Satellite communication ranging and time synchronization method based on IEEE-1588 protocol

Abstract

The application discloses a satellite communication ranging and time synchronization method based on IEEE-1588 protocol. The method comprises the following steps: obtaining a master station time reference and an ephemeris model; exchanging ranging signals carrying session identifiers between the master station and a remote station, capturing corresponding physical layer time stamps at an antenna port, and obtaining a round-trip time stamp set; determining the satellite position at the corresponding moment based on the ephemeris model, and constructing a four-segment asymmetric space geometric propagation path; based on the four-segment asymmetric path and the round-trip time stamp set, solving the one-way distance of the remote station and the clock bias of the remote station; calculating the one-way reverse propagation delay by using the one-way distance of the remote station, and combining the preconfigured internal processing delay to obtain the timing advance; and adjusting the transmission timing of the remote station based on the advance and the clock bias to complete synchronization. The application effectively compensates for dynamic asymmetric errors, and improves the spatial synchronization accuracy and network operation stability of the system.

Description

Technical Field

[0001] This invention relates to the field of satellite communication and navigation technology, and in particular to a satellite communication ranging and time synchronization method based on the IEEE-1588 protocol. Background Technology

[0002] In time-division multiple access (TDMA) communication networks, remote earth stations need to transmit signals within accurately allocated time slots to avoid time slot collisions. High-precision time synchronization and absolute distance measurement are core technical supports for ensuring system spectral efficiency, improving channel throughput, and maintaining the physical time alignment of the entire network.

[0003] Current satellite communication ranging methods generally rely on closed-loop feedback correction mechanisms. The remote station sends ranging bursts in a designated time slot, the master station measures the deviation between the arrival time and the desired time, and then sends the deviation value back to the remote station for adjustment via signaling. Furthermore, existing precise time synchronization technologies applied in satellite links often assume absolute symmetry between uplink and downlink paths and frequently rely on onboard processing payloads for protocol parsing. Under transparent forwarding systems, the high bit error rate of long-distance wireless channels poses a significant challenge to the reliable transmission of time-stamped messages; simultaneously, temperature drift caused by the space environment and natural perturbations of satellite orbits can easily introduce constantly changing dynamic delay errors into the satellite-to-ground link.

[0004] In summary, existing methods struggle to balance high-precision ranging with dynamic compensation for environmental fluctuations in complex space-to-ground links and transparent forwarding architectures, resulting in insufficient adaptability and measurement stability under complex communication conditions. Therefore, it is necessary to investigate a technical solution that can improve the robustness of data interaction and the accuracy of synchronization time in space communication scenarios. Summary of the Invention

[0005] In view of this, a satellite communication ranging and time synchronization method based on the IEEE-1588 protocol is provided to solve the above-mentioned problems of the prior art.

[0006] According to one aspect of this application, a satellite communication ranging and time synchronization method based on the IEEE-1588 protocol includes:

[0007] Obtain the main station's time base and ephemeris model;

[0008] The master station and the remote station exchange ranging signals carrying session identifiers, and capture the corresponding physical layer timestamps at the moment the ranging signals leave and arrive at their respective antenna apertures to obtain a set of round-trip timestamps bound to the session identifiers.

[0009] Based on the ephemeris model, the satellite position at the corresponding time of the round-trip timestamp set is determined, and four asymmetric spatial geometric propagation paths are constructed, consisting of the main station to the satellite, the satellite to the remote station, the remote station to the satellite, and the satellite to the main station.

[0010] Based on four asymmetric spatial geometric propagation paths and round-trip timestamp sets, the one-way distance and clock deviation of the remote station are calculated.

[0011] The one-way back propagation delay from the remote station to the main station via the satellite is calculated based on the one-way distance of the remote station, and the one-way back propagation delay is combined with the pre-configured internal processing delay to obtain the timing advance.

[0012] The remote station's transmission timing is adjusted based on the timing advance and the remote station's clock deviation to achieve synchronization between the remote station and the master station's time reference.

[0013] According to one possible implementation of one aspect of this application, capturing the corresponding physical layer timestamp at the instant the ranging signal leaves and arrives at its respective antenna aperture includes:

[0014] When a trigger event is detected where the ranging signal leaves or arrives at the antenna aperture, the local time counter is frozen in real time by the hardware timestamp capture module to generate a physical layer timestamp corresponding to the time when the trigger event occurs.

[0015] According to one possible implementation of one aspect of this application, the ranging signal is generated and transmitted in the following manner:

[0016] The session identifier and the preset ranging flag are embedded in the frame header of the regular service burst frame, and the regular service burst frame is used as the ranging signal.

[0017] Correspondingly, at the transmitting end, the leading edge of the generated transmit enable signal is used as a trigger event to trigger the hardware timestamp capture module to obtain the physical layer timestamp of the transmitting end;

[0018] At the receiving end, the multiplexed power detection front end is used to perform envelope detection, and the moment when the received signal envelope crosses the preset power threshold is used as a trigger event to trigger the hardware timestamp capture module to obtain the physical layer timestamp of the receiving end.

[0019] According to one possible implementation of one aspect of this application, four asymmetric spatial geometric propagation paths are constructed, consisting of a master station to a satellite, a satellite to a remote station, a remote station to a satellite, and a satellite to a master station, including:

[0020] Using the ephemeris model, the three-dimensional orbital position of the satellite at each time corresponding to the round-trip timestamp set is extracted;

[0021] By utilizing the satellite's three-dimensional orbital position and the known master station position, the communication link from the master station to the remote station is deconstructed into forward and reverse spatial propagation paths of unequal length. A geometric constraint association is then established, which includes the forward and reverse spatial propagation paths and the unidirectional distance to the remote station to be determined.

[0022] According to one possible implementation of one aspect of this application, before calculating the one-way distance between the remote station and the remote station clock offset, the method further includes:

[0023] During the system hold-up time slot of the communication frame structure, the master station transmits and receives self-loopback test signals that are transparently relayed by the satellite.

[0024] The total round-trip propagation time of the self-loop test signal is obtained, and the round-trip spatial geometric delay from the master station to the satellite is deducted using the ephemeris model, as well as the pre-configured inherent radio frequency delay within the master station, to extract the dynamic group delay of the real-time tracked satellite transponder.

[0025] According to one possible implementation of one aspect of this application, the calculation of the one-way distance between the remote station and the remote station clock offset includes:

[0026] Extract the total round-trip time from the round-trip timestamp set, subtract the round-trip spatial geometric delay from the main station to the satellite and twice the satellite transponder dynamic group delay from the total round-trip time, convert the remaining delay into spatial distance and take half of the value to calculate the one-way distance of the remote station.

[0027] Calculate the asymmetric path delay difference between the forward and reverse spatial propagation paths;

[0028] The initial clock skew is constructed using the extracted transmit and receive timestamp difference, and the initial clock skew is compensated using the asymmetric path delay difference to calculate the remote station clock skew.

[0029] According to one possible implementation of one aspect of this application, after adjusting the transmission timing of the remote station based on the timing advance and the remote station clock deviation to complete the synchronization of the remote station's time reference with the master station, the method further includes:

[0030] Based on the synchronized adjustment of the remote station's transmission timing, the transmission burst time jitter of service signals sent from remote stations across the network to the master station will be converged to the preset target accuracy range.

[0031] Based on the convergent transmission burst time jitter, the frame structure of the communication system is adjusted in a cascade manner to compress the time span of the protection time slot between adjacent service time slots.

[0032] According to one possible implementation of one aspect of this application, in the exchange of ranging signals carrying session identifiers between the master station and the remote station, a two-step separation mode is used to process the physical acquisition of timestamps and the transmission of values, including:

[0033] Interacting with event messages that do not carry timestamp values ​​is used to trigger the capture operation of the physical layer timestamp at the physical layer.

[0034] Furthermore, subsequent messages are exchanged after the event message, with each subsequent message carrying the captured physical layer timestamp value. Moreover, the subsequent messages employ a channel coding scheme with higher redundancy than regular service data for error protection.

[0035] According to one possible implementation of one aspect of this application, in an independent operating state where the remote station does not acquire the time signal of the global satellite navigation system, the time reference of its local working clock is locked and maintained in the long term solely by the control command issued by the receiving master station based on the timing advance and the clock deviation of the remote station.

[0036] According to one possible implementation of one aspect of this application, envelope detection is performed using a multiplexed power detection front end, including:

[0037] The low-noise amplifier input of the receiving link shares a broadband envelope detector as a power detection front end. At the moment when the broadband envelope detector detects that the signal burst front edge crosses the preset power threshold, a burst synchronization trigger signal is sent to the demodulator of the communication baseband, and the counting freeze action of the hardware timestamp capture module is triggered in parallel.

[0038] Based on the above technical solutions, the present invention effectively compensates for dynamic asymmetric errors and improves the spatial synchronization accuracy and network operation stability of the system. Attached Figure Description

[0039] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0040] Figure 1 This is a schematic diagram of a satellite communication ranging and time synchronization method based on the IEEE-1588 protocol provided in this application.

[0041] Figure 2 This is a schematic diagram of a process for generating and transmitting ranging signals provided in this application.

[0042] Figure 3 This is a schematic diagram of the process for calculating the one-way distance and clock deviation of the remote station provided in this application.

[0043] Figure 4This application provides a schematic diagram of the timing advance process, which calculates the one-way back propagation delay from the remote station to the main station via the satellite based on the one-way distance of the remote station, and combines the one-way back propagation delay with the pre-configured internal processing delay.

[0044] Figure 5 This is a flowchart illustrating the remote station single-satellite positioning steps that are executed asynchronously in the background, as provided in this application. Detailed Implementation

[0045] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0046] Example 1

[0047] This embodiment provides a satellite communication ranging and time synchronization method based on the IEEE-1588 protocol, such as... Figure 1 As shown, the method may include, but is not limited to, the following steps. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be performed in a different order than that shown here.

[0048] Step 101: Obtain the main station time base and ephemeris model.

[0049] During the system initialization phase, the master station is equipped with a high-precision oven-controlled crystal oscillator and a Global Navigation Satellite System (GNSS) receiver module. By receiving the timing signal from the GNSS, the local oven-controlled crystal oscillator is tamed to output a second pulse signal and a 10MHz frequency reference signal. This physical process keeps the deviation between the master station's internal time and Coordinated Universal Time (UTC) within a very small range, serving as the master station's time reference for the entire network. Simultaneously, orbital element data is loaded in real-time or periodically through an external interface to construct an ephemeris model describing the three-dimensional motion state of geostationary communication satellites, providing reference constraint data for subsequent spatial distance calculations.

[0050] Step 102: Exchange ranging signals carrying session identifiers between the master station and the remote station, and capture the corresponding physical layer timestamps at the moment the ranging signals leave and arrive at their respective antenna apertures to obtain a set of round-trip timestamps bound to the session identifier.

[0051] In satellite communication systems, the master station and remote station exchange air interface signals via transparent transponders. To eliminate the uncertainties and delays introduced by baseband signal processing and channel encoding / decoding, timestamp acquisition is strictly limited to the antenna aperture of the RF front-end. Whenever an RF envelope signal is detected crossing the physical boundary of the antenna aperture, the underlying hardware immediately records the accurate current time. Through a complete round-trip signal exchange between the master and remote stations, key time data for transmission and reception are collected, and a unified session identifier is used to piece together the scattered time data into a set of associated round-trip timestamps.

[0052] An alternative approach involves employing a two-step separation mode to handle the physical acquisition of timestamps and the transmission of numerical data during the exchange of ranging signals carrying session identifiers between the master station and the remote station. Specifically, this includes:

[0053] The event message is exchanged without carrying a timestamp value, which is used to trigger the capture operation of the physical layer timestamp at the physical layer; and subsequent messages are exchanged after the event message, which carry the captured physical layer timestamp value, and the subsequent messages adopt a channel coding scheme with higher redundancy than regular service data for error protection.

[0054] This decoupled architecture decouples high-precision time acquisition from high-reliability data transmission. Physical layer triggering actions are unaffected by bit flips caused by poor channel conditions, while subsequent messages carrying timestamp values ​​penetrate high-error-rate channels through strong error correction coding, preventing ranging data packets from being lost in the spatial link and ensuring the continuity of the time synchronization control loop.

[0055] Step 103: Based on the ephemeris model, determine the satellite position at the time corresponding to the round-trip timestamp set, and construct four asymmetric spatial geometric propagation paths consisting of the main station to the satellite, the satellite to the remote station, the remote station to the satellite, and the satellite to the main station.

[0056] Based on the specific times recorded in the round-trip timestamp set, the three-dimensional orbital coordinates of the satellite at the corresponding time are derived using the ephemeris model. Due to the natural spatial drift phenomenon of geostationary satellites, the satellite's spatial position has objectively shifted during the ascent and descent of the signal. Therefore, abandoning the traditional symmetrical path assumption, and using the derived three-dimensional orbital coordinates and the known geographic coordinates of the master station, an asymmetric topology structure containing four independent segments (forward and reverse links) is established to reconstruct the propagation trajectory of electromagnetic waves in real space.

[0057] Step 104: Based on the four asymmetric spatial geometric propagation paths and the round-trip timestamp set, calculate the one-way distance of the remote station and the clock deviation of the remote station.

[0058] The total round-trip time extracted from the round-trip timestamp set is mapped to the four asymmetric spatial geometric propagation paths mentioned above. After removing known spatial distance segments and inherent satellite relay delays, the true one-way spatial scale from the remote station to the satellite is extracted. Using the asymmetric difference between forward and backward propagation timestamps, mathematical compensation is applied to the transmit and receive timestamp differences, separating the true deviation of the remote station's local clock from the master station's time reference, thus providing an accurate data source for subsequent synchronization operations.

[0059] Step 105: Calculate the one-way back propagation delay from the remote station to the main station via the satellite based on the one-way distance of the remote station, and combine it (one-way back propagation delay) with the pre-configured internal processing delay to obtain the timing advance.

[0060] Specifically, this can be further implemented by calculating the one-way back propagation delay from the remote station to the main station via the satellite based on the one-way distance of the remote station and the reverse path from the satellite to the main station in the four asymmetric spatial geometric propagation paths, and combining the one-way back propagation delay with the pre-configured internal processing delay to obtain the timing advance.

[0061] In a time-division multiple access (TDMA) communication system, the service bursts transmitted by the remote station must fall precisely within the time slot window allocated by the master station. The processor superimposes the calculated one-way distance of the remote station with the geometric distance from the satellite to the master station, converting it into a one-way back propagation delay for the electromagnetic wave traveling in the opposite direction. Based on this, the RF frequency conversion and baseband computation time accurately calibrated during the remote station's manufacturing phase are added to synthesize the final timing advance. This parameter objectively reflects the total end-link lag time from signal generation at the remote station to arrival at the master station's antenna aperture.

[0062] Step 106: Adjust the transmission timing of the remote station based on the timing advance and the remote station clock deviation to complete the synchronization of the time reference between the remote station and the master station.

[0063] The master station transmits the calculated control parameters via the downlink, and the remote station receives and analyzes the timing advance and clock offset. By compensating for the clock offset in the local digital oscillator, the remote station's system time is aligned to Coordinated Universal Time (UTC). Simultaneously, the overall transmit trigger time of the baseband frame structure is shifted forward by a time span equal to the timing advance, ensuring that the RF signal accurately matches the master station's reception timing after traversing the long spatial link.

[0064] In one possible implementation, when the remote station does not receive the time signal from the global satellite navigation system, the local working clock of the remote station periodically locks and calibrates itself by receiving control commands from the master station via the satellite communication link, based on timing advance and remote station clock deviation, to maintain synchronization between the remote station and the master station's time reference.

[0065] When a remote station is deployed in an urban canyon or an extreme environment subject to strong electromagnetic suppression, external time synchronization signals are physically blocked. In this case, the temperature-compensated crystal oscillator inside the remote station receives closed-loop correction commands from the master station via a satellite communication link, realizing periodic correction of local clock frequency drift and ensuring independent networking and reliable time synchronization capabilities in passive environments.

[0066] Example 2

[0067] Based on Embodiment 1 above, the physical layer capture process for timestamps is further described in detail. In one possible implementation, a basic method for timestamp capture based on dedicated time slots and independent messages is provided. Optionally, the ranging signal can be generated and transmitted in the following manner:

[0068] Step 201: Allocate a dedicated ranging time slot in the communication frame structure.

[0069] According to a preset time period, an isolation time window is defined within the superframe structure of the time division multiple access (TDMA) system. This isolation time window, as an independent resource, is separated from the time slot sequence carrying regular user service data. It is used to avoid air interface signal collisions between different network nodes and ensure the availability of the measurement channel.

[0070] Step 202: Within the dedicated ranging time slot, generate and send an independent timestamp event message as a ranging signal.

[0071] The protocol processing unit encapsulates a message data packet with a defined format identifier. The payload area of ​​this data packet does not carry regular service payloads; it only contains the network address, message type identifier, and session identifier for state tracking. The baseband processing unit performs channel coding and modulation operations on this data packet, and the output radio frequency burst signal constitutes the ranging signal required for this stage.

[0072] An optional approach is to capture the corresponding physical layer timestamp at the instant the ranging signal leaves and arrives at its respective antenna aperture, including: when a trigger event is detected that the ranging signal leaves or arrives at the antenna aperture, the local time counter is frozen in real time by a hardware timestamp capture module to generate a physical layer timestamp corresponding to the time when the trigger event occurs.

[0073] The system deploys a high-resolution time-to-digital converter at the RF front end. When electromagnetic wave energy crosses the physical boundary of the antenna feed port, the RF detection circuit generates a level-flipping signal. This level-flipping signal directly serves as a hardware-level interrupt source, driving the time-to-digital converter to latch the current count value of the local time counter driven by a high-stability crystal oscillator. This hardware freeze action avoids the execution delay variables introduced by the operating system kernel scheduling and application layer protocol stack parsing.

[0074] To eliminate the fixed timing offset caused by internal traces and analog components in the RF link, the captured initial timing observations are compensated for using factory parameters. The specific calculations are as follows:

[0075] T corrected =T raw -T RF ;

[0076] Among them, T corrected For the compensated physical layer timestamp, T raw T is the initial time observation output by the time-to-digital converter. RF The inherent group delay of the pre-stored RF components.

[0077] In the transmit direction, the pre-stored RF component inherent group delay includes the fixed latency of the high-power amplifier and up-converter from the baseband digital-to-analog converter to the transmit antenna aperture; in the receive direction, this group delay includes the fixed latency of the low-noise amplifier and down-converter from the receive antenna aperture to the analog-to-digital converter. By subtracting the fixed latency from the initial time observation, the reference cross-section of the timestamp is uniformly extrapolated to the physical aperture of the antenna.

[0078] Example 3

[0079] Unlike Embodiment 2 above, this embodiment provides another implementation method for the ranging signal generation and acquisition stage, reusing regular service burst frames to avoid air interface bandwidth occupation caused by dedicated ranging time slots. The rest of the overall framework structure is the same as in Embodiment 1, and can be referred to the foregoing description, so it will not be repeated here.

[0080] In another possible implementation, such as Figure 2 As shown, ranging signals can be generated and transmitted in the following manner:

[0081] Step 301: Embed the session identifier and the preset ranging flag into the frame header of the regular service burst frame, and use the regular service burst frame as the ranging signal.

[0082] When assembling regular service data packets, the protocol control unit allocates a 2-byte auxiliary information field in the frame header region of the data frame. The data processing process writes the currently active session identifier sequence into this auxiliary information field and defines one bit of the binary register as a ranging flag. When the value of this ranging flag is set to 1, the regular service burst frame representing the current user communication data is given the dual attributes of a time synchronization carrier. Correspondingly, the baseband demodulator at the receiving end performs a logical judgment when parsing the arriving frame header. If the extracted value of the ranging flag is 0, the initial time observation value associated with it is directly removed from the underlying buffer register to prevent invalid redundant data from occupying the microcontroller's calculation cycle.

[0083] In some optional implementations, a periodic sampling and marking strategy is provided for controlling the computational throughput of the central node. A cyclic counting logic is executed according to a given frame interval parameter configuration. Every N regular service burst frames, only one burst frame is assigned a value of 1 for its ranging flag, while the ranging flag of the remaining N-1 burst frames remains at a value of 0. By executing the cyclic counting logic according to the given frame interval parameter configuration, a ranging frame is marked only once after a specified number of frames, thereby adjusting the balance between the timestamp update frequency and the hardware processing computing power consumption.

[0084] Step 302: Accordingly, at the transmitting end, the leading edge of the generated transmit enable signal is used as a trigger event to trigger the hardware timestamp capture module to obtain the physical layer timestamp of the transmitting end.

[0085] When the modulation module completes the waveform generation of the baseband communication signal and prepares to push it to the RF link, the field-programmable gate array (FPGA) synchronously issues a digital control command, namely a transmit enable signal, to control the physical conduction of the RF power amplifier switch. The capture pin of the time-to-digital converter (TD-SCDMA) is configured to directly listen to the rising edge of this transmit enable signal. Since this digital leading edge is a state transition in the pure digital logic domain, it is not affected by the nonlinear distortion and phase noise of analog RF components, and the time jitter of its physical level transition is controlled within 0.2 nanoseconds. The freezing mechanism of the local time counter is driven by this digital leading edge to extract the initial transmission timestamp without delay compensation. Based on the inherent group delay constant of the transmit channel calibrated at the factory, a linear subtraction operation is performed to output the corrected physical layer timestamp.

[0086] Step 303: At the receiving end, envelope detection is performed using a multiplexed power detection front end, and the moment when the received signal envelope crosses a preset power threshold is used as a trigger event to trigger the hardware timestamp capture module to obtain the physical layer timestamp of the receiving end.

[0087] When electromagnetic waves penetrate the space link and reach the receiving antenna at the remote station, the signal processing architecture eliminates the redundant design of adding a separate detection circuit for time synchronization. The baseband burst detector originally used for data demodulation triggering inside the receiving channel is physically deep multiplexed. The power envelope of the input RF signal is monitored in real time by the voltage of the multiplexed power detection front end. The hardware comparator has a preset power threshold reference level higher than the bottom of the current system thermal noise. As the voltage value of the signal burst envelope rises and crosses this reference level, the hardware comparator outputs a level-flipping pulse. The system captures the moment this pulse is generated and defines it as a trigger event, thereby activating the sampling latching action of the receiver's time counter.

[0088] Optionally, envelope detection can be performed using a multiplexed power detection front end, specifically including the following process:

[0089] The low-noise amplifier input of the receiving link shares a broadband envelope detector as a power detection front end. At the moment when the broadband envelope detector detects that the signal burst front edge crosses the preset power threshold, a burst synchronization trigger signal is sent to the demodulator of the communication baseband, and the counting freeze action of the hardware timestamp capture module is triggered in parallel.

[0090] To shorten the transmission delay of physical signals within the receiving device, the probe feed of the broadband envelope detector is directly connected to the input of a low-noise amplifier located immediately behind the receiving antenna feed. Once the signal burst front is confirmed to exceed a preset power threshold, the level-flipping pulse output by the broadband envelope detector is routed to two parallel control links via a physical splitter. The first physical link sends an electrical signal to the communication baseband chip, serving as a burst synchronization trigger signal to initiate frame synchronization and data demodulation / decoding processes; the second physical link is directly connected to the hardware enable pin of the time-to-digital converter, triggering the freezing and acquisition of timestamp values. This front-end shares a circuit topology parallel to the hardwired lines, eliminating the nondeterministic computational lag caused by baseband digital signal processing.

[0091] As an alternative, when the device operates in a low signal-to-noise ratio (SNR) and poorly configured wireless channel, the over-limit decision of a single envelope detection is susceptible to interference from the superposition of random high-frequency thermal noise, resulting in nanosecond-level time determination jitter. A sliding sampling window is defined in the memory region to cache M consecutive physical layer timestamp records captured by hardware. An arithmetic average of the timestamp values ​​within this window is calculated using a multi-frame moving average filter algorithm. The resulting smoothed timestamp is extracted and used to overwrite the original single high-frequency sample value, thereby suppressing the interference caused by random channel fluctuations on the envelope over-limit decision time.

[0092] Example 4

[0093] Based on the above embodiment 1, the decomposition mechanism of the satellite-to-ground propagation path and the calibration of key time delay parameters are further described in detail.

[0094] In one possible implementation, constructing four asymmetric spatial geometric propagation paths consisting of the master station to the satellite, the satellite to the remote station, the remote station to the satellite, and the satellite to the master station may include, but is not limited to, the following steps:

[0095] Step 401: Using the ephemeris model, extract the satellite's three-dimensional orbital position at each time corresponding to the round-trip timestamp set.

[0096] By acquiring ephemeris data and combining it with timestamp sequences obtained from measurements, the spatial position vector of the satellite at a predetermined time is queried and calculated. Due to the influence of perturbations, the spatial position of geostationary satellites exhibits dynamic drift characteristics over time. The three-dimensional orbital positions of satellites at different communication interaction nodes are extracted to provide the basic physical coordinates for subsequent construction of dynamic geometric topology.

[0097] Step 402: Using the satellite's three-dimensional orbital position and the known master station position, the communication link from the master station to the remote station is deconstructed into forward and reverse spatial propagation paths of unequal length, and a geometric constraint association is established that includes the forward and reverse spatial propagation paths and the one-way distance of the remote station to be determined.

[0098] The physical distance between two points is calculated using the spatial coordinates of the master station in the Earth coordinate system and the real-time updated satellite positions. The forward spatial propagation path is obtained by solving for the master station position at the time of signal transmission and the satellite position at the time of signal transit; the reverse spatial propagation path depends on the master station position at the time of reception and the corresponding satellite position at the time of transit. By splitting the independent paths, the calculation residuals caused by satellite displacement under the symmetry assumption are eliminated.

[0099] In the solution process, for geostationary orbit satellite communication scenarios, since the relative linear velocity of the satellite relative to the ground is typically around 3 meters per second, the spatial displacement of the satellite is at the sub-meter level within a typical one-way transmission time of hundreds of milliseconds. The time when the remote station receives the signal is used to replace the actual physical time when the signal travels through the satellite to query the ephemeris model. Under this approximation, the distance from the satellite to the remote station in the forward link and the distance from the remote station to the satellite in the reverse link are considered as scalars of equal length, i.e., the one-way distance of the remote station to be determined. This processing mechanism makes the main process of ranging and time synchronization rely purely on the timestamp and the master station's spatial coordinate input, achieving technical decoupling from the unknown state of the remote station's physical location.

[0100] As an alternative, when the communication network is deployed in a low-to-medium Earth orbit satellite constellation, the satellite's linear velocity can reach 3 to 7 kilometers per second, rendering the aforementioned approximate alternative condition invalid. In this case, the iterative approximation algorithm is switched to, which uses the initial round-trip measurement time to estimate the physical time of the satellite's transit, extracts the ephemeris position at that time to recalculate the space propagation path, and performs iterative iterations until the difference between the two distance calculations converges to within the preset distance determination criteria.

[0101] An alternative approach includes the following steps before calculating the one-way distance between the remote station and the remote station clock offset:

[0102] Step S1: During the system hold-up time slot of the communication frame structure, the master station transmits and receives the self-loopback test signal that is transparently relayed by the satellite.

[0103] To obtain the dynamic group delay parameters affected by temperature field changes in space, the master station scheduling module utilizes the system hold-up time slots reserved in the network management protocol. Within this maintenance time window, a loopback test burst frame is constructed with the destination address pointing to the local device. In full-duplex mode, the master station antenna system, relying on the physical isolation band between the uplink and downlink frequencies, amplifies and converts the transmitted electromagnetic waves via a transparent satellite transponder before directly receiving them locally, thus obtaining the transmission and reception physical times of this self-looping communication cycle.

[0104] Step S2: Obtain the total round-trip propagation time of the self-loop test signal, and use the ephemeris model to deduct the round-trip spatial geometric delay from the master station to the satellite, as well as the pre-configured inherent radio frequency delay within the master station, to extract the dynamic group delay of the real-time tracked satellite transponder.

[0105] Satellite transparent transponders experience internal thermal fluctuations during operation, inducing nanosecond-level signal penetration delay drift. The microprocessor calculates the duration of the loopback test by subtracting the transmission and reception physical times. Linear algebraic operations are then used to remove the two-way spatial geometric propagation delay and the calibration time of electronic components within the ground station from this duration, yielding the pure device dynamic group delay parameters.

[0106] The specific calculations are as follows:

[0107] τ sat =RTT loop -2*d MS_loop / c;

[0108] Where, τ sat For satellite transponder dynamic group delay, RTT loop The total round-trip propagation time of the self-looping test signal, calculated based on the antenna aperture physical layer timestamp, has excluded the inherent delay of the main station's internal RF link. MS_loop Let c be the one-way spatial geometric distance from the master station to the satellite at the self-loop test time calculated by the ephemeris model, and c is the speed of light constant.

[0109] The physical meaning of the above formula is that after the electromagnetic wave undergoes two-way spatial propagation and the transmission and reception processing of the ground radio frequency equipment in the loop path, the remaining time is the dwell time of the signal when it passes through the satellite transparent transponder once.

[0110] The above parameter extraction steps are executed periodically at 20-millisecond intervals, and the calculation results are written to the shared storage area. This transforms the slowly varying heat source interference of the satellite physical platform into known time compensation parameters, providing a reference input for subsequent time synchronization calculations of remote stations across the entire network.

[0111] Example 5

[0112] Based on the preliminary parameter preparation provided in Embodiment 4 above, the specific process of calculating the one-way distance and clock deviation of the remote station is further described in detail, providing a basic method for solving clock error and one-way distance based on algebraic direct calculation. In one possible implementation, such as Figure 3 As shown, this process may specifically include, but is not limited to, the following steps:

[0113] Step 501: Extract the total round-trip time from the round-trip timestamp set. Subtract the round-trip spatial geometric delay from the master station to the satellite and twice the satellite transponder dynamic group delay from the total round-trip time. Since the ranging signal passes through the satellite transponder twice during the complete round trip, the satellite transponder delay contribution removed from the total round-trip time is twice the single group delay value. Convert the remaining delay to spatial distance and take half of the value to calculate the one-way distance of the remote station.

[0114] Specifically, this can be further implemented by extracting the total round-trip time from the round-trip timestamp set, subtracting the round-trip spatial geometric delay from the main station to the satellite and twice the satellite transponder dynamic group delay from the total round-trip time to obtain the remaining delay, multiplying the remaining delay by the speed of light and dividing by 2 to calculate the one-way distance of the remote station.

[0115] Arithmetic operations are performed on the round-trip timestamp set. The difference between the master station's reception time and transmission time is subtracted from the difference between the remote station's transmission time and reception time to obtain the total time spent by the electromagnetic wave traveling to and from the satellite in the space link. This total time includes the propagation time of the four space links between the satellite and ground stations, as well as the two group delay times of the satellite transponder. Algebraic stripping is performed by incorporating known master-to-satellite space information and real-time calibrated group delay data.

[0116] The specific operational logic is expressed as follows:

[0117] d remote =(RTT measured -T MS_MS -2*τ sat )*c / 2;

[0118] Where, d remote For one-way distance from the remote station, RTT measured To measure the total round-trip time, T MS_MS The round-trip spatial geometric delay from the main station to the satellite, τ sat The dynamic group delay of the satellite transponder is given, c is the constant of the speed of light, and the divisor 2 represents the multiple of the constant of the far-end link that is symmetrical in the aforementioned characteristics.

[0119] This operation separates the remote station physical distance variable hidden in the total delay, providing a spatial scale basis for subsequent timing compensation.

[0120] Step 502: Calculate the asymmetric path delay difference between the forward spatial propagation path and the reverse spatial propagation path.

[0121] In classic precise time protocol models, the underlying algorithms often assume bidirectional symmetry in the communication link. However, in geostationary orbit satellite communication scenarios, satellite orbital perturbations cause physical distance discrepancies between the uplink and downlink. To address this objective physical phenomenon, the forward propagation path distance from the master station to the satellite and the reverse propagation path distance from the satellite to the master station are obtained, and the difference between the two is calculated.

[0122] The specific logic for deriving the difference is as follows:

[0123] D diff =(d SM -d MS ) / (2*c);

[0124] Among them, D diff d represents the asymmetric path delay difference. SM d represents the distance along the reverse space propagation path from the satellite to the master station. MS The distance along the forward propagation path from the main station to the satellite is the numerical value, where c is the speed of light constant.

[0125] The asymmetric path delay difference extracted by this formula is used to quantitatively characterize the one-way time transmission error introduced by spatial topological distortion.

[0126] Step 503: Construct the initial clock offset using the extracted transmit and receive timestamp difference, and compensate for the initial clock offset using the asymmetric path delay difference to calculate the remote station clock offset.

[0127] The initial clock bias is calculated using a conventional two-way ranging architecture. This involves calculating the time interval between the remote station's reception time and the master station's transmission time, subtracting the time interval between the master station's reception time and the remote station's transmission time, and then taking the arithmetic mean of the two differences. Since this initial calculation process does not eliminate spatial asymmetry factors, the output results at this stage contain systematic biases.

[0128] Based on this, the asymmetric path delay difference obtained in the previous steps is introduced into the calculation logic to perform superposition compensation.

[0129] The calculation relationship for clock skew is as follows:

[0130] θ final =θ initial +D diff ;

[0131] Where, θ final For the final calculated remote station clock offset, θ initialFor the initial clock skew without eliminating asymmetric errors, D diff This represents the delay difference for asymmetric paths.

[0132] To more intuitively demonstrate the above algebraic calculation logic, a simplified normalized example without a predetermined physical coordinate background is constructed here. Assume that after normalization, the speed of light constant is set to 1.0. The measured total round-trip time is 1.0, the geometrical delay between the master station and the satellite is 0.4, and the satellite transponder dynamic group delay is 0.05. Substituting these parameters into the distance formula, the one-way distance to the remote station is calculated to be (1.0 - 0.4 - 2 * 0.05) * 1.0 / 2 = 0.25.

[0133] Further exploring the asymmetric compensation stage, assuming the forward distance from the master station to the satellite is 0.25 and the reverse distance from the satellite to the master station is 0.15, the asymmetric path delay difference is (0.15-0.25) / (2*1.0)=-0.05. If the initial clock offset obtained using the timestamp sequence is -0.10, after asymmetric numerical compensation calculation, the actual remote station clock offset is -0.10+(-0.05)=-0.15.

[0134] This compensation operation eliminates the systematic errors introduced by the symmetry assumption and achieves accurate separation of spatial and temporal parameters through pure algebraic solutions.

[0135] Example 6

[0136] Unlike the single discrete direct calculation in Embodiment 5 above, this embodiment provides an alternative to the continuous state estimation algorithm based on extended Kalman filtering, and a method for joint state estimation based on extended Kalman filtering.

[0137] In this embodiment, since both the state transition equation and the observation equation are linear, the constructed filter is mathematically equivalent to a standard linear Kalman filter. In practical engineering deployments, if it is necessary to incorporate higher-order nonlinear factors introduced by nonlinear interpolation of the ephemeris model or orbit prediction errors, the filter can be extended to an extended Kalman filter form, while the core state variable definitions and observation matrix structure remain unchanged.

[0138] In one possible implementation, the calculation of the one-way distance and clock offset of the remote station is specifically achieved by constructing an extended Kalman filter for joint state estimation, which may include, but is not limited to, the following steps:

[0139] Step 601: Construct an extended Kalman filter, defining the remote station's clock skew, clock frequency skew, one-way distance of the remote station, and dynamic group delay of the satellite transponder as system state variables.

[0140] During the operation of the communication system, discrete single algebraic operations are upgraded to a state-space model with temporal continuity. A state vector storage area is allocated in memory, and key physical quantities characterizing temporal and spatial attributes are extracted to construct a multidimensional state vector. The system state variables are represented in column vector form as follows:

[0141] X(k)=[θ(k),θ dot (k),d remote (k),τ sat (k)] T ;

[0142] Where X(k) is the system state vector at the k-th observation time, θ(k) is the clock offset of the remote station, and θ dot (k) represents the clock frequency deviation, i.e., the clock drift rate, of the remote station; d remote (k) represents the one-way distance to the remote station, τ sat (k) represents the dynamic group delay of the satellite transponder, and the superscript T indicates the matrix transpose operation.

[0143] Step 602: Using the dynamic equation reflecting the evolution of the local crystal oscillator frequency offset and the drift model reflecting the dynamic change rate of the satellite orbit, establish the state transition matrix of the extended Kalman filter.

[0144] The system state variables are iteratively estimated using an extended Kalman filter, and the output includes state estimates including the one-way distance of the remote station and the clock deviation of the remote station.

[0145] To describe the evolution of the system's state variables over time, a system dynamics model is introduced. This model assumes that the clock deviation follows a second-order polynomial evolution trajectory including a frequency deviation component, and sets the spatial distance and transponder delay as slowly varying parameters with random walk characteristics. The discrete-time state transition equation is expressed as follows:

[0146] X(k+1)=F*X(k)+W(k);

[0147] Where X(k+1) is the predicted system state vector at the (k+1)th observation time, F is the state transition matrix, and W(k) is the system process noise vector.

[0148] During state recursion, the state transition matrix F is set as a 4x4 square matrix with constant coefficients. The first row of this matrix is ​​set to [1,ΔT,0,0], the second row to [0,1,0,0], the third row to [0,0,1,0], and the fourth row to [0,0,0,1]; where ΔT represents the time interval between two adjacent time synchronization ranging interactions.

[0149] The configuration of the covariance matrix of the process noise vector W(k) is constrained by the physical characteristics of the communication equipment itself. Taking a commonly used temperature-compensated crystal oscillator as an example, its clock drift noise power spectral density is typically around 10. -20 Within the numerical range of / Hz; simultaneously, the operation of geostationary satellites is constrained by orbit-keeping actions, and the daily rate of change of their orbital drift rate is controlled within 0.1 degrees. Based on the above objective physical boundary conditions, values ​​are assigned to the covariance matrix parameters to prevent the filter from diverging due to noise input distortion during the iteration process.

[0150] Optionally, the steps may further include:

[0151] The residual delay obtained by subtracting the round-trip spatial geometric delay from the master station to the satellite calculated using the ephemeris model from the total round-trip time is used as the first-dimensional observation input;

[0152] By utilizing the forward and reverse spatial path asymmetry terms between the master station and the satellite, the difference between the transmitted and received timestamps extracted from the round-trip timestamp set is stripped and compensated, generating a clock deviation pseudo-observation value compensated for asymmetric path as the second-dimensional observation input;

[0153] The net transponder delay, obtained by subtracting the two-way spatial geometric delay from the master station to the satellite calculated using the ephemeris model from the total round-trip propagation time of the self-loop test signal, is used as the third-dimensional observation input.

[0154] The system state variables are updated by constructing a joint observation matrix based on the above three-dimensional observation inputs, and the output includes the optimal state estimate including the one-way distance of the remote station and the clock deviation of the remote station.

[0155] To perform physical closed-loop correction on the predicted state, three independent measurement scalars of different dimensions are integrated to construct a composite observation vector. This composite observation vector incorporates direct ranging parameters of the air interface link, timing bias parameters stripped of asymmetric terms, and spatial calibration parameters extracted during system maintenance. The overall measurement and observation equations are expressed as follows:

[0156] Z(k) = H*X(k) + V(k);

[0157] Where Z(k) is the observation vector containing the above three data dimensions, H is the joint observation matrix, and V(k) is the measurement noise vector characterizing the wireless channel fluctuation.

[0158] When establishing the mapping association, the joint observation matrix H is configured as a three-row, four-column numerical structure. The first row corresponds to the spatial mapping of the total round-trip time of the measurement with the state variables, with the element sequence [0,0,2 / c,2], where c represents the constant speed of light, and the coefficient 2 reflects that the signal crosses the transponder twice in the forward and reverse links; the second row corresponds to the temporal mapping of the purified clock bias pseudo-observation value with the state variables, with the element sequence [1,0,0,0]; the third row corresponds to the mapping of the total round-trip propagation time of the self-loop test signal with the state variables, with the element sequence [0,0,0,1], and the coefficient 1 reflects that the signal crosses the transponder only once in the self-loop path.

[0159] The Kalman filter operator is invoked, and the dynamic Kalman gain parameter is calculated using the observation vector Z(k) and the joint observation matrix H. This gain parameter is then used to perform residual feedback correction on the system state prediction variables, thereby outputting the optimal estimated values ​​of each physical variable.

[0160] Optionally, the steps may further include:

[0161] When a subsequent ranging message containing timestamp information is detected to be lost in the wireless channel, the state transition characteristics of the clock frequency deviation in the extended Kalman filter are used to perform an observationless one-step state prediction for a time step, so as to maintain continuous estimation of the remote station clock deviation for a short period of time.

[0162] In electromagnetic environments where satellite communications face fading or interference, the high bit error rate of space links may prevent subsequent data frames carrying numerical feedback information from being completely reassembled. When the protocol stack detector determines that the ranging data packet for the current interaction cycle is missing, it automatically suspends the operation branch that uses composite observation vectors for posterior update. It then switches to the prediction maintenance state, simply acquiring the clock frequency deviation parameter obtained from the convergence of the previous sampling time, and deduces the current prediction state vector by stepping forward according to the state transition matrix F.

[0163] Thus, the physical forward extrapolation without the introduction of new measurements fills the data vacuum caused by one or more channel packet losses, ensuring that the downlink control loop will not experience interruption or failure.

[0164] Example 7

[0165] Based on the optimal distance estimation parameters output by Embodiment 5 or Embodiment 6 above, the physical compensation and synchronization system optimization of the one-way backward propagation delay are further described in detail.

[0166] In one possible implementation, the one-way backpropagation delay from the remote station to the main station via the satellite is calculated based on the one-way distance from the remote station. This one-way backpropagation delay is then combined with a pre-configured internal processing delay to obtain the timing advance, such as... Figure 4As shown, the specific steps may include, but are not limited to, the following:

[0167] Step 701: Using the ephemeris model, determine the reverse geometric distance from the satellite to the main station at the current moment.

[0168] In a time-division multiple access (TDMA) system, to ensure that the radio frequency bursts transmitted by the remote station fall within the time slot window allocated by the master station's receiver, it is necessary to predict the reverse flight trajectory of the signal in advance. By accessing the ephemeris data source and inputting the timestamp reference corresponding to the current scheduling cycle, the spatial three-dimensional coordinates of the satellite at the current communication time are calculated. By calculating the spatial vector magnitude between these satellite three-dimensional coordinates and the known geographic coordinates of the master station, the objective spatial scale of the reverse physical link is obtained.

[0169] Step 702: Combine the one-way distance of the remote station with the reverse geometric distance to obtain the one-way reverse propagation delay, which characterizes the physical time required for a radio signal to travel from the remote station to the main station via the satellite.

[0170] Due to orbital drift of geostationary satellites, there is an asymmetric deviation in the lengths of the uplink and downlink. The one-way distance of the remote station, calculated in the previous step or output by Kalman filtering, is extracted and added to the reverse geometric distance obtained in the preceding step to construct the complete total length of the reverse asymmetric propagation path. This total length is then divided by the constant propagation speed of electromagnetic waves in a vacuum, and the dynamic group delay of the satellite transponder in the current time period is superimposed to complete the physical conversion from spatial distance to propagation time.

[0171] The specific time delay conversion operation relationship is as follows:

[0172] D reverse =(d remote +d SM ) / c+τ sat ;

[0173] Among them, D reverse For one-way backward propagation delay, d remote d represents the one-way distance from the remote station. SM Let τ be the reverse geometric distance from the satellite to the main station, c be the constant of the speed of light, and τ be the constant of the speed of light. sat For satellite transponder dynamic group delay.

[0174] Step 703: Obtain the inherent time consumption of radio frequency and baseband processing inside the remote station as calibrated by the factory, and use it (inherent time consumption of radio frequency and baseband processing inside the remote station) as the pre-configured internal processing delay.

[0175] In physical communication equipment, baseband data packets undergo channel coding and digital modulation before being output and pushed to the antenna aperture via a digital-to-analog converter, up-converter, and high-power amplifier. This process inherently involves a certain signal dwell time. During the hardware calibration phase, testing instruments are used to jointly determine the physical processing dwell time of the transmission link, and this constant is permanently stored in the non-volatile memory of the remote station as a fixed parameter to compensate for local execution delays at the remote station.

[0176] Step 704: Add the one-way back propagation delay to the pre-configured internal processing delay to generate a timing advance for controlling the time boundary of the remote station's transmitted signal.

[0177] The dynamic delay constant caused by spatial propagation and the static delay constant generated by the device hardware link are extracted and linearly added together. The resulting value directly determines the time offset by which the remote station needs to initiate the transmission process in advance relative to the local frame reference.

[0178] The generation relationship of this parameter is as follows:

[0179] TA=D reverse +τ proc ;

[0180] Where TA is the generated timing advance, and D reverse For the one-way backward propagation delay, τ proc Pre-configured internal processing latency.

[0181] To demonstrate the calculation logic of this control parameter, a normalized simplified example is introduced. The following is an independent normalized simplified example; the parameter values ​​do not constitute a continuous calculation relationship with the normalized example in Example 5. Assume that after dimensional unification processing, the current output remote station one-way distance is 0.25, the reverse geometric distance is 0.25, and the speed of light constant is set to 1.0. The extracted satellite transponder dynamic group delay is 0.05, and the calibrated internal processing delay is 0.01. Substituting into the aforementioned formula, the one-way reverse propagation delay is calculated as (0.25+0.25) / 1.0+0.05=0.55. Adding this value to the internal processing delay, the corresponding timing advance is 0.55+0.01=0.56.

[0182] An optional scheme, after adjusting the remote station's transmission timing based on timing advance and remote station clock deviation to complete the synchronization of the remote station's time reference with the master station, also includes the following steps:

[0183] Step S01: Based on the synchronously adjusted transmission timing of the remote station, the transmission burst time jitter of the service signals sent from the remote stations to the master station across the entire network is converged to the preset target accuracy range by repeatedly performing ranging signal exchange and transmission timing adjustment.

[0184] The master station transmits the calculated timing advance and clock offset compensation values ​​to the corresponding remote stations via service channels or independent management signaling. The remote station's baseband control system adjusts the phase of its local direct digital frequency synthesizer based on the received parameters, regulating the timing of the frame-triggered interrupt. Through multiple cycles of closed-loop feedback, the time dispersion of RF pulse groups emitted by multiple remote stations in different geographical locations reaching the master station's receiving antenna aperture is converged, and the drift variance on the time axis is suppressed to the sub-microsecond range.

[0185] Step S02: Based on the converged transmission burst time jitter, the frame structure of the communication system is cascaded and adjusted to compress the time span of the protection time slot between adjacent service time slots.

[0186] In time-division multiple access (TDMA) frame structure planning, a protection buffer time of tens of microseconds is set to prevent sudden overlap and collision of adjacent time slots due to synchronization errors. After the network-wide time jitter converges to the target accuracy range, the time scale within the superframe is re-divided, and the protection buffer time is synchronously compressed to a span of less than 1 microsecond. This system-level adjustment improves the filling density of service data within a fixed bandwidth, increasing the overall throughput capacity of the communication network.

[0187] Example 8

[0188] Based on the same inventive concept as the aforementioned ranging and time synchronization methods, this embodiment provides an implementation method for remote station single-satellite positioning steps executed asynchronously in the background. This embodiment belongs to a non-real-time extended value-added application. In one possible implementation, such as... Figure 5 As shown, this implementation method may specifically include, but is not limited to, the following steps:

[0189] Step 801: During the preset continuous running period, continuously collect multiple sets of remote station one-way distances calculated at different times, as well as the satellite three-dimensional orbital positions provided by the ephemeris model corresponding to each time, and construct a time-series dynamic distance observation sequence.

[0190] In addition to the main thread that performs real-time, second-level time synchronization tasks, a separate background data dumping thread is established. This background data dumping thread pairs and packages the remote station unidirectional distance scalar output from the synchronization loop with the satellite 3D coordinate vectors output from the ephemeris query at the same time, according to a set sampling interval. A preset time span window is used, typically configured to be 6 to 24 hours. Within this continuous runtime period, the background dumping thread pushes all paired data packets into a buffer queue in chronological order of timestamps, forming a time-series dynamic distance observation sequence containing spatial distance and corresponding reference coordinates.

[0191] Step 802: Obtain a pre-constructed geometric constraint model of the Earth's ellipsoid and utilize the spatial positional variations caused by the natural orbital drift of geosynchronous satellites to provide baseline geometric diversity.

[0192] It can be further implemented as follows: obtain a pre-constructed geometric constraint model of the Earth ellipsoid, in which the three-dimensional orbital positions of the satellite at different times in the time-series dynamic distance observation sequence are spatially different due to the natural orbital drift of the geosynchronous satellite, providing baseline geometric diversity for the solution.

[0193] Under a single spatial reference node system, conventional trilateration positioning algorithms face difficulties in solving underdetermined equations. To introduce additional constraints, a standard Earth ellipsoidal geometric constraint model is loaded from a local geographic information database. Simultaneously, utilizing objective physical laws—that geostationary satellites, due to orbital inclination and eccentricity perturbations, project a closed trajectory curve onto the Earth's surface from their nadir—this natural orbital drift continuously changes the coordinate reference point over a spatial span of tens of kilometers over time, transforming the position of a single satellite at different times into multiple virtual reference anchor points distributed in different spatial locations.

[0194] In practical implementation, the geometric constraint model of the Earth ellipsoid is expressed using standard coordinate equations:

[0195] (x 2 +y 2 ) / (R e +h) 2 +z 2 / (R p +h) 2 =1;

[0196] Where x is the X-axis component of the remote station's three-dimensional physical coordinates, y is the Y-axis component of the remote station's three-dimensional physical coordinates, z is the Z-axis component of the remote station's three-dimensional physical coordinates, and R e R is a preset constant representing the Earth's equatorial radius. p is the preset Earth polar radius constant, and h is the elevation data of the remote station.

[0197] For remote station equipment that is not equipped with an elevation sensor, the variable h is assigned the value 0 or the average value of the digital elevation model of the area.

[0198] Step 803: Based on the time-series dynamic distance observation sequence and the geometric constraint model of the Earth ellipsoid, the three-dimensional physical coordinates of the remote station are solved using a nonlinear least squares iterative algorithm.

[0199] Extract each set of data from the time-series dynamic distance observation sequence and establish a three-dimensional spatial distance residual equation between the unknown coordinates of the remote station and the known position of the current satellite. Combine the spatial distance residual equations generated by stacking multiple time points with the aforementioned Earth ellipsoidal geometric constraint model. Using the Gauss-Newton algorithm or other equivalent nonlinear optimization operators, with the satellite nadir coordinates or a certain approximate coordinate as the initial values ​​for iteration, search for the coordinate solution that minimizes the sum of squares of the overall distance observation residuals by repeatedly calculating the Jacobian matrix and updating the step size.

[0200] The x, y, and z components output after iterative convergence constitute the three-dimensional physical coordinates of the remote station. This background positioning process relies on the time-series data, a byproduct of the ranging synchronization main process, and does not require the remote station to be equipped with independent navigation receiving hardware. It provides a rough geographical location estimation capability even when the communication terminal loses external navigation signal coverage.

[0201] Taking a typical geostationary satellite communication scenario as an example, the master station is located at 40°N, 116°E, and the satellite's nominal longitude is 110°E. Due to the satellite's north-south orbital inclination drift, during forward propagation, the satellite is located slightly north of its nominal position, while during reverse propagation, it has drifted slightly south, resulting in a difference of hundreds of meters between the uplink and downlink distances from the master station to the satellite. This method extracts the satellite coordinates at the two time points using an ephemeris model and independently calculates the path distance for each segment, thereby eliminating the impact of this asymmetric error on clock deviation estimation.

[0202] This application reduces the systematic bias caused by the symmetry assumption by deconstructing four independent spatial geometric propagation paths and combining them with the actual one-way reverse physical delay calculation to reduce the timing advance, thus converging the synchronization accuracy of long-distance space communication to the sub-microsecond level. Addressing the internal delay fluctuations caused by temperature variations in space transparent transponders, online dynamic group delay calibration is achieved using self-loopback testing of network management time slots, which can capture and compensate for nanosecond-level thermal drift errors that are otherwise difficult to measure.

[0203] To overcome high bit error rates and sudden noise interference in harsh electromagnetic channels, physical event triggering and high-impact data transmission are decoupled, and a multi-dimensional Kalman filter model integrating crystal oscillator characteristics and orbital drift properties is constructed. This mechanism effectively smooths observation jitter in single ranging measurements, achieving high-precision observation-free extrapolation prediction during data packet loss vacuum periods, and reducing the risk of instantaneous loss of lock-on in the timing link. Furthermore, by embedding the ranging identifier deep within the header of regular service frames and reusing the hardware RF envelope detection port, zero air interface bandwidth occupancy for time synchronization is achieved while suppressing amplitude determination jitter caused by analog circuits.

[0204] Therefore, in extreme navigation signal denial scenarios, this solution can independently maintain a local high-stability time base, accumulate by-product data of the main time synchronization process, and generate a virtual observation baseline by utilizing the natural spatial drift orbit of the satellite, thus giving the passive island terminal the ability to autonomously calculate three-dimensional spatial coordinates.

[0205] It should be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the present invention will not describe the various possible combinations separately.

Claims

1. A satellite communication ranging and time synchronization method based on the IEEE-1588 protocol, characterized in that, include: Obtain the main station's time base and ephemeris model; The master station and the remote station exchange ranging signals carrying session identifiers, and capture the corresponding physical layer timestamps at the moment the ranging signals leave and arrive at their respective antenna apertures to obtain a set of round-trip timestamps bound to the session identifiers. Based on the ephemeris model, the satellite position at the corresponding time of the round-trip timestamp set is determined, and four asymmetric spatial geometric propagation paths are constructed, consisting of the main station to the satellite, the satellite to the remote station, the remote station to the satellite, and the satellite to the main station. Based on four asymmetric spatial geometric propagation paths and round-trip timestamp sets, the one-way distance and clock deviation of the remote station are calculated. The one-way backward propagation delay from the remote station to the main station via the satellite is calculated based on the one-way distance of the remote station, and then combined with the pre-configured internal processing delay to obtain the timing advance. The remote station's transmission timing is adjusted based on the timing advance and the remote station's clock deviation to achieve synchronization between the remote station and the master station's time reference.

2. The method according to claim 1, characterized in that, Capture the corresponding physical layer timestamps at the instants the ranging signal leaves and arrives at its respective antenna aperture, including: When a trigger event is detected where the ranging signal leaves or arrives at the antenna aperture, the local time counter is frozen in real time by the hardware timestamp capture module to generate a physical layer timestamp corresponding to the time when the trigger event occurs.

3. The method according to claim 2, characterized in that, Ranging signals are generated and transmitted in the following manner: The session identifier and the preset ranging flag are embedded in the frame header of the regular service burst frame, and the regular service burst frame is used as the ranging signal. Correspondingly, at the transmitting end, the leading edge of the generated transmit enable signal is used as a trigger event to trigger the hardware timestamp capture module to obtain the physical layer timestamp of the transmitting end; At the receiving end, the multiplexed power detection front end is used to perform envelope detection, and the moment when the received signal envelope crosses the preset power threshold is used as a trigger event to trigger the hardware timestamp capture module to obtain the physical layer timestamp of the receiving end.

4. The method according to claim 1, characterized in that, Construct four asymmetric spatial geometric propagation paths, consisting of the master station to the satellite, the satellite to the remote station, the remote station to the satellite, and the satellite to the master station, including: Using the ephemeris model, the three-dimensional orbital position of the satellite at each time corresponding to the round-trip timestamp set is extracted; By utilizing the satellite's three-dimensional orbital position and the known master station position, the communication link from the master station to the remote station is deconstructed into forward and reverse spatial propagation paths of unequal length. A geometric constraint association is then established, which includes the forward and reverse spatial propagation paths and the unidirectional distance to the remote station to be determined.

5. The method according to claim 4, characterized in that, Before calculating the one-way distance and clock offset of the remote station, the following steps are also included: During the system hold-up time slot of the communication frame structure, the master station transmits and receives self-loopback test signals that are transparently relayed by the satellite. The total round-trip propagation time of the self-loop test signal is obtained, and the round-trip spatial geometric delay from the master station to the satellite is deducted using the ephemeris model, as well as the pre-configured inherent radio frequency delay within the master station, to extract the dynamic group delay of the real-time tracked satellite transponder.

6. The method according to claim 5, characterized in that, The calculation of the one-way distance and clock offset of the remote station includes: Extract the total round-trip time from the round-trip timestamp set, subtract the round-trip spatial geometric delay from the main station to the satellite and twice the satellite transponder dynamic group delay from the total round-trip time, convert the remaining delay into spatial distance and take half of the value to calculate the one-way distance of the remote station. Calculate the asymmetric path delay difference between the forward and reverse spatial propagation paths; The initial clock skew is constructed using the extracted transmit and receive timestamp difference, and the initial clock skew is compensated using the asymmetric path delay difference to calculate the remote station clock skew.

7. The method according to claim 1, characterized in that, After adjusting the remote station's transmission timing based on timing advance and remote station clock deviation to synchronize the remote station's time reference with the master station, the process also includes: Based on the synchronized adjustment of the remote station's transmission timing, the transmission burst time jitter of service signals sent from remote stations across the network to the master station will be converged to the preset target accuracy range. Based on the convergent transmission burst time jitter, the frame structure of the communication system is adjusted in a cascade manner to compress the time span of the protection time slot between adjacent service time slots.

8. The method according to claim 1, characterized in that, In the exchange of ranging signals carrying session identifiers between the master station and the remote station, a two-step separation mode is used to process the physical acquisition of timestamps and the transmission of values, including: Interacting with event messages that do not carry timestamp values ​​is used to trigger the capture operation of the physical layer timestamp at the physical layer. Furthermore, subsequent messages are exchanged after the event message, with each subsequent message carrying the captured physical layer timestamp value. Moreover, the subsequent messages employ a channel coding scheme with higher redundancy than regular service data for error protection.

9. The method according to claim 1, characterized in that, When the remote station does not receive the time signal from the global satellite navigation system, the local working clock of the remote station periodically locks and calibrates itself by receiving control commands from the master station via the satellite communication link, based on timing advance and remote station clock deviation, to maintain the synchronization of the time reference between the remote station and the master station.

10. The method according to claim 3, characterized in that, Envelope detection using a multiplexed power detection front end includes: The low-noise amplifier input of the receiving link shares a broadband envelope detector as a power detection front end. At the moment when the broadband envelope detector detects that the signal burst front edge crosses the preset power threshold, a burst synchronization trigger signal is sent to the demodulator of the communication baseband, and the counting freeze action of the hardware timestamp capture module is triggered in parallel.

Images