CN103501218B - A kind of multicarrier adaptive demodulation method based on resource multiplex - Google Patents

Abstract

Based on a multicarrier adaptive demodulation method for resource multiplex, the key step comprised is: input multi-path serial data buffer memory, variable buffer memory in the middle of each carrier wave, leading head are caught, estimations of position timing offset, variable buffer memory, initial frequency deviation and skew is estimated, RM decoding and carrier track algorithm process in the middle of trigonometrical function interpolation, coordinate conversion, input range and phase data buffer memory, each carrier wave. Present invention achieves the universal design that the burst multi-carrier signal to variable Rate, Multiple modulation mode carries out adaptive demodulation. Solve satellite to carry out serious rain when multi-carrier-wave wireless transmits in Ku/Ka frequency range and decline the needs of problems of the different data transfer demands problem of problem, polymorphic type communication terminal and the anti-interference communication of telstar system multi-carrier.

Description

A multi-carrier adaptive demodulation method based on resource multiplexing

technical field

The invention relates to a multi-carrier adaptive demodulation method based on resource multiplexing.

Background technique

Satellite communication has the characteristics of wide coverage, long communication distance, large communication capacity and good transmission quality, and has become an important means of communication. Due to the increasingly busy satellite communication business and the rapid increase in communication capacity, the radio frequency spectrum is very crowded. In order to solve the problem of shortage of spectrum resources, satellite communication is developing towards the high-frequency band of Ka (20/30GHz) and above with broad prospects. Combined with the diversified requirements of communication types, the application of satellite communication systems in the Ka and above frequency bands has the following problems to be solved: 1) The serious problem of rain attenuation when satellites perform multi-carrier wireless transmission in the Ku/Ka frequency band; 2) How does the communication satellite system meet the different data transmission requirements of multi-type communication terminals; 3) How does the communication satellite system meet the needs of multi-carrier anti-jamming communication. The basis for solving all these is the satellite's efficient multi-carrier adaptive demodulation capability.

Multi-carrier adaptive demodulation not only requires independent adaptive demodulation of each carrier, but also requires each carrier to be able to reuse resources. The resource consumption cannot be much larger than that of a single carrier. Therefore, the performance of multi-carrier adaptive demodulation is directly related to affect the overall performance of the system.

Existing multi-carrier demodulation algorithms are relatively single. For example, Document 1 "Research on the Overall Demodulation Technology of Multi-Carrier Burst Signals" (Li Hui, Xidian University Master's Degree Thesis, 2011) provides an all-digital reception technology for multi-carrier burst signals, including bit synchronization and carrier synchronization ; Literature 2 "Multi-CarrierMulti-RateModemforUniversalFDMA/TDMAsystem" (FumihiroYamashita, 24thAIAAinternationalCommunicationsSatelliteSystemsConference, 2006-5316) gives the design of a QPSK multi-carrier burst demodulator based on resource multiplexing. In Document 1, the multi-carrier burst algorithm is implemented separately, and no specific multiplexing method is given; in Document 2, the multi-carrier burst demodulation algorithm is only applicable to QPSK, and the solution The tuning mode is single.

Contents of the invention

The technical problem solved by the present invention is to overcome the deficiencies of the prior art, provide a multi-carrier adaptive demodulation method based on resource multiplexing, and solve the multi-carrier autonomous demodulation method for satellites in the Ku/Ka frequency band with very small resource consumption. Adapt to the problem of on-board implementation of communication.

The technical solution of the present invention is: a multi-carrier adaptive demodulation method based on resource multiplexing, the steps are as follows:

1) Cache the real part and imaginary part of the multi-channel serial input data after digital splitting into RAM1, and add 1 to the address Addr_in of the input data every time a data is stored, and write the initial parameters of each carrier into FIFO1 in turn ; The initial parameters include the input data address Addr_best to start processing, the number Part_num of output blocks, the numbering Channel_index of each carrier, the capture flag Flag, and the 128-bit sequence Demola used for leading the header;

2) Read the carrier parameters of the first carrier from FIFO1, and judge the value of Flag. If the value of Flag is not 1, it means that it has not been successfully captured, and then go to step 3); if the value of Flag is 1, it means that it has been successfully captured , then go to step 5);

3) Take Addr_best as the starting address, A as the number of carriers, read data from RAM1, make a hard decision after differential calculation, store the hard decision result in Demola, and correlate with the leading header, if Addr_in minus Addr_best is less than twice the number of carriers A, then Flag is set to 1, and Addr_best is set as the starting address of the input sequence when successfully captured, and enters step 4); if Addr_in minus Addr_best is greater than or equal to twice the number of carriers A, set The parameters at this time are stored in FIFO1, jump to step 2) continue to read the parameters of the next carrier;

4) If Addr_in minus Addr_best is greater than M, go to step 5); if Addr_in minus Addr_best is less than or equal to M, store the carrier parameters at this time in FIFO1, and jump to step 2) continue to read the next carrier parameter; of Where N is the number of output symbols after trigonometric function interpolation;

5) With Addr_best as the starting address and A as the interval, read 4N sampling points from RAM1 to obtain the bit timing deviation value;

6) With Addr_best-1 as the starting address and A as the interval, read 4N sampling points from RAM1, and perform trigonometric interpolation on the 4N sampling points according to the bit timing deviation value obtained in step 5, and output N at the same time, store the carrier parameters at this time in FIFO1, and jump to step 2) continue to read the next carrier parameter;

7) Convert the real part and imaginary part data of the N symbols output in step 6) into amplitude and phase data;

8) Store the amplitude and phase data obtained in step 7) into RAM2 according to the carrier number. For each data stored in the nth channel, n=1,2,3...A, add 1 to the input data address Addr_in_mid_n of this channel , and write the initial parameters of each path into FIFO2 in turn; the initial parameters include carrier number Channel_index, carrier frame header indication signal initial_flag (Channel_index), carrier input data address Addr_in_mid_n, initial frequency offset and phase offset estimation flag signal Flag_est, input data address Addr_track to start processing, number of output blocks Part_num, estimated initial frequency offset Fre_move, estimated initial phase offset Phase_offset, coded modulation type Rm_reg;

9) Read out the parameters of the first carrier, judge the value of Flag_est, if the value of Flag_est is not 1, go to step 10); if the value of Flag_est is 1, go to step 13);

10) Determine the value of initial_flag(Channel_index), if the value of initial_flag(Channel_index) is 1, go to step 11), if the value of initial_flag(Channel_index) is not 1, store the current carrier parameters in FIFO, and jump to Go to step 9) continue to read the parameters of the next carrier;

11) With the carrier number Channel_index and the composite address Addr_Channel of the input data address Addr_track as the starting address, read 128-bit amplitude and phase data from RAM2, and perform initial frequency offset and phase offset estimation to obtain the estimated initial frequency offset Fre_move And phase offset Phase_offset;

12) Use Addr_Channel as the starting address to read the amplitude and phase data of the 64-bit coded modulation mode word from RAM2, and read the amplitude and phase data of the 64-bit coded modulation mode word according to the Fre_move and Phase_offset received from step 11). The phase data is subjected to deviation correction processing, and then the data after deviation correction is subjected to hard judgment, and the RM decoding operation is performed on the data after hard judgment to obtain the encoding modulation type RM_reg, and Flag_est is set to 1, and then enter step 13);

13) If Addr_in_mid_n minus Addr_track is greater than N, go to step 14), if Addr_in_mid_n minus Addr_track is less than or equal to N, then store the current carrier parameters in FIFO2, and jump to step 9) Continue to read the next carrier parameter;

14) Use Addr_Channel as the starting address to read the amplitude and phase data of the information data from RAM2, correct the amplitude and phase data of the read input data according to the values of Fre_move and Phase_offset obtained in step 11), and correct the deviation according to RM_reg Different carrier tracking algorithms are used to perform carrier tracking processing on the corrected data, and 128 data are divided into one processing unit to output, and the current carrier parameters are stored in FIFO2, and then skip to step 9) to continue reading The parameters of the next carrier, until the processing of carrier A is completed.

Compared with the prior art, the present invention has the following advantages:

(1) The present invention has good performance in multi-carrier burst adaptive demodulation, and under the influence of frequency deviation, phase deviation, bit timing deviation and white noise, the demodulation loss is less than 0.5 dB.

(2) The resource reuse rate of the present invention is high. Using the Virtex4-55 chip, the multi-carrier adaptive capture and bit synchronization function can be completed with a resource consumption rate of only 14%, and the multi-carrier adaptive carrier synchronization function can be completed with a resource consumption rate of 13%. , slightly larger than the single-carrier module.

(3) The present invention realizes adaptive demodulation of burst multi-carrier signals with variable rate (2-7Mbps) and multi-modulation modes (QPSK, 8PSK and 16APSK). The rate is variable and can be adapted to various modulation methods.

Description of drawings

FIG. 1 is a block diagram of a multi-carrier adaptive demodulation method for resource multiplexing;

Fig. 2 is a transmission signal frame structure;

Figure 3 is a RAM storage allocation diagram.

detailed description

Taking 8 carriers as an example below, the present invention will be further introduced in conjunction with the accompanying drawings.

As shown in Figure 1, the multiplexing of 8 carriers is realized through FIFO, and each channel stores its important intermediate parameters in the FIFO, and which channel to be processed will continue to process on the basis of the read parameters. The parameters to be cached in each channel in FIFO1 are Addr_best (the input data address to start processing), Part_num (the number of output blocks), Channel_index (the number of each carrier, ranging from 0 to 7), Flag (the flag of whether the capture is successful, '1' represents successful capture), Burst_type (burst type, '0' represents data burst, '1' represents ranging burst), Demola (for the 128-bit sequence related to the preamble), the preamble The head is the capture and synchronization sequence in Figure 2.

The overall process is as follows:

1) Read one carrier parameter from FIFO1, and judge the value of Flag. If the value of Flag is not 1, it means that the capture was not successful, and then go to step 2); if the value of Flag is 1, it means that it has been captured successfully, and then go to step 4) ;

2) Take Addr_best as the starting address, 8 as the number of carriers, read data from RAM1, perform a hard decision after differential calculation, store the hard decision result in Demola, and correlate with the leading header, if Addr_in minus Addr_best is less than 16, then Flag is set to 1, Addr_best is set as the starting address of the input sequence when successfully captured, and enters step 3); if Addr_in minus Addr_best is greater than or equal to 16, store the parameters at this time in FIFO1, and jump to step 1) Continue to read the next carrier parameter;

3) If Addr_in minus Addr_best is greater than 512 Go to step 4), if Addr_in minus Addr_best is less than or equal to 512, store the carrier parameters at this time in FIFO1, skip to step 2) and continue to read the next carrier parameter;

4) Take Addr_best as the starting address and 8 as the interval, read 512 sampling points from RAM1, and estimate the bit timing deviation (see the patent "A MAPSK Adaptive Demodulation System");

5) With Addr_best-1 as the starting address and 8 as the interval, read 512 sampling points from the RAM, and perform trigonometric interpolation on the 512 sampling points according to the bit timing deviation value obtained in step 4 (see the patent " A MAPSK adaptive demodulation system") and output in blocks, each block of 128 symbol data, as the burst type of each carrier is fed back to the acquisition and bit synchronization processing part, the number of output symbol blocks of each carrier is determined (data burst: ceil(16352/N)=128, ranging burst: ceil(4256/N)=34), if the carrier of this channel is a data burst and Part_num<128 or the carrier of this channel is a ranging burst And Part_num<34), output 128 symbols, after the output, Part_num is set to Part_num+1, Addr_best is set to Addr_best+512, Burst_type is set to the Burst_type of the current carrier fed back, and the rest remain unchanged. The parameters are stored in FIFO1, and then the parameters of the next carrier are read for processing; if the carrier of this channel is a data burst and Part_num=128, it means that this is the last block of the data burst frame of this carrier, and M1 symbols are output (M1= 16352-N*(Part_num-1)=96), after the output is completed, Addr_best is set to Addr_best+M1′ Channel_index remains unchanged, and the rest are all set to 0. Store the parameters of this carrier in FIFO1, and then read the parameters of the next carrier for processing; if the carrier of this channel is a ranging burst and Part_num=34, it means that this channel is In the last block of the carrier ranging burst frame, M2 symbols are output (M2=4256-N*(Part_num-1)=32). After the output is completed, Addr_best is set to Addr_best+M2′ Channel_index remains unchanged, and the rest are all set to 0, and the parameters of this carrier are stored in FIFO1, and jump to step (1) to read the parameters of the next carrier;

6) Convert the real part and imaginary part data blocks of 128 symbols output in step 5) into amplitude and phase data;

7) Store the amplitude and phase data obtained in step 6) into RAM2 according to the carrier number. For each data stored in the nth channel, n=1,2,3...8, add 1 to the input data address Addr_in_mid_n of this channel , and write the initial parameters of each path into FIFO2 in turn; the initial parameters include carrier number Channel_index, carrier frame header indication signal initial_flag (Channel_index), carrier input data address Addr_in_mid_n, initial frequency offset and phase offset estimation flag signal Flag_est, input data address Addr_track to start processing, number of output blocks Part_num, estimated initial frequency offset Fre_move, estimated initial phase offset Phase_offset, coded modulation type Rm_reg;

8) Read out one carrier parameter, judge the value of Flag_est, if the value of Flag_est is not 1, go to step (9); if the value of Flag_est is 1, go to step (12);

9) Determine the value of initial_flag(n), n is the channel_index of the carrier number read out, if the value of initial_flag(n) is 1, it means that the carrier of this channel has data stored in RAM2, then enter step (10), if The value of initial_flag(n) is not 1, then store the current carrier parameters into FIFO2 (Channel_index is the carrier number of this channel, and the rest are 0), and read the parameters of the next carrier, and return to step (8);

10) With the carrier number Channel_index and the composite address Addr_Channel of the input data address Addr_track as the starting address, read 128 amplitude and phase data from RAM2, and do the initial frequency offset and Phase offset estimation (see the patent "A MAPSK Adaptive Demodulation System") to obtain the estimated initial frequency offset (Fre_move) and phase offset (Phase_offset);

11) With the carrier number Channel_index and the composite address Addr_Channel of the input data address Addr_track as the starting address, read and read 64 amplitude and phase data (the coded modulation mode word in Figure 2) from RAM2, and use the data received from 10) The initial frequency offset and phase offset of these data are corrected, and then the corrected data is hard-judged, and then RM decoding is performed on the hard-judged data to obtain the coded modulation type RM_reg, and Flag_est is set to 1. go to step 12);

12) If Addr_in_mid_n minus Addr_track is greater than 128 (128 is the number of symbol data for an output block), then go to step 13), if Addr_in_mid_n minus Addr_track is less than or equal to 128, then store the current carrier parameters in FIFO2 and jump to step 8) Continue to read the next carrier parameter;

13) Read the amplitude and phase data of the input data from RAM2, use the Fre_move and Phase_offset values obtained in 10) to correct the deviation of these data, and select different carrier tracking algorithms according to the difference of RM_reg (see the patent "A MAPSK Adaptive demodulation system") Carrier tracking processing is performed on the corrected data and output in blocks according to 128 data as a processing unit. As the burst type of the carrier of this channel is fed back to the carrier tracking module, the output of the carrier of this channel The number of data blocks is determined (data burst: ceil(16352/N)=128, ranging burst: ceil(4256/N)=34). If the carrier of this channel is a data burst and Part_num<128 or the carrier of this channel is a ranging burst and Part_num<34), output 128 amplitude and phase data. After the output is completed, Part_num is set to Part_num+1, and Addr_track is set to Addr_track +128, Burst_type is set to the Burst_type of the current carrier that is fed back, and the rest remain unchanged. The parameters of the current carrier are stored in FIFO2, and then the parameters of the next carrier are read for processing; if the current carrier is a data burst and Part_num=128, indicating that this is the last block of the carrier data burst frame, output M1 amplitude and phase data (M1=96), after the output is completed, Addr_track is set to Addr_track+96, Channel_index remains unchanged, and the rest are all set to 0 , store the parameters of the current carrier in FIFO2, and then read the parameters of the next carrier for processing; if the current carrier is a ranging burst and Part_num=34, it means this is the last block of the ranging burst frame of this carrier , output M2 amplitude and phase data (M2=4256-N ^* (Part_num-1)=32), after the output is completed, Addr_track is set to Addr_track+32, Channel_index remains unchanged, the rest are all set to 0, and the current carrier parameters are saved into FIFO2, read the parameters of the next carrier, and jump to step (8) to continue processing.

Parts not described in detail in the present invention belong to the common knowledge of those skilled in the art.

Claims (1)

1. the multicarrier adaptive demodulation method based on resource multiplex, it is characterised in that step is as follows:

1) buffering in RAM1 by real part and the imaginary part of the multi-path serial input data after numeral shunt, often stored in data, the address Addr_in of input data adds 1, and each carrier wave initial parameter is write successively in FIFO1; Described initial parameter comprise start process input data address Addr_best, export the number Part_num of block, the numbering Channel_index of each carrier wave, catch sign of flag, for the 128 bit sequence Demolas relevant with leading head;

2) from FIFO1, read the carrier parameter of the 1st road carrier wave, judge the value of Flag, if the value of Flag is not 1, shows successfully not catch, then enter step 3); If the value of Flag is 1, represents and successfully catch, then enter step 5);

3) take Addr_best as start address, A is variable number, data are read from RAM1, firmly adjudicate after doing difference computing, by hard court verdict stored in Demola, do relevant with leading head, if Addr_in subtracts 2 times that Addr_best is less than variable number A, then Flag is set to the start address that 1, Addr_best is set to list entries when successfully catching, and enters step 4); If Addr_in subtracts 2 times that Addr_best is more than or equal to variable number A, by parameter now stored in FIFO1, jump to step 2) continue to read next road carrier parameter;

4) if Addr_in subtracts Addr_best is greater than M, step 5 is entered); If Addr_in subtracts Addr_best is less than or equal to M, by carrier parameter now stored in FIFO1, jump to step 2) continue to read next road carrier parameter; DescribedWherein N is the output symbol numbers after trigonometrical function interpolation;

5) taking Addr_best as start address, the value of variable number A is interval, reads 4N sampling point from RAM1, obtains position timing offset value;

6) take Addr_best-1 as start address, the value of variable number A is interval, 4N sampling point is read from RAM1, according to step 5) the position timing offset value that obtains, after 4N sampling point is done trigonometrical function interpolation, export N number of symbol, simultaneously by carrier parameter now stored in, in FIFO1, jumping to step 2) continue to read next road carrier parameter;

7) by step 6) in the real part of N number of symbol that exports and imaginary data be converted to amplitude and phase data;

8) by step 7) in the amplitude that obtains and phase data by carrier number stored in RAM2, the n-th tunnel often stored in data, n=1,2,3 ... A, this road input data address Addr_in_mid_n adds 1, and by the initial parameter on each road, write successively in FIFO2; Described initial parameter comprises carrier index Channel_index, carrier wave frame head indicator signal initial_flag (Channel_index), carrier wave input data address Addr_in_mid_n, initial frequency deviation and skew and estimates marking signal Flag_est, starts the input data address Addr_track of process, the number Part_num exporting block, the initial frequency deviation Fre_move estimated, the initial skew Phase_offset estimated, coded modulation type Rm_reg;

9) read first via carrier parameter, judge the value of Flag_est, if the value of Flag_est is not 1, then enter step 10); If the value of Flag_est is 1, then enter step 13);

10) value of initial_flag (Channel_index) is judged, if the value of initial_flag (Channel_index) is 1, then enter step 11), if the value of initial_flag (Channel_index) is not 1, then by current carrier parameter stored in, in FIFO2, jumping to step 9) continue to read next road carrier parameter;

11) taking the compound address Addr_Channel of carrier index Channel_index and input data address Addr_track as start address, 128 amplitudes and phase data is read from RAM2, and carry out initial frequency deviation and skew and estimate, obtain the initial frequency deviation Fre_move that estimates and skew Phase_offset;

12) take Addr_Channel as amplitude and the phase data that start address reads 64 coded modulation pattern words from RAM2, according to from step 11) in Fre_move and Phase_offset that receive the amplitude of 64 the coded modulation pattern words read and phase data are carried out correction process, then the data after correction are adjudicated firmly, data after hard judgement are done RM decoding operation, obtain coded modulation type RM_reg, and Flag_est is set to 1, enter step 13);

13) if Addr_in_mid_n subtracts Addr_track is greater than N, then enter step 14), if Addr_in_mid_n subtracts Addr_track is less than or equal to N, then by current carrier parameter stored in, in FIFO2, jumping to step 9) continue to read next road carrier parameter;

14) take Addr_Channel as amplitude and the phase data that start address reads information data from RAM2, according to step 11) in the value of Fre_move and Phase_offset that obtains the amplitude of the input data read and phase data are rectified a deviation, and the difference according to RM_reg, different carrier track algorithms is selected the data after correction are carried out carrier track process and to be that one piece of processing unit piecemeal exports by 128 data, and by current carrier parameter stored in FIFO2, jump to step 9) continue to read next road carrier parameter, until completing the process to A road carrier wave.