Time hopping signal parameter estimation and multi-hop merging method and system based on genetic algorithm

Abstract

The invention discloses a time hopping signal parameter estimation and multi-hop merging method and system based on a genetic algorithm, and belongs to the field of wireless communication. The system comprises a parameter generation unit, a time hopping solving unit, a frequency offset and phase offset compensation unit and a fitness calculation and reproduction unit. And the time hopping pattern-frequency offset-phase offset parameter encoding and generating unit is used for generating a time hopping pattern parameter, a frequency offset estimation parameter and a phase offset estimation parameter of each hop of signal according to a genetic algorithm encoding rule. By setting the maximum number of iterations and the fitness stagnation threshold, it is ensured that the genetic algorithm converges to the optimal solution within the finite time, and the decoding capacity of the time hopping signal and the overall performance of the system are improved. According to the method, under the condition that the time hopping pattern is unknown, the correct time hopping pattern, frequency offset and phase offset parameters can be searched through the genetic algorithm, the received signals are compensated and combined, phase alignment and coherent combination of multi-hop signals are achieved, remarkable signal-to-noise ratio gain is obtained, and a time hopping communication system can still conduct long-distance reliable communication under the condition of the low signal-to-noise ratio.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology and relates to an optimization method for joint estimation of signal parameters and multi-hop coherent merging of time-hopping communication systems based on genetic algorithms. It is suitable for long-distance reliable communication in low signal-to-noise ratio environments. Background Technology

[0002] In long-distance wireless communication scenarios, signal transmission path loss leads to a significant decrease in the signal-to-noise ratio (SNR) at the receiver, severely affecting communication reliability. Multi-hop coherent combining technology can effectively improve the SNR of the synthesized signal, but it requires strict frequency offset (CFO) compensation and phase offset (CPO) synchronization. Traditional parameter estimation methods suffer from the following technical bottlenecks: (1) high computational complexity of exhaustive search strategies; (2) gradient-based algorithms are prone to getting trapped in local optima; and (3) insufficient accuracy of parameter estimation under low SNR conditions.

[0003] Genetic algorithms, as a parallel global optimization algorithm, achieve collaborative search of multi-dimensional parameters in the solution space through population evolution mechanism and the action of genetic operators. They have advantages such as: (1) swarm intelligence characteristics to avoid local convergence; (2) adaptive adjustment of search step size; and (3) multi-objective optimization capability, and are particularly suitable for solving multi-dimensional parameter joint optimization problems in time-hopping communication systems. Summary of the Invention

[0004] To address the difficulty of multi-hop signal combining in time-hopping communication systems under low signal-to-noise ratio (SNR) environments, this invention aims to provide a method and system for time-hopping signal parameter estimation and multi-hop combining based on a genetic algorithm. Under unknown time-hopping patterns, the genetic algorithm searches for the correct time-hopping pattern, frequency offset, and phase offset parameters, and then compensates and combines the received signals to achieve phase alignment and coherent combining of multi-hop signals. This invention ensures that the algorithm converges to the optimal solution within a finite time by setting a maximum number of iterations and a fitness stagnation threshold, thereby improving the decoding capability of time-hopping signals and the overall system performance. Through signal compensation and coherent combining, a significant SNR gain is achieved, enabling reliable long-distance communication in time-hopping communication systems under low SNR conditions.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] This invention discloses a method for estimating time-hopping signal parameters and multi-hop merging based on a genetic algorithm. The method is characterized by: the transmitting end dividing the communication time into time frames and time slots to generate a time-hopping BPSK modulated signal exhibiting random characteristics in the time domain; the receiving end acquiring the disturbed time-hopping signal through ADC sampling, down-conversion, and matched filtering; quantizing and encoding key time-hopping parameters to form a genotype structure for genetic algorithm optimization; each parameter combination forming a searchable individual covering the multi-dimensional joint estimation space in time-hopping communication; initializing a genetic algorithm population containing multiple parameter combinations according to the encoding format, and setting iteration-related control parameters; and using crossover, mutation, and selection... The algorithm selects individuals to achieve global optimization of time hopping parameters; generates a new generation of time hopping patterns, frequency offset, and phase offset parameter combinations based on the current population; decodes the parameter combination of each individual and sequentially performs frequency offset correction, time hopping extraction, and phase compensation on the received signal; coherently merges the compensated multi-hop signals to form an aggregated signal result for evaluation; selects several individuals with the best performance to enter the next iteration; if the termination condition is not met, the iteration continues; if the termination condition is met, the search ends and the current optimal parameter combination is output; the optimal parameter combination is restored to the specific time hopping pattern, frequency offset, and phase offset values, and the received signal is recompensated; the compensated multi-hop signals are then coherently merged.

[0007] The present invention discloses a method for time-hopping signal parameter estimation and multi-hop coherent merging based on a genetic algorithm, comprising the following steps:

[0008] Step 1: The transmitting end divides the communication time into time frames and time slots, and controls the transmission position of the BPSK modulation signal according to the random time-hopping code, generating a time-hopping BPSK modulation signal with random characteristics in the time domain to enhance anti-interception capability.

[0009] The communication time window is divided into multiple time frames, and each time frame is further divided into multiple time slots. For each information symbol to be transmitted, BPSK modulation processing is performed first. Then, the modulation signal is controlled by the time hopping gating module to ensure that each symbol is transmitted only within the time slot position specified by the time hopping code. The time hopping code determines the time slot position of the signal in each time frame and is randomly generated within a preset range, so that the time hopping code presents randomness in the time domain, thereby generating a time-random BPSK modulation signal. The BPSK modulation signal is expressed as Equation (1):

[0010]

[0011] Where b(m) represents the polarized data, h s (·) represents the impulse response of the shaping filter, f c Where is the carrier frequency, and g(k) is the time slot position corresponding to the time hopping code.

[0012] Step 2: When the time-hopping signal is transmitted in the satellite-to-ground channel, it will be superimposed with interference such as Doppler frequency offset, random time delay, phase shift, and Gaussian white noise. The receiver obtains the disturbed time-hopping signal through ADC sampling, down-conversion, and matched filtering to prepare for subsequent compensation.

[0013] The time-hop BPSK modulated signal generated in step one will be affected by various interferences during its propagation through the channel after being transmitted by the antenna. These interferences mainly include random time delay, Doppler frequency offset, phase shift, and energy attenuation, while being superimposed with Gaussian white noise. The receiver signal model can then be expressed as equation (2):

[0014]

[0015] Where τ represents random delay, f d Doppler frequency shift, Let A represent random phase bias, A represent the energy attenuation coefficient, and w(t) represent noise. Only additive white Gaussian noise (AWGN) is considered.

[0016] The receiver performs ADC sampling, down-conversion, and matched filtering on the signal, resulting in the signal shown in equation (3):

[0017]

[0018] Where Δf represents the frequency offset of the received signal. This indicates the phase offset of the processed signal.

[0019] Step 3: Quantize and encode the key time-hopping parameters—time-hopping pattern, frequency offset, and phase offset—to form a genotype structure for genetic algorithm optimization. Each parameter combination forms a searchable individual, covering the multidimensional joint estimation space in time-hopping communication.

[0020] In the signal processing at the receiver, in order to optimize the time hopping pattern, frequency offset, and phase offset parameters using a genetic algorithm, these parameters need to be encoded first. The number of quantization bits for the time hopping pattern is... The time-jump pattern encoding parameters are The corresponding time jump pattern number N tp =x tp The number of phase-biased quantization bits is The corresponding phase bias coding parameters are The corresponding phase bias is For N f The data set contains a frequency offset parameter, with a frequency offset quantization bit count of D. f The corresponding frequency offset coding parameters are The corresponding frequency offset is Therefore, for each candidate parameter group, the number of bits for its genotype is: Each parameter combination forms a genotype, represented by a joint encoding of time-skip pattern, frequency offset, and phase offset. The encoded genotype signal serves as the input for subsequent genetic algorithms to initialize the population.

[0021] Step 4: Based on the encoding format, initialize a genetic algorithm population containing multiple parameter combinations and set the iteration-related control parameters. This initial population serves as the starting point for subsequent crossover, mutation, and selection processes, achieving global optimization of the jump-time parameters.

[0022] Based on the encoding format in step three, the iteration parameters of the genetic algorithm are set, including the generation counter t=0, the maximum number of iterations T, and the evolutionary stagnation counter t. same =0, the individual fitness convergence threshold H, and the maximum stagnation limit T same Randomly generate N e The combination of group jump time pattern, frequency offset, and phase offset forms the initial population, which serves as the input population for the genetic algorithm and is used for subsequent crossover and mutation operations.

[0023] Step 5: Generate a new generation of time-skipping patterns, frequency offsets, and phase offset parameter combinations based on the current population. Use gene crossover and mutation operations to improve the search breadth. Crossover exchanges partial information in different parameter fragments, while mutation introduces population diversity by randomly flipping individual gene loci.

[0024] Using the population P(t) output in step four, crossover and mutation operations are performed to generate a new generation population P(t+1). The crossover operation selects three loci from each of the three gene segments representing the time-hopping pattern, frequency offset, and phase offset, dividing each segment into six segments. Within each segment, some gene loci are exchanged to generate a new time-hopping pattern-frequency offset-phase offset combination. The mutation operation increases search diversity by randomly flipping a locus within the selected individuals. The new generation population P(t+1) represents the updated set of time-hopping pattern-frequency offset-phase offset parameters, preparing for subsequent decoding compensation and merging.

[0025] Step Six: After decoding the parameter combination of each individual signal, frequency offset correction, time hopping extraction, and phase compensation are performed on the received signal in sequence. The compensated multi-hop signals are then coherently combined to form the aggregated signal result used for evaluation.

[0026] Using the genotype of an individual in the population obtained in step five, a set of time-hopping pattern-frequency offset-phase offset parameters are solved. The frequency offset parameters are then used to compensate the received signal obtained in step two: r(n) = r(n)·e -j·2πfn Then, N is extracted from the received signal using the time-hopping pattern sequence number. f Group data Then, the phase offset parameter is used to compensate for each time jump signal. For N that have completed compensationf The signals are coherently combined, and the combined signal is: The merged signal serves as input for fitness evaluation and iterative updates.

[0027] Step 7: Evaluate the quality of the merged signal results for each individual, assessing the effectiveness of the compensation and merging. Select the best-performing individuals for the next iteration. If the best result does not show significant improvement in multiple iterations, mark it as entering a stagnant state and update the iteration record.

[0028] Based on the merge signal and fitness J obtained in step six, calculate the fitness result corresponding to each individual's genotype. J=|∑sign(r all )·r all | is the fitness function used in the joint estimation algorithm of time-hopping pattern-frequency offset-phase offset in the time-hopping communication system of this invention. It is obtained by evaluating the multi-hop signal r after frequency offset compensation and phase alignment. all The sign function and amplitude are weighted and summed, and the absolute value is taken to measure the phase consistency and energy aggregation effect of each hop signal. This cost function can reflect whether multi-hop signals achieve coherent merging and superposition enhancement after compensation. When the signal phase compensation and frequency offset compensation are accurate, the phase direction of the merged signal is consistent, the superposition energy is maximized, and increasing the value of J indicates that the compensation effect of the combination of hop pattern-frequency offset-phase offset parameters is good and the adaptability is high.

[0029] Sort the fitness results and select the top N with the best performance. e Group of data, the N e The set of data is called the optimal set. The optimal set serves as the parent set for the next iteration. The optimal individual P is recorded. best Fitness results J best (t+1), calculate the difference E = |J| between the current optimal individual fitness result and the previous optimal individual fitness result. best (t+1)-J best (t)|. When the fitness result difference E is less than the threshold value H, the iteration is considered to have stalled, and the stall counter t is set. same Increment by one. Update the iteration counter t so that t = t + 1.

[0030] Step 8: Determine if the current iteration count has reached the maximum limit, or if the fitness result has stagnated for more than the preset number of consecutive iterations. If the termination condition has not been met, continue iterating; if the termination condition has been met, end the search process and output the current optimal parameter combination.

[0031] The decision cycle has reached the maximum number of rounds T or the fitness result has stagnated for more than T rounds. sameIf the cycle does not reach the maximum number of rounds T and the fitness result stagnates and does not exceed T, then... same If the iteration reaches the maximum number of iterations T or the fitness result stagnates for more than T, then return to step three and repeat the iteration. same Then the loop terminates and the optimal combination is output.

[0032] Step Nine: Restore the optimal parameter combination output from Step Eight to the specific time-hopping pattern, frequency offset, and phase offset values, and recompensate the received signal. The compensated multi-hop signal undergoes coherent combining, significantly improving the overall signal-to-noise ratio and enhancing the system's communication reliability in low signal-to-noise ratio environments.

[0033] The optimal combination obtained in step eight is analyzed into time-hopping patterns, frequency offset, and phase offset parameters. These parameters are used to recover the actual time-hopping pattern information of the received signal. Based on the analyzed frequency offset and phase offset parameters, frequency and phase compensation are performed on each time-hopping signal. The compensated multi-hop signals are then coherently combined to obtain an enhanced combined signal. Through compensation and combining processes, the signal-to-noise ratio (SNR) of the combined signal is significantly improved, thereby enhancing the reception performance and reliability of the time-hopping communication system under low SNR conditions.

[0034] This invention discloses a method and system for time-hopping signal parameter estimation and multi-hop coherent merging based on a genetic algorithm. This method is used to achieve joint estimation and signal compensation of time-hopping patterns, frequency offsets, and phase offsets under unknown time-hopping patterns, thereby improving the reliability and anti-interference capability of low signal-to-noise ratio time-hopping communication systems. The system includes a time-hopping pattern-frequency offset-phase offset parameter encoding and generation unit, a genetic population initialization and update unit, a frequency offset compensation and time-hopping de-emitter unit, a phase offset compensation and merging unit, and a fitness calculation and convergence determination unit.

[0035] The time-hopping pattern-frequency offset-phase offset parameter encoding and generation unit is used to generate the time-hopping pattern parameters, frequency offset estimation parameters and phase offset estimation parameters required for each hop signal according to the genetic algorithm encoding rules, and to construct an adaptive multidimensional parameter space for the genetic algorithm search.

[0036] The genetic population initialization and update unit is used to construct the initial population of time-skipping pattern-frequency offset-phase offset parameters. In each iteration, it performs genetic operations such as crossover and mutation based on the fitness results to generate a new generation of parameter combinations, expand the search range, and enhance the global optimization capability.

[0037] The frequency offset compensation and de-hopping unit uses the generated frequency offset parameters to perform frequency offset compensation on the signal. Then, using the generated hopping time pattern parameters, it uses the Doppler scaling formula to calculate the delay time of each hop signal, obtains the start position of each hop signal, and extracts each hop signal.

[0038] The phase offset compensation and merging unit performs phase alignment on the extracted signals for each hop based on the generated phase offset parameters, and simultaneously achieves coherent merging of the signals.

[0039] The fitness calculation and convergence determination unit is used to calculate the fitness of the merged signal corresponding to each group of time-hopping pattern-frequency offset-phase offset parameters, evaluate the quality of the solutions, select the best individuals that meet the preset fitness requirements, and determine whether the termination condition is met. If the upper limit of the number of iterations or the fitness convergence condition is reached, the optimal parameter combination is output; otherwise, the best group is used as the parent to continue the genetic operation and enter the next round of iterative search.

[0040] Beneficial effects:

[0041] 1. The present invention discloses a method and system for estimating time-hopping signal parameters and merging multiple hops based on genetic algorithms. It adopts a genetic algorithm to globally search the time-hopping parameter space, avoiding the high computational complexity of traditional exhaustive search methods and the tendency of gradient search methods to get trapped in local optima. It can achieve rapid joint estimation of multi-dimensional parameters under low signal-to-noise ratio conditions, and significantly improve the parameter acquisition capability of time-hopping communication systems in long-distance and complex environments.

[0042] 2. The genetic algorithm-based method and system for estimating time-hopping signal parameters and merging multiple hops disclosed in this invention, by designing time-hopping pattern-frequency offset-phase offset parameter encoding and a multi-generation breeding mechanism, enables the genetic algorithm to directly perform optimization search based on the fitness function without requiring an analytical expression of the objective function. This avoids the process of differentiating the objective function, reduces the complexity of genetic algorithm design and implementation, and significantly improves the search convergence speed and the ability to obtain the global optimal solution.

[0043] 3. The method and system for estimating time-hopping signal parameters and merging multiple hops based on genetic algorithms disclosed in this invention realizes blind estimation of the time-hopping code under unknown conditions at the receiver by recovering the time-hopping pattern parameters based on genetic algorithms during the time-hopping signal decoding process. This solves the problem of the receiver relying on known time-hopping codes in the prior art, enabling the time-hopping communication system to operate normally in scenarios lacking prior information, and improving the applicability and anti-interference capability of the system.

[0044] 4. The method and system for time-hopping signal parameter estimation and multi-hop merging based on genetic algorithm disclosed in this invention, by performing joint compensation for frequency offset and phase offset of each hop signal and realizing coherent merging of multi-hop signals, the signal-to-noise ratio of the merged signal is higher than that of the single-hop signal. It can significantly reduce the bit error rate under low signal-to-noise ratio conditions, improve the overall performance of the communication system, and enhance the robustness and effectiveness of this invention in real-world scenarios.

[0045] 5. The method and system for hopping signal parameter estimation and multi-hop merging based on genetic algorithm disclosed in this invention balances search time and search quality by setting the maximum number of iterations and fitness stagnation conditions during the multi-hop signal merging process, avoids waste of computing resources, and ensures that the genetic algorithm converges to a satisfactory solution within a finite time. It is suitable for real-time hopping communication scenarios in engineering applications. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a patent process flowchart.

[0048] Figure 2 This is a schematic diagram of a time-hopping communication system transmitting signals.

[0049] Figure 3 Block diagram of a receiver for a time-hopping communication system that assists in the iteration of a genetic algorithm.

[0050] Figure 4 The graph shows the number of time-hopping signals versus the combined signal-to-noise ratio gain.

[0051] Figure 5 This refers to the deviation between the phase prediction value and the true value of the traditional cost function and the fitness function proposed in this embodiment of the invention.

[0052] Figure 6 The graph shows the fitness function as a function of iteration number.

[0053] Figure 7 This is a phase RMSE curve.

[0054] Figure 8 This is a frequency RMSE curve.

[0055] Figure 9 This is a graph showing the bit error rate. Detailed Implementation

[0056] The present invention will be further described and illustrated below with reference to the accompanying drawings and embodiments.

[0057] The flowchart of the time-hopping signal parameter estimation and multi-hop coherent merging method based on genetic algorithm disclosed in this embodiment is as follows: Figure 1 As shown, the specific implementation steps are as follows:

[0058] Step 1: The transmitting end divides the communication time into time frames and time slots, and controls the transmission position of the BPSK modulation signal according to the random time-hopping code, generating a time-hopping BPSK modulation signal with random characteristics in the time domain to enhance anti-interception capability.

[0059] The communication time window is divided into multiple time frames, and each time frame is further divided into multiple time slots, as shown in the diagram below. Figure 2 As shown. For each information symbol to be transmitted, BPSK modulation processing is performed first. Then, the modulation signal is controlled by the time-hopping gating module to ensure that each symbol is transmitted only within the time slot position specified by the time-hopping code. The time-hopping code determines the time slot position of the signal in each time frame and is randomly generated within a preset range, so that the time-hopping code presents randomness in the time domain, thereby generating a time-hopping random BPSK modulation signal. The BPSK modulation signal is expressed as Equation (1):

[0060]

[0061] Where b(m) represents the polarized data, h s (·) represents the impulse response of the shaping filter, f c Where is the carrier frequency, and g(k) is the time slot position corresponding to the time hopping code.

[0062] Step 2: When the time-hopping signal is transmitted in the satellite-to-ground channel, it will be superimposed with interference such as Doppler frequency offset, random time delay, phase shift, and Gaussian white noise. The receiver obtains the disturbed time-hopping signal through ADC sampling, down-conversion, and matched filtering to prepare for subsequent compensation.

[0063] The time-hop BPSK modulated signal generated in step one will be affected by various interferences during its propagation through the channel after being transmitted by the antenna. These interferences mainly include random time delay, Doppler frequency offset, phase shift, and energy attenuation, while being superimposed with Gaussian white noise. The receiver signal model can then be expressed as equation (2):

[0064]

[0065] Where τ represents random delay, f d Doppler frequency shift, Let A represent random phase bias, A represent the energy attenuation coefficient, and w(t) represent noise. Only additive white Gaussian noise (AWGN) is considered.

[0066] Receiver diagram as shown Figure 3 As shown, the signal at the receiving end is sampled by ADC, down-converted, and matched filtered to obtain the signal as shown in equation (3):

[0067]

[0068] Where Δf represents the frequency offset of the received signal. This indicates the phase offset of the processed signal.

[0069] Step 3: Quantize and encode the key time-hopping parameters—time-hopping pattern, frequency offset, and phase offset—to form a genotype structure for genetic algorithm optimization. Each parameter combination forms a searchable individual, covering the multidimensional joint estimation space in time-hopping communication.

[0070] In the signal processing at the receiver, in order to optimize the time hopping pattern, frequency offset, and phase offset parameters using a genetic algorithm, these parameters need to be encoded first. The number of quantization bits for the time hopping pattern is... The time-jump pattern encoding parameters are The corresponding time jump pattern number N tp =x tp The number of phase-biased quantization bits is The corresponding phase bias coding parameters are The corresponding phase bias is For N f The data set contains a frequency offset parameter, with a frequency offset quantization bit count of D. f The corresponding frequency offset coding parameters are The corresponding frequency offset is Therefore, for each candidate parameter group, the number of bits for its genotype is: Each parameter combination forms a genotype, represented by a joint encoding of time-skip pattern, frequency offset, and phase offset. The encoded genotype signal serves as the input for subsequent genetic algorithms to initialize the population.

[0071] Step 4: Based on the encoding format, initialize a genetic algorithm population containing multiple parameter combinations and set the iteration-related control parameters. This initial population serves as the starting point for subsequent crossover, mutation, and selection processes, achieving global optimization of the jump-time parameters.

[0072] Based on the encoding format in step three, the iteration parameters of the genetic algorithm are set, including the generation counter t=0, the maximum number of iterations T, and the evolutionary stagnation counter t. same =0, the individual fitness convergence threshold H, and the maximum stagnation limit T same Randomly generate N e The combination of group jump time pattern, frequency offset, and phase offset forms the initial population, which serves as the input population for the genetic algorithm and is used for subsequent crossover and mutation operations.

[0073] Step 5: Generate a new generation of time-skipping patterns, frequency offsets, and phase offset parameter combinations based on the current population. Use gene crossover and mutation operations to improve the search breadth. Crossover exchanges partial information in different parameter fragments, while mutation introduces population diversity by randomly flipping individual gene loci.

[0074] Using the population P(t) output in step four, crossover and mutation operations are performed to generate a new generation population P(t+1). The crossover operation selects three loci from each of the three gene segments representing the time-hopping pattern, frequency offset, and phase offset, dividing each segment into six segments. Within each segment, some gene loci are exchanged to generate a new time-hopping pattern-frequency offset-phase offset combination. The mutation operation increases search diversity by randomly flipping a locus within the selected individuals. The new generation population P(t+1) represents the updated set of time-hopping pattern-frequency offset-phase offset parameters, preparing for subsequent decoding compensation and merging.

[0075] Step Six: After decoding the parameter combination of each individual signal, frequency offset correction, time hopping extraction, and phase compensation are performed on the received signal in sequence. The compensated multi-hop signals are then coherently combined to form the aggregated signal result used for evaluation.

[0076] Using the genotype of an individual in the population obtained in step five, a set of time-hopping pattern-frequency offset-phase offset parameters are solved. The frequency offset parameters are then used to compensate the received signal obtained in step two: r(n) = r(n)·e -j·2πfn Then, N is extracted from the received signal using the time-hopping pattern sequence number. f Group data Then, the phase offset parameter is used to compensate for each time jump signal. For N that have completed compensation f The signals are coherently combined, and the combined signal is: The merged signal serves as input for fitness evaluation and iterative updates.

[0077] Step 7: Evaluate the quality of the merged signal results for each individual, assessing the effectiveness of the compensation and merging. Select the best-performing individuals for the next iteration. If the best result does not show significant improvement in multiple iterations, mark it as entering a stagnant state and update the iteration record.

[0078] Based on the merge signal and fitness J obtained in step six, calculate the fitness result corresponding to each individual's genotype. J=|∑sign(r all )·r all | is the fitness function used in the joint estimation algorithm of time-hopping pattern-frequency offset-phase offset in the time-hopping communication system of this invention. It is obtained by evaluating the multi-hop signal r after frequency offset compensation and phase alignment. allThe sign function and amplitude are weighted and summed, and the absolute value is taken to measure the phase consistency and energy aggregation effect of each hop signal. This cost function can reflect whether multi-hop signals achieve coherent merging and superposition enhancement after compensation. When the signal phase compensation and frequency offset compensation are accurate, the phase direction of the merged signal is consistent, the superposition energy is maximized, and increasing the value of J indicates that the compensation effect of the combination of hop pattern-frequency offset-phase offset parameters is good and the adaptability is high.

[0079] Sort the fitness results and select the top N with the best performance. e Group of data, the N e The set of data is called the optimal set. The optimal set serves as the parent set for the next iteration. The optimal individual P is recorded. best Fitness results J best (t+1), calculate the difference E = |J| between the current optimal individual fitness result and the previous optimal individual fitness result. best (t+1)-J best (t)|. When the fitness result difference E is less than the threshold value H, the iteration is considered to have stalled, and the stall counter t is set. same Increment by one. Update the iteration counter t so that t = t + 1.

[0080] Step 8: Determine if the current iteration count has reached the maximum limit, or if the fitness result has stagnated for more than the preset number of consecutive iterations. If the termination condition has not been met, continue iterating; if the termination condition has been met, end the search process and output the current optimal parameter combination.

[0081] The decision cycle has reached the maximum number of rounds T or the fitness result has stagnated for more than T rounds. same If the cycle does not reach the maximum number of rounds T and the fitness result stagnates and does not exceed T, then... same If the iteration reaches the maximum number of iterations T or the fitness result stagnates for more than T, then return to step three and repeat the iteration. same Then the loop terminates and the optimal combination is output.

[0082] Step Nine: Restore the optimal parameter combination output from Step Eight to the specific time-hopping pattern, frequency offset, and phase offset values, and recompensate the received signal. The compensated multi-hop signal undergoes coherent combining, significantly improving the overall signal-to-noise ratio and enhancing the system's communication reliability in low signal-to-noise ratio environments.

[0083] The optimal combination obtained in step eight is analyzed into time-hopping patterns, frequency offset, and phase offset parameters. These parameters are used to recover the actual time-hopping pattern information of the received signal. Based on the analyzed frequency offset and phase offset parameters, frequency and phase compensation are performed on each time-hopping signal. The compensated multi-hop signals are then coherently combined to obtain an enhanced combined signal. Through compensation and combining processes, the signal-to-noise ratio (SNR) of the combined signal is significantly improved, thereby enhancing the reception performance and reliability of the time-hopping communication system under low SNR conditions.

[0084] In this embodiment, the signal-to-noise ratio gain is obtained through coherent combining. Figure 4 This demonstrates the gain in signal-to-noise ratio (SNR) after time-hop signal combining compared to the SNR of the single-hop signal before combining. At this time, the combining gain increases with the number of time-hopping signals, satisfying G = 10·lg(M)dB, where M represents the number of time-hopping signals and G represents the signal combining gain.

[0085] Figure 5 The deviations between the phase predictions and actual values ​​of the traditional cost function and the fitness function proposed in this embodiment are shown, wherein the traditional cost function is expressed as follows: The fitness function proposed in the embodiments of the present invention In signal-to-noise ratio Under the given conditions, the proposed fitness function performs significantly better than the traditional cost function based on "absolute value signal-to-noise ratio".

[0086] Figure 6 The study demonstrates how the fitness function changes with the number of iterations. The number of time-hopping signals is 8, and the single-hop signal-to-noise ratio ranges from -6dB to 1dB. The results show that the higher the signal-to-noise ratio, the greater the convergence value of the fitness function. Under different signal-to-noise ratios, convergence occurs within 25 iterations.

[0087] Figure 7 The phase RMSE curve of the signal is shown. Eight time-hopping signals were set, operating within a single-hop signal-to-noise ratio (SNR) of -6dB to 1dB. Each SNR was repeated 100 times. The RMSE index tends to the theoretical limit as the SNR increases under the condition of 5-bit quantization of the phase parameter.

[0088] Figure 8 The frequency RMSE curve of the signal is shown. Eight time-hopping signals are set and run between -6dB and 1dB in single-hop signal-to-noise ratio. Each signal-to-noise ratio is repeated 100 times. The RMSE index decreases as the signal-to-noise ratio increases.

[0089] Figure 9The bit error rate curve of the signal is shown. With 8 time-hop signals, and operating between -6dB and 1dB in single-hop signal-to-noise ratio, the combined signal-to-noise ratio ranges from 3dB to 10dB. The bit error rate of the combined signal is close to the theoretical bit error rate of the BPSK signal.

[0090] The above detailed description further illustrates the purpose, technical solution, and beneficial effects of the invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm, characterized in that: The transmitting end divides the communication time into time frames and time slots, generating a time-hopping BPSK modulated signal with random characteristics in the time domain; the receiving end obtains the disturbed time-hopping signal through ADC sampling, down-conversion, and matched filtering; the key time-hopping parameters are quantized and encoded to form a genotype structure for genetic algorithm optimization; each parameter combination forms a searchable individual, covering the multi-dimensional joint estimation space in time-hopping communication; according to the encoding format, a genetic algorithm population containing multiple parameter combinations is initialized, and iterative control parameters are set; through crossover, mutation, and selection, global optimization of the time-hopping parameters is achieved; based on the current... The previous population generates a new generation of time-hopping patterns, frequency offset, and phase offset parameter combinations. After decoding the parameter combination of each individual, the received signal is sequentially subjected to frequency offset correction, time-hopping extraction, and phase compensation. The compensated multi-hop signals are coherently combined to form an aggregated signal result for evaluation. Several individuals with the best performance are selected to enter the next iteration. If the termination condition is not met, the iteration continues. If the termination condition is met, the search ends and the current optimal parameter combination is output. The optimal parameter combination is restored to the specific time-hopping pattern, frequency offset, and phase offset values, and the received signal is recompensated. The compensated multi-hop signals are then coherently combined.

2. The method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm as described in claim 1, characterized in that: Includes the following steps, Step 1: The transmitting end divides the communication time into time frames and time slots, and controls the transmission position of the BPSK modulated signal according to the random time-hopping code, generating a time-hopping BPSK modulated signal with random characteristics in the time domain; Step 2: When the time-hopping signal is transmitted in the satellite-to-ground channel, it will be superimposed with Doppler frequency offset, random time delay, phase offset and Gaussian white noise interference; the receiver obtains the disturbed time-hopping signal through ADC sampling, down-conversion and matched filtering; Step 3: Quantize and encode the key parameters, time-hopping pattern, frequency offset, and phase offset, to form a genotype structure for genetic algorithm optimization; each set of parameters forms a searchable individual, covering the multidimensional joint estimation space in time-hopping communication; Step 4: According to the encoding format, initialize a genetic algorithm population containing multiple parameter combinations and set the iteration-related control parameters; this initial population serves as the starting point for evolution and is used in subsequent crossover, mutation, and selection processes to achieve global optimization of jump time parameters; Step 5: Generate a new generation of time-skipping patterns, frequency offsets, and phase offset parameter combinations based on the current population, and use gene crossover and mutation operations to improve the search breadth; crossover exchanges some information in different parameter fragments, and mutation introduces population diversity by randomly flipping individual gene loci; Step 6: After decoding the parameter combination of each individual, frequency offset correction, time hopping extraction and phase compensation are performed on the received signal in sequence; the compensated multi-hop signals are coherently combined to form the aggregated signal result for evaluation; Step 7: Evaluate the quality of the merged signal results for each individual to assess the effectiveness of the compensation and merging. Select the best-performing individuals to proceed to the next iteration. If the best result does not show significant improvement in multiple iterations, mark it as entering a stagnant state and update the iteration record. Step 8: Determine whether the current iteration count has reached the maximum limit, or whether the fitness results have stagnated for more than the preset number of consecutive iterations; If the termination condition is not met, continue iterating; if the termination condition is met, end the search process and output the current optimal parameter combination. Step 9: Restore the optimal parameter combination output from Step 8 to the specific time hopping pattern, frequency offset, and phase offset values, and recompensate the received signal; The compensated multi-hop signals are then coherently combined.

3. The method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm as described in claim 2, characterized in that: The implementation method for step one is as follows: The communication time window is divided into multiple time frames, and each time frame is further divided into multiple time slots; for each information symbol to be transmitted, BPSK modulation processing is performed first; Then, the modulation signal is controlled by the time-hopping gating module to ensure that each symbol is transmitted only within the time slot position specified by the time-hopping code; the time-hopping code determines the time slot position of the signal in each time frame and is randomly generated within a preset range, so that the time-hopping code presents randomness in the time domain, thereby generating a time-hopping random BPSK modulation signal; the BPSK modulation signal is expressed as Equation (1): Where b(m) represents the polarized data, h s (·) represents the impulse response of the shaping filter, f c Where is the carrier frequency, and g(k) is the time slot position corresponding to the time hopping code.

4. The method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm as described in claim 3, characterized in that: The second step is implemented as follows: The time-hop BPSK modulated signal generated in step one will be affected by various interferences when it propagates through the channel after being transmitted by the antenna. These interferences mainly include random time delay, Doppler frequency offset, phase shift and energy attenuation, and are also superimposed with Gaussian white noise. At this time, the model of the receiver signal is expressed as Equation (2): Where τ represents random delay, f d Doppler frequency shift, Let A represent random phase bias, A represent the energy attenuation coefficient, and w(t) represent noise; only additive white Gaussian noise (AWGN) is considered. The receiver performs ADC sampling, down-conversion, and matched filtering on the signal, resulting in the signal shown in equation (3): Where Δf represents the frequency offset of the received signal. This indicates the phase offset of the processed signal.

5. The method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm as described in claim 4, characterized in that: The method for implementing step three is as follows: In the signal processing at the receiving end, in order to optimize the time hopping pattern, frequency offset, and phase offset parameters using a genetic algorithm, these parameters need to be encoded first; the number of quantization bits for the time hopping pattern is... The jump pattern encoding parameter is x tp , The corresponding time jump pattern number N tp =x tp The number of phase-biased quantization bits is The corresponding phase bias coding parameters are The corresponding phase bias is For N f The data set contains a frequency offset parameter, with a frequency offset quantization bit count of D. f The corresponding frequency offset coding parameter is x. f , The corresponding frequency offset is Therefore, for each candidate parameter group, the number of bits for its genotype is: Each parameter combination forms a genotype, represented by a time-skip pattern-frequency offset-phase offset joint encoding; the encoded genotype signal serves as the input for subsequent genetic algorithms to initialize the population.

6. The method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm as described in claim 5, characterized in that: Step four is implemented as follows: Based on the encoding format in step three, the iteration parameters of the genetic algorithm are set, including the generation counter t=0, the maximum number of iterations T, and the evolutionary stagnation counter t. same =0, the individual fitness convergence threshold H, and the maximum stagnation limit T same Randomly generate N e The combination of group jump time pattern, frequency offset, and phase offset forms the initial population, which serves as the input population for the genetic algorithm and is used for subsequent crossover and mutation operations.

7. The method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm as described in claim 6, characterized in that: Step five is implemented as follows: Using the population P(t) output in step four, crossover and mutation operations are performed to generate a new generation population P(t+1). The crossover operation selects three loci in each of the three gene segments of time-hopping pattern, frequency offset, and phase offset, divides the three gene segments into six segments, randomly selects each segment, and exchanges some gene loci to generate a new time-hopping pattern-frequency offset-phase offset combination. The mutation operation increases the search diversity by randomly flipping a locus in the selected individuals. The new generation population P(t+1) represents the updated time-hopping pattern-frequency offset-phase offset parameter set, which is prepared for subsequent decoding compensation and merging.

8. The method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm as described in claim 7, characterized in that: Step six is ​​implemented as follows: Using the genotype of an individual in the population obtained in step five, a set of time-hopping pattern-frequency offset-phase offset parameters are solved. The frequency offset parameters are then used to compensate the received signal obtained in step two: r(n) = r(n)·e -j·2πfn Subsequently, N is extracted from the received signal using the time-hopping pattern sequence number. f Group data Then, the phase offset parameter is used to compensate for each time jump signal. For N that have completed compensation f The signals are coherently combined, and the combined signal is: The merged signal serves as input for fitness evaluation and iterative updates.

9. The method for time-hopping signal parameter estimation and multi-hop coherent merging based on genetic algorithm as described in claim 8, characterized in that: Step seven is implemented as follows: Based on the merge signal and fitness J obtained in step six, calculate the fitness result corresponding to the genotype of each individual; J=|∑sign(r all )·r all | is the fitness function used in the joint estimation algorithm of time-hopping pattern-frequency offset-phase offset in the time-hopping communication system of this invention. It is obtained by evaluating the multi-hop signal r after frequency offset compensation and phase alignment. all The sign function and amplitude are weighted and summed, and the absolute value is taken to measure the phase consistency and energy aggregation effect of each jump signal; This cost function can reflect whether multi-hop signals achieve coherent combining and superposition enhancement after compensation. When the signal phase compensation and frequency offset compensation are accurate, the phase direction of the combined signal is consistent, and the superposition energy is maximized. Increasing the value of J indicates that the compensation effect of this combination of time-hopping pattern-frequency offset-phase offset parameters is good and the adaptability is high. Sort the fitness results and select the top N with the best performance. e Group of data, the N e The set of data is called the optimal set; the optimal set is used as the parent set for the next iteration; the best individual P is recorded. best Fitness results J best (t+1), calculate the difference E = |J| between the current optimal individual fitness result and the previous optimal individual fitness result. best (t+1)-J best (t)|;When the fitness result difference E is less than the threshold value H, the iteration is considered to have stalled, and the stall counter t is set. same Increment by one; update the iteration counter t so that t = t + 1; Step eight is implemented as follows: The decision cycle has reached the maximum number of rounds T or the fitness result has stagnated for more than T rounds. same If the cycle does not reach the maximum number of rounds T and the fitness result stagnates and does not exceed T, then... same If the iteration reaches the maximum number of iterations T or the fitness result stagnates for more than T, then return to step three and repeat the iteration. same The loop terminates and the optimal combination is output. Step nine is implemented as follows: The optimal combination obtained in step eight is analyzed into time hopping pattern, frequency offset, and phase offset parameters. The actual time hopping pattern information of the received signal is recovered using these parameters. Based on the analyzed frequency offset and phase offset parameters, frequency compensation and phase compensation are performed on each time hopping signal. The compensated multi-hop signals are coherently combined to obtain the enhanced combined signal. Through compensation and merging processing, the signal-to-noise ratio of the merged signal is significantly improved, thereby improving the receiving performance and reliability of the time-hopping communication system under low signal-to-noise ratio conditions.

10. A genetic algorithm-based system for estimating time-hopping signal parameters and coherently combining multiple hops, used to implement the genetic algorithm-based method for estimating time-hopping signal parameters and coherently combining multiple hops as described in claims 2, 3, 4, 5, 6, 7, 8, or 9, characterized in that: It includes a time-hopping pattern-frequency offset-phase offset parameter encoding and generation unit, a genetic population initialization and update unit, a frequency offset compensation and time-hopping resolution unit, a phase offset compensation and merging unit, and a fitness calculation and convergence determination unit; The hop pattern-frequency offset-phase offset parameter encoding and generation unit is used to generate the hop pattern parameters, frequency offset estimation parameters and phase offset estimation parameters required for each hop signal according to the genetic algorithm encoding rules, and to construct an adaptive multidimensional parameter space for the genetic algorithm search; The genetic population initialization and update unit is used to construct the initial population of time-skipping pattern-frequency offset-phase offset parameters, and to perform genetic operations based on the fitness results in each iteration to generate a new generation of parameter combinations, expand the search range, and enhance the global optimization capability; genetic operations include crossover and mutation; The frequency offset compensation and de-hopping unit uses the generated frequency offset parameters to perform frequency offset compensation on the signal, and then uses the generated hopping time pattern parameters to calculate the delay time of each hop signal using the Doppler expansion formula to obtain the start position of each hop signal and extract each hop signal. The phase offset compensation and merging unit performs phase alignment on the extracted each-hop signal according to the generated phase offset parameters, and simultaneously achieves coherent merging of the signals. The fitness calculation and convergence determination unit is used to calculate the fitness of the combined signal corresponding to each group of time-hopping pattern-frequency offset-phase offset parameters, evaluate the quality of the solution, select the preferred individuals that meet the preset fitness requirements, and determine whether the termination condition is met. If the upper limit of the number of iterations or the fitness convergence condition is reached, the optimal parameter combination is output; otherwise, the preferred group is used as the parent to continue the genetic operation and enter the next round of iterative search.

Images