A multi-satellite formation time synchronization method based on micro-satellite radio frequency measurement

Abstract

This invention discloses a multi-satellite formation time synchronization method based on microsatellite radio frequency measurement, comprising: (1) obtaining time difference measurement results using a multi-satellite radio frequency measurement method based on time division multiplexing and ADS-TWR; (2) each satellite calculates the on-board time of the satellite in the launch state based on its own time difference with the launch state satellite, and achieves phase correction by direct setting; (3) by calculating the difference between the two time difference values, each receiving state satellite obtains the frequency difference between itself and the launch state satellite, and modifies the frequency control word of the NCO to achieve frequency correction; (4) after the time synchronization of the current time slot is completed, it waits to enter the next time slot and performs time synchronization for the next launch state satellite; (5) the time synchronization of each state satellite is completed in sequence to achieve inter-satellite time synchronization of the formation. This invention effectively solves the time synchronization problem under the time division multiplexing system, has strong scalability, and provides a unified high-precision time reference for relative navigation of multi-satellite formations of microsatellites.

Description

Technical Field

[0001] This invention belongs to the field of satellite formation time synchronization, and in particular relates to a multi-satellite formation time synchronization method based on microsatellite radio frequency measurement. Background Technology

[0002] As microsatellite formations gradually become a hot topic in the aerospace field, high-precision inter-satellite measurements, as a key supporting technology for relative navigation in microsatellite formations, have received increasing attention. Among these, inter-satellite time synchronization can be established based on inter-satellite time difference measurements, thereby providing a unified time reference for navigation calculations using the measured inter-satellite distances.

[0003] Due to the severe shortage of spectrum resources in the space environment, and the limitations of power consumption and size of microsatellites, frequency division multiple access (FDMA) and code division multiple access (CDMA) measurement systems, which have poor scalability, are unsuitable for multi-satellite formation. Time division multiple access (TDMA) systems can effectively solve the scalability problem; typical examples are the inter-satellite link design of the Global Positioning System (GPS) IIR / IIF and the BeiDou navigation constellation.

[0004] Time synchronization methods need to consider two aspects: the first is the time difference compensation method used, i.e., how to achieve local time adjustment of nodes; the second is the time synchronization architecture of satellite formation network based on the time difference compensation method.

[0005] There are currently several methods for clock regulation, mainly including voltage-controlled oscillators (VCOs), numerically controlled oscillators (NCOs), and digitally controlled oscillators (DCOs).

[0006] The principle of a VCO is to compensate for time differences between nodes by adjusting the input frequency of the VCO based on the input voltage value. The clock output of a VCO oscillator has a good duty cycle, but the stability of analog VCO devices is poor and easily affected by the environment.

[0007] The DCO module consists of two parts: a cyclic-controlled stage (CCS) for generating the clock, and a coarse-tuning delay stage (CTDS) and a fine-tuning delay stage (FTDS) for adjusting the phase. CTDS is responsible for coarse phase adjustment, while FTDS is responsible for fine phase adjustment. The advantage of DCO phase adjustment is its high precision, but its disadvantage is that it requires extremely high accuracy in line delay time, thus it is mostly used in application-specific integrated circuits (ASICs). The NCO consists of an adder and a register, generating the clock signal through a ROM-based lookup table. NCO time difference compensation can be achieved by directly adjusting the clock frequency control word and phase control word, which is easily implemented on an FPGA. However, NCO time difference compensation also has a significant problem: due to the periodic phase overflow of the NCO in the time domain, the clock generated by the NCO has large frequency spurious signals, severely degrading the frequency domain characteristics of the compensated clock. There are two main existing sensor network time synchronization architectures: hierarchical tree-based clock synchronization network architecture and distributed clock network architecture.

[0008] In a hierarchical tree clock network, a node is assigned as the root node, and the remaining nodes are assigned to different levels based on factors such as the communication range of the sensor network. Clock synchronization in the hierarchical tree is performed layer by layer; that is, nodes in the first level synchronize with the root node first, then nodes in the second level synchronize with the first level, and so on. It's easy to see that in this architecture, the root node's clock serves as the synchronization reference for the entire network; therefore, the accuracy of the root node's clock determines the overall clock synchronization accuracy of the network. At the same time, the error in the hierarchical tree synchronization method accumulates layer by layer. The higher the level of the node, the worse the clock synchronization accuracy, and if any node fails to synchronize, all its subsequent child nodes will also fail to synchronize. The hierarchical tree clock synchronization network architecture has poor scalability. As the number of satellites in a multi-satellite constellation increases, the number of levels in the hierarchical tree structure also increases, which will introduce larger time synchronization errors for nodes with a large number of levels.

[0009] Distributed clock synchronization networks can effectively overcome the above-mentioned shortcomings. The Consensus clock synchronization network is a typical example of a distributed clock synchronization network. Instead of using any single node's clock as a reference clock, the Consensus network constructs a virtual synchronization clock through communication and measurement between nodes, to which all nodes synchronize. The synchronization clock in this network architecture is determined by the local clocks of all nodes within the formation. A significant problem with the Consensus network architecture is that if the timing of phase corrections performed by each node to the virtual clock is inconsistent, it introduces a large synchronization error. For time-division multiplexing satellite formations that synchronize their own time, it is difficult to meet these conditions without ground station assistance. Furthermore, the clock failure of any node in the Consensus network, such as continuous clock jumps, will prevent all nodes from synchronizing with the virtual clock. The Consensus clock synchronization network also suffers from poor scalability; increasing the number of satellites in a multi-satellite formation increases the probability of satellite failure, leading to clock synchronization failures within the Consensus network. Summary of the Invention

[0010] This invention provides a multi-satellite formation time synchronization method based on microsatellite radio frequency measurement. After synchronization, the clock error is reduced to about 1 nanosecond. It has the advantages of strong scalability, short synchronization period, no accumulation of synchronization error, accuracy not limited by a single node, minimal impact of clock failure of any satellite in the formation on formation time synchronization, and independent time synchronization of each node.

[0011] A multi-satellite formation time synchronization method based on microsatellite radio frequency measurements, characterized by the following steps:

[0012] (1) The time difference measurement results between the satellite in the launch state and the satellite in the receiving state are obtained by adopting a multi-satellite radio frequency measurement method based on time division system and asymmetric bilateral two-way method;

[0013] (2) Each receiving satellite calculates the on-board time of the launching satellite based on the time difference between itself and the launching satellite, and performs phase correction by directly setting the data.

[0014] (3) By calculating the difference between the time difference values ​​before and after, each receiving satellite obtains the frequency difference between itself and the satellite in the transmitting state, and modifies the frequency control word of the NCO to achieve frequency correction.

[0015] (4) After the current time slot is synchronized, wait to enter the next time slot and synchronize the time of the next satellite in the launch status.

[0016] (5) The time synchronization of each status satellite is completed in sequence, and the inter-satellite time synchronization of the formation is finally achieved.

[0017] The specific process of step (1) is as follows:

[0018] Assume there are K satellites in the formation. Each satellite in the formation and each time slot in a single measurement cycle is assigned a number, S1 to SK. Each satellite occupies a time slot with the same number as itself and only transmits signals in its own time slot. It only receives signals in other time slots, thus completing the transmission and reception of signals in sequence. Three signal transmissions between any two satellites in the formation complete one asymmetric bilateral two-way measurement.

[0019] The formula for measuring the time difference between two satellites is:

[0020] ΔT=T A (t4)-R AT / cT B (t3)

[0021] =[T A (t3)-T B (t3)]+[T A (t4)-T A (t3)-R AT / c]

[0022] Among them, T A (t3), T B (t3) represents the signal transmission times of satellites A and B during the second measurement in the asymmetric bilateral two-way measurement process, respectively. A (t4) represents the signal reception time of satellite A, R AT Here, c represents the interstellar distance measurement, and c represents the speed of light.

[0023] In the formula, T A (t3)-T B (t3) represents the time difference between the two satellites, while T A (t4)-T A (t3)-R AT / c represents the error term for time difference measurement.

[0024] In step (2), the clock of the receiving satellite is kept in sync with the clock of the transmitting satellite by direct setting of the data. Different transmitting satellites are switched in different time slots, while the other receiving satellites undergo a phase correction.

[0025] In step (3), the formula for calculating the frequency difference between the two satellites is:

[0026]

[0027] In the formula, Δf(i) is the frequency difference at time i, ΔT(i) is the time difference measurement result at time i, and T updatef is the duration of the time synchronization period. local To receive the satellite's local clock frequency;

[0028] The frequency difference between corresponding nodes is calculated by measuring the time difference between two consecutive measurements, and then fed back to the NCO's frequency control word to achieve frequency correction.

[0029] In step (3), the synchronization error is calculated after frequency correction. The expression for the synchronization error is:

[0030]

[0031] In the formula, T error K represents the maximum synchronization error. A and K B These represent the ratios of the onboard clock source frequency to the nominal frequency of satellite A and satellite B, respectively, where f0 is the nominal frequency.

[0032] In step (3), the Dither algorithm and low-pass filter LPF are used to optimize the on-board clock generated by the NCO;

[0033] The Dither algorithm is implemented using a 36-bit pseudocode generator, whose expression is as follows:

[0034] F(x) = 1 + x 11 +x 36

[0035] The pseudocode generated by the pseudocode generator has pseudo-randomness. The generated pseudocode sequences are then integrated into a sequence ranging from (0 to 2^3). 36 The random number is generated by -1), and then the generated random number is normalized for whitening of NCO spurious signals. After whitening the NCO output by the Dither algorithm, the generated noise is suppressed by a low-pass filter (LPF).

[0036] The low-pass filter (LPF) is a finite-length unit impulse response filter with length M, taking x(n) as input and y(n) as output, and is expressed by the following difference equation.

[0037]

[0038] Among them, b k This represents the set of filter coefficients, using a 5th-order filter, as shown in the following equation.

[0039] H(z) = -0.0232 + 0.2017z -1 +0.4029z -2 +0.4029z -3 +0.2017z -4 +-0.0232z-5 .

[0040] The specific process of step (5) is as follows:

[0041] Assume there are K satellites in the formation, numbered S1, S2...SK, and the K satellites enter the launch state sequentially;

[0042] When S1 enters the launch state, the remaining K-1 satellites in the formation measure their time difference with S1, and thus synchronize with S1 as the master node; when the time slot of satellite S1 ends, satellite S2 enters the launch state, and S2 becomes the new master node. The remaining K-1 satellites synchronize with S2, and so on, to achieve inter-satellite time synchronization of the formation.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. The method of the present invention achieves time synchronization by phase correction and frequency correction after obtaining the time difference measurement results. At the same time, it effectively solves the time synchronization problem under the time division multiplexing system, has the advantage of strong scalability, is suitable for large-scale formations, and provides a unified high-precision time reference for relative navigation of multiple microsatellite formations.

[0045] 2. This invention addresses the issue of significant frequency spurious emissions in the clock generated by the NCO by employing the Dither algorithm and LPF to optimize the on-board clock generated by the NCO. This compensation method is implemented on the same hardware platform as the measurement system, greatly simplifying system design. Attached Figure Description

[0046] Figure 1 This is a flowchart of a multi-satellite formation time synchronization method based on microsatellite radio frequency measurement according to the present invention;

[0047] Figure 2 This is a schematic diagram of the time-division multiplexing system and ADS-TWR multi-satellite measurement scheme in an embodiment of the present invention;

[0048] Figure 3 This is a diagram of the NCO optimization scheme in an embodiment of the present invention;

[0049] Figure 4 This is a schematic diagram of the time synchronization network architecture in an embodiment of the present invention;

[0050] Figure 5 This is a block diagram of the overall design for time synchronization between two satellites in an embodiment of the present invention;

[0051] Figure 6 This is a physical image of the experimental platform of this invention;

[0052] Figure 7 This is a diagram illustrating the NCO optimization effect in an embodiment of the present invention;

[0053] Figure 8 This is a diagram showing the synchronization results before and after frequency correction in an embodiment of the present invention;

[0054] Figure 9 This is a diagram showing the synchronization results of multi-node experiments in an embodiment of the present invention. Detailed Implementation

[0055] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be noted that the embodiments described below are intended to facilitate the understanding of the present invention and do not constitute any limitation thereof.

[0056] like Figure 1 As shown, a multi-satellite formation time synchronization method based on microsatellite radio frequency measurements includes the following steps:

[0057] S101, based on the time division system and ADS-TWR multi-satellite radio frequency measurement scheme, obtains the time difference measurement results between the transmitting satellite and the receiving satellite.

[0058] like Figure 2 As shown, assuming there are K satellites in the formation, each satellite and each time slot in a single measurement cycle is assigned a number from 1 to K. Each satellite occupies a time slot with the same number as itself and only transmits signals in its assigned time slot, receiving signals in other time slots, thus completing the transmission and reception of signals sequentially.

[0059] The formula for measuring the time difference between two satellites is:

[0060] ΔT=T A (t4)-R AT / cT B (t3)

[0061] =[T A (t3)-T B (t3)]+[T A (t4)-T A (t3)-R AT / c]

[0062] Where T A (t3), T B (t3) represents the signal transmission times of satellites A and B during the second measurement in the asymmetric bilateral two-way measurement process, respectively. A (t4) represents the signal reception time of satellite A, R AT Here, c represents the interstellar distance measurement, and c represents the speed of light.

[0063] In the formula T A (t3)-T B (t3) represents the time difference between the two satellites, while TA (t4)-T A (t3)-R AT / c represents the error term for time difference measurement.

[0064] S102, all satellites in the receiving state obtain the time difference between themselves and the satellites in the launching state, and then make their own clocks consistent with the launching state clocks by directly setting the data. Different launching state satellites are switched in each time slot, while the other receiving state satellites perform a phase correction.

[0065] S103, by calculating the difference between the time difference values ​​before and after, each receiving satellite obtains the frequency difference between itself and the satellite in the transmitting state, and modifies the NCO's frequency control word to achieve frequency correction.

[0066] This invention employs FPGA to implement inter-satellite joint measurements; therefore, using the NCO method for time difference compensation best suits the hardware and software system design of this invention, and the NCO method can overcome the shortcomings of VCO and DCO. To address the issue of significant frequency spurious signals in the clock generated by the NCO, a spurious power whitening algorithm (Dither algorithm) and a low-pass filter (LPF) are used to optimize the on-board clock generated by the NCO, such as... Figure 3 As shown, it exhibits good stability.

[0067] The frequency correction parameter is calculated as follows:

[0068]

[0069] In the formula, Δf(i) is the frequency difference at time i, ΔT(i) is the time difference measurement result at time i, and T update f is the duration of the time synchronization period. local In order to receive the satellite's local clock frequency, the frequency difference between the corresponding nodes can be calculated from the results of two consecutive time difference measurements, and then fed back to the NCO's frequency control word.

[0070] Error sources for time synchronization include frequency source error, time difference measurement error, and NCO frequency control word accuracy error.

[0071] The expression for the frequency-corrected synchronization error is:

[0072]

[0073] In the formula, T error K represents the maximum synchronization error. A and K B These represent the ratios of the onboard clock source frequency to the nominal frequency of satellite A and satellite B, respectively, where f0 is the nominal frequency.

[0074] To address the issue of excessive spectral spurious activity in the NCO, and the fact that its distribution is periodically dependent, a spurious power whitening algorithm is employed. This algorithm whitens the spurious activity by introducing a random dither at the NCO output, thus distributing the spurious power evenly across the entire spectrum. The Dither algorithm is implemented using a pseudocode generator; the pseudocode generator used in this invention is 36-bit. The expression for the pseudocode generator of the Dither algorithm selected in this invention is shown below.

[0075] F(x) = 1 + x 11 +x 36

[0076] Because the pseudocode generated by the pseudocode generator has pseudo-randomness, the generated pseudocode sequences can be integrated into a range of (0~2). 36 The generated random number is then normalized and can be used for whitening stray numbers.

[0077] Whitening the NCO output using the Dither algorithm results in a significantly louder output clock. Therefore, a low-pass filter can be used to suppress this noise. Since ideal filters are non-causal and physically impossible to implement, this invention employs a Finite Impulse Response (FIR) filter to design the required low-pass filter. An FIR filter of length M with x(n) as input and y(n) as output can be represented by the following difference equation.

[0078]

[0079] Among them, b k This represents the set of filter coefficients. In the frequency domain, its system function can be expressed as H(z) = h(0) + h(1)z -1 +h(2)z -2 +...+h(M-2)z- (M-2) +h(M-1)z- (M-1)

[0080] The filter coefficients used in this invention are shown in the following formula.

[0081] H(z) = -0.0232 + 0.2017z -1 +0.4029z -2 +0.4029z -3 +0.2017z -4 +-0.0232z -5

[0082] This invention employs a 5th-order filter design. Higher filter orders generally result in better performance, but also increase resource consumption and design complexity. The filter designed in this paper is sufficient to effectively filter out high-frequency components and noise.

[0083] Adjusting the frequency and phase words of the NCO can effectively regulate the satellite's onboard clock, thereby compensating for time differences. The Dither algorithm and LPF can further address the issues of excessive spectral spurious activity and periodic phase overflow in the NCO, thus improving clock quality.

[0084] S104: After the current time slot time synchronization is completed, wait to enter the next time slot to synchronize the time of the next satellite in launch status.

[0085] S105, based on the time synchronization scheme, finally implements and verifies a multi-satellite time synchronization method.

[0086] According to the above formula, assuming a time slot of 5 seconds, a frequency source stability of 0.01 ppb, a frequency source accuracy of 0.01 ppm, and a time synchronization accuracy of about 1 nanosecond, the time synchronization accuracy can be achieved.

[0087] Considering the advantages and disadvantages of hierarchical tree network architecture and Consensus network architecture, and the multi-satellite measurement scheme upon which this invention is based, the clock synchronization method of this invention is as follows: Figure 4 As shown, the measurement process can be described as follows: Considering that in a multi-satellite measurement scheme, each satellite only launches within its own time slot, and is in a receiving state during the remaining time slots, assuming there are five satellites in the formation, numbered S1, S2, S3, S4, and S5, the five satellites sequentially enter the launch state. When S1 enters the launch state, the remaining four satellites in the formation measure their own time difference with S1 using a joint measurement method, and thus synchronize with S1 as the master node. When satellite S1's time slot ends, satellite S2 enters the launch state, and S2 becomes the new master node. The remaining four satellites synchronize their time with S2, and so on, achieving inter-satellite time synchronization of the formation.

[0088] The following describes the hardware and software implementation design. The time synchronization module mainly includes four functional modules: control, time difference compensation, phase correction, and frequency correction. The overall design block diagram for synchronizing two satellites within the formation is shown below. Figure 5 As shown.

[0089] Joint measurements between two satellites within a formation can determine the time difference between the two satellite transmitters, or in other words, the phase difference between the two transmitters. Furthermore, since both the onboard timing module and the transmitter module are implemented within a single FPGA, we can easily obtain the phase difference between them. and A satellite in the process of launching transmits the phase difference between its onboard time and the transmitter time to the receiver on the other side. The receiver then uses this phase difference to determine the time. and The time difference between the time modules on the two stars can be calculated.

[0090] The control module controls the satellites in the formation to perform phase and frequency corrections in each receiving time slot, maintaining a maximum time synchronization error of approximately 1 nanosecond. The phase correction module, after calculating the on-board time difference between the two satellites using the ADS-TWR module, transmits a preset signal to the on-board clock module to complete the phase correction. The frequency correction module calculates the frequency parameters and adjusts the NCO frequency control word based on the time difference and time interval between two consecutive measurements.

[0091] The derivation of this invention will be verified in detail below with reference to actual measurements and accompanying drawings, but this invention is not limited to the implementation scenarios shown below.

[0092] Experimental scenarios such as Figure 6 As shown, the joint measurement devices are interconnected via RF cables and attenuators. Before the experiment, the system's hardware delay and the RF cable transmission speed need to be calibrated, with calibration results of 3.294 μs and 20,660,000 m / s, respectively.

[0093] First, the effect of NCO clock optimization was tested, and the test results are as follows: Figure 7 As shown, the optimization of the NCO effectively solves the problems of NCO spectral spurious emissions and periodic overflow, thereby effectively improving the clock quality output by the onboard clock module.

[0094] The time slot is set to 5 seconds, and the carrier-to-noise ratio is 70 dB / Hz. Figure 8 Figure (a) shows a comparison of time synchronization accuracy before and after frequency correction. It can be seen that without frequency correction, the error between the two phase corrections can reach the level of hundreds of nanoseconds, which is much greater than the synchronization result after frequency correction. Figure 8 (b) shows the time synchronization results after frequency correction separately. It can be seen that after frequency correction, the maximum error between the two phase corrections is only about 1 nanosecond, which is consistent with the theoretical analysis conclusions above.

[0095] In the multi-node experiment, the time slot was set to 5 seconds, and the carrier-to-noise ratio was 70 dBHz. Since each satellite synchronized independently, the results of multi-node time synchronization were similar to those of two-node time synchronization. To more intuitively demonstrate the formation time synchronization error when satellites with different serial numbers act as master nodes, a magnified view of the multi-node time synchronization within the time slots when satellites with different serial numbers were in transmitter mode was shown. The results are as follows: Figure 9As shown, in a multi-node scenario, when satellites in the formation successively become master nodes, the system can still function well and maintain extremely high synchronization accuracy.

[0096] This invention proposes a multi-satellite formation time synchronization method. Based on a time-division multiplexing (TDM) system and an ADS-TWR method, a multi-satellite radio frequency measurement scheme is used to obtain time difference measurement results for multi-satellite formation time synchronization, which is verified through experimental testing. Simultaneously, to address the issue of significant frequency spurious emissions from the NCO-generated clock, the Dither algorithm and LPF are employed to optimize the onboard clock generated by the NCO. Experimental results show that the time synchronization accuracy error of this invention is maintained within 1 nanosecond, verifying the correctness of the scheme. This invention effectively solves the time synchronization problem under a TDM system, has the advantage of strong scalability, is suitable for large-scale formations, and provides a unified high-precision time reference for relative navigation of multi-satellite formations of microsatellites.

[0097] The embodiments described above provide a detailed explanation of the technical solutions and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, additions, and equivalent substitutions made within the scope of the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for time synchronization of multi-satellite formation based on micro-satellite radio frequency measurement, characterized in that, Includes the following steps: (1) The time difference measurement results between the satellite in the launch state and the satellite in the receiving state are obtained by adopting a multi-satellite radio frequency measurement method based on time division system and asymmetric bilateral two-way method; (2) Each receiving satellite calculates the onboard time of the launching satellite based on the time difference between itself and the launching satellite, and performs phase correction by directly setting the data. (3) By calculating the difference between the two time differences, each receiving satellite obtains the frequency difference between itself and the transmitting satellite, and modifies the frequency control word of the NCO to achieve frequency correction; the Dither algorithm and low-pass filter LPF are used to optimize the on-board clock generated by the NCO; Dither's algorithm is implemented using a pseudocode generator with 36 bits, and its expression is as follows: The pseudocode generated by the pseudocode generator has pseudo-randomness. The generated pseudocode sequences are then integrated into a sequence ranging from (0 to...). The random numbers generated are then normalized for whitening NCO spurious signals. After whitening the NCO output using the Dither algorithm, the generated noise is suppressed using a low-pass filter (LPF). The low-pass filter LPF uses For input, The length of the output is M A finite-length unit impulse response filter is expressed by the following difference equation. in, This represents the set of filter coefficients, using a 5th-order filter, as shown in the following equation. (4) After the current time slot is synchronized, wait to enter the next time slot and synchronize the time of the next satellite in the launch state; (5) Time synchronization of each status satellite is completed in sequence, and finally the inter-satellite time synchronization of the formation is achieved.

2. The multi-satellite formation time synchronization method based on micro-satellite radio frequency measurement according to claim 1, characterized in that, The specific process of step (1) is as follows: Assume there are K satellites in the formation. Each satellite in the formation and each time slot in a single measurement cycle is assigned a number, S1~SK. Each satellite occupies a time slot with the same number as itself and only transmits signals in its own time slot. It only receives signals in other time slots, thus completing the transmission and reception of signals in sequence. Three signal transmissions between any two satellites in the formation complete one asymmetric bilateral two-way measurement. The formula for measuring the time difference between two satellites is: in, , These represent the signal transmission times of satellites A and B during the second measurement in the asymmetric bilateral two-way measurement process. The time when satellite A receives the signal. This is a measurement of inter-satellite distance. The speed of light; wherein is the two-star time difference, and is the error term for the time difference measurement.

3. The multi-satellite formation time synchronization method based on micro-satellite radio frequency measurement of claim 1, wherein, In step (2), the clock of the receiving satellite is kept in sync with the clock of the transmitting satellite by direct setting of the data. Different transmitting satellites are switched in different time slots, while the other receiving satellites undergo a phase correction.

4. The multi-satellite formation time synchronization method based on micro-satellite radio frequency measurement of claim 1, wherein, In step (3), the formula for calculating the frequency difference between the two satellites is: In the formula, for Frequency difference at time, for The time difference measurement results at any given moment. For the duration of the time synchronization period, To receive the satellite's local clock frequency; By measuring the time difference between two consecutive measurements, the frequency difference between the corresponding nodes is calculated and fed back to the NCO's frequency control word to achieve frequency correction.

5. The multi-satellite formation time synchronization method based on micro-satellite radio frequency measurement of claim 4, wherein, In step (3), the synchronization error is calculated after frequency correction. The expression for the synchronization error is: In the formula, Indicates the maximum synchronization error. and These represent the ratios of the onboard clock source frequency to the nominal frequency for satellites A and B, respectively. This is the nominal frequency.

6. The multi-satellite formation time synchronization method based on microsatellite radio frequency measurement according to claim 1, characterized in that, The specific process of step (5) is as follows: Assume there are K satellites in the formation, numbered S1, S2...SK, and the K satellites enter the launch state sequentially; When S1 enters the launch state, the remaining K-1 satellites in the formation measure their time difference with S1, and thus synchronize with S1 as the master node; when the time slot of satellite S1 ends, satellite S2 enters the launch state, and S2 becomes the new master node. The remaining K-1 satellites synchronize with S2, and so on, to achieve inter-satellite time synchronization of the formation.

Images