A multi-application method and device based on double-channel parallel reception in satellite communication

Abstract

The application discloses a multi-application method and device based on double-channel parallel receiving in satellite communication, wherein the method configures a first receiving channel and a second receiving channel as a full-bandwidth receiving mode or a double-satellite communication mode according to an application scenario; in the full-bandwidth receiving mode, according to a total bandwidth of a downlink working frequency band and a maximum processing bandwidth of a single baseband processing unit, the continuous downlink working frequency band is divided into adjacent and non-overlapping first and second sub-bands, and local oscillator frequencies of the first and second receiving channels are respectively set as center frequencies of the first and second sub-bands, so as to parallelly receive, down-convert and sample corresponding sub-band signals through the first and second receiving channels. In the case that the processing bandwidth of the single baseband processing unit is smaller than the complete downlink working bandwidth, the complete downlink working frequency band can be simultaneously utilized through double-channel parallel receiving, time alignment and digital domain splicing, so as to improve the downlink receiving bandwidth and the spectrum utilization rate.

Description

Technical Field

[0001] This invention relates to the field of satellite communication technology, and in particular to a multi-application method and apparatus based on dual-channel parallel reception in satellite communication. Background Technology

[0002] With the rapid deployment of low-Earth orbit satellite internet systems such as the GW constellation and Qianfan constellation in China, satellite communication terminals, especially handheld terminals, portable IoT terminals, and vehicle-mounted terminals, are facing increasing demands for communication speeds and higher requirements for network continuity. In satellite communication systems, improving downlink speeds, efficiently utilizing spectrum resources, and ensuring uninterrupted service during satellite handovers have become urgent issues that need to be addressed on the terminal side.

[0003] In scenarios such as rapid distribution of remote sensing satellite data, high-definition video-on-demand on the internet, and large file downloads, downlink traffic is often much greater than uplink traffic. However, to control cost, power consumption, and size, the maximum processing bandwidth of the terminal's baseband module is usually less than the full downlink operating bandwidth supported by the RF front-end and antenna. For example, the RF front-end can support 80MHz bandwidth, while a single channel of the baseband module can only process 40MHz bandwidth, causing the terminal to be unable to utilize the full downlink spectrum simultaneously, making it difficult to achieve the corresponding channel capacity in downlink speed.

[0004] To address the aforementioned issues, existing solutions typically include replacing the baseband module with one that has a larger processing bandwidth, using frequency sweeping or frequency hopping to receive different frequency bands in a time-division multiplexing manner, or using subband selective filtering to receive only a portion of the full operating frequency band. However, replacing the baseband module with a high-bandwidth one increases hardware cost, power consumption, and terminal size; frequency sweeping or frequency hopping sacrifices real-time throughput; and subband selective filtering prevents some spectrum from being utilized simultaneously.

[0005] In low-Earth orbit satellite constellations, the overpass visibility time of a single satellite is typically only 5 to 15 minutes, requiring terminals to frequently switch between adjacent satellites. Traditional single-channel terminals usually employ a "disconnect-then-reconnect" switching method, meaning that the search for and access to the next overpass satellite only begins after the signal of the currently serving satellite is lost or the link quality deteriorates. During the search, time-frequency synchronization, and network registration periods, there may be network interruptions of several seconds to tens of seconds, thus affecting real-time services such as video calls, remote control, and IoT data backhaul.

[0006] Carrier aggregation technology can aggregate multiple component carriers to improve data transmission rate. However, carrier aggregation usually relies on network-side configuration and scheduling, and is mainly aimed at the aggregation of multiple component carriers. It is difficult to directly solve the problem that the terminal side cannot process continuous and complete downlink working frequency bands at the same time due to the limited bandwidth of single-channel baseband processing. It is also difficult to directly solve the problem of switching between low-orbit satellites after connection.

[0007] Chinese utility model patent CN219181509U discloses a shortwave multi-channel parallel receiving and processing system. This system employs multiple receiving channels to receive signals across the entire shortwave band. Each receiving channel filters and amplifies a portion of the band, and the multiple receiving channels operate in parallel to cover the entire shortwave band. The multi-channel digital processing module uses multiple analog-to-digital converters to sample the parallel multi-channel signals and utilizes an FPGA device for multi-channel digital down-conversion processing, outputting multiple wideband digital down-converted signals covering the entire shortwave band. This system is primarily used for shortwave signal reconnaissance or monitoring to reduce the dynamic range requirements of receiving devices for shortwave wideband reception.

[0008] Chinese invention patent CN112953539B discloses a method for implementing a large-scale planar array multi-beamforming system. This method amplifies, mixes, and bandpass-filters the signal received by the planar antenna array to obtain multiple beam signals of different frequencies, which are then multiplexed and combined. A lens is then used to simulate beamforming on the multiplexed signal. After electro-optical conversion and fiber optic transmission to a remote location, the multiplexed signal undergoes photoelectric conversion, down-conversion, analog-to-digital conversion, demultiplexing, and digital beamforming in sequence, thereby reducing the complexity of the large-scale planar array multi-beamforming system.

[0009] Chinese invention patent CN118381553B discloses a satellite communication method, system, and apparatus based on auxiliary cells. When the terminal supports carrier aggregation, this scheme acquires the terminal's location information and determines the intersection area of ​​the frequency division duplex (FDM) cell and time division duplex (TDM) cell corresponding to the satellite base station. When the terminal is located in the intersection area, an auxiliary cell confirmation command is generated to confirm the auxiliary TDM cell. This enables the terminal and the satellite base station to establish a TDM signal channel when preset judgment conditions are met, and data is transmitted through the FDM and TDM signal channels to improve the data transmission rate of the satellite communication network.

[0010] US Patent 6865238B2 discloses a system and method for combining multiple satellite channels into a virtual composite channel. This scheme divides the original signal into multiple sub-channels on the transmitting side according to a power-segmentation or content-segmentation method, allowing these sub-channels to be transmitted through one or more satellite channels. The receiving terminal simultaneously receives the multiple sub-channels, determines the relative time delay between each sub-channel using a timing signal on a separate carrier or superimposed on an information carrier, and combines the multiple sub-channels into the original signal after time synchronization.

[0011] While existing technologies involve solutions such as multi-channel parallel reception, beam signal multiplexing, auxiliary cell carrier aggregation, and synchronous merging of multiple satellite sub-channels, multi-channel parallel reception still faces engineering challenges when the single-channel baseband processing bandwidth of a satellite communication terminal is less than the full downlink operating bandwidth. These challenges include local oscillator frequency deviation, sampling clock asynchrony, amplitude and phase inconsistency between channels, adjacent channel leakage at sub-band boundaries, and digital domain splicing delay misalignment. During LEO satellite handover, further solutions are needed for pre-searching for the next passing satellite, time-frequency synchronization, network registration, handover threshold judgment, and rapid switching of service data paths. Therefore, a technical solution is required that can improve the full downlink spectrum utilization under limited baseband processing bandwidth and maintain service continuity during LEO satellite handover using dual channels. Summary of the Invention

[0012] The purpose of this invention is to overcome at least some of the shortcomings of the prior art and provide a multi-application method and apparatus for satellite communication based on dual-channel parallel reception. When the maximum processing bandwidth of a single channel of the baseband module is less than the full downlink operating bandwidth supported by the radio frequency front-end and antenna, dual independent receiving channels are used to perform parallel reception and digital domain splicing of continuous downlink operating frequency bands to improve downlink spectrum utilization and real-time throughput. At the same time, the dual independent receiving channels are used to track the current serving satellite and the next passing satellite respectively, and the time-frequency synchronization and network registration of the next passing satellite are completed in advance while the current service link remains operational, so as to reduce service interruption caused by low-orbit satellite passing handover.

[0013] The technical solution of the present invention is as follows: The present invention provides a multi-application method based on dual-channel parallel reception in satellite communication, which is applied to a satellite communication terminal. The satellite communication terminal is provided with a first receiving channel, a second receiving channel, a baseband controller and a digital domain merging module. The first receiving channel and the second receiving channel respectively include independently configured receiving antennas, radio frequency front-ends, mixers, filters, amplifiers, analog-to-digital converters and baseband processing units, and the first receiving channel and the second receiving channel share the same reference clock source.

[0014] Furthermore, depending on the application scenario, the first receiving channel and the second receiving channel are configured as either full-bandwidth receiving mode or dual-satellite communication mode.

[0015] Furthermore, in the full-bandwidth reception mode, based on the total bandwidth of the downlink operating frequency band and the maximum processing bandwidth of a single baseband processing unit, the continuous downlink operating frequency band is divided into a first sub-band and a second sub-band with sequentially arranged frequencies that do not overlap. The first sub-band and the second sub-band are directly connected at the boundary frequency, or a guard interval is set between the first sub-band and the second sub-band. The local oscillator frequencies of the first receiving channel and the second receiving channel are respectively set as the center frequencies of the first sub-band and the second sub-band, so that the corresponding sub-band signals can be received, down-converted, and sampled in parallel through the first receiving channel and the second receiving channel. During the system initialization phase, pre-programmed signals are injected into the first receiving channel and the second receiving channel. Knowing the calibration signal, determine the amplitude calibration coefficient and phase calibration coefficient between the two receiving channels, and perform digital domain compensation on at least one sub-band signal according to the amplitude calibration coefficient and the phase calibration coefficient during normal reception; detect the frame header or pilot sequence of the two sub-band signals respectively, determine the arrival time difference of the two sub-band signals, and adjust the read pointer of the FIFO buffer according to the arrival time difference to align the two sub-band data in time; when the guard interval is set between the first sub-band and the second sub-band, remove the data corresponding to the guard interval between the first sub-band and the second sub-band, and splice the time-aligned two sub-band data in ascending frequency order to obtain a merged data stream corresponding to the downlink operating frequency band.

[0016] Furthermore, in the dual-satellite communication mode, the first receiving channel is configured to track the currently serving satellite, and the second receiving channel is configured to search for and track the next passing satellite based on the low-orbit satellite ephemeris. While the first receiving channel continues to carry service data, carrier acquisition, time-frequency synchronization, and network registration with the next passing satellite are completed through the second receiving channel. When the second receiving channel meets the handover preparation conditions and the currently serving satellite meets the handover execution conditions, the service data path is switched from the first receiving channel to the second receiving channel, and the first receiving channel is switched to standby mode or scanning mode.

[0017] Furthermore, in the full-bandwidth reception mode, the reference clock source is a temperature-compensated crystal oscillator or a temperature-controlled crystal oscillator. The reference clock source is connected to the sampling clock input terminals of the two analog-to-digital converters and the reference input terminals of the two phase-locked loops respectively through a low-skew clock buffer. The two phase-locked loops generate a first local oscillator signal and a second local oscillator signal according to the reference clock source. The frequency difference between the second local oscillator signal and the first local oscillator signal is equal to the maximum processing bandwidth of a single baseband processing unit. When the phase difference drift between the two local oscillator signals exceeds a preset threshold, the phase-locked loop is triggered to recalibrate or undergo digital domain phase compensation.

[0018] Further, determining the amplitude calibration coefficient and the phase calibration coefficient includes: simultaneously injecting a single-carrier calibration signal located at the two sub-band boundary frequencies into the first receiving channel and the second receiving channel through a built-in calibration signal source, and acquiring the I / Q data output by the two receiving channels; determining the amplitude ratio based on the ratio of the root mean square amplitudes of the two I / Q data, and determining the phase offset based on the average phase angle of the conjugate product of one I / Q data and the other I / Q data; storing the calibration coefficients, including the amplitude ratio and the phase offset, together with the temperature value during calibration in a non-volatile memory after cyclic redundancy check; and performing amplitude scaling and phase rotation on one of the I / Q data channels during normal reception based on the calibration coefficients and a preset temperature compensation table.

[0019] Furthermore, the guard interval is set between the first sub-band and the second sub-band; adjacent sub-band signals are suppressed by radio frequency filters of the two receiving channels, and out-of-band leakage components are filtered out by digital filters after analog-to-digital conversion; the baseband controller dynamically adjusts the size of the guard interval according to the modulation and coding scheme in the downlink control signaling, and determines the data to be removed during splicing according to the guard interval.

[0020] Further, determining the arrival time difference of the two sub-band signals includes: performing sliding correlation detection on the two sub-band signals using a unique word or PN sequence in the downlink frame header, and recording the first sampling point number and the second sampling point number corresponding to the correlation peak; performing parabolic interpolation or sinc interpolation near the correlation peak, and performing median filtering on the time difference measurement results of multiple consecutive frames to obtain the arrival time difference; the baseband controller determines the starting offset of the FIFO read pointer relative to the write pointer of the first arriving channel according to ΔN=round(Δt×R_bit), where ΔN is the number of bits that need to be delayed, Δt is the arrival time difference, and R_bit is the bit rate of the corresponding sub-band data.

[0021] Furthermore, a subband identifier field is appended to the frame header of the two subband data streams to distinguish between the low-frequency and high-frequency subbands. The digital domain merging module writes the low-frequency and high-frequency subband data into the merging FIFO in ascending order of frequency according to the subband identifier field, and renumbers the concatenated data frame with a continuity counter. Cyclic redundancy check is performed on the merged data stream. When it is determined from the subband identifier field that an error has occurred in the data of one of the subbands, a selective retransmission request is initiated only for the subband with the error.

[0022] Furthermore, the satellite communication terminal calculates the arrival time and azimuth of the next transiting satellite based on the pre-stored low-orbit satellite ephemeris, and activates the second receiving channel when the remaining visibility time of the currently serving satellite is less than or equal to a preset time; the second receiving channel determines the initial Doppler frequency offset based on the low-orbit satellite ephemeris, and sequentially completes carrier frequency acquisition and tracking, code synchronization and frame synchronization, physical broadcast channel decoding, random access, connection establishment, and network registration.

[0023] Furthermore, a pre-synchronization initiation threshold, a handover readiness threshold, and a handover execution threshold are used for handover determination: when the received signal strength of the currently serving satellite is lower than the pre-synchronization initiation threshold or its remaining visibility time is less than a preset time, pre-synchronization of the second receiving channel is initiated; when the signal-to-noise ratio of the second receiving channel is higher than the handover readiness threshold and network registration is completed, it is determined that the second receiving channel meets the handover readiness conditions; when the signal-to-noise ratio of the currently serving satellite is lower than the handover execution threshold and remains below the preset hold time, it is determined that the handover execution conditions are met; after the handover is completed, a handover is allowed only when the signal quality of the currently serving satellite recovers to the handover return threshold and remains below the preset hold time.

[0024] Furthermore, switching the service data path from the first receiving channel to the second receiving channel includes: switching the data path selector connected to the upper-layer interface from the baseband output of the first receiving channel to the baseband output of the second receiving channel by writing to a hardware register; switching the RF front-end of the first receiving channel to standby mode or scanning mode, and updating the IP routing table and MAC address mapping in the network protocol stack; within the overlapping window where the current serving satellite and the next overtaking satellite simultaneously cover the satellite communication terminal, dynamically adjusting the allocation ratio of the two data streams according to the channel quality of the two receiving channels, and reordering the two received data streams according to the sequence number at the IP layer or PDCP layer.

[0025] The present invention also provides a multi-application device based on dual-channel parallel reception in satellite communication, for executing the above-mentioned multi-application method based on dual-channel parallel reception in satellite communication, including a first receiving channel, a second receiving channel, a reference clock source, a baseband controller, a calibration module, a frame synchronization module, a FIFO buffer, a digital domain merging module, and a data path selector.

[0026] Furthermore, the first receiving channel and the second receiving channel each include independently configured receiving antennas, RF front-ends, mixers, filters, amplifiers, analog-to-digital converters, and baseband processing units; the reference clock source is connected to the analog-to-digital converters and local oscillator generation circuits of the two receiving channels respectively, and is used to provide the two receiving channels with the same clock source; the calibration module is used to determine the amplitude calibration coefficient and phase calibration coefficient between the two receiving channels based on the known calibration signals injected into the two receiving channels, and to perform digital domain compensation on at least one received data channel; the frame synchronization module is used to determine the arrival time difference based on the frame header or pilot sequence in the two received data channels, and the baseband controller is used to adjust the read pointer of the FIFO buffer based on the arrival time difference.

[0027] Furthermore, the digital domain merging module is used to remove the data corresponding to the guard interval between two adjacent subbands in full-bandwidth reception mode, and to concatenate the two time-aligned subband data into a merged data stream in ascending frequency order; the baseband controller is also used to enable the first receiving channel to track the currently serving satellite in dual-satellite communication mode, enable the second receiving channel to search for and track the next passing satellite according to the low-orbit satellite ephemeris, and control the data path selector to switch the service data path from the first receiving channel to the second receiving channel when the second receiving channel completes time-frequency synchronization and network registration and meets the switching conditions.

[0028] The beneficial effects of this invention are as follows: When the processing bandwidth of a single baseband processing unit is less than the full downlink operating bandwidth, by using dual-channel parallel reception, time alignment, and digital domain splicing, the entire downlink operating frequency band can be utilized simultaneously, improving downlink receiving bandwidth, spectrum utilization, and real-time throughput; by sharing a reference clock source, local oscillator synchronization, amplitude and phase calibration, frame synchronization detection, and FIFO delay compensation, the impact of sampling clock deviation, channel amplitude and phase differences, and processing delay differences on digital domain splicing can be reduced; by using RF filtering, guard intervals, and digital domain auxiliary filtering, adjacent channel leakage at subband boundaries can be reduced; the same dual-channel receiving architecture can be used for full downlink bandwidth reception of the same satellite, and can also be used for... Independent tracking and service handover of two low-Earth orbit (LEO) satellites improves the reuse of hardware resources. While the first receiving channel continues to carry service data, the second receiving channel pre-completes the search, time-frequency synchronization, and network registration for the next passing satellite, and switches the service data path after the handover conditions are met. This reduces service interruptions caused by the "disconnect first, reconnect later" approach and improves network continuity during LEO satellite handover. By using pre-synchronization start threshold, handover readiness threshold, handover execution threshold, hold time, and back-switch threshold for handover judgment, frequent handovers when the signal quality of the two satellites is similar can be reduced. Furthermore, downlink throughput can be improved within the overlapping coverage window of the two satellites through data aggregation and packet reordering. Attached Figure Description

[0029] Figure 1 This is a flowchart of a multi-application method based on dual-channel parallel reception in satellite communication according to an embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram of downlink operating frequency band allocation and multi-channel parallel reception in a multi-application method based on dual-channel parallel reception in satellite communication according to an embodiment of the present invention.

[0031] Figure 3 This is a system structure block diagram of a multi-application device based on dual-channel parallel reception in satellite communication according to an embodiment of the present invention.

[0032] Figure 4 This is a schematic diagram of dual-channel local oscillator, clock synchronization, and calibration in a multi-application method based on dual-channel parallel reception in satellite communication according to an embodiment of the present invention.

[0033] Figure 5 This is a schematic diagram illustrating an application scenario of simultaneous dual-satellite communication and seamless service switching in a multi-application method based on dual-channel parallel reception in satellite communication according to an embodiment of the present invention. Detailed Implementation

[0034] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0035] Please refer to the following: Figures 1 to 5 This invention provides a multi-application method and apparatus for satellite communication based on dual-channel parallel reception. It can be applied to handheld terminals, portable IoT terminals, vehicle-mounted terminals, airborne terminals, or shipborne terminals, and can be used for rapid distribution of remote sensing satellite data, internet high-definition video-on-demand, large file downloads, video calls, remote control, and IoT data backhaul. This invention divides a continuous downlink operating frequency band into multiple sub-bands using frequency division, without spatial segmentation, beamforming, or phased array antennas. Multiple sub-bands are received in parallel by multiple independent receiving channels on the terminal side, and time alignment and frequency stitching are performed in the digital domain. This method employs a time-division receiving approach, using parallel simultaneous reception rather than frequency sweeping or frequency hopping, and does not rely on special carrier aggregation configurations on the satellite or base station side.

[0036] like Figure 1 and Figure 3As shown, the satellite communication terminal includes a first receiving channel, a second receiving channel, a reference clock source, a baseband controller, a calibration module, a frame synchronization module, a FIFO buffer, a digital domain merging module, and a data path selector. The first and second receiving channels each include independently configured receiving antennas, RF front-ends, mixers, filters, amplifiers, analog-to-digital converters, and baseband processing units. The maximum processing bandwidth of the baseband processing unit in each receiving channel is adapted to the processing capability of the baseband module, and the two receiving channels share the same reference clock source.

[0037] The reference clock source is connected to the analog-to-digital converter and local oscillator generation circuit of each of the two receiving channels to provide the same clock source to the two receiving channels; the calibration module is used to determine the amplitude calibration coefficient and phase calibration coefficient between the two receiving channels based on the known calibration signal injected into the two receiving channels, and to perform digital domain compensation on at least one received data channel; the frame synchronization module is used to determine the arrival time difference based on the frame header or pilot sequence in the two received data channels; the baseband controller is used to adjust the read pointer of the FIFO buffer based on the arrival time difference.

[0038] The digital domain merging module is used to remove data corresponding to the guard interval between two adjacent subbands in full-bandwidth reception mode, and to concatenate the time-aligned data from the two subbands into a merged data stream in ascending frequency order. The baseband controller is also used in dual-satellite communication mode to enable the first receiving channel to track the currently serving satellite, and the second receiving channel to search for and track the next passing satellite based on the low-Earth orbit satellite ephemeris. When the second receiving channel completes time-frequency synchronization and network registration and meets the switching conditions, the controller controls the data path selector to switch the service data path from the first receiving channel to the second receiving channel.

[0039] The baseband controller configures the first and second receiving channels to either full-bandwidth receiving mode or dual-satellite communication mode based on the application scenario. Full-bandwidth receiving mode is used when the single-channel processing bandwidth of the baseband module is less than the full downlink operating bandwidth, allowing parallel processing of adjacent subbands through the two receiving channels to restore the full bandwidth data stream. Dual-satellite communication mode enables the two receiving channels to communicate independently with two low-Earth orbit satellites, supporting service switching (connection before disconnection) and data aggregation within the overlapping coverage window of the two satellites.

[0040] Example 1: Rapid distribution of remote sensing satellite data: This embodiment corresponds to Figures 1 to 4 This example uses the rapid distribution of low-Earth orbit (LEO) remote sensing satellite data as an application scenario. A LEO remote sensing satellite captures a high-resolution SAR image, with a data size of 800MB, which needs to be directly transmitted to a ground-based authorized terminal within 8 minutes of its overhead transit. In this embodiment, the downlink operating frequency band is 2.12GHz–2.20GHz, with a total bandwidth of 80MHz, and the maximum single-channel processing bandwidth of the baseband module is 40MHz.

[0041] Step 1: System initialization and dual independent receive channel configuration: After the terminal is powered on, the baseband controller reads the pre-stored configuration parameters: downlink operating frequency start frequency F_low=2.12GHz, end frequency F_high=2.20GHz, total bandwidth BW_total=80MHz, and maximum single-channel baseband processing bandwidth BW_ch=40MHz. Based on the total bandwidth and the maximum single-channel baseband processing bandwidth, the required number of parallel channels N=2 is determined, and the first and second receiving channels are configured in full-bandwidth receiving mode.

[0042] like Figure 2 As shown, the complete downlink operating frequency band is divided into a continuous and non-overlapping first sub-band and a second sub-band. The first sub-band is F_low to F_low+BW_ch, i.e., 2.12GHz to 2.16GHz, with a center frequency f_c1=2.14GHz; the second sub-band is F_high-BW_ch to F_high, i.e., 2.16GHz to 2.20GHz, with a center frequency f_c2=2.18GHz. The two sub-bands can be adjacent at the boundary frequency of 2.16GHz, or a guard interval of 200kHz to 500kHz can be set at the boundary frequency.

[0043] The first and second receiving channels each have independent receiving antennas, RF front-ends, mixers, filters, amplifiers, analog-to-digital converters, and baseband processing units. Each baseband processing unit has a maximum processing bandwidth of 40MHz. The first receiving channel processes the first sub-band, and the second receiving channel processes the second sub-band. Both receiving channels share the same high-precision reference clock source.

[0044] Step 2: Dual-channel local oscillator, clock, amplitude, phase, and fixed delay calibration: like Figure 4 As shown, the two receiving channels share the same high-precision temperature-compensated crystal oscillator (TCXO) or oven-controlled crystal oscillator (OCXO) as the reference clock source, ensuring that the sampling clocks of the two analog-to-digital converters (ADCs) originate from the same source. In the high-precision clock allocation method, the frequency stability of the main reference clock source is better than ±0.1ppm, and the phase noise is better than -145dBc / Hz@1kHz. The main reference clock source is fanned out simultaneously to the sampling clock inputs of the two ADCs and the reference inputs of the two phase-locked loops (PLLs) via a low-skew clock buffer. The low-skew clock buffer can be a TI LMK04832 or an ADI HMC7044, with a channel skew of less than 50ps, aligning the sampling clock edges of the two ADCs and ensuring that the phase detection frequencies of the two PLLs originate from the same source.

[0045] The local oscillator frequency of the first receiving channel is set to the center frequency of the first sub-band, 2.14 GHz, and the local oscillator frequency of the second receiving channel is set to the center frequency of the second sub-band, 2.18 GHz, enabling each receiving channel to downconvert the RF signal of its corresponding sub-band to zero intermediate frequency (IF) or low IF. The two local oscillator signals are generated by frequency division using the same phase-locked loop (PLL) circuit, or by two synchronous PLLs using the same reference clock. By configuring the division ratio, LO2 - LO1 = BW_ch, meaning the frequency difference between the second and first local oscillator signals equals the maximum processing bandwidth of a single baseband processing unit. After PLL locking detection is completed, the phase difference between the two local oscillator signals is monitored by a digital phase detection circuit; when the phase difference drift exceeds a preset threshold, such as ±5°, PLL recalibration or digital domain phase compensation is triggered.

[0046] During the terminal factory calibration phase or the system power-on self-test phase, a known calibration signal is simultaneously injected into the RF front-end input terminals of both receiving channels via a built-in calibration signal source. The calibration signal can be a single-carrier signal or a PN sequence. In one calibration method, a single-carrier calibration signal with a frequency located at the two sub-band boundary frequency fc and a power of -30dBm±0.5dB is generated by the DDS; the two receiving channels simultaneously receive the calibration signal, and the baseband processing unit acquires two I / Q data streams with a duration of not less than 1ms, calculating the amplitude calibration coefficient and phase calibration coefficient using the following formula.

[0047] The formula for calculating the amplitude calibration coefficient is: Amplitude_Ratio =RMS(|I1+jQ1|) / RMS(|I2+jQ2|)(1) The formula for calculating the phase calibration coefficient is: Phase_Offset =mean(angle((I1+jQ1)×conj(I2+jQ2)))(2) The amplitude ratio α is determined based on the root mean square ratio of the amplitudes of the two I / Q data streams, and the phase offset Δφ is determined based on the average phase angle of the conjugate product of one I / Q data stream and the other I / Q data stream. The calibration coefficients, including the amplitude ratio α and the phase offset Δφ, are stored in non-volatile memory, such as the calibration partition of SPI Flash, after being checked by a 32-bit CRC check, along with a calibration timestamp and temperature sensor readings. During normal operation, the baseband processing unit reads the calibration coefficients, multiplies the data from the second receiving channel by 1 / α for amplitude scaling, and multiplies by exp(-jΔφ) for phase rotation. Temperature drift of the calibration coefficients is corrected according to a preset temperature compensation table, ensuring digital domain compensation accuracy of amplitude better than 0.1 dB and phase better than 1°.

[0048] During the initial calibration phase, the total group delay difference τ_diff from the RF input to the baseband digital output of the two receiving channels is also measured. A wideband calibration signal, which can be a linear frequency modulated signal or a PN sequence, is simultaneously injected into both receiving channels. After simultaneous acquisition by both channels, the delay difference is determined through cross-correlation calculation. The calculation formula is as follows: τ_diff = argmax_τ |corr(s1(t),s2(t+τ))|(3) The total group delay difference τ_diff includes the group delay difference of the RF front-end filter, the pipeline delay difference of the analog-to-digital converter, and the delay difference of the digital front-end processing. After storing the τ_diff calibration value, the fixed delay difference is used as the initial bias value for FIFO read pointer compensation during normal reception, and the slow delay fluctuations caused by temperature changes and device aging are dynamically compensated.

[0049] In the initial configuration of this embodiment, the two receiving channels share a TCXO reference clock source with a frequency of 40MHz and a frequency stability of ±0.1ppm. The phase-locked loop of the first receiving channel is set to LO1=2.14GHz, and the phase-locked loop of the second receiving channel is set to LO2=2.18GHz. A single-carrier calibration signal with a frequency of 2.16GHz and a power of -30dBm is injected through the RF test interface. Both receiving channels receive the signal simultaneously. The baseband processing unit measures the I / Q amplitude and phase of the two channels, calculates the amplitude calibration coefficients α1 and α2 and the phase calibration coefficients θ1 and θ2, and stores them in non-volatile memory. During normal operation, the calibration coefficients are read in real time for compensation.

[0050] Step 3: Sub-band boundary protection and adjacent channel leakage suppression: The RF front-end filters of both receiving channels exhibit a steep roll-off characteristic at the subband boundary frequency of 2.16 GHz. In a typical configuration, the suppression at the boundary frequency is no less than 40 dB, ensuring that each receiving channel only transmits the target subband signal. A guard interval of 200 kHz to 500 kHz is set between the two subbands, and a digital low-pass filter or digital band-pass filter is used in the digital domain after analog-to-digital conversion to filter out out-of-band leakage components, ensuring that the two subband signals do not overlap in the frequency domain.

[0051] The two RF filters have matched group delay characteristics within their respective passbands, with the group delay difference between the two filters being less than 1 / 20 of the symbol period; for a 2MHz symbol rate, the group delay difference is less than 25ns. The RF filters can be SAW filter pairs with matched group delay characteristics, such as two adjacent SAW devices on the same wafer; alternatively, a group delay equalizer can be used in the digital domain to compensate for the group delay difference and avoid phase discontinuities at the splicing point.

[0052] For suppressing the adjacent channel leakage ratio (ACLR) at the subband boundary frequency fc±Δf / 2, the suppression degree of the first receiving channel for the second subband signal is no less than 50dB, and the suppression degree of the second receiving channel for the first subband signal is no less than 50dB, ensuring that the mutual leakage power between subbands is more than 50dB lower than the signal power, resulting in an additional EVM degradation of less than 0.5%. The stopband attenuation of the digital domain FIR filter is no less than 80dB to further suppress the adjacent channel leakage remaining after analog filtering.

[0053] The guard interval Δf is dynamically adjusted based on the modulation scheme and signal-to-noise ratio of the actual received signal. When the received signal is QPSK or low-order modulation, the guard interval can be reduced to 100kHz; when the received signal is 64QAM or higher-order modulation, the guard interval increases to 500kHz. The baseband controller parses the modulation and coding scheme (MCS) information in the downlink control signaling and automatically adjusts the guard interval based on the MCS information.

[0054] When stitching together data in the digital domain, the number of sampling points corresponding to the guard interval Δf is determined according to the following formula: N_guard =Δf / (BW_ch / N_fft) (4) In this process, the data stream of the first receiving channel retains only the demodulated data corresponding to the frequency band [fc-BW_ch / 2, fc-Δf / 2], and the data stream of the second receiving channel retains only the demodulated data corresponding to the frequency band [fc+Δf / 2, fc+BW_ch / 2]. The removed data is indicated by the idle padding bits in the frame format. The concatenated data streams are renumbered sequentially, and the continuity counter is updated in the frame header, enabling the upper-layer protocol stack to continuously reassemble the data.

[0055] In this embodiment, a guard interval of Δf = 300kHz is set at the subband boundary of 2.16GHz. The actual range of the first subband is 2.12GHz to 2.15985GHz, and the actual range of the second subband is 2.16015GHz to 2.20GHz. The RF front-end SAW filters of both receiving channels have a suppression level of not less than 45dB at 2.16GHz. After analog-to-digital conversion, the first receiving channel uses an FIR low-pass filter with an order of 128 and a cutoff frequency of 19.85MHz, and the second receiving channel uses an FIR low-pass filter with an order of 128 and a cutoff frequency of 20.15MHz.

[0056] Step 4: Parallel reception and down-conversion of sub-band signals: The first receiving antenna receives an RF signal containing the complete 80MHz downlink signal. The RF front-end filter of the first receiving channel extracts the first sub-band signal, amplifies it through a low-noise amplifier (LNA), and mixes it with the first local oscillator signal to generate a zero-IF or low-IF I / Q signal. The second receiving antenna synchronously receives the complete downlink frequency band RF signal. The RF front-end filter of the second receiving channel extracts the second sub-band signal, amplifies it through a low-noise amplifier, and mixes it with the second local oscillator signal to generate a second zero-IF or low-IF I / Q signal. Both signals are then sampled by their respective analog-to-digital converters (ADCs) using the same sampling clock after passing through a programmable gain amplifier (PGA) and a low-pass filter. The sampling rate of the ADCs is not less than 80Msps, and 81.92Msps can be used.

[0057] In this embodiment, the first receiving channel filters out the sub-band signal from 2.12 GHz to 2.15985 GHz, amplifies it with a low-noise amplifier, and mixes it with LO1 = 2.14 GHz to generate a zero-IF I / Q signal with a bandwidth of approximately 39.85 MHz. This signal is then sampled by an analog-to-digital converter (ADC) at a sampling rate of 81.92 Msps after passing through a programmable gain amplifier and a low-pass filter. The second receiving channel extracts the sub-band signal from 2.16015 GHz to 2.20 GHz, mixes it with LO2 = 2.18 GHz, and similarly processes it through a programmable gain amplifier, a low-pass filter, and an ADC to output the second digital baseband signal. The sampling clocks of the two ADCs are driven by the same reference clock source through a clock buffer; in the receiving configuration of this embodiment, the sampling clock skew is no greater than 100 ps.

[0058] Step 5: Baseband processing, frame synchronization alignment, and FIFO dynamic delay compensation: The two digital baseband signals are fed into two independent processing units of the baseband module. Each processing unit performs channel estimation, equalization, demodulation, and decoding, and outputs the corresponding bit stream and frame synchronization information. Since the processing delays of the two receiving channels may differ, frame synchronization detection is performed on the two digital baseband signals before digital domain splicing.

[0059] Two frame synchronization detection modules perform sliding correlation detection using the unique word UW or PN sequence in the downlink frame header. When the correlation peak is greater than 0.9 and the sidelobe is less than 0.3, the sampling point numbers N1 and N2 of the two frame headers are recorded. The arrival time difference between the two frame headers is determined by the following formula: Δt = |N1-N2|×Ts(5) Where Ts is the sampling period of the analog-to-digital converter; at a sampling rate of 81.92 Msps, Ts is approximately 12.2 ns. To improve measurement accuracy, three-point parabolic interpolation or sinc interpolation is performed near the relevant peak to achieve a sub-sampling point level of time difference measurement accuracy of less than 0.1 Ts. To avoid abnormal results caused by false detections or missed detections in the frame header, median filtering is applied to the time difference measurement results of M consecutive frames, where M is not less than 5.

[0060] In the baseband processing pipeline of each processing channel, a depth-configurable FIFO buffer is set between the demodulated bitstream and the digital domain merging module. The typical depth of the FIFO buffer is 2048 bits, which can buffer a maximum delay difference of approximately 25 μs at a bit rate of 80 Mbps. The write pointer of the FIFO is driven by the demodulated output clock of the corresponding channel, and the read pointer is driven by a unified time base corresponding to the reference time after frame synchronization.

[0061] The baseband controller applies an additional delay to the FIFO read pointer of the first arriving channel based on the measured arrival time difference Δt. The number of bits that the start address of the read pointer lags behind the write pointer is determined by the following formula: ΔN=round(Δt×R_bit)(6) Where ΔN is the number of bits that need to be delayed, and R_bit is the bit rate of the corresponding subband data. Data arriving at the channel first is buffered in the FIFO by Δt before being read out, ensuring that the two data streams are aligned in the time domain with an alignment accuracy better than ±1 bit cycle. The baseband controller continuously monitors the FIFO's water level line, i.e., the difference between the write pointer and the read pointer; when temperature drift or clock jitter causes the water level line to deviate beyond a preset threshold, such as ±5 bits, the read pointer is automatically fine-tuned for dynamic compensation. The fixed total group delay difference τ_diff obtained from the initial calibration is used as the initial bias for read pointer compensation.

[0062] Step 6: Digital domain frequency splicing, integrity verification and error handling Each processing channel outputs a bitstream with a 2-bit subband identifier field appended to the frame header, where 00 indicates a low-frequency subband and 01 indicates a high-frequency subband. Even if the data processing order of the two receiving channels changes, the digital domain merging module can distinguish the corresponding subband data based on the subband identifier field.

[0063] The digital domain merging module writes low-frequency and high-frequency subband data into the merging FIFO in ascending frequency order according to the subband identifier field. The first bit stream corresponds to the low-frequency band data from 2.12GHz to 2.15985GHz, and the second bit stream corresponds to the high-frequency band data from 2.16015GHz to 2.20GHz. During splicing, the data corresponding to the 300kHz guard interval is removed, and the continuity counter of the data frame is renumbered. The merging FIFO outputs a serial data stream corresponding to the complete 80MHz downlink operating frequency band, and the merged data stream is sent to the upper-layer protocol stack.

[0064] After concatenation, the merged data stream is checked frame by frame using CRC-24 verification. If the CRC check passes, the merged data stream is sent to the upper-layer protocol stack normally; if the CRC check fails, recovery is performed according to the error type. If the sub-band identifier field indicates a data error in a single sub-band, a selective retransmission request (ARQ) is initiated only for the erroneous sub-band, while the data in the normal sub-band remains unchanged; alternatively, forward error correction (FEC) can be used to recover the erroneous sub-band. If the error is caused by FIFO overflow or underload within the digital domain merging module, a synchronization realignment process for the two receiving channels is triggered. Selective retransmission of only erroneous sub-bands can save approximately 50% of retransmission bandwidth overhead compared to full-bandwidth retransmission.

[0065] Step 7: Full-bandwidth receive performance verification: In the channel synchronization performance test, a TCXO reference clock with a frequency stability of ±0.1ppm and a TILMK04832 clock buffer were used. The measured sampling clock skew of the two analog-to-digital converters was 38ps. During the initial calibration phase, the amplitude ratio between the channels was measured to be α=1.03, corresponding to a difference of approximately 0.26dB, and the phase offset Δφ=-2.8°. After digital domain compensation, the amplitude consistency of the two I / Q signals was no greater than 0.15dB, and the phase consistency was no greater than ±1.5°.

[0066] In the splicing accuracy test with SNR=18dB and QPSK modulation, the arrival time difference Δt between the two frame headers was 3.2 sampling periods, approximately 39ns. After FIFO read pointer compensation, the alignment accuracy was no greater than 0.08 sampling periods. For a symbol rate of 2MHz, this alignment time was no greater than 40ns, corresponding to no greater than 3.2 sampling periods. The BER of the merged data stream after splicing was 1.2×10⁻⁶. -6 The ideal single-channel 80MHz full-bandwidth receiver has a BER of 1.1×10⁻⁶. -6 The equivalent SNR loss corresponding to the difference between the two is less than 0.08dB.

[0067] With a guard interval configuration of Δf = 300kHz, the effective bandwidth is 80MHz - 0.3MHz = 79.7MHz, and the spectral efficiency is 79.7 / 80 = 99.625%. Using a dynamic guard interval and adjusting Δf to 100kHz during QPSK modulation, the spectral efficiency is 79.9 / 80 = 99.875%. Single-channel 40MHz reception utilizes only 40MHz of the total 80MHz bandwidth, resulting in a spectral efficiency of 50%.

[0068] When two sub-bands are adjacent at the boundary frequency and no guard interval is set, the two receiving channels can simultaneously receive the full 80MHz downlink operating bandwidth. Compared to the single-channel mode that can only receive 40MHz, the downlink receiving bandwidth is extended from 40MHz to 80MHz, the spectrum utilization is increased from 50% to 100%, and the theoretical downlink rate is increased by 100%.

[0069] In single-channel reception mode, the terminal only receives the 40MHz sub-band, with an equivalent downlink rate of R, for example, 80Mbps. 800MB of data needs to be transmitted in two overpasses, 400MB each time. The terminal needs to complete the reception after at least two overpasses, resulting in a total time exceeding 16 minutes. In dual-channel parallel reception mode, the terminal simultaneously receives two 40MHz sub-bands, with an equivalent downlink rate of 2R, or 160Mbps. This allows for the reception of 800MB of remote sensing data within a single 8-minute overpass window.

[0070] Under general performance specifications, using a high-precision reference clock source with a frequency stability of ±0.1ppm and a low-skew clock distribution network, the sampling clock skew of the two analog-to-digital converters can be kept to no more than 50ps. After calibration, the local oscillator phase coherence deviation is no more than ±3°, the inter-channel amplitude consistency is no more than 0.2dB, and the phase consistency is no more than ±2°. When the SNR is not less than 15dB, after frame synchronization alignment, FIFO delay compensation, and frequency splicing, the alignment accuracy is no more than one symbol period, and the equivalent signal-to-noise ratio loss corresponding to the difference between the BER of the combined data stream and the BER of the ideal single-channel full-bandwidth reception is less than 0.1dB.

[0071] Example 2: Simultaneous communication and seamless handover between two satellites: This embodiment corresponds to Figure 1 , Figure 3 and Figure 5 This technology is applied to low-Earth orbit (LEO) satellite internet terminals, such as airborne and shipborne terminals. In a LEO satellite constellation, the overpass visibility time of a single satellite is typically 5–15 minutes, and the overlap coverage window between two adjacent satellites is typically 1–5 minutes. The baseband controller configures the two receiving channels in dual-satellite communication mode.

[0072] Step 1: Dual-star independent channel configuration and ephemeris-assisted pre-synchronization: like Figure 5 As shown, the first receiving channel is configured as a service channel, continuously tracking the downlink signal of the currently serving satellite A and carrying real-time service data such as video calls and data backhaul; the second receiving channel is configured as a scanning channel, pre-searching for and tracking the downlink beacon or broadcast channel of the next satellite B that is about to pass overhead. The two receiving channels independently complete frequency tracking, time-frequency synchronization, and signal demodulation, respectively.

[0073] The terminal calculates the arrival time and azimuth of the next transiting satellite B based on the pre-stored LEO constellation ephemeris. The LEO constellation ephemeris can be either two lines of orbital parameters (TLE) or a precise ephemeris. When more than half of the current serving satellite A's transit time has passed, or when its remaining visibility time (T_remain) is no more than 3 minutes, the baseband controller initiates the pre-synchronization process for the second receiving channel.

[0074] The second receiving channel determines the initial Doppler frequency offset based on the ephemeris of low-Earth orbit satellites. The ephemeris-aided initial Doppler frequency offset estimation accuracy is better than ±2kHz, reducing the carrier frequency search range of the second receiving channel from ±150kHz to ±10kHz, and shortening the frequency acquisition time from the second level to the hundreds of milliseconds level.

[0075] After the second receiving channel initiates pre-synchronization, the following access preparation processes are completed sequentially: (a) carrier frequency acquisition and tracking with Doppler pre-compensation is performed in less than 50ms; (b) code synchronization and frame synchronization are completed using the primary synchronization signal (PSS) and secondary synchronization signal (SSS) in the downlink synchronization channel in less than 100ms; (c) the physical broadcast channel (PBCH) is decoded to obtain system information such as the system frame number (SFN) and bandwidth configuration in less than 160ms; (d) the random access procedure, including PRACH preamble transmission, random access response reception, and RRC connection establishment, is completed in less than 500ms; (e) authentication and IP address allocation are completed when necessary in less than 1s. The entire access preparation process can be completed within 1-2 seconds, which is less than the remaining visible time window of 3-5 minutes for low-Earth orbit satellites. During the access preparation process performed by the second receiving channel, the first receiving channel continuously carries service data.

[0076] Step 2: Three-level switching threshold and hysteresis protection: The handover decision-making process employs three thresholds: a pre-synchronization initiation threshold, a handover readiness threshold, and a handover execution threshold. The pre-synchronization initiation threshold is triggered when the RSSI of the currently serving satellite A is lower than Th_high (e.g., -85dBm), or when the remaining coverage time of satellite A is less than 3 minutes. The handover readiness threshold is triggered when the SNR of the second receiving channel is higher than Th_ready (e.g., 10dB), and the second receiving channel has completed network registration. The target link is confirmed to be available when these conditions are met. The handover execution threshold is triggered when the SNR of satellite A is lower than Th_low (e.g., 3dB), and the SNR remains below the threshold for a period greater than T_hold (e.g., 2 seconds).

[0077] After the handover is executed, a handover is only allowed when the signal quality of satellite A recovers to the handover threshold and remains above T_hold. The handover threshold can be set higher than the pre-synchronization start threshold Δhysteresis. For example, if the RSSI of satellite A is higher than -82dBm, which is 3dB higher than the pre-synchronization start threshold of -85dBm, and remains above it for more than 2 seconds, this three-level handover threshold, hold time, and handover threshold reduce ping-pong handovers that occur when the signal quality of the two satellites is close.

[0078] Step 3: Switching Business Data Paths When the second receiving channel meets the handover preparation conditions and the currently serving satellite A meets the handover execution conditions, the baseband controller performs a hardware-level service data path handover: (a) by writing to the hardware register, the data path selector connected to the upper-layer interface is switched from the baseband output of the first receiving channel to the baseband output of the second receiving channel. The upper-layer interface can be an Ethernet MAC layer interface or a PCIe interface. The typical time for this operation is less than 10 μs; (b) the RF front-end of the first receiving channel is switched to standby mode or scanning mode to release its hardware resources. The time is less than 50 μs; (c) the IP routing table and MAC address mapping in the network protocol stack are updated. The time is less than 100 μs. The total handover execution time is less than 1 ms, and the jitter increment of the upper-layer TCP / UDP service is less than 0.5 ms.

[0079] After switching, the first receiving channel enters sleep mode, standby mode, or scan mode, preparing for subsequent searches and switching. In deep sleep mode, the power consumption of the idle channel is reduced by more than 80%, for example, from 2.5W to less than 0.5W; compared with the normal operating mode, the power consumption of the idle channel can be reduced to less than 20%.

[0080] In one switching determination method of this embodiment, when the RSSI or SNR of satellite A is lower than a preset switching threshold, for example, when the SNR is less than 5dB, the terminal controller issues a switching command to switch the service data stream from the first receiving channel to the second receiving channel that has completed synchronization and registration with satellite B. The first receiving channel stops receiving data from satellite A and enters standby mode or scanning mode.

[0081] In the handover latency test, the time taken for the baseband output selector to write to the 16-bit hardware register via the SPI interface was 8μs, the time taken for the RF front-end of the first receiving channel to switch to power-off or standby mode via the PMIC voltage rail was 35μs, the time taken for the IP routing table software update was 80μs, and the time taken for the MAC address mapping to be written via the hardware CAM table was 15μs. The total handover execution time was 8+35+80+15=138μs, which is less than 1ms.

[0082] In the laboratory, the process of gradually weakening satellite A signal and gradually strengthening satellite B signal was simulated using an attenuator, with a total of 1000 dual-satellite handovers performed. Before each handover, the second receiving channel completed pre-synchronization and network registration, and the number of service data packet losses in the 1000 handovers was 0. In contrast, when using a single-channel handover with a disconnect-then-reconnect approach, the average number of packet losses in 1000 handovers was approximately 15200, corresponding to a search and registration time of approximately 3.8 seconds and a data rate of 4 Mbps, with a packet loss rate of approximately 0.38%.

[0083] In the boundary region where the SNR difference between satellite A and satellite B is less than 1 dB, a hysteresis threshold Δhysteresis = 3 dB and a hold time T_hold = 2 s are set. During 100 consecutive attempts to enter the signal proximity scenario, ping-pong handovers occur 0 times. When hysteresis protection is disabled and Δhysteresis = 0 dB and T_hold = 0 s are set, ping-pong handovers occur 73 times out of 100 tests.

[0084] Step 4: Data aggregation and flow control within the double-star overlapping coverage window: Within the overlapping window where both satellite A and satellite B simultaneously cover the terminal, two receiving channels receive data from both satellites at the same time. At the IP or PDCP layer, the two received data streams are reordered based on the packet sequence number (SN) and the corresponding satellite identifier to eliminate packet out-of-order issues caused by the different transmission paths of the two satellites.

[0085] The baseband controller dynamically adjusts the data allocation ratio between the two receiving channels based on the instantaneous channel capacity of each channel. The instantaneous channel capacity is determined through CQI channel quality indication feedback. When the channel quality of satellite A is better than that of satellite B, satellite A is allocated more data in the downlink scheduling request. Due to the difference in channel quality between the two receiving channels within the overlapping window, the aggregated downlink throughput can reach 1.5 to 1.8 times that of the single-satellite mode. In embodiments where the channel conditions of the two receiving channels are comparable, for example, when the overlapping coverage window is 2 minutes, the aggregated downlink throughput can approach twice that of the single-satellite mode, making it suitable for large-volume burst transmissions within the overlapping window.

[0086] In this embodiment, the total execution time of the dual-satellite handover is less than 1ms, including hardware handover and protocol stack update. When the second receiving channel has completed synchronization and network registration before handover, the packet loss rate during handover can be 0, the TCP / UDP service interruption time can be 0ms, and the handover jitter is less than 0.5ms. A comparison with traditional solutions is shown in the table below.

[0087]

[0088] Please refer to the following: Figure 3 The present invention also provides a multi-application device based on dual-channel parallel reception in satellite communication for executing the above-described method. The device includes a first receiving channel, a second receiving channel, a reference clock source, a baseband controller, a calibration module, a frame synchronization module, a FIFO buffer, a digital domain merging module, and a data path selector.

[0089] The first and second receiving channels each include independently configured receiving antennas, RF front-ends, mixers, filters, amplifiers, analog-to-digital converters (ADCs), and baseband processing units. Reference clock sources are connected to the ADCs and local oscillator generation circuits of both receiving channels, providing a common clock source to both channels. The calibration module determines amplitude and phase calibration coefficients based on known calibration signals injected into the two receiving channels and performs digital domain compensation on at least one received data stream. The frame synchronization module determines the arrival time difference based on the frame headers or pilot sequences in the two received data streams, and the baseband controller adjusts the read pointer of the FIFO buffer based on the arrival time difference.

[0090] In full-bandwidth reception mode, the digital domain merging module removes the data corresponding to the guard interval between two adjacent subbands and concatenates the time-aligned data from the two subbands into a merged data stream in ascending frequency order. In dual-satellite communication mode, the baseband controller controls the first receiving channel to track the currently serving satellite, and controls the second receiving channel to search for and track the next passing satellite based on the low-Earth orbit satellite ephemeris. When the second receiving channel completes time-frequency synchronization and network registration and meets the switching conditions, the controller controls the data path selector to switch the service data path from the first receiving channel to the second receiving channel.

[0091] This invention is not limited to two receiving channels. Based on the total bandwidth of the complete downlink operating frequency band and the maximum processing bandwidth of a single baseband processing unit, three, four, or more independent receiving channels can be set in the terminal, dividing the complete downlink operating frequency band into N consecutive and non-overlapping sub-bands. The bandwidth of each sub-band is equal to or slightly less than the maximum processing bandwidth of a single baseband processing unit and is allocated to the corresponding receiving channel. By increasing the number of receiving channels, the downlink operating bandwidth that the terminal can receive simultaneously can be expanded.

[0092] The aforementioned full-bandwidth reception and dual-satellite communication are both handled by the terminal-side dual-channel hardware, baseband controller, and digital domain processing, without requiring special carrier aggregation configuration, scheduling, or MAC layer aggregation from the satellite or base station side. While maintaining the existing single-channel processing bandwidth of the baseband module, the downlink reception bandwidth can be expanded by adding an independent reception channel and combining digital domain processing, without replacing the entire existing baseband module with a higher bandwidth one. This can be used for upgrading already deployed terminals.

[0093] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. All modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multi-application method based on dual-channel parallel reception in satellite communication, applied to a satellite communication terminal, wherein the satellite communication terminal is provided with a first receiving channel, a second receiving channel, a baseband controller, and a digital domain merging module, wherein the first receiving channel and the second receiving channel respectively include independently configured receiving antennas, radio frequency front-ends, mixers, filters, amplifiers, analog-to-digital converters, and baseband processing units, and the first receiving channel and the second receiving channel share the same reference clock source, characterized in that, include: Depending on the application scenario, the first receiving channel and the second receiving channel are configured as either full-bandwidth receiving mode or dual-satellite communication mode. In the full-bandwidth reception mode, based on the total bandwidth of the downlink operating frequency band and the maximum processing bandwidth of a single baseband processing unit, the continuous downlink operating frequency band is divided into a first sub-band and a second sub-band with sequentially arranged frequencies that do not overlap. The first sub-band and the second sub-band are directly connected at the boundary frequency, or a guard interval is set between the first sub-band and the second sub-band. The local oscillator frequencies of the first receiving channel and the second receiving channel are respectively set as the center frequencies of the first sub-band and the second sub-band, so that the corresponding sub-band signals can be received, down-converted, and sampled in parallel through the first receiving channel and the second receiving channel. During system initialization, known calibration signals are injected into the first and second receiving channels to determine the amplitude and phase calibration coefficients between the two receiving channels. During normal reception, digital domain compensation is performed on at least one sub-band signal based on the amplitude and phase calibration coefficients. The frame headers or pilot sequences of the two sub-band signals are detected respectively to determine the arrival time difference between the two sub-band signals. The read pointer of the FIFO buffer is adjusted according to the arrival time difference to align the two sub-band data in time. When the guard interval is set between the first and second sub-bands, the data corresponding to the guard interval between the first and second sub-bands is removed, and the time-aligned two sub-band data are spliced ​​in ascending frequency order to obtain a merged data stream corresponding to the downlink operating frequency band. In the dual-satellite communication mode, the first receiving channel is configured to track the currently serving satellite, and the second receiving channel is configured to search for and track the next passing satellite based on the low-orbit satellite ephemeris. While the first receiving channel continues to carry service data, the second receiving channel completes carrier acquisition, time-frequency synchronization, and network registration with the next passing satellite. When the second receiving channel meets the handover preparation conditions and the currently serving satellite meets the handover execution conditions, the service data path is switched from the first receiving channel to the second receiving channel, and the first receiving channel is switched to standby mode or scanning mode.

2. The multi-application method based on dual-channel parallel reception in satellite communication according to claim 1, characterized in that, In the full-bandwidth reception mode, the reference clock source is a temperature-compensated crystal oscillator or a temperature-controlled crystal oscillator. The reference clock source is connected to the sampling clock input terminals of the two analog-to-digital converters and the reference input terminals of the two phase-locked loops through a low-skew clock buffer. The two phase-locked loops generate a first local oscillator signal and a second local oscillator signal according to the reference clock source. The frequency difference between the second local oscillator signal and the first local oscillator signal is equal to the maximum processing bandwidth of a single baseband processing unit. When the phase difference between the two local oscillator signals drifts beyond a preset threshold, the phase-locked loop is triggered to recalibrate or undergo digital domain phase compensation.

3. The multi-application method based on dual-channel parallel reception in satellite communication according to claim 2, characterized in that, Determining the amplitude calibration coefficient and the phase calibration coefficient includes: simultaneously injecting a single-carrier calibration signal located at the two sub-band boundary frequencies into the first receiving channel and the second receiving channel through a built-in calibration signal source, and acquiring the I / Q data output by the two receiving channels; determining the amplitude ratio based on the root mean square ratio of the amplitudes of the two I / Q data, and determining the phase offset based on the average phase angle of the conjugate product of one I / Q data and the other I / Q data; storing the calibration coefficients, including the amplitude ratio and the phase offset, together with the temperature value during calibration in a non-volatile memory after cyclic redundancy check; and performing amplitude scaling and phase rotation on one of the I / Q data channels during normal reception based on the calibration coefficients and a preset temperature compensation table.

4. The multi-application method based on dual-channel parallel reception in satellite communication according to claim 1, characterized in that, The guard interval is set between the first sub-band and the second sub-band; the radio frequency filters of the two receiving channels suppress the signals of adjacent sub-bands, and the out-of-band leakage components are filtered out by the digital filter after analog-to-digital conversion; the baseband controller dynamically adjusts the size of the guard interval according to the modulation and coding scheme in the downlink control signaling, and determines the data to be removed during splicing according to the guard interval.

5. The multi-application method based on dual-channel parallel reception in satellite communication according to claim 3, characterized in that, Determining the arrival time difference between two sub-band signals includes: performing sliding correlation detection on the two sub-band signals using a unique word or PN sequence in the downlink frame header, and recording the first sampling point number and the second sampling point number corresponding to the correlation peak; performing parabolic interpolation or sinc interpolation near the correlation peak, and performing median filtering on the time difference measurement results of multiple consecutive frames to obtain the arrival time difference; the baseband controller determines the starting offset of the FIFO read pointer relative to the write pointer of the first arriving channel according to ΔN=round(Δt×R_bit), where ΔN is the number of bits that need to be delayed, Δt is the arrival time difference, and R_bit is the bit rate of the corresponding sub-band data.

6. The multi-application method based on dual-channel parallel reception in satellite communication according to claim 4 or 5, characterized in that, Subband identifier fields, used to distinguish between low-frequency and high-frequency subbands, are appended to the frame headers of the two subband data streams. The digital domain merging module writes the low-frequency and high-frequency subband data into the merging FIFO in ascending order of frequency according to the subband identifier fields, and renumbers the concatenated data frames with a continuity counter. Cyclic redundancy check is performed on the merged data stream. When it is determined from the subband identifier fields that an error has occurred in the data of one of the subbands, a selective retransmission request is initiated only for the subband with the error.

7. The multi-application method based on dual-channel parallel reception in satellite communication according to claim 1, characterized in that, The satellite communication terminal calculates the arrival time and azimuth of the next transiting satellite based on the pre-stored low-orbit satellite ephemeris, and activates the second receiving channel when the remaining visible time of the currently serving satellite is less than or equal to a preset time. The second receiving channel determines the initial Doppler frequency offset based on the low-orbit satellite ephemeris, and sequentially completes carrier frequency acquisition and tracking, code synchronization and frame synchronization, physical broadcast channel decoding, random access, connection establishment, and network registration.

8. The multi-application method based on dual-channel parallel reception in satellite communication according to claim 7, characterized in that, The handover decision is made using a pre-synchronization start threshold, a handover preparation ready threshold, and a handover execution threshold: when the received signal strength of the currently serving satellite is lower than the pre-synchronization start threshold or its remaining visible time is less than a preset time, the pre-synchronization of the second receiving channel is initiated. When the signal-to-noise ratio of the second receiving channel is higher than the handover readiness threshold and network registration is completed, it is determined that the second receiving channel meets the handover readiness conditions; when the signal-to-noise ratio of the current serving satellite is lower than the handover execution threshold and continues to exceed the preset hold time, it is determined that the handover execution conditions are met; after the handover is completed, a handover is allowed only when the signal quality of the current serving satellite recovers to the handover back threshold and continues to exceed the preset hold time.

9. The multi-application method based on dual-channel parallel reception in satellite communication according to claim 8, characterized in that, Switching the service data path from the first receiving channel to the second receiving channel includes: switching the data path selector connected to the upper-layer interface from the baseband output of the first receiving channel to the baseband output of the second receiving channel by writing to a hardware register; switching the RF front-end of the first receiving channel to standby mode or scanning mode, and updating the IP routing table and MAC address mapping in the network protocol stack; within the overlapping window where the current serving satellite and the next overtaking satellite simultaneously cover the satellite communication terminal, dynamically adjusting the allocation ratio of the two data streams according to the channel quality of the two receiving channels, and reordering the two received data streams according to the sequence number at the IP layer or PDCP layer.

10. A multi-application device based on dual-channel parallel reception in satellite communication, characterized in that, The method for performing a multi-application method based on dual-channel parallel reception in satellite communication as described in any one of claims 1 to 9 includes a first receiving channel, a second receiving channel, a reference clock source, a baseband controller, a calibration module, a frame synchronization module, a FIFO buffer, a digital domain merging module, and a data path selector. The first receiving channel and the second receiving channel each include independently configured receiving antennas, radio frequency front-ends, mixers, filters, amplifiers, analog-to-digital converters, and baseband processing units; The reference clock source is connected to the analog-to-digital converter and the local oscillator generation circuit of the two receiving channels respectively, and is used to provide the two receiving channels with the same clock source; The calibration module is used to determine the amplitude calibration coefficient and phase calibration coefficient between the two receiving channels based on the known calibration signal injected into the two receiving channels, and to perform digital domain compensation on at least one received data channel. The frame synchronization module is used to determine the arrival time difference based on the frame header or pilot sequence in the two received data streams, and the baseband controller is used to adjust the read pointer of the FIFO buffer based on the arrival time difference. The digital domain merging module is used to remove the data corresponding to the guard interval between two adjacent subbands in full-bandwidth reception mode, and to splice the time-aligned two subband data into a merged data stream in ascending frequency order; The baseband controller is also used in dual-satellite communication mode to enable the first receiving channel to track the currently serving satellite, enable the second receiving channel to search for and track the next passing satellite according to the low-orbit satellite ephemeris, and control the data path selector to switch the service data path from the first receiving channel to the second receiving channel when the second receiving channel completes time-frequency synchronization and network registration and meets the switching conditions.

Images