Method and apparatus for detecting dkab signals in a satellite communication system

Abstract

The application discloses a kind of detection methods of DKAB signal in satellite communication system, comprising the following steps: low-pass filter;Filtering to input signal using 22 order FIR low-pass filter;Extraction;Peak detection;The highest signal-to-noise ratio position is detected by sliding window, and the timing offset of the integer multiple of the input 4 groups of data is calculated;Timing adjustment;Time offset statistics;Symbol demodulation;Hard decision;Frequency offset estimation;Each segment burst has a reference symbol before, and the symbol record after inverse modulation is obtained;After removing modulation symbol, the radian is accumulated according to column to modulation symbol, and the frequency offset value is calculated;Decoding;Frequency offset statistics.The algorithm defined by the application, i.e.low-pass filter plus 1 / 4 extraction, can realize 4 times downsampling, set search window w to traverse the decimal time offset of entire subframe, and the peak detection method based on signal-to-noise ratio, frequency offset estimation and frequency offset statistics are used to realize high-performance detection of DKAB signal, which can exclude the influence of noise and is more accurate.

Description

Technical Field

[0001] This invention relates to the field of satellite communication system technology, specifically to a method for detecting DKAB signals in a satellite communication system. Background Technology

[0002] The European satellite communication standard GMR (GEO-Mobile Radio Interface) was developed under the leadership of the European Telecommunications Standards Institute (ETSI). It evolved from the GSM technology framework and was designed specifically for geostationary orbit (GEO) satellites. In its early stages, it focused on voice calls and low-speed data services (such as SMS) and was used in systems such as Thuraya (Middle East) and TerreStar (North America).

[0003] Compared to continuously transmitting service data, DKAB reduces resource consumption and optimizes satellite system capacity through a minimal burst structure (containing only a small number of symbols).

[0004] The DKAB signal employs π / 4-DBPSK (Differential Binary Phase Shift Keying) modulation, combined with a root-raised cosine filter with a roll-off factor of 0.35 for pulse shaping. This modulation method exhibits resistance to frequency offset, adapting to the Doppler shift effects in satellite channels. It consists of two KAB (Keep-Alive Burst) signals, has no unique words, a small number of symbols, and a simple structure.

[0005] For example, Chinese patent CN 116015412A discloses a discontinuous communication method for a satellite mobile communication system. It improves the existing DKAB burst format by transferring the background noise characteristic information carried by DKAB to the NT3 channel for periodic transmission. The DKAB channel only carries unique words and power control information. It uses sliding correlation technology based on unique words to obtain accurate timing, and calculates the frequency offset by using the phase difference between two KAB bursts. It can track satellite networks using DKAB when the symbol signal-to-noise ratio is low, and the DKAB synchronization performance is good.

[0006] Chinese patent CN 116015412A discloses a demodulation method for DKAB in a satellite mobile communication system, including TOA detection, timing synchronization, symbol recovery, soft information extraction, and frequency offset estimation. After the TOA detection module detects the presence of burst signals and their approximate arrival times, the timing module corrects the timing and outputs the accurate arrival time. After symbol recovery, soft information is extracted and sent to the decoding module, while the system frequency offset is estimated and sent to the frequency correction module, thus achieving DKAB demodulation. However, this method has low detection performance, and the bit error rate significantly affects the accuracy of frequency offset estimation.

[0007] Existing DKAB signal detection methods suffer from high implementation complexity, inferior detection performance compared to other logical channels, or require protocol modifications to achieve high performance. Summary of the Invention

[0008] The purpose of this invention is to provide a method and apparatus for detecting DKAB signals in a satellite communication system. Based on the DKAB signal structure given by GMR-1, after downsampling the received signal, this invention utilizes a custom algorithm—namely, low-pass filtering plus 1 / 4 decimation to achieve 4x downsampling; a search window w to traverse the entire subframe using fractional time offset traversal; a peak detection method based on signal-to-noise ratio; frequency offset estimation and frequency offset statistics; etc.—to achieve high-performance detection of DKAB signals. This invention can eliminate the influence of noise and achieve higher accuracy.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a method and apparatus for detecting DKAB signals in a satellite communication system, comprising the following steps:

[0010] Step S1, low-pass filtering;

[0011] A 22nd-order FIR low-pass filter is used to filter the input signal;

[0012] Step S2, extraction;

[0013] One sampling point is extracted for every 4 points, and the sampling starts from the starting point and is traversed; from the time offsets corresponding to 0, 1 / 4, 1 / 2, and 3 / 4 decimals, 4 sets of data with a sampling rate of 23.4k are extracted and output.

[0014] Step S3, peak detection;

[0015] The position with the highest signal-to-noise ratio is detected by sliding window, and the timing offset of the four sets of input data is calculated as an integer multiple.

[0016] Step S4, timed adjustment;

[0017] For the four sets of highest signal-to-noise ratio positions and integer multiple timing offsets obtained in step S3, search for the maximum signal-to-noise ratio, its corresponding timing offset, and the corresponding set of extracted data. Adjust the timing and use it as the input for subsequent modules.

[0018] Step S5, time-biased statistics;

[0019] Calculate the time offset value of the subframe based on the data from step S4;

[0020] Step S6, symbol demodulation;

[0021] Extract the data symbols from the DKAB burst, calculate the differential phase, and calculate the likelihood ratio after inverse rotation by π / 4.

[0022] Step S7, hard decision;

[0023] In the likelihood ratio, values ​​less than 0 are bit 1, and all others are 0.

[0024] Step S8, frequency offset estimation;

[0025] Using the transmit bit information obtained in step S7, transmit symbols are modulated according to the protocol. Each burst segment is preceded by a reference symbol to obtain the demodulated symbol. After removing the modulation symbols, the modulation symbols are accumulated column by column to calculate the frequency offset value.

[0026] Step S9, Decode;

[0027] Based on continuous sets of DKAB soft information, likelihood ratio concatenation is extracted, and after deinterleaving, Gray decoding is used to obtain the decoding result.

[0028] Step S10, frequency offset statistics;

[0029] Store a filtered frequency offset value. The frequency offset value estimated for each subframe is filtered and recorded as the new frequency offset value. The filter first calculates the difference between the current estimated frequency offset value and the historical frequency offset value. If the result exceeds the set deviation threshold, no filtering is performed; otherwise, filtering is performed. Finally, the frequency offset value of the current subframe is obtained.

[0030] More preferably, in step S3, peak detection specifically includes:

[0031] Step 1: Set up the sliding window The total length of the cable window is... When performing blind detection, a sliding window W needs to be set to traverse the entire subframe;

[0032] Step 2: Calculate the total power of the signal in the entire subframe;

[0033] Right now ,in This represents the total signal power of the entire subframe. express The row vector of the DKAB signal has 117 symbol periods, represented as... ; for The conjugate transpose of the signal, the average power of which is expressed as ;

[0034] Step 3: Calculate the received signal power of the two bursts at each search location;

[0035] Right now ;in, The received signal power of the two bursts at each search location, for The row vector, express The conjugate transpose of; the burst average power is expressed as ;

[0036] Step 4: For burst average power Its corresponding average noise power Represented as: ;

[0037] Step 5: The average power of the actual burst signal is expressed as... The corresponding signal-to-noise ratio for ;

[0038] Step 6: Search Maximum value, corresponding index ,use Two dots on the left and right -1 is used for linear interpolation, resulting in a final timing offset. for: );

[0039] in, This indicates an index. The corresponding signal-to-noise ratio, i.e. Maximum value; Indicates index The corresponding signal-to-noise ratio; Indicates index The corresponding signal-to-noise ratio; This means rounding off;

[0040] A further preferred option is to replace Step5 with: Step5':burst, where the actual average signal power is expressed as... , ;

[0041] Replace Step6 with: Step6': Search Maximum value, corresponding index ,use Two dots on the left and right -1 is used for linear interpolation, and the final timing offset is:

[0042] );in, Indicates index The corresponding average power of the actual burst signal; Indicates index The corresponding average power of the actual burst signal; Indicates index The corresponding average power of the actual burst signal.

[0043] More preferably, in step S4, the signal-to-noise ratio of the four groups is... and the corresponding integer multiple timing offset are Search signal-to-noise ratio The maximum value, and its corresponding timing offset is The corresponding set of data after a 4x extraction, with adjusted timing. This will then be used as input for subsequent modules.

[0044] More preferably, in step S5, the time offset of the subframe The calculation formula is: / 4, of which Traverse from the starting point. .

[0045] More preferably, in step S6, the data symbol representation extracted from the DKAB burst is... Calculate the differential phase and reverse rotation. The likelihood ratio was then calculated. for: ;in Indicates the differential phase value; This represents the final LLR value; It refers to the department that seeks truth.

[0046] More preferably, in step S8, the demodulated symbol is recorded for: Remove modulation symbols After; among them, This indicates the data symbols in the DKAB burst mentioned above. * indicates finding the conjugate of elements in the matrix, and for modulation symbols... Calculate the radians by summing the values ​​in each column. , ,in and For matrix The first and second lines, Represents the phase difference; the frequency offset value is calculated. for: .

[0047] More preferably, in step S10, the frequency offset value of the current subframe for: ,in, This represents the current estimated value. This represents the filter coefficients.

[0048] A further preferred approach is to replace step S7, the hard decision, with:

[0049] Step S7', buffer; buffer 6 consecutive subframes. Symbol, denoted as ;

[0050] Replace step S8 with: step S8', frequency offset estimation. When the CRC check is correct, the correctly decoded result is decoded, interleaved, and modulated according to the protocol to generate the transmission symbols for each subframe. Each subframe is processed as follows:

[0051] Each burst segment is preceded by a reference symbol, and the symbol after demodulation is denoted as: .

[0052] Remove modulation symbols , right Calculate radians by summing the columns: ;

[0053] The frequency offset value was calculated. for: .

[0054] The present invention also provides an apparatus for detecting DKAB signals in the above-mentioned satellite communication system, comprising a low-pass filtering module, a decimation module, a peak detection module, a timing adjustment module, a time-frequency statistics module, a symbol demodulation module, a hard decision module, a frequency offset estimation module, a decoding module, and a frequency offset statistics module;

[0055] The low-pass filter module is used to filter the input signal using a 22nd-order FIR low-pass filter.

[0056] The extraction module is used to extract one sampling point every four points, and the extraction starts from the starting point and iterates through the points. It extracts and outputs four sets of data with a sampling rate of 23.4k from the time offsets corresponding to multiples of 0, 1 / 4, 1 / 2, and 3 / 4.

[0057] The peak detection module is used to detect the position of the highest signal-to-noise ratio through a sliding window and calculate the timing offset as an integer multiple of the four sets of input data.

[0058] The timing adjustment module uses the peak detection module to obtain four sets of the highest signal-to-noise ratio positions and integer multiple timing offsets, searches for the maximum signal-to-noise ratio value and its corresponding timing offset and the corresponding set of extracted data, and adjusts the timing as input for subsequent modules.

[0059] The time-frequency statistics module is used to calculate the time offset value of the subframe based on the data from the timing adjustment module;

[0060] The symbol demodulation module is used to extract data symbols from the DKAB burst, calculate the differential phase, and calculate the likelihood ratio after inverse rotation by π / 4.

[0061] The hard decision module is used to record bit 1 for values ​​less than 0 in the likelihood ratio and 0 for others;

[0062] The frequency offset estimation module is used to use the transmitted bit information obtained by the hard decision module to modulate the transmitted symbols according to the protocol. Each burst segment is preceded by a reference symbol to obtain the demodulated symbol. After removing the modulation symbols, the modulation symbols are accumulated column by column to calculate the radian value and obtain the frequency offset value.

[0063] The decoding module is used to extract likelihood ratio concatenation based on multiple sets of continuous DKAB soft information, and after deinterleaving, Gray decoding is used to obtain the decoding result;

[0064] The frequency offset statistics module is used to store a filtered frequency offset value. The frequency offset value estimated for each subframe is filtered and recorded as a new frequency offset value. The filter first calculates the difference between the current estimated frequency offset value and the historical frequency offset value. If the result exceeds the set deviation threshold, no filtering is performed; otherwise, filtering is performed. Finally, the frequency offset value of the current subframe is obtained.

[0065] Compared with the prior art, the beneficial effects of the present invention are:

[0066] 1. Through extensive experimental analysis, this invention demonstrates that the detection performance of DKAB signals is determined by the accuracy of timing and frequency offset estimation, as well as the bit error rate. Therefore, based on this, this invention addresses the DKAB signal structure given in GMR-1. After downsampling the received signal, it employs a custom algorithm—namely, low-pass filtering plus 1 / 4 decimation to achieve 4x downsampling, setting a search window w to traverse the entire subframe using fractional time offset traversal, peak detection based on signal-to-noise ratio, frequency offset estimation and statistics, etc.—to achieve high-performance detection of DKAB signals. This invention can eliminate the influence of noise, resulting in higher accuracy.

[0067] 2. The low-pass filtering plus 1 / 4 decimation of the present invention can achieve 4x downsampling, and traversing the starting point can provide input data for the subsequent measurement module to find the optimal fractional sampling point, thereby improving accuracy.

[0068] 3. The peak detection method of this invention uses a sliding window to detect the position of the highest signal-to-noise ratio and confirms the timing of integer multiples of the four sets of input data. Compared with the power detection method, this method can eliminate the influence of noise and has higher accuracy.

[0069] 4. The timing adjustment of this invention selects the data sampled from four sets of detection data by the optimal fractional multiple based on the signal-to-noise ratio, which can further eliminate the influence of noise and thus improve accuracy.

[0070] 5. The symbol demodulation and hard decision of this invention obtain the transmitted information bits by calculating the differential phase; the frequency offset estimation obtains the transmitted symbols by demodulation, thereby removing the influence of information bits from the received symbols and achieving the purpose of calculating the frequency offset, which has higher accuracy and reliability.

[0071] 6. The frequency offset statistics of the present invention achieve smoothing of frequency offset values ​​through Kalman filtering, and by setting a deviation threshold, unreliable frequency offset estimates are eliminated, thereby increasing the reliability of the frequency offset filter values.

[0072] 7. In Embodiment 3 of the present invention, the correct information bits are obtained after soft demodulation decoding and CRC verification, which increases the correctness of the inversion symbol and thus increases the correctness of frequency offset estimation.

[0073] 8. The principle of this invention is simple, easy to implement, and has low complexity. Its detection performance is higher than that of other logical channels, and it does not require modification of the protocol to achieve high performance. Its application range is very wide.

[0074] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0075] Figure 1 This is a flowchart of the main process of Embodiment 1 of the present invention;

[0076] Figure 2 This is a schematic diagram showing that useful signals exist at two burst points in the DKAB signal of the present invention;

[0077] Figure 3 This is the DKAB signal bit information modulation table of the present invention;

[0078] Figure 4 This is a flowchart of Embodiment 3 of the present invention;

[0079] Figure 5 This is a structural block diagram of Embodiment 4 of the present invention.

[0080] The module includes: 1. Low-pass filter module; 2. Decimation module; 3. Peak detection module; 4. Timing adjustment module; 5. Time-frequency statistics module; 6. Symbol demodulation module; 7. Hard decision module; 8. Frequency offset estimation module; 9. Decoding module; and 10. Frequency offset statistics module. Detailed Implementation

[0081] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0082] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0083] The DKAB (Dual Keep-Alive Burst) channel is a key control channel in GMR-1 (GEO-Mobile Radio Interface-1) used to maintain the link's active state, primarily for managing silent periods in voice services. Its design goal is to reduce transmit power when users are not transmitting voice messages, while maintaining communication link stability and avoiding frequent connection rebuilds.

[0084] When the user is in voice silence, the vocoder will trigger the terminal or gateway to send a DKAB burst signal to maintain link activation with extremely low power consumption and avoid disconnection due to prolonged lack of data.

[0085] Example 1

[0086] To address the aforementioned technical problems, this invention, through extensive experimental analysis, concludes that the detection performance of DKAB signals is determined by the accuracy of timing and frequency offset estimation, as well as the bit error rate. Therefore, this invention provides a method for detecting DKAB signals in a satellite communication system, such as... Figure 1 As shown, it includes the following steps:

[0087] Step S1, low-pass filtering;

[0088] A 22nd-order FIR low-pass filter is used to filter the input signal;

[0089] Step S2, extraction;

[0090] Every four points are sampled to achieve downsampling from 93.6k to 23.4k. The starting point is selected for traversal. From the time offsets corresponding to 0, 1 / 4, 1 / 2, and 3 / 4 decimal times, extract the output as 4 groups of data with a sampling rate of 23.4k.

[0091] This invention utilizes low-pass filtering and 1 / 4 decimation to achieve 4x downsampling, and traversing the starting point provides input data for subsequent measurement modules to find the optimal fractional sampling point.

[0092] Step S3, peak detection;

[0093] The position with the highest signal-to-noise ratio is detected by a sliding window, and the timing offset, which is an integer multiple of the four sets of input data, is calculated.

[0094] The DKAB signal has 117 symbol periods, which can be represented as Only two burst segments contain useful signals, such as... Figure 2 As shown, the starting point of the burst is determined by the known variable p. Each burst segment has 5 sampling points, and the symbol length between two bursts is fixed. The symbols of two burst segments can be represented as: , Let its index be . .

[0095] Based on this structure, the peak detection method of the present invention using signal-to-noise ratio detection has the following specific steps:

[0096] Step 1: Set up the sliding window The total length of the cable window is... When performing blind detection, a sliding window W needs to be set to traverse the entire subframe;

[0097] Step 2: Calculate the total power of the signal in the entire subframe;

[0098] Right now ,in This represents the total signal power of the entire subframe. express The row vector of the DKAB signal has 117 symbol periods, represented as... ; for The conjugate transpose of the signal, the average power of which is expressed as ;

[0099] Step 3: Calculate the received signal power of the two bursts at each search location;

[0100] Right now ;in, The received signal power of the two bursts at each search location, for The row vector, express The conjugate transpose of; the burst average power is expressed as ;

[0101] Step 4: For burst average power Its corresponding average noise power Represented as: ;

[0102] Step 5: The average power of the actual burst signal is expressed as... The corresponding signal-to-noise ratio for ;

[0103] Step 6: Search Maximum value, corresponding index ,use Two dots on the left and right -1 is used for linear interpolation, resulting in a final timing offset. for: );

[0104] in, This indicates an index. The corresponding signal-to-noise ratio, i.e. Maximum value; Indicates index The corresponding signal-to-noise ratio; index The corresponding signal-to-noise ratio; This means rounding off;

[0105] The peak detection method of this invention confirms the timing of the four sets of input data as integer multiples by detecting the position of the highest signal-to-noise ratio through a sliding window. Compared with the power detection method, this method can eliminate the influence of noise and has higher accuracy.

[0106] Step S4, timed adjustment;

[0107] The signal-to-noise ratios of the four groups obtained in step S3 are: and the corresponding integer multiple timing offset are Search signal-to-noise ratio The maximum value, and its corresponding timing offset is The corresponding set of data after a 4x extraction, with adjusted timing. This will then be used as input for subsequent modules.

[0108] The timing adjustment of this invention can select the optimal fractional sampling data from four sets of detection data based on the signal-to-noise ratio.

[0109] Step S5, time-biased statistics;

[0110] Calculate the time offset value of the subframe based on the data from step S4;

[0111] Subframe time offset The calculation formula is: / 4, of which Traverse from the starting point. .

[0112] Step S6, symbol demodulation;

[0113] Extract the data symbols from the DKAB burst and represent them as follows: .

[0114] Calculate differential phase, reverse rotation The likelihood ratio was then calculated. for: ;in Indicates the differential phase value; This represents the final LLR value; Indicates the Department of Seeking Truth

[0115] Step S7, hard decision;

[0116] Likelihood Values ​​less than 0 are bit 1, and others are 0;

[0117] Step S8, frequency offset estimation;

[0118] Using the transmit bit information obtained in step S7, transmit symbols are modulated according to the protocol. Each burst segment is preceded by a reference symbol, resulting in the demodulated symbol record. for: Remove modulation symbols Afterwards, among them, This indicates the data symbols in the DKAB burst mentioned above. * indicates finding the conjugate of elements in the matrix, and for modulation symbols... Calculate the radians by summing the values ​​in each column. , ,in and For matrix The first and second lines, Represents the phase difference; the frequency offset value is calculated. for: .

[0119] like Figure 3 As shown, the DKAB signal bit information modulation table of the present invention, k in the table is the bit index; frequency offset estimation is obtained by demodulating the transmitted symbol, thereby removing the influence of information bits from the received symbol to achieve the purpose of calculating frequency offset.

[0120] Step S9, Decode;

[0121] Based on continuous multiple sets (preferably 6 sets) of DKAB soft information, the likelihood ratio is extracted. The cascaded code, after deinterleaving, is Gray-coded to obtain the CRC decoding result.

[0122] Step S10, frequency offset statistics;

[0123] Store a filtered frequency offset value. The frequency offset value estimated for each subframe is filtered and recorded as the new frequency offset value. The filter first calculates the difference between the current estimated frequency offset value and the historical frequency offset value. If the result exceeds the set deviation threshold, no filtering is performed; otherwise, filtering is performed. Finally, the frequency offset value of the current subframe is obtained.

[0124] In step S10, the frequency offset value of the current subframe for: ,in, This represents the current estimated value. This represents the filter coefficients.

[0125] This invention is based on the DKAB signal structure given by GMR-1. After downsampling the received signal, a custom algorithm is used to achieve high-performance detection of the DKAB signal. The frequency offset statistics of this invention achieve smoothing of the frequency offset value through Kalman filtering, and by setting the deviation threshold, unreliable frequency offset estimates are eliminated, thereby increasing the reliability of the frequency offset filter value.

[0126] Example 2

[0127] The difference between Example 2 and Example 1 is that Step5 is replaced with: Step5':burst. The actual average signal power is expressed as follows: , ;

[0128] Replace Step6 with: Step6': Search Maximum value, corresponding index ,use Two dots on the left and right -1 is used for linear interpolation, and the final timing offset is:

[0129] ) ;in, Indicates index The corresponding average power of the actual burst signal; Indicates index The corresponding average power of the actual burst signal; Indicates index The corresponding average power of the actual burst signal.

[0130] In this embodiment 2, peak detection can directly use signal power without calculating SNR. That is, by directly calculating the actual signal power, the influence of noise is eliminated, thereby achieving the purpose of peak detection.

[0131] Example 3

[0132] like Figure 4 As shown, the difference between Embodiment 3 and Embodiment 1 lies in the following: the frequency offset estimation process is modified, and frequency offset estimation is performed every 6 consecutive DKAB subframes. After symbol demodulation, instead of performing a hard decision to obtain 0 and 1 bits, the LLR is buffered, and the S symbols of 6 consecutive subframes are buffered. The buffered modulation symbols are only used for frequency offset estimation after the CRC check is correct. The bit information used for demodulation is the 0 and 1 bits of the correctly decoded payload information.

[0133] Specifically, replace step S7, the hard decision, with:

[0134] Step S7', buffer; buffer 6 consecutive subframes. Symbol, denoted as ;

[0135] Replace step S8 with:

[0136] Step S8', frequency offset estimation: When the CRC check is correct, the correctly decoded result is decoded, interleaved, and modulated according to the protocol to generate the transmission symbols for each subframe. Each subframe is processed as follows:

[0137] Each burst segment is preceded by a reference symbol, and the symbol after demodulation is denoted as: .

[0138] Remove modulation symbols , right Calculate radians by summing the columns: ;

[0139] The frequency offset value was calculated. for: .

[0140] In this embodiment 3, the correct information bits can be obtained through soft demodulation decoding and CRC verification, which increases the correctness of the inversion symbol and thus increases the accuracy of frequency offset estimation.

[0141] Example 4

[0142] The present invention also provides an apparatus for detecting DKAB signals in the above-described satellite communication system, such as... Figure 5 As shown, it includes a low-pass filter module 1, a decimation module 2, a peak detection module 3, a timing adjustment module 4, a time-frequency statistics module 5, a symbol demodulation module 6, a hard decision module 7, a frequency offset estimation module 8, a decoding module 9, and a frequency offset statistics module 10.

[0143] Low-pass filter module 1 is used to filter the input signal using a 22nd-order FIR low-pass filter;

[0144] Extraction module 2 is used to extract one sampling point every 4 points, and the extraction starts from the starting point and iterates through the data. It extracts and outputs 4 sets of data with a sampling rate of 23.4k from the time offsets corresponding to multiples of 0, 1 / 4, 1 / 2, and 3 / 4.

[0145] Peak detection module 3 is used to detect the position of the highest signal-to-noise ratio through a sliding window and calculate the timing offset of integer multiples of the four sets of input data;

[0146] The timing adjustment module 4 uses the peak detection module 3 to obtain 4 sets of the highest signal-to-noise ratio positions and integer multiple timing offsets, searches for the maximum signal-to-noise ratio and its corresponding timing offset and the corresponding set of extracted data, adjusts the timing and uses it as the input for subsequent modules.

[0147] The time-frequency statistics module 5 is used to calculate the time offset value of the subframe based on the data from the timing adjustment module 4;

[0148] Symbol demodulation module 6 is used to extract data symbols from the DKAB burst, calculate the differential phase, and calculate the likelihood ratio after inverse rotation by π / 4.

[0149] Hard decision module 7 is used to record bit 1 for values ​​less than 0 in the likelihood ratio and 0 for others;

[0150] The frequency offset estimation module 8 uses the transmitted bit information obtained by the hard decision module 7 to modulate the transmitted symbols according to the protocol. Each burst segment is preceded by a reference symbol to obtain the demodulated symbol. After removing the modulation symbols, the modulation symbols are accumulated column by column to calculate the frequency offset value.

[0151] Decoding module 9 is used to extract likelihood ratio concatenation based on multiple consecutive sets of DKAB soft information, and after deinterleaving, Gray decoding is used to obtain the decoding result;

[0152] The frequency offset statistics module 10 is used to store a filtered frequency offset value. The frequency offset value estimated for each subframe is filtered and recorded as the new frequency offset value. The filter first calculates the difference between the current estimated frequency offset value and the historical frequency offset value. If the result exceeds the set deviation threshold, no filtering is performed; otherwise, filtering is performed. Finally, the frequency offset value of the current subframe is obtained.

[0153] This invention addresses the problems of existing DKAB signal detection methods, such as high implementation complexity, inferior detection performance compared to other logical channels, or the need for protocol modifications to achieve high performance. For the DKAB signal structure given by GMR-1, after downsampling the received signal, this invention employs a custom algorithm: low-pass filtering plus 1 / 4 decimation to achieve 4x downsampling; a fractional time offset traversal method using a search window w to traverse the entire subframe; a peak detection method based on signal-to-noise ratio; frequency offset estimation; and frequency offset statistics. This invention eliminates the influence of noise, resulting in higher accuracy. Furthermore, it smooths the frequency offset value using Kalman filtering and eliminates unreliable frequency offset estimates by setting a deviation threshold, increasing the reliability of the frequency offset filter value.

[0154] In the description of this invention, it should be understood that the indicated orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing this invention and simplifying the description, and is not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0155] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0156] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for detecting DKAB signals in a satellite communication system, characterized in that: Includes the following steps: Step S1, low-pass filtering; A 22nd-order FIR low-pass filter is used to filter the input signal; Step S2, extraction; One sampling point is extracted for every 4 points, and the sampling starts from the starting point and is traversed; from the time offsets corresponding to 0, 1 / 4, 1 / 2, and 3 / 4 decimals, 4 sets of data with a sampling rate of 23.4k are extracted and output. Step S3, peak detection; The position with the highest signal-to-noise ratio is detected by sliding window, and the timing offset of the four sets of input data is calculated as an integer multiple. Step S4, timed adjustment; For the four sets of highest signal-to-noise ratio positions and integer multiple timing offsets obtained in step S3, search for the maximum signal-to-noise ratio, its corresponding timing offset, and the corresponding set of extracted data. Adjust the timing and use it as the input for subsequent modules. Step S5, time-biased statistics; Calculate the time offset value of the subframe based on the data from step S4; Step S6, symbol demodulation; Extract the data symbols from the DKAB burst, calculate the differential phase, and calculate the likelihood ratio after inverse rotation by π / 4. Step S7, hard decision; In the likelihood ratio, values ​​less than 0 are bit 1, and all others are 0. Step S8, frequency offset estimation; Using the transmit bit information obtained in step S7, the transmit symbols are modulated according to the protocol. Each burst segment is preceded by a reference symbol to obtain the demodulated symbol record. After removing the modulation symbols, the frequency offset value is calculated by summing the values ​​of the modulation symbols column by column in radians. Step S9, Decode; Based on continuous sets of DKAB soft information, likelihood ratio concatenation is extracted, and after deinterleaving, Gray decoding is used to obtain the decoding result. Step S10, frequency offset statistics; Store a filtered frequency offset value. The frequency offset value estimated for each subframe is filtered and recorded as the new frequency offset value. The filter first calculates the difference between the current estimated frequency offset value and the historical frequency offset value. If the result exceeds the set deviation threshold, no filtering is performed; otherwise, filtering is performed. Finally, the frequency offset value of the current subframe is obtained.

2. The method for detecting DKAB signals in a satellite communication system according to claim 1, characterized in that: In step S3, peak detection specifically includes: Step 1: Set up the sliding window The total length of the cable window is... When performing blind detection, a sliding window W needs to be set to traverse the entire subframe; Step 2: Calculate the total power of the signal in the entire subframe; Right now ,in This represents the total signal power of the entire subframe. express The row vector of the DKAB signal has 117 symbol periods, represented as... ; for The conjugate transpose of the signal, the average power of which is expressed as ; Step 3: Calculate the received signal power of the two bursts at each search location; Right now ;in, The received signal power of the two bursts at each search location, for The row vector, express The conjugate transpose of; the burst average power is expressed as ; Step 4: For burst average power Its corresponding average noise power Represented as: ; Step 5: The average power of the actual burst signal is expressed as... The corresponding signal-to-noise ratio for ; Step 6: Search Maximum value, corresponding index ,use Two dots on the left and right -1 is used for linear interpolation, resulting in a final timing offset. for: ); in, This indicates an index. The corresponding signal-to-noise ratio, i.e. Maximum value; Indicates index The corresponding signal-to-noise ratio; Indicates index The corresponding signal-to-noise ratio; It means to round off.

3. The method for detecting DKAB signals in a satellite communication system according to claim 2, characterized in that: Replace Step5 with: Step5':burst. The actual average signal power is represented as follows: , ; Replace Step6 with: Step6':search Maximum value, corresponding index ,use Two dots on the left and right -1 is used for linear interpolation, and the final timing offset is: )； in, Indicates index The corresponding average power of the actual burst signal; Indicates index The corresponding average power of the actual burst signal; Indicates index The corresponding average power of the actual burst signal.

4. The method for detecting DKAB signals in a satellite communication system according to claim 2, characterized in that: In step S4, the signal-to-noise ratios of the four groups are: and the corresponding integer multiple timing offset are Search signal-to-noise ratio The maximum value, and its corresponding timing offset is The corresponding set of data after a 4x extraction, with adjusted timing. This will then be used as input for subsequent modules.

5. The method for detecting DKAB signals in a satellite communication system according to claim 4, characterized in that: In step S5, the time offset of the subframe The calculation formula is: / 4, of which Traverse from the starting point. .

6. The method for detecting DKAB signals in a satellite communication system according to claim 4, characterized in that: In step S6, the data symbol representation extracted from the DKAB burst is as follows: ; Calculate differential phase, reverse rotation The likelihood ratio was then calculated. for: ;in Indicates the differential phase value; This represents the final LLR value; It refers to the department that seeks truth.

7. The method for detecting DKAB signals in a satellite communication system according to claim 6, characterized in that: In step S8, the demodulated symbol is recorded. for: Remove modulation symbols After; among them, This indicates the data symbols in the DKAB burst mentioned above. * indicates finding the conjugate of elements in the matrix, and for modulation symbols... Calculate the radians by summing the values ​​in each column. , ,in and For matrix The first and second lines, Represents the phase difference; the frequency offset value is calculated. for: .

8. The method for detecting DKAB signals in a satellite communication system according to claim 7, characterized in that: In step S10, the frequency offset value of the current subframe for: ,in, This represents the current estimated value. This represents the filter coefficients.

9. The method for detecting DKAB signals in a satellite communication system according to claim 7, characterized in that: Replace step S7, the hard decision, with: Step S7', caching; Buffering 6 consecutive subframes symbol, The data symbol in the DKAB burst in the above paragraph is denoted as... ; Replace step S8 with: step S8', frequency offset estimation. When the CRC check is correct, the correctly decoded result is decoded, interleaved, and modulated according to the protocol to generate the transmission symbols for each subframe. Each subframe is processed as follows: Each burst segment is preceded by a reference symbol, and the symbol after demodulation is denoted as: ; Remove modulation symbols , right Calculate radians by summing the columns: ; The frequency offset value was calculated. for: .

10. An apparatus for detecting DKAB signals in a satellite communication system as described in any one of claims 1-9, characterized in that: It includes a low-pass filter module, a decimation module, a peak detection module, a timing adjustment module, a time-frequency statistics module, a symbol demodulation module, a hard decision module, a frequency offset estimation module, a decoding module, and a frequency offset statistics module; The low-pass filter module is used to filter the input signal using a 22nd-order FIR low-pass filter. The extraction module is used to extract one sampling point every four points, and the extraction starts from the starting point and iterates through the points. It extracts and outputs four sets of data with a sampling rate of 23.4k from the time offsets corresponding to multiples of 0, 1 / 4, 1 / 2, and 3 / 4. The peak detection module is used to detect the position of the highest signal-to-noise ratio through a sliding window and calculate the timing offset as an integer multiple of the four sets of input data. The timing adjustment module uses the peak detection module to obtain four sets of the highest signal-to-noise ratio positions and integer multiple timing offsets, searches for the maximum signal-to-noise ratio value and its corresponding timing offset and the corresponding set of extracted data, and adjusts the timing as input for subsequent modules. The time-frequency statistics module is used to calculate the time offset value of the subframe based on the data from the timing adjustment module; The symbol demodulation module is used to extract data symbols from the DKAB burst, calculate the differential phase, and calculate the likelihood ratio after inverse rotation by π / 4. The hard decision module is used to record bit 1 for values ​​less than 0 in the likelihood ratio and 0 for others; The frequency offset estimation module is used to use the transmitted bit information obtained by the hard decision module to modulate the transmitted symbols according to the protocol. Each burst segment is preceded by a reference symbol to obtain the demodulated symbol record. After removing the modulation symbols, the frequency offset value is calculated by summing the values ​​of the modulation symbols column by column in radians. The decoding module is used to extract likelihood ratio concatenation based on multiple sets of continuous DKAB soft information, and after deinterleaving, Gray decoding is used to obtain the decoding result; The frequency offset statistics module is used to store a filtered frequency offset value. The frequency offset value estimated for each subframe is filtered and recorded as a new frequency offset value. The filter first calculates the difference between the current estimated frequency offset value and the historical frequency offset value. If the result exceeds the set deviation threshold, no filtering is performed; otherwise, filtering is performed. Finally, the frequency offset value of the current subframe is obtained.

Images