A segmented frequency conversion planning method for radio frequency signals

Abstract

The present invention discloses a segmented frequency conversion planning method for radio frequency signals, belonging to the field of radio frequency signal processing, including: performing down-conversion on the radio frequency time-domain signal, extracting the joint distribution of the radio frequency time-domain signal, and performing spectrum analysis to determine the frequency coverage range of the radio frequency signal; establishing an objective optimization function in the process of sub-band division of the radio frequency signal; randomly dividing the radio frequency signal into N sub-bands based on the frequency coverage range of the radio frequency signal, and using the crossover and mutation of the genetic algorithm to N optimize the sub-bands, and output the optimal continuous sub-bands; based on the output optimal continuous sub-bands, calculate the frequency conversion control signals of the sub-bands in the optimal continuous sub-bands. This solution can realize the segmented frequency conversion planning of complex radio frequency signals to meet subsequent complex spectrum requirements. Improve the signal processing ability of the overall system, reduce spurious interference, and adapt to complex and changeable radio frequency application scenarios.

Description

Technical Field

[0001] The present invention relates to the field of radio frequency signal processing, and specifically to a segmented frequency conversion planning method for radio frequency signals. Background Art

[0002] In many fields such as modern communication and radar detection, the frequency conversion of radio frequency signals plays a crucial role. Traditional frequency conversion methods often adopt a single frequency conversion mode. For example, a fixed local oscillator frequency is directly used to mix with the input radio frequency signal. In the face of complex spectrum requirements and high-dynamic-range signal processing, this method has problems such as decreased spectrum purity, serious spurious interference, and inability to flexibly adapt to efficient processing in different frequency bands. With the development of technology, the requirements for the frequency conversion accuracy, efficiency, and adaptability of radio frequency signals are increasing day by day. There is an urgent need for an innovative frequency conversion planning method to meet diverse requirements. Summary of the Invention

[0003] In view of the above deficiencies in the prior art, the present invention provides a segmented frequency conversion planning method for radio frequency signals, which uses target optimization and a deep genetic algorithm to achieve the segmented frequency conversion planning of radio frequency signals and improve the performance of radio frequency signal frequency conversion processing.

[0004] In order to achieve the above invention purpose, the technical solution adopted by the present invention is as follows:

[0005] Provide a segmented frequency conversion planning method for radio frequency signals, which includes:

[0006] Step S1: Down-convert the radio frequency time-domain signal, extract the joint distribution of the radio frequency time-domain signal, and perform spectrum analysis to determine the frequency coverage range of the radio frequency signal;

[0007] Step S2: Establish an objective optimization function in the process of sub-band division of the radio frequency signal;

[0008] Step S3: Randomly divide the radio frequency signal into N sub-bands based on the frequency coverage range of the radio frequency signal, and use the crossover and mutation of the genetic algorithm to N optimize the

[0009] sub-bands, and output the optimal continuous sub-bands;

[0010] Step S4: Calculate the frequency conversion control signal of the sub-bands in the optimal continuous sub-bands based on the output optimal continuous sub-bands.

[0011] Step S11: Down-convert the radio frequency time-domain signal and extract the joint distribution of the radio frequency time-domain signal based on the Fourier transform ;

[0012] ;

[0013] Wherein, f is the signal frequency, t is the time domain of the signal, is the Fourier transform, is the carrier frequency, j is the imaginary unit, is the Fourier convolution, is the transfer function of the reconfigurable filter bank, e is the natural constant;

[0014] Step S12: Based on the joint distribution perform spectrum analysis on the radio frequency time domain signal to determine the frequency coverage range of the radio frequency signal.

[0015] Furthermore, the target optimization function is specifically:

[0016] ;

[0017] Wherein, n is the number of the sub-band, is the N set of divided sub-bands, is the set of bandwidths of all sub-bands, n is the rated bandwidth of the th n sub-band, is the transmission power of the n th sub-band, is the frequency band power weight coefficient, K is the rectangular coefficient of the filter, is the intermediate frequency of the subsequent signal processing link, is the n th preset local oscillator frequency of the sub-band, is the intermediate frequency bandwidth, is the phase noise power spectral density of the local oscillator source, is the signal-to-noise ratio threshold, is the minimum frequency interval ratio, is the n th N termination frequency of the sub-band, is the number of sub-bands, NThe termination frequency of the sub-band, as a constraint condition, d for the integral operation, represents the minimum value function.

[0018] Furthermore, step S3 includes:

[0019] Step S31: Based on the frequency coverage range of the radio frequency signal , and randomly divide the radio frequency signal into N continuous sub-bands according to the performance requirements of the system;

[0020] ;

[0021] Among them, is the frequency range of the N th sub-band, is the starting frequency of the N th sub-band, is the termination frequency of the N th sub-band, and ;

[0022] After the sub-band division, the transfer function of the reconfigurable filter bank is;

[0023] ;

[0024] Among them, Q is the dynamic adjustment coefficient, is the carrier frequency of the n th sub-band;

[0025] Step S32: Input the N continuous sub-bands currently divided into the target optimization function to calculate the current fitness value , is the scaling factor, e is the natural constant;

[0026] Step S33: Use the ideal bandwidth of each sub-band currently divided as a genetic gene to construct the chromosome encoding of the genetic algorithm , , is the N th genetic gene;

[0027] Step S34: Based on the chromosome encoding , calculate the bandwidth difference between any two genetic genes, and select the genetic gene combination with the smallest bandwidth difference ;

[0028] ;

[0029] Wherein, are respectively the two genetic genes with the smallest bandwidth difference in the chromosome encoding R , and are respectively the numbers of the two genetic genes with the smallest difference;

[0030] Step S35: According to the encoding positions of the genetic gene combination within the chromosome encoding , exchange the encoding positions between the genetic genes to form a chromosome encoding after a set of genetic gene crossover transformations, and divide the corresponding to the chromosome encoding into N sub - frequency bands and input them into the target optimization function to calculate the fitness value ;

[0031] Step S36: Compare the magnitudes of the fitness value and the fitness value ;

[0032] If , then execute steps S37 - S310;

[0033] If , then execute step S311;

[0034] Step S37: Use the smallest genetic gene in the chromosome encoding as the mutation basis, and perform mutation according to the set mutation rate to obtain the mutated genetic gene , and the remaining genetic genes after mutation are to obtain the mutated chromosome encoding , is the genetic gene in the chromosome encoding , k is the number of the genetic gene in the chromosome encoding ;

[0035] Step S38: Divide the corresponding to the chromosome encoding into N sub - frequency bands and input them into the target optimization function to calculate the fitness value ;

[0036] Step S39: Return to step S37, and use the smallest genetic gene in the chromosome encoding As the basis for mutation, steps S37 - S38 are executed to calculate the fitness value ;

[0037] Step S310: Until after M times of genetic gene mutations, M fitness values are obtained, and M fitness values are screened out, and the minimum fitness value in them;

[0038] If , the chromosome encoding corresponding to the minimum fitness value is output, and the sub - frequency bands corresponding to the chromosome encoding are used as the planned optimal continuous sub - frequency bands; N ;

[0039] If , the sub - frequency bands corresponding to the chromosome encoding N are used as the planned optimal continuous sub - frequency bands;

[0040] Step S311: Screen out the bandwidth combination with the second - smallest bandwidth difference in the chromosome encoding , and , are the two genetic genes with the second - smallest bandwidth difference in the chromosome encoding , are the numbers of the two genetic genes with the second - smallest bandwidth difference in the chromosome encoding ; And step S35 is executed to form the chromosome encoding after the second genetic gene crossover transformation, and the fitness value is calculated;

[0041] Step S312: Compare the size between the fitness value and the fitness value ;

[0042] If , step S37 - S310 are executed using the chromosome encoding ;

[0043] If , return to step S311, and continue the crossover transformation of genetic genes in the chromosome encoding until the optimal continuous sub - frequency bands are output;

[0044] Step S313: If the chromosome encoding After all the genetic genes have been cross-transformed and it is still impossible to output the optimal continuous sub-band, return to step S31 and randomly divide again N continuous sub-bands, and execute steps S31 - S312.

[0045] Further, the frequency conversion control signal ;

[0046] ;

[0047] Among them, is the gain coefficient, is the bandwidth of the n th sub-band in the optimal continuous sub-band, is the frequency of the n th sub-band in the optimal continuous sub-band, e is the natural constant.

[0048] The beneficial effects of the present invention are as follows: This solution can achieve segmented frequency conversion planning for complex radio frequency signals to adapt to subsequent complex spectrum requirements. By constructing an objective optimization function to optimize the frequency segmentation of radio frequency signals, the optimal band power and band continuity are followed during the frequency segmentation process, while meeting the signal noise requirements. At the same time, the divided bands are optimized based on the genetic algorithm, increasing the accuracy of the segmented frequency conversion planning of radio frequency signals and making the frequency conversion planning more efficient, achieving precise frequency conversion for radio frequency signals in different bands, improving the signal processing ability of the overall system, reducing spurious interference, and adapting to complex and changing radio frequency application scenarios. Description of the Drawings

[0049] Figure 1 is the flowchart of the segmented frequency conversion planning method for radio frequency signals. Detailed Embodiments

[0050] The following describes the detailed embodiments of the present invention to facilitate those skilled in the art to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the detailed embodiments. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions created using the concept of the present invention are within the scope of protection.

[0051] As Figure 1 shown, a segmented frequency conversion planning method for radio frequency signals includes:

[0052] Step S1: Perform down-conversion on the radio frequency time-domain signal, extract the joint distribution of the radio frequency time-domain signal, and perform spectrum analysis to determine the frequency coverage range of the radio frequency signal. Step S1 specifically includes:

[0053] Step S11: Down-convert the RF time-domain signal and extract the joint distribution of the RF time-domain signal based on Fourier transform ;

[0054] ;

[0055] wherein, f is the signal frequency, t is the time domain of the signal, is the Fourier transform, is the carrier frequency, j is the imaginary unit, is the Fourier convolution, is the transfer function of the reconfigurable filter bank, e is the natural constant;

[0056] Step S12: Perform spectrum analysis on the RF time-domain signal based on the joint distribution to determine the frequency coverage range of the RF signal ; In this embodiment, by extracting the joint distribution of the RF time-domain signal , a spectrum diagram of the RF signal can be drawn during spectrum analysis, and the frequency-domain diagram can provide clear frequency characteristics of the RF signal.

[0057] Step S2: Establish the objective optimization function in the sub-band division process of the RF signal ;

[0058] ;

[0059] wherein, n is the sub-band number, is the set of N sub-bands divided, is the set of bandwidths of all sub-bands, is the rated bandwidth of the n th sub-band, is the transmit power of the n th sub-band, is the starting frequency of the n th sub-band, is the band power weight coefficient, is the band continuity weight coefficient. Under general conditions, the weight coefficients are taken, , K is the rectangular coefficient of the filter, is the intermediate frequency of the subsequent signal processing link, is the nThe local oscillator frequency preset for each sub-band, is the IF bandwidth, is the phase noise power spectral density of the local oscillator source, is the noise power spectral density of the sub-band, is the signal-to-noise ratio threshold, is the minimum frequency interval ratio, For the n The stop frequency of each sub-band, N is the number of sub-bands, For the N The stop frequency of each sub-band, As constraints, d For the integral operation, Represents the minimum value function.

[0060] Step S3: Randomly divide the radio frequency signal into N sub-bands, and use the crossover and mutation of genetic algorithms to N The sub-bands are optimized and the optimal continuous sub-bands are output. Step S3 specifically includes:

[0061] Step S31: Based on the frequency coverage of the radio frequency signal , and randomly divide the RF signal into N A continuous sub-band;

[0062] ;

[0063] in, For the N The frequency range of the sub-bands, For the N The starting frequency of each sub-band, For the N The stop frequency of the sub-band, and ;

[0064] After sub-band division, the transfer function of the filter bank can be reconstructed for;

[0065] ;

[0066] in, Q is the dynamic adjustment coefficient, For the n The reconfigurable filter bank can dynamically change the filtering characteristics to adapt to different signals or environmental changes, and can meet the needs of variable frequency signals by adjusting parameters or structures while maintaining stability.

[0067] Step S32: divide the currently dividedN A series of consecutive sub - frequency bands Input the target optimization function and calculate the current fitness value , is the scaling factor, e and \(e\) is the natural constant;

[0068] Step S33: Use the ideal bandwidth of each currently divided sub - frequency band as a genetic gene to construct the chromosome encoding of the genetic algorithm , , is the N th genetic gene;

[0069] Step S34: Based on the chromosome encoding , calculate the bandwidth difference between any two genetic genes, and select the genetic gene combination with the smallest bandwidth difference ;

[0070] ;

[0071] Among them, are respectively the two genetic genes with the smallest bandwidth difference in the chromosome encoding R , are respectively the numbers of the two genetic genes with the smallest difference;

[0072] Step S35: According to the encoding positions of the genetic gene combination in the chromosome encoding , exchange the encoding positions between the genetic genes to form a chromosome encoding after genetic gene crossover transformation , and input the sub - frequency bands corresponding to the chromosome encoding N into the target optimization function to calculate the fitness value ;

[0073] Step S36: Compare the sizes of the fitness values and ;

[0074] If , then execute steps S37 - S310;

[0075] If , then execute step S311;

[0076] The genetic algorithm logic adopted in this embodiment first performs crossover variation of genetic genes to make a large-scale change in chromosome coding, and then performs genetic mutation to achieve a small-scale change in chromosome coding, ensuring that the optimal continuous sub-band can be output quickly.

[0077] Step S37: Use the smallest genetic gene in the chromosome coding as the basis for mutation, and perform mutation according to the set mutation rate . Generally, the mutation rate is taken as 0.1 to obtain the mutated genetic gene . In order to ensure that the overall frequency range of the RF signal remains unchanged, the mutation value needs to be evenly distributed to other genetic genes to achieve the effect of overall genetic gene mutation. After mutation, the remaining genetic genes are , and the mutated chromosome coding is obtained. is the genetic gene in the chromosome coding , k is the number of the genetic gene in the chromosome coding ;

[0078] Step S38: Input the sub-bands correspondingly divided from the chromosome coding N into the target optimization function to calculate the fitness value ;

[0079] Step S39: Return to Step S37, use the smallest genetic gene in the chromosome coding as the basis for mutation, and execute Steps S37 - S38 to calculate the fitness value ;

[0080] Step S310: Until after M times of genetic gene mutation, M fitness values are obtained, and M fitness values are screened out, and the minimum fitness value in them is obtained;

[0081] If , then output the chromosome coding corresponding to the minimum fitness value , and use the sub-bands correspondingly divided from the chromosome coding N as the planned optimal continuous sub-band;

[0082] If , then use the chromosome coding The correspondingly divided N sub - frequency bands are used as the optimal continuous sub - frequency bands planned;

[0083] Step S311: Select the bandwidth combination with the second - smallest bandwidth difference in the chromosome encoding and, and , are the two genetic genes with the second - smallest bandwidth difference in the chromosome encoding ; are the numbers of the two genetic genes with the second - smallest bandwidth difference in the chromosome encoding ; and execute Step S35 to form the chromosome encoding after the second genetic gene crossover transformation , and calculate the fitness value ;

[0084] Step S312: Compare the magnitudes of the fitness value and the fitness value ;

[0085] If , then use the chromosome encoding to execute Steps S37 - S310;

[0086] If , then return to Step S311 and continue the crossover transformation of genetic genes in the chromosome encoding until the optimal continuous sub - frequency bands are output;

[0087] Step S313: If after all the genetic genes in the chromosome encoding have been crossover - transformed and the optimal continuous sub - frequency bands still cannot be output, then return to Step S31, randomly re - divide N continuous sub - frequency bands, and execute Steps S31 - S312.

[0088] Step S4: Based on the output optimal continuous sub - frequency bands, calculate the frequency conversion control signal of the sub - frequency bands in the optimal continuous sub - frequency bands ;

[0089] ;

[0090] wherein, is the gain coefficient, is the bandwidth of the n th sub - frequency band in the optimal continuous sub - frequency bands, is the frequency of the n th sub - frequency band in the optimal continuous sub - frequency bands, e is the natural constant.

[0091] ​The present invention can achieve segmented frequency conversion planning for complex radio frequency signals to meet subsequent complex spectrum requirements. By constructing an objective optimization function to optimize the frequency segmentation of radio frequency signals, the optimal frequency band power and frequency band continuity are followed during the frequency segmentation process, while meeting the signal noise requirements. At the same time, based on the genetic algorithm, the divided frequency bands are optimized, which increases the accuracy of the segmented frequency conversion planning of radio frequency signals and makes the frequency conversion planning more efficient, realizes precise frequency conversion for radio frequency signals in different frequency bands, improves the signal processing ability of the overall system, reduces spurious interference, and adapts to complex and changeable radio frequency application scenarios.

Claims

1. A method for segmented frequency conversion planning of radio frequency signals, characterized in that Including: Step S1: Down-convert the RF time-domain signal, extract the joint distribution of the RF time-domain signal, and perform spectrum analysis to determine the frequency coverage range of the RF signal; Step S2: Establish an objective optimization function in the process of sub-band division of the RF signal; Step S3: Randomly divide the radio frequency signals based on the frequency coverage range of the radio frequency signals N into sub-bands, and use the crossover and mutation of the genetic algorithm to N optimize the sub-bands and output the optimal continuous sub-bands; Step S4: Based on the output optimal continuous sub-bands, calculate the frequency conversion control signals of the sub-bands in the optimal continuous sub-bands; The target optimization function Specifically: ； Wherein, n is the number of the sub - frequency band, is the divided N set of sub - frequency bands, is the set of bandwidths of all sub - frequency bands, is the rated bandwidth of the n th sub - frequency band, is the transmission power of the n th sub - frequency band, is the starting frequency of the n th sub - frequency band, is the frequency band power weight coefficient, is the frequency band continuity weight coefficient, K is the rectangular coefficient of the filter, is the intermediate frequency of the subsequent signal processing link, is the n th preset local oscillator frequency of the sub - frequency band, is the intermediate frequency bandwidth, is the phase noise power spectral density of the local oscillator source, is the noise power spectral density of the sub - frequency band, is the signal - to - noise ratio threshold, is the minimum frequency interval ratio, is the n th termination frequency of the sub - frequency band, N is the number of sub - frequency bands, is the N th termination frequency of the sub - frequency band, is the constraint condition, d is the integral operation, represents the minimum - value function; The specific content of the said Step S3 includes: Step S31: Based on the frequency coverage range of the radio frequency signal , and randomly divide the radio frequency signal into N continuous sub-bands according to the performance requirements of the system; ； in, For the N The frequency range of the sub-bands, For the N The starting frequency of each sub-band, For the N The stop frequency of the sub-band, and ; After sub-band division, the transfer function of the reconfigurable filter bank is; ； Among them, Q is the dynamic adjustment coefficient, is the n carrier frequency of the Step S32: Divide the currently divided N consecutive sub - frequency bands and input them into the target optimization function to calculate the current fitness value , where is the scaling factor, e and is the natural constant; Step S33: Using the ideal bandwidth of each currently divided sub-band as a genetic gene, construct the chromosome encoding of the genetic algorithm , , is the N th genetic gene; Step S34: Based on the chromosome coding , calculate the bandwidth difference between any two genetic genes, and screen the genetic gene combination with the smallest bandwidth difference ; ； Among them, are respectively the chromosome codes R of the two genetic genes with the smallest bandwidth difference in which are respectively the numbers of the two genetic genes with the smallest difference; Step S35: According to the genetic gene combination At the coding position within the chromosome coding Exchange the coding positions between the genetic genes To form a chromosome coding after genetic gene crossover transformation And divide the corresponding chromosome coding Into the N Sub - frequency bands and input them into the target optimization function To calculate the fitness value ; Step S36: Compare the fitness values with the fitness value to determine the magnitude relationship; If , then perform steps S37 - S310; If , then execute step S311; Step S37: Take the smallest genetic gene in the chromosome encoding as the basis for mutation, and perform mutation according to the set mutation rate to obtain the mutated genetic gene . After mutation, the remaining genetic genes are , obtaining the mutated chromosome encoding . is the genetic gene in the chromosome encoding , and k is the number of the genetic gene in the chromosome encoding .​ Step S38: Input the chromosome encoding corresponding to the N sub-bands divided into the target optimization function to calculate the fitness value ; Step S39: Return to Step S37, and use the smallest genetic gene in the chromosome encoding as the basis for mutation, and execute Steps S37 - S38 to calculate the fitness value ; Step S310: Until after M times of genetic mutations, M fitness values are obtained, and M fitness values are screened to find the minimum fitness value ; If , then output the chromosome encoding corresponding to the minimum fitness value , and use the sub-bands corresponding to the chromosome encoding to be divided as the planned optimal continuous sub-bands; N sub-bands as the planned optimal continuous sub-bands;​ If , then the chromosome coding correspondingly divided N sub-bands are used as the planned optimal continuous sub-bands; Step S311: Screen out the chromosome encoding The bandwidth combination with the second smallest bandwidth difference among , and , is the chromosome encoding The two genetic genes with the second smallest bandwidth difference among the chromosome encodings are the chromosome encoding The numbers of the two genetic genes with the second smallest bandwidth difference among the chromosome encodings; and execute step S35 to form the chromosome encoding after the second genetic gene crossover transformation , and calculate the fitness value ; Step S312: Compare the fitness values with the fitness value to determine the magnitude If , then use chromosome encoding to execute steps S37 - S310; If , return to step S311 and continue the crossover transformation of genetic genes in the chromosome encoding until the optimal continuous sub-band is output; Step S313: If after all the genetic genes in the chromosome encoding have been cross-transformed and the optimal continuous sub-band still cannot be output, then return to step S31 to randomly divide N a number of continuous sub-bands again and execute steps S31 - S312; The frequency conversion control signal ; ； Among them, is the gain coefficient, is the bandwidth of the n th sub-band in the optimal continuous sub-band, is the frequency of the n th sub-band in the optimal continuous sub-band, e is the natural constant.

2. The segmented frequency conversion planning method for radio frequency signals according to claim 1, wherein The said Step S1 includes: Step S11: Perform down-conversion on the radio frequency time-domain signal and extract the joint distribution of the radio frequency time-domain signal based on Fourier transform ; ； wherein, f is the signal frequency, t is the time domain of the signal, is the Fourier transform, is the carrier frequency, j is the imaginary unit, is the Fourier convolution, is the transfer function of the reconfigurable filter bank, e is the natural constant; Step S12: Based on the joint distribution Perform spectral analysis on the radio frequency time-domain signal to determine the frequency coverage range of the radio frequency signal .

Images