Satellite navigation signal weak signal tracking loop switching method

Abstract

The invention relates to a satellite navigation signal weak signal tracking loop switching method. The method comprises the following steps: designing a loop filter of a double-frequency phase-locked loop, a loop filter of a double-frequency frequency-locked loop and a loop filter of a pilot component single-frequency frequency-locked loop based on a Kalman filtering algorithm; constructing observed quantity and state quantity of a Kalman filter based on phase discrimination, frequency discrimination and code discrimination results of the B2a signal and the B1I signal; the state quantity output by the loop filter is fed back to a local carrier generator and a composite spreading code generator for error correction; in the filtering process, the adopted loop filter is switched according to the carrier-to-noise ratio, the phase locking detection index and the frequency locking detection index of the two signals, the corresponding loop filter is adaptively selected to complete parameter estimation, loop locking is further carried out, and signal tracking and positioning are achieved.

Description

Technical Field

[0001] This invention relates to the field of navigation signal acquisition and tracking technology, and in particular to a method for switching the tracking loop of weak satellite navigation signals. Background Technology

[0002] As navigation applications become increasingly diverse, user navigation receivers frequently encounter complex environments, which can easily lead to signal strength instability. Fluctuations in satellite navigation signal strength can cause receiver loop lock-up, resulting in the inability to provide positioning. With the expanding scope of navigation applications, higher demands are placed on receiver positioning accuracy under complex environmental conditions. Loop lock-up can prevent receiving equipment from acquiring conventional satellite navigation signals, leading to a decline in navigation service capabilities. Therefore, research on the impact of navigation signal power variations on receiver loops is imperative.

[0003] Dual-frequency or multi-frequency receivers have become a current development trend. Multi-frequency receivers offer several advantages: First, observations at multiple frequencies can be used to estimate ionospheric delay in real time, improving positioning accuracy, and enabling rapid resolution of carrier phase ambiguity in precision positioning systems. Second, frequency diversity enhances the receiver's security and robustness; even when one frequency band is disturbed, other signals are still highly likely to continue operating. Third, providing multiple frequency options can meet the needs of different receiver users, facilitating widespread system application and market penetration.

[0004] In the baseband signal processing flow of navigation receivers, the design of the tracking loop is the most crucial and has received the most extensive research, encompassing various aspects such as loop filter design and loop switching strategies. Regarding track loop filter design, some researchers have employed Kalman filters to replace phase-locked loops and frequency-locked loops for loop feedback, thereby improving the loop's tracking accuracy for carrier and code phases. Other researchers have modeled the control system for Kalman loop filters and proposed a complete set of design standards for parameters such as the filter's noise covariance based on this model. In terms of tracking loop switching, the loop carrier-to-noise ratio (CNR) is the most common decision metric. Furthermore, based on the CNR, methods such as switching the tracking loop's noise bandwidth and the Kalman filter's noise covariance matrix parameters have been developed to improve carrier phase tracking accuracy.

[0005] With the continuous improvement of satellite navigation system construction, multi-frequency tracking loop processing algorithms have become one of the hot research topics in the field of satellite navigation. For example, using GPS L1 and L2C frequency signals, based on the carrier Doppler correlation between the frequencies, the least squares algorithm is used to combine the phase detection results of the carrier phase and code phase of the two frequencies to perform loop feedback.

[0006] Multi-frequency joint processing methods for satellite navigation signals are increasingly being applied to GNSS (Global Navigation Satellite System) construction to improve satellite navigation and positioning capabilities in complex environments. For satellite navigation receivers, alternating strong and weak signals can easily cause a receiver that was initially locked to lose its lock on a particular satellite signal. How to quickly reacquire the lost signal directly affects the accuracy and continuity of subsequent positioning results. Therefore, it is necessary to analyze the impact of satellite navigation signal power variations on the receiver loop and design relevant adaptive switching or auxiliary loops to improve the rapid acquisition and positioning capabilities of satellite navigation receivers. Summary of the Invention

[0007] Based on the above analysis, the present invention aims to provide a method for switching the weak signal tracking loop of satellite navigation signals, in order to solve the problem that existing satellite navigation receivers are prone to signal loss due to the alternating changes in signal strength.

[0008] This invention provides a method for switching the weak signal tracking loop of a satellite navigation signal. The method involves acquiring B1I and B2a signals, processing them to obtain the leading, immediate, and lagging intermediate frequency (IF) signals of the I branch and the leading, immediate, and lagging IF signals of the Q branch for each signal, respectively. The leading, immediate, and lagging IF signals of the I and Q branches of each signal are then subjected to code correlation operations with a composite spreading code to obtain multi-path correlation data. The multi-path correlation data is then coherently integrated. When a preset coherent integration time is reached, the coherent integration result corresponding to each signal is input to a phase detector, a frequency detector, and a code detector for phase, frequency, and code detection operations.

[0009] Loop filters based on the Kalman filter algorithm are designed for dual-frequency phase-locked loop, dual-frequency frequency-locked loop, and pilot component single-frequency frequency-locked loop; the observations and state variables of the Kalman filter are constructed based on the phase discrimination, frequency discrimination, and code discrimination results of the B2a and B1I signals; the state variables output by the loop filter are fed back to the local carrier generator and the composite spreading code generator for error correction.

[0010] During the filtering process, the loop filter used is switched according to the carrier-to-noise ratio, phase-lock detection index, and frequency-lock detection index of the two signals. The appropriate loop filter is adaptively selected to complete parameter estimation and then loop locking, thereby realizing signal tracking and positioning.

[0011] Furthermore, the preset coherent integration time includes: coherent integration time t1 and coherent integration time t2, where t1 = 20ms and t2 ranges from 120ms to 140ms.

[0012] Furthermore, the phase-locked detection index is the threshold range set by the dual-frequency phase-locked loop based on the coherent integral values ​​of the instantaneous signals of the I and Q branches to determine whether the phase tracking of the B1I and B2a signals is normal; the frequency-locked detection index is the threshold range set by the dual-frequency frequency-locked loop based on the coherent integral values ​​of the leading and lagging signals of the I and Q branches to determine whether the frequency tracking of the B1I and B2a signals is normal, and the pilot component single-frequency frequency-locked loop based on the coherent integral values ​​of the leading and lagging signals of the I and Q branches to determine whether the frequency tracking of the B2a signal is normal.

[0013] Furthermore, the loop filter is switched based on the carrier-to-noise ratio, phase-locked detection index, and frequency-locked detection index of the two signals, including:

[0014] When the phase pull-in of any frequency point in the dual-frequency locked loop is normal, the dual-frequency phase-locked loop pull-in begins. When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1. The tracking process includes:

[0015] When the phase tracking of the dual-frequency phase-locked loop (PLL) is normal during the coherent integration time t1 and the carrier-to-noise ratio (CNR) is less than the lower limit of the CNR threshold set by the dual-frequency PLL, the dual-frequency PLL will track for a coherent integration time t2. When the phase tracking of the dual-frequency PLL during the coherent integration time t1 is abnormal and the CNR is greater than the lower limit of the CNR threshold set by the dual-frequency frequency-locked loop, the dual-frequency frequency-locked loop will track for a coherent integration time t1. When the phase tracking of the dual-frequency PLL during the coherent integration time t1 is abnormal or the CNR is less than the lower limit of the CNR threshold set by the dual-frequency frequency-locked loop, the pilot component frequency-locked loop will track for a coherent integration time t2.

[0016] Furthermore, when the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, which also includes:

[0017] When a phase tracking anomaly occurs during tracking by the dual-frequency phase-locked loop at a coherent integration time of t2, the pilot component phase-locked loop is switched to track at a coherent integration time of t2. Otherwise, when the tracking phase is normal during tracking by the dual-frequency phase-locked loop at a coherent integration time of t2 and the signal carrier-to-noise ratio is greater than the upper limit of the carrier-to-noise ratio threshold set by the dual-frequency phase-locked loop, the dual-frequency phase-locked loop is switched to track at a coherent integration time of t1.

[0018] Furthermore, when the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, which also includes:

[0019] When the frequency tracking of the pilot component frequency-locked loop is normal or the carrier-to-noise ratio is greater than the upper limit of the carrier-to-noise ratio threshold set by the dual-frequency frequency-locked loop, the dual-frequency frequency-locked loop will enter the tracking mode and ...

[0020] Furthermore, when the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, which also includes:

[0021] When the frequency tracking of the dual-frequency phase-locked loop is normal and the carrier-to-noise ratio (CNR) is greater than the upper limit of the CNR threshold set by the dual-frequency phase-locked loop, the dual-frequency phase-locked loop tracks for t1 coherent integration time. When the frequency tracking of the dual-frequency phase-locked loop is abnormal or the CNR is less than the lower limit of the CNR threshold set by the dual-frequency phase-locked loop, the pilot component phase-locked loop tracks for t2 coherent integration time.

[0022] Furthermore, for the dual-frequency phase-locked loop (PLL), when the carrier-to-noise ratio (CNR) of the B2a signal is greater than the lower limit of the CNR threshold set by the PLL, the B2a signal is selected as the main tracking signal. When the CNR of the B2a signal is less than the lower limit of the CNR threshold set by the PLL and the CNR of the B1I signal is greater than the lower limit of the CNR threshold set by the PLL, the B1I signal is selected as the main tracking signal. The dual-frequency PLL selects the carrier phase error corresponding to the main tracking signal output by the phase detector, the code phase error corresponding to the main tracking signal output by the code detector, the frequency error and frequency change rate error corresponding to the main tracking signal output by the frequency discriminator, the inter-frequency carrier phase delay of the B2a and B1I signals output by the phase detector, and the inter-frequency code phase delay of the B2a and B1I signals output by the code discriminator as state variables. The carrier phase error corresponding to the main tracking signal output by the phase detector, the code phase error corresponding to the main tracking signal output by the code discriminator, and the frequency error corresponding to the main tracking signal output by the frequency discriminator are selected as observations.

[0023] Furthermore, for the pilot component single-frequency frequency-locked loop, the B2a signal is selected as the main tracking signal, and the code phase error, frequency error, and frequency change rate error of the B2a signal output by the code discriminator, are taken as state variables; the code phase error and frequency error of the B2a signal output by the code discriminator and the frequency error of the B2a signal output by the frequency discriminator are taken as observations.

[0024] Furthermore, for the dual-frequency locking loop, when the carrier-to-noise ratio (CNR) of the B2a signal is greater than the lower limit of the CNR threshold set by the dual-frequency locking loop, the B2a signal is selected as the main tracking signal. When the CNR of the B2a signal is less than the lower limit of the CNR threshold set by the dual-frequency locking loop and the CNR of the B1I signal is greater than the lower limit of the CNR threshold set by the dual-frequency locking loop, the B1I signal is selected as the main tracking signal. The dual-frequency locking loop selects the frequency error and frequency change rate error of the main tracking signal output by the frequency discriminator, the code phase error of the main tracking signal output by the code discriminator, and the inter-frequency code phase delay of the B2a and B1I signals output by the code discriminator as state variables. The frequency error of the main tracking signal output by the frequency discriminator and the code phase errors of the B2a and B1I signals output by the code discriminator are selected as observations.

[0025] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0026] 1. This invention designs a signal tracking loop filter, a carrier loop discriminator, and a code detector. Based on the carrier-to-noise ratio estimation, it designs tracking loop parameters and switches the tracking mode at any time according to the monitoring of the loop status. It has higher sensitivity and phase and frequency locking accuracy, and realizes loop locking and satellite acquisition and tracking.

[0027] 2. The present invention provides a weak signal tracking loop switching method for satellite navigation signals. It utilizes the characteristics of existing novel dual-frequency signals, proposes an asymmetric dual-frequency tracking algorithm, designs a highly sensitive adaptive loop, and realizes the tracking of dual-frequency navigation satellite signals with different modulation methods, initial phases, and signal powers, as well as the acquisition and tracking under weak signal conditions, thereby improving navigation and positioning accuracy.

[0028] 3. The receiver of this invention is designed with pilot signal frequency-locked loop tracking in weak signal conditions. The pilot signal of the satellite signal refers to a signal that does not contain data bits or only contains fixed second-order codes. It is not affected by the data bit mutation during long-term coherent integration, thus improving the loop sensitivity.

[0029] 4. This invention designs three loop filter Kalman equations: dual-frequency phase-locked loop, dual-frequency frequency-locked loop, and pilot component single-frequency frequency-locked loop. It also proposes an adaptive loop switching strategy based on satellite carrier-to-noise ratio, phase-locking index, and frequency-locking index, to achieve a loop tracking algorithm with higher sensitivity and phase and frequency locking accuracy compared to single-frequency signal processing of B1I signals.

[0030] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0031] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0032] Figure 1 This is a diagram of a dual-frequency signal tracking loop structure for a weak signal tracking loop switching method in satellite navigation;

[0033] Figure 2 This is a schematic diagram of adaptive loop switching in a weak signal tracking loop switching method for satellite navigation signals. Detailed Implementation

[0034] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0035] A specific embodiment of the present invention discloses a method for switching the tracking loop of a weak satellite navigation signal, comprising steps S1-S3. The method is as follows: Figure 1 and Figure 2 As shown.

[0036] S1. Acquire B1I and B2a signals, process B1I and B2a signals to obtain the intermediate frequency signals of the I branch leading, instantaneous and lagging for each signal and the intermediate frequency signals of the Q branch leading, instantaneous and lagging for each signal.

[0037] Specifically, the B1I and B2a signals received by the Beidou antenna are acquired, and the signals are down-converted and demodulated to obtain the I-branch and Q-branch intermediate frequency signals. The I-branch and Q-branch intermediate frequency signals are correlated with the carrier generated by the carrier generator to remove the local carrier, and then the leading, immediate and lagging I-branch and Q-branch intermediate frequency signals are generated.

[0038] like Figure 1 As shown, for the 1ms B1I and B2a signals, local carrier stripping and code correlation operations of the composite spreading code are performed simultaneously. The local carrier is generated by the carrier generator based on the intermediate frequency of the signal sampling device and the carrier Doppler obtained during the bit synchronization process.

[0039] The B2a signal can be replaced with the B1C signal.

[0040] Optionally, the multi-frequency point combination can select B1C and B1I signals. Utilizing the homology between B1C and B1I signals, joint and coordinated acquisition can be achieved. In weak signal mode, B1C plays an auxiliary role in B1I signal acquisition.

[0041] S2. Perform code correlation operations on the I and Q branches of each signal for the leading, instantaneous, and lagging intermediate frequency signals with the composite spreading code to obtain multi-channel correlation data. Perform coherent integration on the multi-channel correlation data. When the preset coherent integration time is reached, input the coherent integration result corresponding to each signal into the phase detector, frequency detector, and code detector for phase detection, frequency detection, and code detection operations.

[0042] Specifically, the composite spreading code is generated by the code generator based on the spreading code and the phase of the second-level code of the signal. During the correlation operation, in addition to the instantaneous branch, code correlation operations are also required for the lead and lag branches, respectively, for code phase error identification of the code loop; Figure 1 The mid-carrier loop discriminator includes a frequency discriminator and a phase discriminator; it performs coherent integration on the correlation calculation result of 1ms, and when the coherent integration time is reached, it performs frequency discrimination, phase discrimination and code discrimination operations on the B1I signal and the B2a signal, respectively.

[0043] Specifically, an arctangent phase detector is used to implement the phase detection operation of the loop. For the B1I signal, a two-quadrant arctangent phase detector is used; for the B2a signal, a four-quadrant arctangent phase detector is used to increase its pull-in range. The carrier phase error output by the phase detector can be represented as... in, and Let represent the coherent integral values ​​of the in-phase and quadrature components of the instantaneous branch of the B1I signal, respectively. and These represent the coherent integral values ​​of the in-phase and quadrature components on the instantaneous branch of the pilot component B2a, respectively.

[0044] The code detector uses the lead-lag power method. When the code phase of the instantaneous tributary signal lags behind the actual satellite signal code phase, or when the signal energy of the leading tributary is greater than that of the lagging tributary, or when the code phase error value obtained by the code detector is greater than zero, the code phase of the instantaneous tributary is adjusted forward, and vice versa.

[0045] The preset coherent integration time includes: coherent integration time t1 and coherent integration time t2, where t1 = 20ms and t2 ranges from 120ms to 140ms.

[0046] Optionally, under weak signal conditions, the loop needs to perform coherent integration for hundreds of milliseconds to maintain stable tracking. Therefore, the coherent integration time can be further extended according to the setting of the loop tracking accuracy.

[0047] S3. Based on the Kalman filter algorithm, design loop filters for dual-frequency phase-locked loop, dual-frequency frequency-locked loop, and pilot component single-frequency frequency-locked loop; construct the observations and state variables of the Kalman filter based on the phase discrimination, frequency discrimination, and code discrimination results of the B2a and B1I signals; feed back the state variables output by the loop filter to the local carrier generator and composite spreading code generator for error correction; during the filtering process, switch the loop filter used according to the carrier-to-noise ratio, phase-lock detection index, and frequency-lock detection index of the two signals, adaptively select the appropriate loop filter to complete parameter estimation and then perform loop locking to achieve signal tracking and positioning.

[0048] Specifically, an adaptive loop switching strategy is employed to switch the loop filters of a dual-frequency phase-locked loop, a dual-frequency frequency-locked loop, and a pilot component single-frequency frequency-locked loop, aiming to achieve loop locking and satellite acquisition and tracking. The adaptive loop switching strategy is based on the carrier-to-noise ratio (CNR) of the two frequencies, the loop phase-locking index, and the frequency-locking index. At any given time, only one type of loop filter operates. The CNR is calculated using the wide-bandwidth / narrow-band power ratio method and extrapolated using the coherent integral value of the loop under different noise bandwidths.

[0049] Specifically, since the correlation peak of the B2a signal is narrower and the tracking accuracy is higher, the B2a signal is selected as the primary tracking signal.

[0050] For a dual-frequency phase-locked loop (PLL), when the carrier-to-noise ratio (CNR) of signal B2a is greater than the lower limit of the CNR threshold set by the PLL, signal B2a is selected as the main tracking signal. When the CNR of signal B2a is less than the lower limit of the CNR threshold set by the PLL and the CNR of signal B1I is greater than the lower limit of the CNR threshold set by the PLL, signal B1I is selected as the main tracking signal. The PLL selects the carrier phase error corresponding to the main tracking signal output by the phase detector, the code phase error corresponding to the main tracking signal output by the code detector, the frequency error and frequency change rate error corresponding to the main tracking signal output by the frequency discriminator, the inter-frequency carrier phase delay of signals B2a and B1I output by the phase detector, and the inter-frequency code phase delay of signals B2a and B1I output by the code discriminator to generate state variables through state transition equations. The carrier phase error corresponding to the main tracking signal output by the phase detector, the code phase error corresponding to the main tracking signal output by the code discriminator, and the frequency error corresponding to the main tracking signal output by the frequency discriminator are selected as observations.

[0051] For a single-frequency frequency-locked loop for the pilot component, the B2a signal is selected as the main tracking signal. The code phase error of the B2a signal output by the code discriminator, the frequency error of the B2a signal output by the frequency discriminator, and the frequency change rate error are used to generate state variables through the state transition equation. The code phase error of the B2a signal output by the code discriminator and the frequency error of the B2a signal output by the frequency discriminator are taken as observations.

[0052] For a dual-frequency locking loop, when the carrier-to-noise ratio (CNR) of the B2a signal is greater than the lower limit of the CNR threshold set by the dual-frequency locking loop, the B2a signal is selected as the main tracking signal. When the CNR of the B2a signal is less than the lower limit of the CNR threshold set by the dual-frequency locking loop and the CNR of the B1I signal is greater than the lower limit of the CNR threshold set by the dual-frequency locking loop, the B1I signal is selected as the main tracking signal. The dual-frequency locking loop selects the frequency error and frequency change rate error of the main tracking signal output by the frequency discriminator, the code phase error of the main tracking signal output by the code discriminator, and the inter-frequency code phase delay of the B2a and B1I signals output by the code discriminator to generate state variables through state transition equations. The frequency error of the main tracking signal output by the frequency discriminator and the code phase errors of the B2a and B1I signals output by the code discriminator are selected as observations.

[0053] The measurement noise covariance matrix of the loop filter is designed from the measurement noise of the carrier phase error, carrier frequency error, and spreading code phase error of the dual-frequency signal. The process noise covariance matrix consists of carrier phase noise, carrier frequency noise, dynamic noise, code phase noise, inter-frequency delay noise, and inter-code delay noise. These are related to factors such as crystal oscillator noise and carrier frequency, and can be set according to empirical values.

[0054] The state transition equation is established using an acceleration model. Taking carrier phase error as an example, its state transition equation can be expressed as: Where k is the Kalman filter timestamp. T represents the carrier phase error of the main tracking signal, Δf represents the frequency error of the main tracking signal, Δa represents the rate of change error of the frequency of the main tracking signal, and T represents the carrier phase error of the main tracking signal. coh This represents the coherent integration time. The establishment of the remaining error equations is similar to the principle of carrier phase error.

[0055] The code phase error can be expressed as:

[0056] In the formula, Δτ represents the code phase error, and R code The standard spreading code rate is represented by Δf, which represents the frequency error of the main tracking signal. carrier T represents the standard carrier frequency. coh This represents the time of coherent integration.

[0057] For a dual-frequency phase-locked loop, the state transition equation composed of the state variables is as follows:

[0058] Where, ΔI p and ΔI c These represent the carrier phase delay and spreading code phase delay between signals B1I and B2a, respectively. Δf represents the carrier phase error of the main tracking signal, Δa represents the frequency error of the main tracking signal, Δτ represents the code phase error, and T represents the code phase error. coh This represents the time of coherent integration.

[0059] For the observation equation of a dual-frequency phase-locked loop, the error observations output by the phase detector and the code detector are the average values ​​of the state variables over the coherent integration time.

[0060] The carrier phase observation equation is:

[0061]

[0062] in, This represents the carrier phase error observation at time k+1. Δf represents the carrier phase error of the main tracking signal at time k. k Δa represents the frequency error of the main tracking signal at time k. k T represents the error in the rate of change of the frequency of the main tracking signal. coh This represents the time of coherent integration.

[0063] The code phase observation equation is:

[0064]

[0065] in, This represents the observed code phase error at time k+1. R represents the carrier phase error of the main tracking signal at time k. code Δf represents the standard spreading code rate. k f represents the frequency error of the main tracking signal at time k. carrier Indicates the standard carrier frequency, Δa k T represents the error in the rate of change of the frequency of the main tracking signal. coh This represents the time of coherent integration.

[0066] The observation equations for the B1I and B2a signals are as follows:

[0067]

[0068] in, The observed value of the carrier phase error of the B1I signal. The observed value of the phase error of the B1I signal code; The carrier phase error observation value for signal B2a. The observed value of the phase error of the B2a signal code, f B1I and f B2a These are the carrier frequencies of the B1I and B2a signals, respectively; R B1I and R B2aThese represent the spreading code rates of the B1I and B2a signals, respectively. Δf represents the carrier phase error of the corresponding signal, Δa represents the frequency error of the corresponding signal, Δτ represents the code phase error of the corresponding signal, and ΔI represents the frequency change rate error of the corresponding signal. p and ΔI c These are represented as the carrier phase delay and spreading code phase delay between signals B1I and B2a, respectively.

[0069] In complex urban scenarios or when the receiver is in motion, the tracking loop may lose lock, thus requiring the design of a dual-frequency locked loop using Kalman equations. In the frequency-locked loop, carrier phase state estimation is unnecessary; therefore, the state variables degenerate into four: the frequency error of the main tracking signal, the frequency change rate error of the main tracking signal, the code phase error, and the spreading code phase delay. The observable changes are the frequency error output by the frequency discriminator and the code phase error output by the code discriminator at the two frequency points.

[0070] The state transition equation is:

[0071]

[0072] The observation equation is:

[0073]

[0074] The carrier frequency error of the B1I signal is observed. This represents the observed carrier frequency error value for the B2a signal.

[0075] Under weak signal conditions, tracking of the B1I signal may not be possible. However, since the pilot component of the B2a signal does not contain data components, the loop sensitivity can be improved through long-term coherent integration. Therefore, based on the two dual-frequency tracking loop filters, a single-frequency locked tracking loop for the pilot component of the B2a signal was further designed. In this case, the loop filter only addresses the tracking error of the pilot component of the B2a signal.

[0076] The state variables are estimated. The state transition equation for the frequency-locked loop of the pilot component B2a is:

[0077]

[0078] The observation equation for the frequency-locked loop of the B2a pilot component is:

[0079]

[0080] The meanings of the variables are the same as those described above.

[0081] The measurement noise covariance matrix in the Kalman equation is designed from the measurement noise of the carrier phase error, carrier frequency error, and spreading code phase error at two frequency points. The measurement noise of the carrier phase error is related to the carrier-to-noise ratio and the coherence integration time. The formula for the measurement noise of the carrier phase error is: Where C / N0 is the carrier-to-noise ratio.

[0082] The measurement noise of carrier frequency error can be derived from the measurement noise of carrier phase error: The measurement noise of code phase error is related to the signal carrier-to-noise ratio and code loop width: d is the width of the code ring.

[0083] The measurement noise covariance matrix of the Kalman filter used in the tracking loop can be expressed as follows:

[0084]

[0085]

[0086] Here, FLL represents dual-frequency frequency-locked loop, and PLL represents dual-frequency phase-locked loop.

[0087] The process noise covariance matrix Q of the Kalman filter C Carrier phase noise Carrier frequency noise q f Dynamic noise q a , code phase sound q c Inter-frequency delay noise and inter-symbol delay noise The components are related to factors such as crystal oscillator noise and carrier frequency, and can be designed based on empirical values. The noise covariance matrices for the continuous signal measurement process of the three tracking loops are as follows:

[0088]

[0089] The state variable covariance matrix P only needs to be set to a large value during initialization. As the algorithm iterates, this matrix will gradually converge to a stable value. The initialization methods for the three loop covariance matrices are as follows:

[0090] The parameters correspond to the initial errors of the three loop states.

[0091] Based on the Kalman filter observation equations, state transition equations, and corresponding measurement noise covariance matrices, process noise covariance matrices, and state variable covariance matrices for different types of tracking loops, the loop filtering implementation process is as follows:

[0092]

[0093]

[0094] Where A is the state transition equation, H is the observation matrix, and Z is the observation vector. First, the state variables at the current time are predicted using the state transition equation. Then, the predicted state variables are corrected using the observed values ​​and the observation equation to obtain the optimal estimate of the final loop error state.

[0095] The phase-locked detection index is the threshold range set by the dual-frequency phase-locked loop based on the coherent integral values ​​of the instantaneous signals of the I and Q branches to determine whether the phase tracking of the B1I and B2a signals is normal. The frequency-locked detection index is the threshold range set by the dual-frequency frequency-locked loop based on the coherent integral values ​​of the leading and lagging signals of the I and Q branches to determine whether the frequency tracking of the B1I and B2a signals is normal, and the pilot component single-frequency frequency-locked loop based on the coherent integral values ​​of the leading and lagging signals of the I and Q branches to determine whether the frequency tracking of the B2a signal is normal.

[0096] Specifically, this establishes the threshold range for determining whether phase tracking and frequency tracking are normal for the B1I and B2a signals. For example... Figure 2 As shown, normal phase tracking means that the phase of the signal tracked by the dual-frequency phase-locked loop is within the normal tracking threshold range of its phase; normal frequency tracking of the dual-frequency frequency-locked loop means that the frequency of the signal tracked by the dual-frequency frequency-locked loop is within the normal tracking threshold range of its frequency; normal frequency tracking of the pilot component single-frequency frequency-locked loop means that the frequency of the signal tracked by the pilot component single-frequency frequency-locked loop is within the normal tracking threshold range of its frequency.

[0097] The loop filter is switched based on the carrier-to-noise ratio, phase-lock detection index, and frequency-lock detection index of the two signals, including:

[0098] When the phase pull-in of any frequency point in the dual-frequency locked loop is normal, the dual-frequency phase-locked loop pull-in begins. When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1. The tracking process includes:

[0099] When the phase tracking of the dual-frequency phase-locked loop (PLL) is normal during the coherent integration time t1 and the carrier-to-noise ratio (CNR) is less than the lower limit of the CNR threshold set by the dual-frequency PLL, the dual-frequency PLL will track for a coherent integration time t2. When the phase tracking of the dual-frequency PLL during the coherent integration time t1 is abnormal and the CNR is greater than the lower limit of the CNR threshold set by the dual-frequency frequency-locked loop, the dual-frequency frequency-locked loop will track for a coherent integration time t1. When the phase tracking of the dual-frequency PLL during the coherent integration time t1 is abnormal or the CNR is less than the lower limit of the CNR threshold set by the dual-frequency frequency-locked loop, the pilot component frequency-locked loop will track for a coherent integration time t2.

[0100] When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, and also includes:

[0101] When a phase tracking anomaly occurs during tracking by the dual-frequency phase-locked loop at a coherent integration time of t2, the pilot component phase-locked loop is switched to track at a coherent integration time of t2. Otherwise, when the tracking phase is normal during tracking by the dual-frequency phase-locked loop at a coherent integration time of t2 and the signal carrier-to-noise ratio is greater than the upper limit of the carrier-to-noise ratio threshold set by the dual-frequency phase-locked loop, the dual-frequency phase-locked loop is switched to track at a coherent integration time of t1.

[0102] When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, and also includes:

[0103] When the frequency tracking of the pilot component frequency-locked loop is normal or the carrier-to-noise ratio is greater than the upper limit of the carrier-to-noise ratio threshold set by the dual-frequency frequency-locked loop, the dual-frequency frequency-locked loop will enter the tracking mode and ...

[0104] When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, and also includes:

[0105] When the frequency tracking of the dual-frequency phase-locked loop is normal and the carrier-to-noise ratio (CNR) is greater than the upper limit of the CNR threshold set by the dual-frequency phase-locked loop, the dual-frequency phase-locked loop tracks for t1 coherent integration time. When the frequency tracking of the dual-frequency phase-locked loop is abnormal or the CNR is less than the lower limit of the CNR threshold set by the dual-frequency phase-locked loop, the pilot component phase-locked loop tracks for t2 coherent integration time.

[0106] Specifically, the coherent integration time of the loop is related to the signal structure, tracking accuracy, and computational load. Considering these three factors, the optimal integration time of 20ms and 120ms is adopted in this embodiment. Under the condition that the computational conditions permit, the integration time of 120ms can be appropriately extended to 130ms or 140ms.

[0107] Specifically, this embodiment of the invention starts with a dual-frequency locked loop (PLL) because the designed FFT discriminator has higher accuracy, thus starting with frequency locking and gradually transitioning to phase locking. The ultimate goal is to obtain both frequency and phase. First, a PLL can accurately track the carrier phase of a signal, making it suitable for precision positioning systems based on carrier phase observations. Second, compared to the differential bit demodulation method of a PLL, a PLL directly demodulates bits based on I-path observations, resulting in a lower bit error rate. However, a PLL is much more vulnerable than a PLL. When multipath signals or ionospheric scintillation cause rapid changes in carrier phase observations, it can easily lead to PLL malfunction or even loss of lock. Furthermore, under weak signal conditions, the variance of the carrier phase observations increases, and because the pull-in range of the phase detector is relatively smaller, it will lose lock faster than a PLL. Therefore, this embodiment employs a tracking method that switches between PLL and PLL modes. When an increase in carrier phase variance is detected (determining an abnormal phase tracking), the receiver switches from a PLL to a PLL, and then switches back to PLL mode when the signal is good.

[0108] When the frequency and phase input of the dual-frequency locked loop (DFLL) are normal, tracking begins with a 20ms coherent integration time DFLL. If phase tracking is normal and the carrier-to-noise ratio (CNR) is less than the lower bound of the CNR threshold set by the DFLL, it indicates the signal is slightly weak. The DFLL is then used to track the signal at a slightly longer coherent integration time t2. If phase tracking is normal and the CNR is greater than the upper bound of the CNR threshold set by the DFLL, the signal strength is sufficient, and an excessively long integration time is unnecessary. Therefore, the DFLL is switched back to a 20ms coherent integration time for signal tracking. If phase tracking during the 20ms coherent integration time is abnormal and the CNR is less than the lower bound of the CNR threshold set by the DFLL, the pilot component DFLL is used to track the signal at a coherent integration time t2. This is because the pilot component of the B2a signal does not contain data components, and long-term coherent integration can improve loop sensitivity. Similarly, if the frequency tracking of the signal under a 20ms coherent integration time using the dual-frequency locking loop fails or the carrier-to-noise ratio (CNR) is lower than the lower limit of the CNR threshold set by the dual-frequency locking loop, the pilot component frequency locking loop will be used to track the signal under a 2ms coherent integration time. When the pilot component frequency locking loop finds that the frequency tracking is normal, or the CNR is higher than the upper limit of the CNR threshold set by the dual-frequency locking loop, it indicates that the signal has recovered somewhat, and the system will switch back to the dual-frequency locking loop to track the signal under a 20ms coherent integration time.

[0109] Compared with existing technologies, the signal tracking loop filter, carrier loop discriminator, and code detector provided in this embodiment, and the tracking loop parameters designed based on carrier-to-noise ratio estimation, switch the tracking mode in real time according to the monitoring of the loop state, which has higher sensitivity and phase and frequency locking accuracy, realizing loop locking and satellite acquisition and tracking. This embodiment also provides a weak signal tracking loop switching method for satellite navigation signals. Utilizing the characteristics of existing novel dual-frequency signals, it proposes an asymmetric dual-frequency tracking algorithm, designs a highly sensitive adaptive loop, and realizes the tracking of dual-frequency navigation satellite signals with different modulation methods, initial phases, and signal powers, as well as acquisition and tracking under weak signal conditions, thereby improving navigation and positioning accuracy. In this embodiment, the receiver is designed with a pilot signal frequency-locked loop for tracking under weak signal conditions. The pilot signal of the satellite signal refers to a signal that does not contain data bits or only contains fixed second-order codes, and is not affected by data bit mutations during long-term coherent integration, thus improving loop sensitivity. This embodiment designs three loop filter Kalman equations: dual-frequency phase-locked loop, dual-frequency frequency-locked loop, and pilot component single-frequency frequency-locked loop. It also proposes an adaptive loop switching strategy based on satellite carrier-to-noise ratio, phase-locking index, and frequency-locking index to achieve a loop tracking algorithm with higher sensitivity and phase and frequency locking accuracy compared to single-frequency signal processing of B1I signals.

[0110] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0111] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for switching tracking loops for weak satellite navigation signals, characterized in that, The B1I and B2a signals are acquired. After processing the B1I and B2a signals, the leading, instantaneous, and lagging intermediate frequency (IF) signals of the I branch and the leading, instantaneous, and lagging IF signals of the Q branch corresponding to each signal are obtained. The leading, instantaneous, and lagging IF signals of the I and Q branches of each signal are respectively subjected to code correlation operation with the composite spreading code to obtain the data after multi-channel correlation operation. The data after multi-channel correlation operation are respectively subjected to coherent integration. When the preset coherent integration time is reached, the coherent integration result corresponding to each signal is input to the phase detector, frequency detector, and code detector for phase detection, frequency detection, and code detection operations. Loop filters based on the Kalman filter algorithm are designed for dual-frequency phase-locked loop, dual-frequency frequency-locked loop, and pilot component single-frequency frequency-locked loop. Based on the phase discrimination, frequency discrimination, and code discrimination results of the B2a and B1I signals, the observations and state variables of the Kalman filter are constructed; the state variables output by the loop filter are fed back to the local carrier generator and the composite spreading code generator for error correction. During the filtering process, the loop filter used is switched according to the carrier-to-noise ratio, phase-lock detection index, and frequency-lock detection index of the two signals. The appropriate loop filter is adaptively selected to complete parameter estimation and then loop locking, thereby realizing signal tracking and positioning.

2. The tracking loop switching method according to claim 1, characterized in that, The preset coherent integration time includes: coherent integration time t1 and coherent integration time t2, where t1 = 20ms and t2 ranges from 120ms to 140ms.

3. The tracking loop switching method according to claim 1, characterized in that, The phase-locked detection index is the threshold range set by the dual-frequency phase-locked loop based on the coherent integral values ​​of the instantaneous signals of the I and Q branches to determine whether the phase tracking of the B1I and B2a signals is normal. The frequency-locked detection index is the threshold range set by the dual-frequency frequency-locked loop based on the coherent integral values ​​of the leading and lagging signals of the I and Q branches to determine whether the frequency tracking of the B1I and B2a signals is normal, and the pilot component single-frequency frequency-locked loop based on the coherent integral values ​​of the leading and lagging signals of the I and Q branches to determine whether the frequency tracking of the B2a signal is normal.

4. The tracking loop switching method according to claim 3, characterized in that, The loop filter is switched based on the carrier-to-noise ratio, phase-lock detection index, and frequency-lock detection index of the two signals, including: When the phase pull-in of any frequency point in the dual-frequency locked loop is normal, the dual-frequency phase-locked loop pull-in begins. When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1. The tracking process includes: When the phase tracking of the dual-frequency phase-locked loop (PLL) is normal during the coherent integration time t1 and the carrier-to-noise ratio (CNR) is less than the lower limit of the CNR threshold set by the dual-frequency PLL, the dual-frequency PLL will track for a coherent integration time t2. When the phase tracking of the dual-frequency PLL during the coherent integration time t1 is abnormal and the CNR is greater than the lower limit of the CNR threshold set by the dual-frequency frequency-locked loop, the dual-frequency frequency-locked loop will track for a coherent integration time t1. When the phase tracking of the dual-frequency PLL during the coherent integration time t1 is abnormal or the CNR is less than the lower limit of the CNR threshold set by the dual-frequency frequency-locked loop, the pilot component frequency-locked loop will track for a coherent integration time t2.

5. The tracking loop switching method according to claim 4, characterized in that, When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, and also includes: When a phase tracking anomaly occurs during tracking by the dual-frequency phase-locked loop at a coherent integration time of t2, the pilot component phase-locked loop is switched to track at a coherent integration time of t2. Otherwise, when the tracking phase is normal during tracking by the dual-frequency phase-locked loop at a coherent integration time of t2 and the signal carrier-to-noise ratio is greater than the upper limit of the carrier-to-noise ratio threshold set by the dual-frequency phase-locked loop, the dual-frequency phase-locked loop is switched to track at a coherent integration time of t1.

6. The tracking loop switching method according to claim 5, characterized in that, When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, and also includes: When the frequency tracking of the pilot component frequency-locked loop is normal or the carrier-to-noise ratio is greater than the upper limit of the carrier-to-noise ratio threshold set by the dual-frequency frequency-locked loop, the dual-frequency frequency-locked loop will enter the tracking mode and ...

7. The tracking loop switching method according to claim 6, characterized in that, When the phase pull-in of the dual-frequency phase-locked loop is normal, the dual-frequency phase-locked loop starts tracking at a coherent integration time of t1, and also includes: When the frequency tracking of the dual-frequency phase-locked loop is normal and the carrier-to-noise ratio (CNR) is greater than the upper limit of the CNR threshold set by the dual-frequency phase-locked loop, the dual-frequency phase-locked loop tracks for t1 coherent integration time. When the frequency tracking of the dual-frequency phase-locked loop is abnormal or the CNR is less than the lower limit of the CNR threshold set by the dual-frequency phase-locked loop, the pilot component phase-locked loop tracks for t2 coherent integration time.

8. The tracking loop switching method according to claim 1, characterized in that, For a dual-frequency phase-locked loop (PLL), when the carrier-to-noise ratio (CNR) of signal B2a is greater than the lower limit of the CNR threshold set by the PLL, signal B2a is selected as the main tracking signal. When the CNR of signal B2a is less than the lower limit of the CNR threshold set by the PLL and the CNR of signal B1I is greater than the lower limit of the CNR threshold set by the PLL, signal B1I is selected as the main tracking signal. The PLL selects the carrier phase error corresponding to the main tracking signal output by the phase detector, the code phase error corresponding to the main tracking signal output by the code detector, the frequency error and frequency change rate error corresponding to the main tracking signal output by the frequency discriminator, the inter-frequency carrier phase delay of signals B2a and B1I output by the phase detector, and the inter-frequency code phase delay of signals B2a and B1I output by the code discriminator as state variables. The carrier phase error corresponding to the main tracking signal output by the phase detector, the code phase error corresponding to the main tracking signal output by the code discriminator, and the frequency error corresponding to the main tracking signal output by the frequency discriminator are selected as observations.

9. The tracking loop switching method according to claim 1, characterized in that, For the pilot component single-frequency frequency-locked loop, the B2a signal is selected as the main tracking signal. The code phase error of the B2a signal output by the code discriminator, the frequency error of the B2a signal output by the frequency discriminator, and the frequency change rate error are taken as state variables. The code phase error of the B2a signal output by the code discriminator and the frequency error of the B2a signal output by the frequency discriminator are taken as observations.

10. The tracking loop switching method according to claim 1, characterized in that, For a dual-frequency locking loop, when the carrier-to-noise ratio (CNR) of the B2a signal is greater than the lower limit of the CNR threshold set by the dual-frequency locking loop, the B2a signal is selected as the main tracking signal. When the CNR of the B2a signal is less than the lower limit of the CNR threshold set by the dual-frequency locking loop and the CNR of the B1I signal is greater than the lower limit of the CNR threshold set by the dual-frequency locking loop, the B1I signal is selected as the main tracking signal. The dual-frequency locking loop selects the frequency error and frequency change rate error of the main tracking signal output by the frequency discriminator, the code phase error of the main tracking signal output by the code discriminator, and the inter-frequency code phase delay of the B2a and B1I signals output by the code discriminator as state variables. The frequency error of the main tracking signal output by the frequency discriminator and the code phase errors of the B2a and B1I signals output by the code discriminator are selected as observations.

Images