation, in nanoseconds (ns). Figure 7 As can be seen, without sampling offset compensation, the arrival time estimation error slowly decreases from approximately 0.92 ns under low signal-to-noise ratio (SNR) conditions and stabilizes at approximately 0.82 ns under medium-to-high SNR conditions, exhibiting a clear lower limit for the estimation error. This error is mainly caused by the cumulative timing drift between symbols due to sampling offset and the additional shifts in the positions of the upper and lower chirped spectral peaks, which are difficult to eliminate by increasing the SNR. After sampling offset compensation, the arrival time estimation error decreases rapidly with increasing SNR, gradually decreasing from approximately 0.41 ns at -5 dB to 0.1 ns at 7 dB. This result demonstrates that sampling offset compensation can not only improve the interpolation accuracy of the spectral peaks of a single chirped symbol but also correct the cumulative timing drift between different chirped symbols, providing spectral peak observations that satisfy the ideal model for joint upper and lower chirped solutions.

[0082] Figure 8 This paper presents a comparison of the accuracy of a precise signal phase estimation method and a traditional combined upper and lower chirp estimation method. Simulations were performed with two crystal oscillator frequency offset conditions: 1 ppm and 2 ppm. Figure 8It is evident that the error of the traditional chirp-plus-chirp joint estimation method gradually decreases with increasing signal-to-noise ratio (SNR), but stabilizes at approximately 0.41 ns and 0.82 ns under medium-to-high SNR conditions, respectively. Furthermore, the larger the crystal oscillator offset, the more pronounced the residual arrival time estimation error. After employing the signal phase fine estimation method of this invention, the estimation curves under the two crystal oscillator offset conditions essentially overlap, and the estimation error decreases rapidly with increasing SNR. When the SNR reaches approximately 5 dB, the error has decreased to around 0.02 ns; as the SNR further increases, the estimation error gradually approaches zero. These results demonstrate that, after completing sampling offset compensation and carrier frequency offset correction, this invention can utilize the high sensitivity of the signal phase to small time changes to further correct the residual timing error in the chirp-plus-chirp joint coarse estimation results, and exhibits good robustness to crystal oscillator frequency offsets of different magnitudes.

[0083] Figure 9 The table shows the overall performance simulation results of the wireless time-of-arrival estimation method based on chirped signal phase according to the present invention. The horizontal axis represents the signal-to-noise ratio and crystal oscillator frequency offset, respectively, and the vertical axis represents the root mean square error of the final time-of-arrival estimation. Figure 9 As can be seen, within the range of crystal oscillator frequency offset from 0 ppm to 2 ppm, the final arrival time estimation error decreases significantly with increasing signal-to-noise ratio (SNR). However, under the same SNR conditions, the estimation error changes relatively little with crystal frequency offset. When the SNR is low, the signal phase observation is significantly affected by noise, resulting in a final arrival time estimation error on the sub-nanosecond scale. As the SNR increases, the accuracy of sampling offset estimation, spectral peak interpolation, and phase extraction improves simultaneously, leading to a rapid decrease in the final estimation error, which gradually enters the picosecond range under medium-to-high SNR conditions. Simulation results demonstrate that this invention can effectively suppress the influence of crystal oscillator frequency offset on chirped signal arrival time estimation, achieving stable and high-precision wireless arrival time estimation.

[0084] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0085] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A picosecond-level wireless time-of-arrival estimation method based on chirped signal phase, characterized in that... ,include: The receiving end performs chirped synchronization sequence acquisition on the received digital baseband signal and uses the sliding dechirped Fourier transform search method to detect spectral peaks. When stable and closely spaced significant spectral peaks appear in multiple consecutive processing windows or multiple consecutive chirped symbols, it is determined that the synchronization sequence has been acquired, and an integer sampling point-level acquisition timing estimate is obtained. A sampling offset observation is constructed using the phase difference between multiple consecutive chirped symbols. The observations of multiple symbols are coherently accumulated or averaged to obtain a sampling frequency offset estimate. Based on the sampling frequency offset estimate, a phase compensation factor is constructed for each sampling point and each symbol to compensate for the residual secondary phase term and the cumulative timing drift between symbols in the dechirped signal. Fourier transforms are performed on the compensated upper-chirped and lower-chirped dechirped signals respectively to obtain the integer positions of the upper-chirped and lower-chirped spectral peaks. Peak frequency points and their adjacent frequency points are selected near the integer spectral peaks. The fractional spectral peak corrections for upper-chirped and lower-chirped signals are obtained using the three-point spectral peak interpolation method. Based on the characteristics that the carrier frequency offset causes the upper-chirped and lower-chirped spectral peaks to shift in the same direction and the timing offset causes the upper-chirped and lower-chirped spectral peaks to shift in the opposite direction, a joint solution is performed to separate the carrier frequency offset and the arrival time offset, and a coarse estimate of the arrival time at the subsampling point level is obtained. The dechirped signal used for fine estimation is moved to the vicinity of zero frequency, coherently accumulated, and the phase observation value is extracted. The period integer corresponding to the phase is determined using the coarse arrival time estimate to eliminate the period ambiguity of the phase. The residual timing correction is calculated based on the mapping relationship between phase change and timing change. The residual timing correction is fused with the coarse arrival time estimate to obtain the final arrival time estimate.

2. The picosecond-level wireless time-of-arrival estimation method based on chirped signal phase according to claim 1, characterized in that, The chirping synchronization sequence includes multiple consecutive up-chirped symbols and at least one down-chirped symbol, wherein the instantaneous frequency of the up-chirped symbol increases with time, and the instantaneous frequency of the down-chirped symbol decreases with time.

3. The picosecond-level wireless time-of-arrival estimation method based on chirped signal phase according to claim 1, characterized in that, In the chirped synchronization sequence capture, for each candidate window, the received signal within the window is multiplied by the local lower chirped reference signal to complete the dechirping process of the received upper chirped signal, resulting in a dechirped symbol. The dechirped symbol is then subjected to a Fourier transform to obtain the amplitude spectrum. The spectral peak positions of multiple consecutive windows or multiple consecutive upper chirped symbols are detected. When multiple consecutive spectral peaks exceed a set threshold and the peak position changes are within the allowable range, it is determined that a synchronization sequence has been captured.

4. The picosecond-level wireless time-of-arrival estimation method based on chirped signal phase according to claim 1, characterized in that, The sampling offset observation is constructed by performing conjugate multiplication and coherent accumulation on the dechirped results of two adjacent chirped symbols in the synchronization sequence, extracting the phase of the conjugate multiplication result, and calibrating it in combination with the chirped symbol length, the acquisition timing estimate, and the frequency doubling relationship between the system carrier and the sampling frequency to obtain the estimated sampling frequency offset value.

5. The picosecond-level wireless time-of-arrival estimation method based on chirped signal phase according to claim 1, characterized in that, The phase compensation factor includes an upper chirp compensation factor and a lower chirp compensation factor, which are used to compensate for the residual quadratic phase term generated by time scale scaling and the cumulative timing drift term formed by sampling offset between different chirp symbols, respectively. After performing phase rotation compensation on the dechirped signal, the upper and lower chirped dechirped signals are transformed from signals containing residual quadratic phase into complex single-tone signals.

6. The picosecond-level wireless time-of-arrival estimation method based on chirped signal phase according to claim 1, characterized in that, In the three-point peak interpolation method, the peak frequency and its left and right adjacent frequencies are selected. The fractional peak correction is calculated based on the complex spectrum values ​​of the three frequencies. The corrected upper chirped peak position is the integer peak position minus the fractional peak correction. The corrected lower chirped peak position is the integer peak position minus the fractional peak correction. The subsampling point level arrival time coarse estimation is calculated based on the difference between the corrected upper chirped peak position and the corrected lower chirped peak position.

7. The picosecond-level wireless time-of-arrival estimation method based on chirped signal phase according to claim 1, characterized in that, In the fine estimation of the signal phase, the chirped symbol is selected from the synchronization sequence as the fine estimation symbol of the signal phase. According to the spectral peak position corresponding to the coarse estimation of arrival time, the dechirped signal used for fine estimation is moved to the vicinity of zero frequency. The near-zero frequency signal is coherently accumulated and the phase is extracted. The corresponding predicted phase is calculated using the coarse estimation of arrival time. The predicted phase is compared with the observed phase to determine the period integer corresponding to the observed phase. After the period fuzzy determination is completed, the continuous residual phase between the observed phase and the coarsely estimated predicted phase is calculated. The residual timing deviation is estimated based on the residual phase to obtain the residual timing correction amount.

8. A wireless time-of-arrival estimation system based on the phase of a chirped signal, characterized in that, include: A digital signal processor for generating digital baseband synchronization sequences and performing digital signal processing for time-of-arrival estimation; A transmitting subsystem is used to convert the digital baseband synchronization sequence into an analog signal and transmit it via an antenna; the transmitting subsystem includes an interpolation filter, a digital up-converter, a digital-to-analog converter, a power amplifier, and an output filter; A receiving subsystem is used to convert the wireless signal received by the antenna into a digital baseband sampling sequence and input it into the digital signal processor; the receiving subsystem includes a receiving filter, a low-noise amplifier, an analog-to-digital converter, a digital down-converter, and a decimation filter; The reference clock and phase-locked loop module includes a system phase-locked loop, a transmit phase-locked loop, and a receive phase-locked loop. The reference crystal drives the transmit and receive phase-locked loops through the system phase-locked loop, providing a clock or local oscillator reference for the digital-to-analog converter, analog-to-digital converter, digital mixer, and digital signal processor.

9. The wireless time-of-arrival estimation system based on chirped signal phase according to claim 8, characterized in that, The digital signal processor operates the chirped spread spectrum baseband module; the synchronization sequence is processed sequentially by an interpolation filter and a digital up-converter, then converted into an analog signal by a digital-to-analog converter, and finally transmitted by an antenna after passing through a power amplifier and an output filter.

10. The wireless time-of-arrival estimation system based on chirped signal phase according to claim 8, characterized in that, In the receiving subsystem, the wireless signal received by the antenna is sequentially passed through a receiving filter, a low-noise amplifier, and an analog-to-digital converter to form a digital sampling signal; the digital sampling signal is then passed through a digital down-converter and a decimation filter to obtain a digital baseband sampling sequence, which is then input into the digital signal processor.

Images