A soft bit joint demodulation and despreading method based on GMSK

Abstract

The present invention relates to the technical field of demodulation and despreading of direct sequence spread spectrum system, and in particular to a soft bit joint demodulation and despreading method based on GMSK, comprising the following steps: grouping a received data sequence according to the number of oversampling points based on the spread multiple; using a Gold pseudo-random sequence to generate a local spread spectrum code, and multiplying the spread spectrum code with a corresponding NCO signal to obtain a local modulation sequence; using an effective sampling point signal to multiply the received data sequence with a local modulation sequence and its conjugate sequence to obtain a corresponding correlation sequence; accumulating and integrating the two multiplication results according to the sequence grouping length, respectively, to obtain a correlation integral value of each code element and a local reference signal, and calculating the average value of the correlation integral value; calculating the power of the two average values, and outputting the power difference as the final demodulation and despreading soft bit data. Compared with the implementation scheme of demodulation before despreading of conventional FSK-type modulation, the present invention effectively improves the demodulator performance.

Description

Technical Field

[0001] The invention relates to the technical field of demodulation and despreading of a direct sequence spread spectrum system, and in particular to a soft bit joint demodulation and despreading method based on GMSK. Background Art

[0002] GMSK spread spectrum communication technology is a GMSK modulation communication method based on direct sequence spread spectrum. Spread spectrum communication is a communication method that uses a radio frequency signal with a much wider bandwidth than the original signal itself. It has the advantages of strong anti-interference ability, low interception rate, and signal shielding. Because it can adapt to complex electromagnetic environments and ensure normal communication, it is widely used in military and civilian communications. Direct sequence spread spectrum is currently the most widely used spread spectrum technology. It uses a pseudo-random code sequence and the information stream to be sent for modulo-two addition or direct multiplication, and then generates a spread spectrum sequence with a bandwidth much larger than the original sequence bandwidth according to the spread spectrum factor. The receiving end uses the same spread spectrum code for reception and despreading recovery.

[0003] GMSK (Gaussian Minimum Shift Keying) adds Gaussian filtering preprocessing to the baseband on the basis of MSK, so that the phase of the modulated signal is continuous and changes smoothly at the symbol switching moment. It is a constant envelope modulation signal. GMSK has the characteristics of continuous phase, no sudden change in carrier phase when the symbol alternates, fast signal power attenuation outside the main lobe, and narrow intermediate frequency signal bandwidth, so it has also been widely used in the communication field.

[0004] Conventional GMSK receivers usually adopt the method of demodulation before despreading. First, the local modulated signal and the received signal are correlated to obtain the demodulated value, and then the spread spectrum code and the demodulated data are integrated to obtain the final despread data. For the demodulated and despread data finally obtained, the analog quantity can be directly output by soft decision, or the demodulator output waveform can be quantized at multiple levels and sent to the decoder for decoding, such as the Viterbi decoding algorithm based on the maximum likelihood algorithm; or the demodulation result can be directly judged by hard decision according to the decision threshold and then output as 0 or 1.

[0005] For the GMSK receiver described above, compared with the joint demodulation and despreading structure, in terms of the amount of computation, the signal processing process must be demodulated before despreading, and there is an extra level of correlation integral operation and division truncation operation, so it occupies more resources, and because of the two truncation outputs, part of the spread spectrum gain is lost for spread spectrum communication. In addition, if the hard decision algorithm is used, the code elements used by the decoder for decoding are all 0 or 1, which loses the expected coding gain. Generally speaking, although the computational complexity of soft decision decoding is increased, it is about 2dB better than hard decision decoding in performance. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention discloses a soft bit joint demodulation and despreading method based on GMSK, which not only ensures better demodulation performance of the receiver, but also saves computing resources to a great extent.

[0007] The present invention is achieved through the following technical solutions:

[0008] A GMSK-based soft bit joint demodulation and despreading method comprises the following steps:

[0009] The received data sequence is grouped according to the number of oversampling points based on the spread spectrum multiple; the Gold pseudo-random sequence is used to generate a local spread spectrum code, and the spread spectrum code is multiplied with the corresponding NCO signal to obtain a local modulation sequence; the received data sequence is multiplied with the local modulation sequence and its conjugate sequence respectively using the effective sampling point signal to obtain the corresponding correlation sequence; the two multiplication results are accumulated and integrated according to the sequence grouping length to obtain the correlation integral value of each code element and the local reference signal, and the average value of the correlation integral value is calculated; the power of the two average values ​​is calculated, and the output power difference is used as the final demodulated and spread soft bit data.

[0010] Further, in order to better implement the present invention, the GMSK-based soft bit joint demodulation and despreading method specifically includes the following steps:

[0011] The data sequence obtained after front-end preprocessing is expressed as r(n)=r i (n)+jr g (n), the sequence length is L. Among them, r i (n) and r q (n) are the I and Q components of the received signal, n is the number of sampling points, n=0, 1, 2, ..., L-1, and j is an imaginary unit.

[0012] The received data sequence is oversampled according to the number of points based on the spreading factor, i.e., length N. g The number of groups after grouping can be expressed as

[0013] in, n sf is the configurable spreading factor, k s is the sampling multiple, n sf =1,2,3,4...,k s = 8. The length L of the received sequence r(n) must be the packet length N g The receiver determines the number of sampling points to be demodulated according to the number of received bytes.

[0014] The modulation signal is generated using the local spreading code. The specific implementation process is:

[0015] 1) In the direct sequence spread spectrum system, two m sequences are used to generate the Gold sequence as the local pseudo-random code s(n);

[0016] 2) changing the phase change according to the 0 or 1 value of the current pseudo-random code s(n);

[0017] 3) Using the changed phase value Get current NCO data and And serve as the I and Q components of the local modulation sequence.

[0018] To better implement the present invention, further, the step 1) includes the following steps:

[0019] 1-1) Construct two m-sequences m1(n) and m2(n) according to the initial value of the polynomial, both of which have the same length and the same code rate;

[0020] 1-2) The local pseudo-random code s(n) sequence is formed by multiplying the selected pairs. Represents modulo-2 addition.

[0021] To better implement the present invention, further, the step 3) includes the following steps:

[0022] 3-1) For the generation of ROM data, the range of 0 to 2π is divided into 65536 phases, also known as phase quantization, and the result of the trigonometric function is quantized to 16 bits, including a sign bit, to generate data for all quadrants.

[0023] 3-2) According to different modulation indices, the modulation phase corresponding to a code element or symbol can be expressed as That is, for GFSK or FSK with modulation index h = 1, one symbol corresponds to phase π, while for GMSK or MSK with modulation index h = 0.5, one symbol corresponds to phase π / 2. Correspondingly, the phase accumulation value is The modulation modes are also different due to the 8-fold oversampling relationship. In GFSK or FSK modulation mode, In GMSK or MSK modulation mode,

[0024] 3-3) The NCO signal is generated by the data stored in the local ROM table and the phase accumulation value By controlling the phase change and changing the ROM table access address to generate the corresponding I and Q components, the final local modulation sequence can be expressed as

[0025] The received data sequence is multiplied with the local modulation sequence and its conjugate sequence by using the effective sampling point signal to obtain the corresponding correlation sequence. The specific implementation process is:

[0026] 1) The reference basis for the multiplication of the correlation sequence is the effective sampling point signal, and its rate is determined by the code rate set by the receiving system, and the code rate is R b =9600×2 n b / s, n is a non-negative integer.

[0027] 2) The calculation expression of the related sequence is shown as follows:

[0028]

[0029] According to the sequence group length N g The two multiplication results are accumulated and integrated respectively to obtain the correlation integral value of each symbol and the local reference signal, and the average value of the correlation integral value is calculated. The symbol integral value is based on the number of symbol sampling points (i.e., the packet length N g ) as the unit to obtain the integral value, the calculation process is shown as follows:

[0030]

[0031] The power of M1(n) and M2(n) is calculated, and the output power difference is used as the final demodulated and despread soft bit data. The calculation process is shown in the following formula:

[0032]

[0033] The present invention has the following beneficial effects:

[0034] 1. Since the present invention adopts a joint demodulation and despreading algorithm structure, the cumulative integration and data truncation operations in the calculation process are reduced, the integrity of the received signal is retained to a great extent, which is beneficial to subsequent decoding.

[0035] 2. The present invention integrates the same calculation processes in the demodulation and despreading processes, mainly the multiplication, addition and division processes related to integration, which reduces the calculation complexity and saves the resources required for implementation.

[0036] 3. The present invention combines the soft bit decision algorithm with joint demodulation and despreading, fully utilizes the information of the channel transmission signal, directly sends it to the decoder for decoding processing, reduces the impact of misjudgment after demodulation, and greatly improves the performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 It is a structural block diagram of the GMSK spread spectrum transmitter of the present invention.

[0039] Figure 2 This is a block diagram of the GMSK receiver system of the present invention.

[0040] Figure 3 The present invention is a flow chart of the soft bit joint demodulation and despreading method based on GMSK.

[0041] Figure 4 This is a circuit diagram for implementing the Gold pseudo-random sequence generator of the present invention.

[0042] Figure 5 The present invention is a circuit diagram for realizing the soft bit joint demodulation and despreading method based on GMSK. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] In view of the GMSK spread spectrum communication receiver described in the background technology, it is found in the implementation process that the structure of the traditional GMSK receiver that demodulates first and then despreads is not conducive to the correct decoding of the receiver due to repeated calculations, truncation errors, loss of precision, etc., and the bit error rate is high. In addition, since the GMSK joint demodulation and despreading output is soft bit information, the relevant log-likelihood ratio information is also required in the Viterbi decoding process, and the relevant information between the received code elements of the joint demodulation and despreading results can be fully utilized to improve the demodulation performance of the receiver and reduce the system bit error rate.

[0045] like Figures 1 to 5The embodiment of the present invention provides a method for implementing soft bit joint demodulation and despreading based on GMSK, which usually requires a transmitter device and a receiver device to jointly complete the communication function of the system. In the embodiment, the transmitter performs direct sequence spread spectrum and GMSK modulation on the coded sequence, and transmits the modulated signal through radio frequency; and the receiver captures and synchronizes the modulated signal of the transmitter, and then completes demodulation, despreading and decoding.

[0046] In the embodiment of the present invention, the structural block diagram of the GMSK spread spectrum transmitter is as follows: Figure 1 As shown. The original data sequence is processed by channel coding and modulo-two addition with the local spread spectrum code sequence to obtain the spread spectrum sequence, which is then modulated by GMSK. The impulse response of the Gaussian filter can be expressed as:

[0047]

[0048] In the formula, B is the 3dB bandwidth of the Gaussian filter.

[0049] Then the step response of the Gaussian filter is:

[0050]

[0051] In the formula,

[0052] The spread spectrum sequence passes through the Gaussian filter and enters the integrator for integration, and the phase function θ(t) can be obtained, which can be expressed as a n The input data is not returned to zero.

[0053] Then the GMSK modulated signal is:

[0054]

[0055] From the above formula, it can be found that θ(t) is determined by the data sequence after spread spectrum, and the baseband signals cosθ(t) and sinθ(t) of the orthogonal branches can be obtained, which are finally multiplied with the carrier and added to obtain the GMSK transmission signal.

[0056] In an embodiment of the present invention, Figure 2 The implementation flow chart of the GMSK receiver is given, and the specific implementation steps are as follows:

[0057] First, the ADC sampled and quantized data is digitally down-converted, which can be obtained by multiplying the received signal sequence x(n) with a rotating complex vector, that is, The rotated complex vector can be generated by the NCO, which generates two orthogonal signals based on the Cordic algorithm design.

[0058] Secondly, in order to reduce the data sampling rate, the down-converted signal is digitally low-pass filtered and decimated. The filtering is to remove the high-frequency components in the signal to prevent aliasing; and the decimation is to retain the sampling points with an interval of M in the original signal, which is M-fold downsampling.

[0059] Furthermore, the local synchronization header data is correlated with the down-sampled signal to capture the frame header position for subsequent tracking and correction of frequency deviation.

[0060] Furthermore, the frequency offset is estimated according to the correlation of the training sequence, and the phase of the subsequent received signal is continuously adjusted.

[0061] It is worth noting that the joint demodulation and despreading method proposed in the present invention uses the phase-adjusted data as the input signal, and the output signal is the soft bit data after demodulation and despreading.

[0062] Finally, the soft bit data is descrambled, decoded, etc. Based on the above steps, a complete receiver embodiment can be constructed, and the embodiment of the present invention is described in detail below.

[0063] Figure 3 A flowchart of a GMSK-based soft bit joint demodulation and despreading method provided by an embodiment of the present invention is given. As a reference, the receiver can execute the following steps: Figure 3 The joint joint modulation and expansion method shown in the figure comprises:

[0064] Step 301: Group the received data sequence according to the number of oversampling points based on the spreading factor.

[0065] The embodiment of the present invention firstly follows Figure 2 The signal is processed by the process shown in the figure to obtain a complex signal sequence r(n)=r i (n)+jr q (n) is the input of joint demodulation and despreading, and its length is L, r i (n) and r q (n) are the I and Q components of the received signal respectively, n is the number of sampling points, n=0, 1, 2, ..., L-1.

[0066] Specifically, the length of the sequence after grouping is N g , which can be expressed as where n sf is the configurable spreading factor, k s is the sampling multiple, n sf =1,2,3,4...,k s = 8. Number of groups after grouping G is a positive integer. Therefore, the length L of the received sequence r(n) must be the packet length N gThe receiver determines the number of sampling points to be demodulated according to the number of received bytes.

[0067] Step 302: Generate a local spreading code using a pseudo-random sequence, and multiply the spreading code with a corresponding NCO signal to obtain a local modulation sequence.

[0068] The embodiment of the present invention adopts a direct sequence spread spectrum system, uses two m sequences to generate a Gold sequence as a local pseudo-random code s(n), then changes the phase change according to the 0 or 1 value of the current pseudo-random code s(n), and finally uses the changed phase value Get current NCO data and And serve as the I and Q components of the local modulation sequence.

[0069] Specifically, first, two m-sequences m1(n) and m2(n) are constructed based on the initial value of the polynomial, and the two sequences have the same length and the same code rate. Figure 4 The Gold sequence generation circuit diagram shown in the figure, the local pseudo-random code s(n) sequence is formed by multiplying the preferred pairs. Figure 4 In the embodiment, two sequences m1(n) and m2(n) are generated by 31 registers performing shift operations and XOR operations according to their initial states and valid enable signals, and the two results are XORed to generate the final pseudo-random sequence signal.

[0070] In the embodiment of the present invention, the NCO signal is generated by the data stored in the local ROM table, and the phase accumulation value is By controlling the phase change and changing the ROM table access address to generate the corresponding I and Q components, the final local modulation sequence can be expressed as

[0071] Specifically, for the generation of ROM data, the range of 0 to 2π is divided into 65536 phases, also known as phase quantization, and the result of the trigonometric function is quantized to 16 bits, including a sign bit, to generate data for all quadrants.

[0072] According to different modulation indices, the modulation phase corresponding to a code element or symbol can be expressed as That is, for GFSK or FSK with modulation index h = 1, one symbol corresponds to phase π, while for GMSK or MSK with modulation index h = 0.5, one symbol corresponds to phase π / 2. Correspondingly, the phase accumulation value is The modulation modes are also different due to the 8-fold oversampling relationship. In GFSK or FSK modulation mode, In GMSK or MSK modulation mode,

[0073] Furthermore, in order to save resource area during the implementation process, the area of ​​the ROM table should be as small as possible. In addition, according to the above-mentioned fixed phase change, the corresponding trigonometric function value is also determined. Moreover, by using the symmetry of the trigonometric function within 0 to 2π, taking the data within the range of 0 to π / 2 as the reference, the data in other ranges can be obtained in sequence according to the phase relationship. The specific corresponding ROM table data is shown in Table 1.

[0074] Table 1 ROM table storage data

[0075] Address MSB(I) LSB(Q) 0 32767 0 1 32137 6392 2 30272 12539 3 27244 18204 4 23169 23169 5 18204 27244 6 12539 30272 7 6392 32137 8 0 32767

[0076] Step 303: using the effective sampling point signal, multiplying the received data sequence with the local modulation sequence and its conjugate sequence respectively to obtain a corresponding correlation sequence;

[0077] In this step, the calculation of the correlation value is mainly obtained by complex multiplication, and the calculation formula of the correlation sequence is:

[0078]

[0079] In the embodiment of the present invention, the calculation of the correlation sequence requires complex multiplication and conjugate complex multiplication at the same time. In order to save resources, considering the correlation between the real and imaginary parts of the complex multiplication and the conjugate complex multiplication, it can be simplified to four multipliers and four adders to complete the calculation of the real and imaginary parts of the results.

[0080] Specifically, the reference basis for the multiplication of the correlation sequence is the effective sampling point signal, and its rate is determined by the code rate set by the receiving system, and the code rate is R b =9600×2 n b / s, n is a non-negative integer.

[0081] Step 304: Group the sequence according to the length N g Accumulate the integrals of the two multiplication results respectively to obtain the correlation integral value of each code element and the local reference signal, and calculate the average value of the correlation integral value;

[0082] The symbol integral value is based on the number of symbol sampling points (i.e., the packet length N g ) as the unit to obtain the integral value, the calculation process is shown as follows:

[0083]

[0084] Step 305: Calculate the power of the two average values, and output the power difference as the final demodulated and despread soft bit data.

[0085] The power of M1(n) and M2(n) is calculated, and the output power difference is used as the final demodulated and despread soft bit data. The calculation process is shown in the following formula:

[0086]

[0087] In an embodiment of the present invention, Figure 5 The circuit diagram for implementing the soft bit joint demodulation and despreading method is provided. Figure 5 , the data of the I and Q branches in the received signal are passed through 4 multipliers and 4 adders to obtain the I and Q data of the related signal, and then the integrals are accumulated and the power value and power difference are obtained. Among them, the Gold sequence generator is controlled by the effective signal of the received data, and the spread spectrum code determines the accumulation of the phase. In this embodiment, the addition and subtraction of the address is controlled to read the cos and sin data in the ROM. For the throughput of data input and output, according to the packet length N g As the processing unit, each N g The input data of each sampling point outputs a valid soft bit data, which is determined by the spreading factor.

[0088] Compared with the conventional FSK modulation scheme of first demodulating and then despreading, the present invention effectively improves the demodulator performance; at the same time, the present invention can also be applied to FSK modulation systems with different modulation indexes, such as BFSK, MSK, GFSK, etc.; in addition, the circuit design scheme is optimized based on the ASIC implementation requirements, which greatly saves resource area.

[0089] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A soft bit joint demodulation and despreading method based on GMSK, characterized in that: The steps include: Step 1: Group the received data sequence according to the number of oversampling points based on the spreading multiple; Step 2: Use the Gold pseudo-random sequence to generate a local spreading code, and multiply the spreading code with the corresponding NCO signal to obtain a local modulation sequence; Step 3: Use the effective sampling point signal to multiply the received data sequence with the local modulation sequence and its conjugate sequence to obtain the corresponding correlation sequence; Step 4: Accumulate and integrate the two multiplication results according to the sequence group length to obtain the correlation integral value between each code element and the local modulation sequence, and calculate the average value of the correlation integral value; Step 5: Calculate the power of the two average values, and output the power difference as the final demodulated and despread soft bit data; Repeat the above steps until all received data are processed.

2. The GMSK-based soft bit joint demodulation and despreading method according to claim 1, characterized in that: The received data sequence in step 1 is a complex signal sequence, and its length is , expressed as , and They are the I and Q components of the received signal respectively. is the number of sampling points, , Is an imaginary unit.

3. The GMSK-based soft bit joint demodulation and despreading method according to claim 2, characterized in that: The length of the sequence after grouping in step 1 is , expressed as ,in is a configurable spreading factor, is the sampling multiple, , ; Number of groups after grouping , Is a positive integer.

4. The GMSK-based soft bit joint demodulation and despreading method according to claim 3, characterized in that: The received data sequence Length Must be the length of the group The receiver determines the number of sampling points to be demodulated according to the number of received bytes.

5. The GMSK-based soft bit joint demodulation and despreading method according to claim 4, characterized in that: The specific process in step 2 is: Step 201: Use two m sequences to generate a Gold sequence as a local pseudo-random code ; Step 202: changing the phase change according to the current pseudo-random code 0 / 1 value; Step 203: Using the changed phase value Get current NCO data and , and used as the I and Q components of the local modulation sequence.

6. The GMSK-based soft bit joint demodulation and despreading method according to claim 1 or 5, characterized in that: The GMSK spread spectrum system is direct sequence spread spectrum, which uses a pseudo-random sequence to encode the original transmission sequence; the pseudo-random sequence uses a Gold sequence, which is obtained by multiplying two m sequences with equal code lengths and the same code rate.

7. The GMSK-based soft bit joint demodulation and despreading method according to claim 5, characterized in that: The NCO signal in step 203 is generated by the data stored in the local ROM table, and the phase accumulation value is By controlling the phase change and changing the ROM table access address to generate the corresponding I and Q components, the final local modulation sequence can be expressed as ( ).

8. The GMSK-based soft bit joint demodulation and despreading method according to claim 7, characterized in that: According to different modulation indices, the modulation phase corresponding to a code element or symbol can be expressed as Given; that is, for the modulation index For GFSK or FSK, one symbol corresponds to the phase , while for the modulation index For GMSK or MSK, one symbol corresponds to the phase , then the corresponding phase accumulation value The modulation modes are also different due to the 8-fold oversampling relationship. In GFSK or FSK modulation mode, , while in GMSK or MSK modulation mode there is .

9. The GMSK-based soft bit joint demodulation and despreading method according to claim 1, characterized in that: The rate of the effective sampling point signal in step 3 is determined by the bit rate set by the receiving system, and the bit rate is , A non-negative integer.

10. The GMSK-based soft bit joint demodulation and despreading method according to claim 7, characterized in that: The calculation process of the correlation sequence in step 3 is as follows: ； The code element integral value in step 4 is obtained by taking the code element sampling points as the unit, and the calculation process is shown in the following formula: ； The soft bit data in step 5 is the power difference of the average value, and the calculation process is shown in the following formula: ； in To receive the data sequence, is the local modulation sequence, for The conjugated sequence of and denote related sequences and conjugated related sequences, respectively; Indicates the number of sampling points, is the length of the sequence after grouping, and denote the average of the correlation integral and the average of the conjugate correlation integral, respectively; , They are The I and Q components, and They are I and Q components of the and They are and The power value, express and The power difference.

Images