A jungle channel-oriented GFDM effective capacity optimization method and system

Abstract

The application relates to the technical field of wireless communication, and particularly discloses a GFDM effective capacity optimization method and system for a jungle channel, the method comprising the following steps: obtaining a raw binary data stream at a transmitting end, pre-processing and demodulating a signal received at a receiving end to obtain a demodulated binary data stream, and calculating a bit error rate based on the raw binary data stream and the demodulated binary data stream; calculating effective capacity based on the bit error rate, taking the effective capacity as a utility function, and step-by-step optimizing GFDM waveform design parameters by applying a particle swarm algorithm, wherein the GFDM waveform design parameters comprise a roll-off coefficient and a subcarrier number; and the method can optimize the GFDM effective capacity under the jungle channel, and obtain the best effective capacity and corresponding subcarrier number and pulse shaping filter coefficient.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, and particularly relates to a GFDM effective capacity optimization method and system for a jungle channel. BACKGROUND

[0002] Forest fires can bring serious influence to the natural environment and economic construction, so it is necessary to prevent them by all means and minimize the adverse effects of forest fires. At the same time, in recent years, there have been some cases of missing persons who need rescue while exploring in forest mountain areas. The occurrence of these situations and disasters needs to use communication equipment to ensure that it can better provide information services and location protection for the fire brigade and search and rescue personnel when carrying out forest mountain rescue. In addition, considering the construction cost and other problems, the communication mode in the forest mainly adopts wireless communication. However, due to the problem of tree leaf shielding, the propagation loss of radio waves in the forest is very large, and the reliable and real-time transmission of communication cannot be guaranteed. Therefore, it is necessary to study the reliable and real-time transmission of forest or jungle wireless communication.

[0003] How to overcome the negative effects of jungle propagation environment and ensure real-time and reliable transmission is the main problem faced by jungle communication. Effective capacity is the maximum data transmission rate that the channel can stably support under the specified quality of service constraint, and is the core index for measuring reliable and real-time transmission. There are few studies on the effective capacity of new multi-carrier GFDM, and there is no report on the effective capacity of GFDM for jungle communication. How to optimize the design of GFDM parameters to reduce the bit error rate and increase the effective capacity according to the channel characteristics of jungle communication is a problem worthy of further study. SUMMARY

[0004] The present application aims to provide a GFDM effective capacity optimization method and system for a jungle channel to solve the problems in the background.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0006] A GFDM effective capacity optimization method for a jungle channel, the method comprising:

[0007] Obtaining a raw binary data stream at a transmitting end, pre-processing and demodulating a signal received at a receiving end to obtain a demodulated binary data stream, and calculating a bit error rate based on the raw binary data stream and the demodulated binary data stream;

[0008] Calculating an effective capacity based on the bit error rate, and taking the effective capacity as a utility function, and step-by-step optimizing GFDM waveform design parameters by applying a particle swarm algorithm, wherein the GFDM waveform design parameters include a roll-off coefficient and a subcarrier number;

[0009] The parameter optimization step specifically comprises:

[0010] Taking the effective capacity as the fitness function of the particle swarm algorithm, setting the roll-off coefficient as the decision variable, and through the particle swarm algorithm for iterative optimization in the preset roll-off coefficient value interval, the optimal roll-off coefficient is obtained;

[0011] Taking the obtained optimal roll-off coefficient as a fixed parameter, taking the effective capacity as the fitness function, setting the subcarrier number as the decision variable, and through the particle swarm algorithm for iterative optimization in the preset subcarrier number value interval, the optimal subcarrier number is obtained.

[0012] As a further scheme of the present application, the step of obtaining the original binary data stream of the transmitting end, and pre-processing and demodulating the signal received by the receiving end to obtain the demodulated binary data stream specifically comprises:

[0013] The transmitting end modulates the input binary data stream by M-QAM to obtain a complex symbol matrix; and performs pulse shaping on the complex symbol matrix by a raised cosine roll-off filter to generate a time domain signal;

[0014] The time domain signal is added with a cyclic prefix to obtain a sending signal;

[0015] The sending signal is transmitted through a jungle channel, and a receiving signal is obtained at the receiving end;

[0016] The receiving end performs matched filtering and equalization processing on the receiving signal to recover a complex symbol estimation matrix, and performs demodulation to obtain the demodulated binary data stream.

[0017] As a further scheme of the present application, the bit error rate calculation formula is:

[0018] ;

[0019] Wherein, BER is the bit error rate, b tx is the number of sending bits, b rx is the number of receiving bits, N is the total number of bits, µ is the number of bits per symbol.

[0020] As a further scheme of the present application, the effective capacity C(θ) The calculation formula is:

[0021] ;

[0022] ;

[0023] In the formula, θ is a QoS index, BER is a system error rate, M is a QAM modulation order, fs is a sampling rate, a is a roll-off factor.

[0024] As a further scheme of the present application, the step of iteratively optimizing by the particle swarm algorithm comprises:

[0025] initializing the particle swarm, setting the particle size, the maximum number of iterations, the inertia weight, the learning factor, and randomly generating the initial position and speed of each particle in the solution space corresponding to the decision variable;

[0026] calculating the fitness value corresponding to the current position of each particle;

[0027] updating the individual historical optimal position of each particle and the global historical optimal position of the entire particle swarm according to the fitness value;

[0028] updating the speed and position of each particle according to the speed update formula and the position update formula;

[0029] judging whether the iteration termination condition is met, and if so, terminating the iteration and outputting the global historical optimal position as the optimal solution, and if not, returning to the fitness calculation and entering the next iteration.

[0030] As a further scheme of the present application, before applying the particle swarm algorithm for optimization, a jungle channel model is constructed; the jungle channel model simulates multipath fading by Rayleigh distribution and calculates path loss by a jungle path loss empirical model PL , the calculation formula of the path loss PL includes:

[0031] ;

[0032] wherein, PL_dθ is a basic path loss measured at a reference distance dθ , is an actual transmission distance, d is a path loss index related to signal frequency n , tree trunk diameter f , and vegetation density TD . Dc

[0033] As a further scheme of the present application, the calculation formula of the path loss index n is:

[0034] ;

[0035] wherein, is a tree trunk diameter,​Dc is a vegetation density, k 1 is a basic correction coefficient determined by frequency, k 2 is a basic path loss index.

[0036] The application also provides a GFDM effective capacity optimization system for a jungle channel, which comprises:

[0037] a performance test module, configured to obtain original binary data stream of a transmitting end, pre-process and demodulate a signal received by a receiving end to obtain demodulated binary data stream, and calculate a bit error rate based on the original binary data stream and the demodulated binary data stream;

[0038] a parameter optimization module, configured to calculate effective capacity based on the bit error rate, take the effective capacity as an utility function, and stepwise optimize GFDM waveform design parameters by using a particle swarm algorithm, wherein the GFDM waveform design parameters comprise a roll-off coefficient and a subcarrier number;

[0039] The parameter optimization step specifically comprises:

[0040] taking the effective capacity as a fitness function of the particle swarm algorithm, taking the roll-off coefficient as a decision variable, and iteratively optimizing the roll-off coefficient in a preset roll-off coefficient value interval by using the particle swarm algorithm to obtain an optimal roll-off coefficient;

[0041] taking the obtained optimal roll-off coefficient as a fixed parameter, taking the effective capacity as a fitness function, taking the subcarrier number as a decision variable, and iteratively optimizing the subcarrier number in a preset subcarrier number value interval by using the particle swarm algorithm to obtain an optimal subcarrier number.

[0042] Compared with the prior art, the application has the beneficial effects that: based on various influences on signal transmission in a jungle environment, the application uses GFDM technology to construct an integrated communication solution that combines an adaptive utility function and dynamic parameter optimization, the solution uses a particle swarm algorithm to optimize and solve key parameters of a GFDM system, and finally obtains optimal parameter configuration of the system, thereby effectively improving overall communication quality of radio waves in a jungle scenario.

[0043] The application can carry out targeted parameter design according to different performance requirements, can provide accurate position information for personnel and equipment by relying on high-precision ranging and positioning technology of GFDM signals, and can simultaneously realize efficient data transmission and stable environment sensing, thereby comprehensively improving communication and sensing comprehensive performance of the GFDM system in a complex jungle environment and achieving the goal of optimal parameter configuration of the GFDM communication system in multiple scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application.

[0045] Figure 1 Different values of the utility function provided by the embodiments of the present application a The numerical trend chart of the utility function under different values.

[0046] Figure 2 The numerical trend chart of the utility function under different values of K provided by the embodiments of the present application. DETAILED DESCRIPTION

[0047] In order to make the technical problems to be solved by the present application, the technical solutions and the beneficial effects more clearly, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not used to limit the present application.

[0048] In the embodiments of the present application, a GFDM effective capacity optimization method for jungle channel, the method comprises:

[0049] Obtaining the original binary data stream of the transmitting end, pre-processing and demodulating the signal received by the receiving end to obtain the demodulated binary data stream, and calculating the bit error rate based on the original binary data stream and the demodulated binary data stream;

[0050] Calculating the effective capacity based on the bit error rate, and taking the effective capacity as the utility function, and applying the particle swarm algorithm to step-by-step optimize the GFDM waveform design parameters, the GFDM waveform design parameters including the roll-off coefficient and the subcarrier number;

[0051] The parameter optimization step specifically includes:

[0052] Taking the effective capacity as the fitness function of the particle swarm algorithm, taking the roll-off coefficient as the decision variable, and performing iterative optimization in the preset roll-off coefficient value interval by the particle swarm algorithm to obtain the optimal roll-off coefficient;

[0053] Taking the obtained optimal roll-off coefficient as a fixed parameter, taking the effective capacity as the fitness function, taking the subcarrier number as the decision variable, and performing iterative optimization in the preset subcarrier number value interval by the particle swarm algorithm to obtain the optimal subcarrier number.

[0054] In the present embodiment, a performance test module is built in the GFDM jungle communication system, and the receiving end will test the bit error rate of the GFDM jungle communication system according to the known pilot signal sent by the transmitting end, which is used for subsequent construction of other evaluation indexes.

[0055] The number of subcarriers of a GFDM system is a core parameter for balancing transmission performance (throughput, anti-fading, positioning accuracy), system latency, implementation complexity and adaptability to specific scenarios such as jungle, and directly determines the frequency domain resource division granularity and overall communication sensing efficiency.

[0056] In the GFDM system for jungle communication, the number of subcarriers is dynamically related to the bit error rate. In the jungle environment, there is serious multipath propagation and frequency selective fading. When the number of subcarriers is appropriately increased, the system can transmit signals on more subcarriers, reduce the probability of affecting a single subcarrier by fading through the effect of spectral diversity, and thus improve the bit error rate. However, the GFDM system itself gives up the strict orthogonality of subcarriers. If the number of subcarriers is too large, the frequency interval between subcarriers is reduced, the time-frequency characteristics of the jungle channel will further amplify the interference between subcarriers, and the sensitivity of the system to frequency offset and phase noise will also be improved, ultimately leading to a significant increase in the bit error rate. Conversely, if the number of subcarriers is too small, the bandwidth of a single subcarrier is increased, which is more easily affected by the frequency selective fading of the jungle, and the spectral resource utilization is insufficient, which also causes the bit error rate to be at a high level.

[0057] As a preferred embodiment of the present application, the step of obtaining the original binary data stream of the transmitting end, pre-processing and demodulating the signal received by the receiving end to obtain the demodulated binary data stream specifically comprises:

[0058] The transmitting end modulates the input binary data stream by M-QAM to obtain a complex symbol matrix, and performs pulse shaping on the complex symbol matrix by a root-raised cosine roll-off filter to generate a time domain signal.

[0059] A cyclic prefix is added to the time domain signal to obtain a transmission signal.

[0060] The transmission signal is transmitted through a jungle channel, and a receiving signal is obtained at the receiving end.

[0061] The receiving end performs matched filtering and equalization processing on the receiving signal to recover a complex symbol estimation matrix, and demodulates the complex symbol estimation matrix to obtain a demodulated binary data stream.

[0062] In this embodiment, at the transmitting side, the input binary data stream is first mapped by M-QAM modulation to obtain a complex symbol (C represents a complex number set, K represents the number of subcarriers, and M represents the number of subcarriers). x t .

[0063] The frequency response function of the root-raised cosine roll-off filter is:

[0064] ;​

[0065] wherein T s is the symbol period, R s is the symbol transmission rate, and , f is the frequency, and a is the roll-off factor, and ;

[0066] In the jungle communication environment, the received signal is: ;

[0067] wherein, represents the time-domain data of the signal received by the receiving end, x t represents the unit impulse response model of the jungle communication channel, represents the noise, and conforms to the Gaussian noise with a mean of 0 and a variance of ;

[0068] At the sending end, let the original symbol sequence to be transmitted be , wherein is the sequence length, and is used to represent the symbol length corresponding to the cyclic prefix, then the symbol after the completion of the cyclic prefix addition is:

[0069] ;

[0070] Assuming that the impulse response of the channel is , and the length is L h , then the signal received by the receiving end after the addition of the cyclic prefix can be represented as:

[0071] ;

[0072] In the above formula, is the additive noise existing in the channel; the cyclic prefix added in advance cp , the core role of which is to avoid the inter-symbol interference (ISI) caused by the inter-symbol overlap between x n - l and x n , so as to guarantee the integrity and demodulation of the received signal.

[0073] At the receiving end, the signal is first preliminarily processed through matching filtering and frequency domain equalization (if necessary) to restore the received signal D to a complex symbol ; then the complex symbol sequence is subjected to​​​​ The M-QAM demodulation operation is performed to restore the corresponding binary data stream.

[0074] As a preferred embodiment of the present application, the error rate calculation formula is:

[0075] ;

[0076] Wherein, BER is the error rate, b tx is the number of transmitted bits, b rx is the number of received bits, N is the total number of bits, µ is the number of bits per symbol (determined by the M-QAM modulation order), i denotes the number from 1 to N .

[0077] In this embodiment, the error rate of the system is BER Then, the bit difference between the original binary data stream b tx of the transmitting end and the binary data stream b rx obtained by demodulation of the receiving end is calculated, and the core calculation logic is the ratio of the number of error bits to the total number of bits.

[0078] For the error rate BER, the roll-off factor a plays a role by affecting the inter-symbol interference; a When the roll-off factor is small, the tail length of the raised cosine roll-off filter impulse response is long, the inter-symbol interference is large, the receiving end is easily disturbed by adjacent symbols, and the error rate is high; a When the roll-off factor is large, the tail is short, the inter-symbol interference is reduced, but at the same time the bandwidth increases, which may cause the signal-to-noise ratio to decrease, and thus the error rate to increase; therefore, according to the actual channel conditions, a suitable a value needs to be selected to balance between inter-symbol interference and signal-to-noise ratio to reduce the error rate;

[0079] The symbol transmission rate R s is determined by the number of subcarriers K , symbol period and the proportion of cyclic prefix β, and the calculation formula is as follows:

[0080] ;

[0081] In the formula, is the ratio of the cyclic prefix time T CP to the symbol period T S .

[0082] As a preferred embodiment of the present invention, the effective capacity C(θ) The calculation formula is:

[0083] ;

[0084] ;

[0085] In the formula, θ For QoS index, BER For the system bit error rate, M This is the QAM modulation order. fs Sampling rate, a This is the roll-off factor.

[0086] In this embodiment, effective capacity is the maximum constant arrival rate that the service rate can support under the constraint of a specified QoS index θ. Effective capacity quantifies the upper limit of the service load that the system can stably support, and is a precise description of the system's carrying capacity under QoS constraints. It serves as a bridge to balance service randomness and service reliability.

[0087] Effective capacity exhibits a monotonic relationship with bit error rate (BER). As BER decreases, effective capacity continuously increases. Effective capacity is an indicator that describes the maximum constant data arrival rate that a random service system can stably support under statistical QoS (Quality of Service) constraints. Essentially, effective capacity is an extension of traditional channel capacity: traditional channel capacity only describes the theoretical maximum rate without QoS constraints, while effective capacity combines the randomness of the system's service rate (such as time-varying fading of the wireless channel) with the QoS requirements in practical applications (such as latency limits and packet loss rate thresholds), reflecting the actual available capacity of the system while meeting specified QoS requirements.

[0088] Effective capacity decreases monotonically with increasing bit error rate. In communication systems, an increase in bit error rate directly reduces effective capacity. To improve effective capacity, it is necessary to reduce bit error rate (e.g., by optimizing modulation, coding, and channel conditions).

[0089] Effective capacity quantifies the trade-off between QoS constraints and system capacity: the service rate of real-world systems (such as wireless communication and network transmission) fluctuates randomly (for example, in jungle communication, the transmission rate of a GFDM system changes due to multipath fading), while application scenarios often have strict QoS requirements; effective capacity introduces a QoS index. ( The larger the value of θ, the stricter the QoS constraints. This quantifies the inverse relationship between the strength of QoS constraints and the maximum rate that the system can support: when QoS requirements are more stringent (θ increases), the effective capacity decreases; conversely, relaxing QoS constraints can increase the effective capacity.

[0090] In the field of communication, network, etc., effective capacity is the core basis for resource allocation and parameter design:

[0091] In the jungle GFDM system, the effective capacity formula can be used to calculate the effective capacity under different configurations by combining the bit error rate, subcarrier number, roll-off factor and other parameters, so as to select the optimal parameter combination that can meet the QoS requirement and transmission rate requirement at the same time.

[0092] In network scheduling, effective capacity can help determine the bandwidth allocation strategy of different service flows (such as real-time voice and non-real-time data), and ensure that the QoS constraint of high-priority services is met.

[0093] Effective capacity can evaluate the actual performance of a random service system: traditional capacity indicators (such as Shannon capacity) are theoretical upper limits under ideal unconstrained conditions, and cannot reflect the performance of the system in the actual random environment.

[0094] Effective capacity is based on the statistical characteristics (such as stationarity and correlation) of service rate, and is closer to the actual scenario: for example, the service rate of a wireless channel is time-varying, and effective capacity can accurately describe the maximum transmission rate of the channel when ensuring that the data packet delay is less than 10ms with 99.9% probability, providing a more practical standard for system performance evaluation.

[0095] As a preferred embodiment of the present application, before applying the particle swarm algorithm for optimization, a jungle channel model is constructed; the jungle channel model simulates multipath fading through Rayleigh distribution and calculates path loss PL through a jungle path loss empirical model, and the calculation formula of the path loss PL includes:

[0096] ;

[0097] wherein, PL_dθ is the basic path loss measured at a reference distance dθ , is the actual transmission distance, d is the path loss exponent related to the signal frequency n , tree trunk diameter f , and vegetation density TD . Dc The calculation formula of the path loss exponent

[0098] is: n

[0099] ;

[0100] wherein, is the tree trunk diameter, Dc is the vegetation density, k 1 is a basic correction coefficient determined by the frequency,​k 2 is the basic path loss index.

[0101] In this embodiment, the jungle channel is a typical complex variable-parameter wireless channel. Its characteristics are jointly determined by factors such as jungle vegetation obstruction, terrain undulation, and electromagnetic environment. Its core features include severe path loss influenced by multiple factors, frequency-band dependent propagation modes, significant multipath fading, and limited communication distance. Constructing a GFDM communication system in the jungle channel can be achieved by using a superposition of Rayleigh distribution and an empirical model of jungle path loss for simulation. The Rayleigh power spectral density function is:

[0102] ;

[0103] in, For channel gain, This represents the channel variance.

[0104] The jungle path loss empirical model is derived from empirical fitting of experimental data. The formula for calculating path loss PL is:

[0105] ;

[0106] ;

[0107] ;

[0108] ;

[0109] ;

[0110] In the formula, f The signal frequency (GHz) The transmission distance (m) The diameter of the tree trunk (cm) Dc For vegetation density, FD For vegetation cover, refer to distance. Basic path loss index k 2 = 2.27.

[0111] In a preferred embodiment of the present invention, the iterative optimization step using the particle swarm optimization algorithm includes:

[0112] Initialize the particle swarm, set the particle size, maximum number of iterations, inertia weight, and learning factor, and randomly generate the initial position and velocity of each particle in the solution space of the corresponding decision variables;

[0113] Calculate the fitness value corresponding to the current position of each particle;

[0114] Update the individual historical best position of each particle and the global historical best position of the entire particle swarm based on the fitness value.

[0115] Update the velocity and position of each particle according to the velocity update formula and the position update formula;

[0116] Determine whether the iteration termination condition is met. If it is met, terminate the iteration and output the global historical best position as the optimal solution. If it is not met, return to fitness calculation and enter the next round of iteration.

[0117] In this embodiment, for the GFDM jungle communication system, while keeping other waveform parameters constant, the particle swarm optimization algorithm is integrated into the communication system. The number of subcarriers and the roll-off coefficient are optimized using a utility function. The effective capacity is used as the fitness function of the particle swarm optimization algorithm, and the roll-off coefficient is set as the decision variable, with its value range limited to (…). a min , a max )( a min and a max (The minimum and maximum values ​​of the custom roll-off coefficients are defined). By continuously adjusting the values ​​of the decision variables, the fitness function converges to the optimal value.

[0118] The solution process for optimizing the roll-off factor of the pulse shaping filter is as follows:

[0119] First, determine the core parameters of the algorithm, including the particle swarm size (i.e., the total number of particles, taken as 50), the maximum number of iterations (taken as 100), the inertia weight (taken as [0.3 0.8]), and the learning factor. c 1 and c 2 (all set to 1.5), boundary range of velocity and position, parameter settings:

[0120] K is the number of subcarriers, and the simulation value is set to 256; cp is the cyclic prefix, and the simulation value is set to 0.2. a max The simulation value is set to 0.95 as the maximum roll-off factor; a min The minimum roll-off factor is set to 0.05 in the simulation; the signal-to-noise ratio (SNR) is set to 25 dB in the simulation; the number of modulation bits (mu) is set to 4 in the simulation; the number of sub-symbols (M) is set to 15 in the simulation; and the trunk diameter (TD) is set to 0.1 in the simulation. Dc The simulated value for vegetation density is set to 2.6; FD The simulated value for vegetation cover is 75; the simulated value for QoS index is 1. d For the transmission distance, the simulation value is set to 600 (m);f For signal frequency, the simulation value is set to 0.87GHz; fs For sampling rate, the simulation value is set to 5MHz.

[0121] In the solution space of the problem to be optimized, an initial position (corresponding to a candidate solution of the problem) and an initial velocity are randomly generated for each particle, and the initial position of each particle is set as its individual optimal position p b The optimal one is selected from the individual optimal positions of all particles as the global optimal position of the entire population g b .

[0122] For the current position of each particle, the fitness value is calculated by substituting it into the objective function of the problem.

[0123] For each particle, the current fitness value is compared with the historical p b fitness value, and if the current value is better, the p b of the particle is updated to the current position; then the p b of all particles are traversed p b , and if the fitness value of a certain g b is better than the fitness value of the current g b , the p b of the population is updated to the position of the

[0124] The velocity and position of each particle are updated as follows:

[0125] Velocity update: ;

[0126] Position update: ;

[0127] The particle swarm algorithm updates the velocity and position of each particle through iteration, guiding the particles to approach the individual optimal p b and the global optimal g b , and finally converges to the optimal solution of the problem. The velocity determines the direction and step size of the particle movement, and the position corresponds to the candidate solution of the problem. The update of the two is the core part of PSO, which needs to strictly follow the classical formula and combine with boundary constraints.

[0128] In the particle swarm optimization algorithm, the velocity update of each particle is a weighted combination of inertia, individual learning and social learning, and the position update is a direct superposition of the velocity, which must be constrained by the boundary to ensure that the particle is in the solution space. This update logic embodies the core idea of group cooperation and balances the exploratory and convergent nature of optimization through random numbers and weights, which is the core of PSO optimization.

[0129] Check if the termination condition is met: one is to reach the preset maximum number of iterations, and the other is g b The corresponding fitness value has met the accuracy requirements of the problem (e.g., the change in fitness of consecutive generations is less than a threshold value). If it is met, stop iteration, if it is not met, return to the "fitness calculation" step and enter the next iteration.

[0130] After the iteration is terminated, the final g b position is output as the optimal solution of the problem, and its corresponding fitness value can be output simultaneously to complete the entire optimization process and obtain the final value of the utility function F after optimization.

[0131] The roll-off factor and time-frequency locality of the filter directly determine the anti-interference ability of the system: a higher roll-off factor can provide steeper out-of-band attenuation, effectively suppressing the inter-carrier interference caused by dense multipath in the jungle environment, but will expand the time-domain response of the symbol, exacerbating the inter-symbol interference in the time-varying channel; a lower roll-off factor can compress the time-domain expansion and reduce the impact of Doppler spread, but will relax the spectral limit, leading to an increase in adjacent channel interference. The length of the filter is also crucial, a longer filter can provide better spectral concentration to resist the frequency-selective fading of the jungle environment, but will increase the system delay and amplify the impact of channel time variation; a shorter filter responds quickly and is more robust to channel changes, but at the expense of frequency resolution and interference suppression ability. In particular, under the condition of fast time variation unique to the jungle channel, the filter needs to have good localization characteristics in both time and frequency domains to balance the dual challenges of time delay spread and Doppler spread. Therefore, the optimization of filter coefficients must consider the multipath strength, time variation rate and interference characteristics of the jungle channel, and find the best balance between spectral efficiency, interference suppression and time variation robustness by adjusting the coefficients, so as to minimize the system bit error rate. The effective capacity and the bit error rate are in a monotonic decreasing relationship, and then the maximum value of the effective capacity is obtained.

[0132] The numerical trend of the utility function under different a values is observed through MATLAB simulation, as shown in Figure 1 . Figure 1 The fitness value of the best individual of each generation is recorded aThe value and its corresponding optimal utility function value. With the increase of the number of iterations, the curve gradually converges, and the utility function as a whole shows a downward trend, indicating that the optimization is successfully achieved. After optimization, the final solution of the system is a opt =0.0527, and the corresponding utility function is the effective capacity F=4.893721. The results show that when the roll-off coefficient of the pulse filter is 5.27% in the GFDM parameter design, the optimal effective capacity can be obtained in the jungle communication system.

[0133] Optimization of the number of subcarriers K:

[0134] Take the optimal solution obtained in the above step of optimizing the roll-off coefficient when the best utility function is reached a opt ; the value of a opt As a constant waveform parameter in the GFDM jungle communication system, the rest of the constant waveform parameters remain unchanged in the given configuration. The roll-off coefficient of the pulse shaping filter is set to 0.0527. The rest of the parameters are set as follows:

[0135] The optimal roll-off coefficient is 0.0527; the cyclic prefix cp is set to 0.2; K max The maximum number of subcarriers is 512; K min The minimum number of subcarriers is 16; the signal-to-noise ratio SNR is set to 25(dB), the modulation bit number mu is set to 4; the number of sub-symbols M is set to 15; the tree trunk diameter TD is set to 0.1; Dc The vegetation density is set to 2.6; FD The vegetation coverage is set to 75; the QoS index θ is set to 1; d The transmission distance is set to 600(m); f The signal frequency is set to 0.87GHz; fs The sampling rate is set to 5MHz.

[0136] For the same utility function, the particle swarm optimization algorithm optimizes the number of subcarriers K, and the value range is limited to K min , K max ] (where K min is the minimum value of the number of subcarriers defined by the user, K max is the maximum value of the number of subcarriers defined by the user).

[0137] First, the algorithm core parameters are determined, including the particle swarm size (i.e. the total number of particles, 50), the maximum number of iterations (100), the inertia weight ([0.3 0.8]), the learning factor (both set to 1.5), the boundary range of speed and position; then, in the solution space of the problem to be optimized, the initial position (corresponding to the candidate solution of the problem) and the initial speed of each particle are randomly generated, and the initial position of each particle is set as its individual optimal position c 1and c 2 (both set to 1.5), the boundary range of speed and position; then, in the solution space of the problem to be optimized, the initial position (corresponding to the candidate solution of the problem) and the initial speed of each particle are randomly generated, and the initial position of each particle is set as its individual optimal position , and the best one is selected from the individual optimal positions of all particles as the global optimal position of the entire population g b .

[0138] For the current position of each particle, the objective function of the problem is substituted to calculate its fitness value.

[0139] For each particle, the current fitness value is compared with the historical p b fitness value, and if the current value is better, the p b of the particle is updated to the current position; then the p b of all particles are traversed, and if the fitness value of a p b is better than the fitness value of the current g b , the g b of the population is updated to the position of the p b .

[0140] Update the speed and position of each particle:

[0141] Speed update: ;

[0142] Position update: ;

[0143] Check if the termination condition is met: one is to reach the preset maximum number of iterations, and the other is g b the corresponding fitness value has met the accuracy requirement of the problem (such as the change in fitness value for consecutive generations being less than a threshold value). If it is met, the iteration is stopped, and if it is not met, the "fitness calculation" step is returned to enter the next iteration.

[0144] After the iteration is terminated, the final g bThe position is output as the optimal solution to the problem, and its corresponding fitness value can be output simultaneously to complete the entire optimization process and obtain the final value of the optimized utility function F.

[0145] In jungle channel environments, the relationship between the number of subcarriers and the bit error rate (BER) in GFDM systems exhibits unique complexity. Jungle channels are characterized by severe multipath fading, dense scatterers, and high time-varying characteristics, leading to a nonlinear impact of subcarrier number variations on the BER. When the number of subcarriers is low, a wider subcarrier spacing, while able to resist frequency-selective fading, still generates severe inter-symbol interference under dense multipath conditions, resulting in a persistently high BER. As the number of subcarriers increases, a narrower subcarrier spacing allows for more precise spectrum segmentation, better adapting to the complex frequency response variations in jungle channels and improving BER performance through frequency diversity. However, once the number of subcarriers exceeds a certain threshold, extremely narrow subcarrier spacing makes the system highly sensitive to Doppler shift and phase noise common in jungle environments. Simultaneously, dense subcarrier configurations exacerbate inter-carrier interference under strong multipath conditions, causing the BER to rise again. Furthermore, severe shadowing and vegetation obstruction in jungle channels introduce slow fading, requiring the system to consider adaptive power adjustment during subcarrier allocation, further impacting the relationship between the BER and the number of subcarriers. In jungle channels, there exists an optimal range of subcarrier numbers within which the system achieves the best balance between frequency diversity gain and interference suppression, resulting in the lowest bit error rate. Effective capacity and bit error rate have a monotonically decreasing relationship, leading to the maximum effective capacity.

[0146] The numerical trend of the utility function under different K values ​​was observed through MATLAB simulation, such as... Figure 2 As shown. Figure 2 The K value and its corresponding optimal utility function value for each generation of the best individual were recorded. As the number of iterations increased, the curve gradually converged, and the utility function showed an overall decreasing trend, indicating successful optimization. After optimization, the final solution of the system was obtained as follows: K opt =18, corresponding to the utility function, i.e., the effective capacity F=5.703782. This result shows that when the pulse filter roll-off factor is 5.27% and the number of subcarriers is 18 in the GFDM parameter design, the optimal effective capacity can be obtained in the jungle communication system.

[0147] This invention also provides a GFDM effective capacity optimization system for jungle channels, the system comprising:

[0148] The performance testing module is used to acquire the original binary data stream from the transmitter, preprocess and demodulate the signal received by the receiver to obtain the demodulated binary data stream, and calculate the bit error rate based on the original binary data stream and the demodulated binary data stream.

[0149] a parameter optimization module, configured to calculate effective capacity based on the bit error rate, and apply a particle swarm algorithm to stepwise optimize GFDM waveform design parameters including a roll-off coefficient and a subcarrier number, with the effective capacity as a utility function;

[0150] The parameter optimization step specifically includes:

[0151] The roll-off coefficient is set as a decision variable, and the particle swarm algorithm is used for iterative optimization in a preset roll-off coefficient value interval to obtain an optimal roll-off coefficient, with the effective capacity as a fitness function of the particle swarm algorithm.

[0152] The obtained optimal roll-off coefficient is taken as a fixed parameter, the effective capacity is taken as a fitness function, the subcarrier number is set as a decision variable, and the particle swarm algorithm is used for iterative optimization in a preset subcarrier number value interval to obtain an optimal subcarrier number.

[0153] The above merely describes preferred embodiments of the present application but should not be taken in a limiting sense, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall fall within the scope of the present application.

Claims

1. A method for optimizing effective capacity of GFDM for jungle channel, characterized in that, The method comprises: acquiring a binary data stream at a transmitting end, pre-processing and demodulating a signal received at a receiving end to obtain a demodulated binary data stream, and calculating a bit error rate based on the binary data stream at the transmitting end and the demodulated binary data stream; calculating an effective capacity based on the bit error rate, taking the effective capacity as a utility function, and stepwise optimizing GFDM waveform design parameters by using a particle swarm algorithm, wherein the GFDM waveform design parameters comprise a roll-off coefficient and a subcarrier number; the parameter optimization step specifically comprises: taking the effective capacity as a fitness function of the particle swarm algorithm, taking the roll-off coefficient as a decision variable, and iteratively optimizing the roll-off coefficient in a preset roll-off coefficient value range by using the particle swarm algorithm to obtain an optimal roll-off coefficient; taking the obtained optimal roll-off coefficient as a fixed parameter, taking the effective capacity as a fitness function, taking the subcarrier number as a decision variable, and iteratively optimizing the subcarrier number in a preset subcarrier number value range by using the particle swarm algorithm to obtain an optimal subcarrier number. 2.The method of claim 1, wherein The step of acquiring a binary data stream at a transmitting end, pre-processing and demodulating a signal received at a receiving end to obtain a demodulated binary data stream specifically comprises: modulating an input binary data stream at the transmitting end by M-QAM to obtain a complex symbol matrix; performing pulse shaping on the complex symbol matrix by a raised cosine roll-off filter to generate a time domain signal; adding a cyclic prefix to the time domain signal to obtain a transmission signal; after the transmission signal is transmitted through a jungle channel, a receiving end obtains a received signal; the receiving end performs matched filtering and equalization processing on the received signal to recover a complex symbol estimation matrix, and demodulates the complex symbol estimation matrix to obtain a demodulated binary data stream.

3. The method of Claim 1, wherein The bit error rate calculation formula is: ； wherein, BER BER is the bit error rate, b tx N is the number of transmitted bits, b rx N is the number of received bits, N N is the total number of bits, µ N is the number of bits per symbol.

4. The method of Claim 1, wherein the effective capacity C(θ) The calculation formula is: ； ； In the formula, θ is a QoS index, BER is a system error rate, M is a QAM modulation order, fs is a sampling rate, a is a roll-off factor.

5. The method of Claim 1, wherein The step of iteratively optimizing by using the particle swarm algorithm comprises: initializing a particle swarm, setting a particle size, a maximum iteration number, an inertia weight, a learning factor, and randomly generating an initial position and a speed of each particle in a solution space corresponding to a decision variable; calculating a fitness value corresponding to a current position of each particle; updating an individual historical optimal position of each particle and a global historical optimal position of the entire particle swarm according to the fitness value; updating a speed and a position of each particle according to a speed updating formula and a position updating formula; judging whether an iteration termination condition is met, and if the iteration termination condition is met, terminating the iteration and outputting the global historical optimal position as an optimal solution, and if the iteration termination condition is not met, returning to fitness calculation and entering a next round of iteration.

6. The method of Claim 1, wherein Before the particle swarm algorithm is applied for optimization, a jungle channel model is constructed; the jungle channel model simulates multipath fading by using a Rayleigh distribution, and calculates a path loss PL by using a jungle path loss empirical model, and a calculation formula of the path loss PL is comprises: ； in, PL_dθ To be at the reference distance dθ The base path loss measured at the location, d denoted as the actual transmission distance, and n as the path loss index related to the signal frequency f, trunk diameter TD, and vegetation density Dc.

7. The method of Claim 1, wherein A calculation formula of the path loss index n is: ； wherein, D is the trunk diameter, Dc D is the vegetation density, k 1 is a base correction factor determined by frequency, k 2 is a base path loss exponent.

8. A system for jungle channel oriented GFDM effective capacity optimization, for implementing the method of any one of claims 1-7, characterized in that, The system comprises: a performance test module configured to acquire a binary data stream at a transmitting end, pre-process and demodulate a signal received at a receiving end to obtain a demodulated binary data stream, and calculate a bit error rate based on the binary data stream at the transmitting end and the demodulated binary data stream. The parameter optimization module is configured to calculate an effective capacity based on the bit error rate, take the effective capacity as a utility function, and step by step optimize GFDM waveform design parameters including a roll-off coefficient and a subcarrier number by using a particle swarm algorithm. The parameter optimization step specifically includes: Taking the effective capacity as a fitness function of the particle swarm algorithm, setting the roll-off coefficient as a decision variable, and performing iterative optimization in a preset roll-off coefficient value interval by using the particle swarm algorithm to obtain an optimal roll-off coefficient. Taking the obtained optimal roll-off coefficient as a fixed parameter, taking the effective capacity as a fitness function, setting the subcarrier number as a decision variable, and performing iterative optimization in a preset subcarrier number value interval by using the particle swarm algorithm to obtain an optimal subcarrier number.

Images