5G network intelligent optimization method and system

Abstract

The invention relates to the technical field of 5G networks, in particular to a 5G network intelligent optimization method and system, and the method comprises the steps: obtaining 5G network signal data, recognizing an abnormal power spectral density region, carrying out the filtering extraction of the abnormal power spectral density region, and separating interference signals, interference types including narrowband interference, broadband interference and directional interference; extracting features from the obtained interference signals, and performing interference type identification according to a random forest algorithm; starting a corresponding anti-interference means according to the interference type, and generating operation state data in real time; a bee colony algorithm is adopted to simulate bee behaviors for iterative search, and a local optimal resource scheduling strategy is determined; and executing the resource scheduling strategy, feeding back an execution effect, and restarting the bee colony algorithm to determine a new resource scheduling strategy if the execution effect does not reach a set expectation. Therefore, the problems of lack of dynamic adaptive adjustment, single anti-interference means, lack of cross-base station collaboration and the like in the anti-interference aspect in the prior art are solved.

Description

Technical Field

[0001] The present application relates to the field of 5G network technology, and in particular to a 5G network intelligent optimization method and system. Background Art

[0002] With the large-scale commercial deployment of 5G networks, the conflict between the development of high-frequency spectrum resources and the complex electromagnetic environment has become increasingly prominent. 5G networks offer high bandwidth, low latency, and a large number of connections, but they also face various interference challenges, including narrowband interference (such as single-frequency interference from adjacent wireless devices), wideband interference (such as radar pulses and intermodulation products of communication systems), and directional interference (such as malicious signal suppression in specific areas). This interference can lead to degraded signal quality, increased bit error rates, and even network service interruptions, seriously impacting the reliability of 5G in scenarios such as the Internet of Vehicles, the Industrial Internet, and smart cities.

[0003] Traditional anti-interference technologies, such as fixed spread spectrum and static beamforming, lack dynamic perception of the interference environment and the ability to adapt adaptively, making them limited in effectiveness in complex and changing interference scenarios. For example, fixed spread spectrum sequences are easily intercepted and jammed by the enemy, and static beamforming cannot track changes in the interference direction in real time. Furthermore, the high-density deployment and spectrum resource sharing of 5G networks mean that the anti-interference strategies of a single base station may trigger resource conflicts between neighboring stations, necessitating cross-base station coordination mechanisms to improve overall network performance. Summary of the Invention

[0004] This application provides a 5G network intelligent optimization method and system to solve the problems of the existing technology in 5G anti-interference, such as lack of dynamic adaptive adjustment, single anti-interference means, and lack of cross-base station collaboration.

[0005] The first embodiment of the present application provides a 5G network intelligent optimization method, including the following steps: A1: Acquire 5G network signal data, convert the time-domain signal into a frequency-domain spectrum using a fast Fourier transform (FFT), compare the spectrum with the 5G standard signal's spectrum template, identify areas of abnormal power spectrum density, and perform filtering and extraction on these areas to isolate the interference signal. A2: Extracting features from the obtained interference signal and using the features to identify interference types using a random forest algorithm. The interference types include narrowband interference, broadband interference, and directional interference. A3: Activate corresponding anti-interference measures based on the interference type and generate operating status data in real time. If the interference type is narrowband interference, generate a chaotic coding sequence based on the improved logistic map to increase the spread spectrum gain, and update the chaotic system parameters in real time through quantum key distribution technology. If the interference type is broadband interference, activate the adaptive frequency hopping mode, dynamically adjust the frequency hopping period according to the interference situation, establish a frequency hopping coordination mechanism with adjacent base stations, and ensure the trusted interaction of frequency hopping information through blockchain technology. If the interference type is directional interference, suppress it based on beam space filtering. A4: Using a bee swarm algorithm to receive the operating status data, setting an objective function of the bee swarm algorithm based on current network requirements and anti-interference goals, simulating bee behavior to iteratively search and determine a locally optimal resource scheduling strategy; A5: Execute the resource scheduling strategy and provide feedback on the execution effect. If the execution effect does not meet the set expectations, restart the bee swarm algorithm to determine a new resource scheduling strategy.

[0006] In a preferred embodiment, 5G network signal data is obtained, and the time domain signal is converted into a frequency domain spectrum using a fast Fourier transform (FFT). The spectrum template of the 5G standard signal is compared to identify abnormal power spectrum density areas, and the abnormal power spectrum density areas are filtered and extracted to separate the interference signal. The specific steps are as follows: B1. Collect a time domain signal sequence through the RF front end of the 5G base station, preprocess the time domain signal sequence, limit the bandwidth to the 5G operating frequency band, and obtain target signal data x(f); B2. Convert the target signal data into a frequency domain representation by fast Fourier transform (FFT), and calculate the power spectrum density based on the transformed result to obtain a power spectrum estimate. The specific formula of Fourier transform is: , , where N is the number of FFT points and k is the index of the discrete signal in the frequency domain. is a complex exponential function that weights the time domain signal; the power spectrum estimation formula is: ; B3. Construct a 5G standard spectrum template and calculate the deviation between the power spectrum estimate and the template spectrum ,in, is a template spectrum, and if the deviation is greater than the target threshold, the region is determined to be abnormal; B4. Design of notch filter , filter and extract the abnormal area, then convert the filtered frequency domain signal back to the time domain according to the inverse FFT method, reconstruct the interference signal, and obtain the separated interference signal, where the filter extraction formula is: , the inverse FFT formula is .

[0007] In a preferred embodiment, features are extracted from the obtained interference signal, and the features are used to identify interference types according to a random forest algorithm. The interference types include narrowband interference, broadband interference, and directional interference. The specific steps are as follows: C1, the interference signal After preprocessing, extract the signal's time domain, frequency domain, time-frequency domain, and spatial domain features respectively. Concatenate the extracted feature vectors into a multidimensional feature vector. Through feature importance analysis, select the M most discriminative features to form a comprehensive feature vector x. The time domain features include signal variance, duty cycle, and crest factor; the frequency domain features include center frequency, spectral entropy, and normalized bandwidth; the time-frequency domain includes a time-frequency graph; and the spatial domain features are the angle of arrival (AoA) of the interference signal estimated using the MUSIC algorithm using the base station's multi-antenna array. C2. Input the comprehensive feature vector x into the trained random forest classification model and output the classification results, which include narrowband interference, broadband interference, and directional interference. The classification formula is: , where c is the value of the interference type, T is the number of decision trees, is the classification result of the t-th tree for sample x, .

[0008] In a preferred embodiment, if the interference type is narrowband interference, a chaotic coding sequence is generated according to an improved logistic map to increase the spread spectrum gain, and the chaotic system parameters are updated in real time through quantum key distribution technology. The specific steps are as follows: D1. Transmit polarized photons through quantum channels to generate the original key , the original key Divided into three parts, mapped to the initial value of the chaotic system , control parameter r and disturbance coefficient k, each Re-execute QKD in milliseconds and update parameters synchronously to ensure that the key space complexity reaches above; D2. Use improved Logistic mapping to iteratively generate chaotic sequences , where the generation formula is: ,in, is the chaos control parameter, is the perturbation coefficient, the initial value Randomly generated by quantum key distribution, the chaotic sequence Take the first N values and perform binarization processing, mapping the continuous values into bipolar code elements c(n); D3, multiply the data signal d(t) by the chaotic sequence c(n) to realize direct sequence spread spectrum. The specific formula is: , where t is the time variable, representing the state of the signal at different times, is the pulse shaping function, is the chip period.

[0009] In a preferred embodiment, if the interference type is broadband interference, an adaptive frequency hopping mode is activated, the frequency hopping period is dynamically adjusted according to the interference situation, and a frequency hopping coordination mechanism is established with adjacent base stations. Blockchain technology is used to ensure the trusted interaction of frequency hopping information. The specific steps are as follows: E1. Calculate the frequency modulation period: Scan the entire 5G operating frequency band and generate an interference heat map , filter the frequency bands whose interference power does not exceed the threshold, obtain the available frequency band A, and dynamically adjust the initial frequency hopping period according to the interference intensity , ,in, is the base period, is the adjustment factor, B is the total bandwidth; E2. Generate frequency hopping sequence and frequency selection: Generate frequency hopping sequence using improved Tent mapping , the chaotic sequence Mapped to the available frequency band A, the mapping function is ,in is the lower limit of the available frequency band, is the frequency interval; E3. Blockchain collaboration: Adjacent base stations form a consortium chain network. Each base station acts as a node and runs a blockchain client. Each base station encapsulates the current frequency hopping parameters (frequency hopping sequence, frequency hopping period, available frequency band) as a transaction and writes it into the blockchain through the PBFT consensus algorithm. Each base station queries the frequency hopping information of the neighboring station through the smart contract and executes the interference avoidance algorithm. The specific formula is: ,in, is the weight between base stations, and Overlap is the frequency band overlap.

[0010] In a preferred embodiment, if the interference type is directional interference, suppression is performed according to beam space filtering, and the specific steps are as follows: F1. Accurately locate the interference direction: The MUSIC algorithm is used again to estimate the arrival angle of the interference signal with higher resolution, with an accuracy of less than 1°, to determine the specific incident direction of the interference. F2. Beamforming matrix design: Design the beamforming matrix based on the base station antenna array structure and the specific incident direction of the interference. , so that the matrix forms a null in the interference direction and maintains gain in the target signal direction. The null constraint optimization formula is: , where w is the beamforming vector, is the beam space steering vector; F3, signal weighting processing: Multiply the received signal with the designed beamforming matrix W, perform weighted summation on the signals received by different antennas, enhance the target signal, suppress interference signals from specific directions, and effectively suppress directional interference.

[0011] In a preferred embodiment, a bee swarm algorithm is used to receive the operating status data, an objective function of the bee swarm algorithm is set according to the current network demand and anti-interference target, and an iterative search is performed to simulate the bee behavior to determine the local optimal resource scheduling strategy. The specific steps are as follows: G1. Objective function definition: The objective function of the swarm algorithm is set according to the current network requirements and anti-interference goals. Let the throughput be T, the delay be D, and the interference intensity be I. The objective function is ,in, 、 、 is the weight coefficient, and ; G2. Initialize the bee colony: Randomly generate N individual worker bees, each of which represents a resource scheduling strategy, forming the initial solution space. The resource scheduling strategy includes parameters such as channel allocation, power control, and beamforming. G3, Iterative Search and Local Optimization: This method simulates the foraging behavior of bees, where leader, follower, and scout bees collaborate to perform iterative search. After multiple rounds of iterations, when the objective function value no longer increases significantly or reaches the preset number of iterations, the currently optimal resource scheduling strategy is output as the local optimal solution.

[0012] In a preferred embodiment, the resource scheduling strategy is executed and the execution effect is fed back. If the execution effect does not meet the set expectations, the bee swarm algorithm is restarted to determine a new resource scheduling strategy, including: The network control center encapsulates the optimal strategy parameters (channel allocation scheme, power control value, beamforming weight) determined by the swarm algorithm into control signaling and sends it to each device via the downlink control channel. The device then adjusts its hardware configuration in real time based on the received parameters. For example, the terminal adjusts the uplink transmit power and modulation coding scheme, the base station adjusts the phase and amplitude weights of the antenna array, and dynamically allocates physical resource blocks and time-frequency resources. The device collects key performance indicators in real time and reports the monitoring results every 20ms. The deviation between the actual performance and the expected target is calculated based on the evaluation function. If the deviation is greater than the target threshold, it is determined that the performance has not met the expectations and the swarm algorithm is restarted. ,in, is the weight of the i-th indicator, is the actual value, is the target value; The bee swarm algorithm adjusts the parameters of the bee swarm algorithm according to the last execution result, initializes the bee swarm with the historical optimal solution, and accelerates convergence.

[0013] The second aspect of the present application provides a 5G network intelligent optimization system, including: a 5G interference data separation module, an interference type identification module, an anti-interference module, and a data scheduling module, wherein the 5G interference data separation module is used to obtain 5G network signal data, use fast Fourier transform (FFT) to convert the time domain signal into a frequency domain spectrum, compare the spectrum template of the 5G standard signal, identify the abnormal power spectrum density area, filter and extract the abnormal power spectrum density area, and separate the interference signal; the interference type identification module is used to extract features from the obtained interference signal, and identify the interference type according to the random forest algorithm, wherein the interference type includes narrowband interference, broadband interference, and directional interference; the anti-interference module is used to start the corresponding anti-interference means according to the interference type, and generate operation status data in real time, wherein, if the interference type If the interference is narrowband interference, a chaotic coding sequence is generated according to the improved Logistic mapping to increase the spread spectrum gain, and the chaotic system parameters are updated in real time through quantum key distribution technology; if the interference type is broadband interference, an adaptive frequency hopping mode is started, the frequency hopping period is dynamically adjusted according to the interference situation, and a frequency hopping coordination mechanism is established with adjacent base stations, and the trusted interaction of frequency hopping information is ensured through blockchain technology; if the interference type is directional interference, it is suppressed according to beam space filtering; the data scheduling module is used to receive the operating status data using a swarm algorithm, set the objective function of the swarm algorithm according to the current network demand and anti-interference target, simulate the iterative search of bee behavior, and determine the local optimal resource scheduling strategy; the device executes the resource scheduling strategy and feeds back the execution effect. If the execution effect does not meet the set expectations, the swarm algorithm is restarted to determine a new resource scheduling strategy.

[0014] The third aspect of the present application provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are configured to execute a 5G network intelligent optimization method as described in the above embodiment.

[0015] The beneficial effects achieved by adopting the above-mentioned present invention are as follows: By converting the time-domain signal into a frequency-domain spectrum using FFT and comparing it with the 5G standard spectrum template, the system can accurately locate abnormal power areas and improve the accuracy of interference separation. The random forest algorithm is used to fuse the time-frequency and spatial characteristics of the interference signal, overcoming the limitations of a single feature and significantly improving the accuracy of interference type identification. Different anti-interference measures are adopted according to the interference type. For narrowband interference, a dual security mechanism of chaotic spread spectrum and quantum key is used. For wideband interference, global avoidance using dynamic frequency hopping and blockchain collaboration is adopted. For directional interference, spatial nulling suppression using beam space filtering is adopted, comprehensively enhancing the adaptability and coordination of anti-interference measures. A local optimal strategy is formulated using a bee swarm algorithm to achieve intelligent optimization of resource scheduling. This solves the technical problems of existing technologies in 5G anti-interference, such as the lack of dynamic adaptive adjustment, single anti-interference measures, and lack of cross-base station coordination.

[0016] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 This is a flowchart of a 5G network intelligent optimization method provided according to an embodiment of the present application; Figure 2 A flowchart of interference signal acquisition according to an embodiment of the present application; Figure 3 A flowchart of interference type classification according to an embodiment of the present application; Figure 4 A specific flow chart of the intelligent optimization method provided according to an embodiment of the present application; Figure 5 A schematic diagram of the composition of a 5G network intelligent optimization system provided according to an embodiment of the present application; Figure 6 A schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0019] The following describes a 5G network intelligent optimization method and system of an embodiment of the present application with reference to the accompanying drawings. In response to the problems mentioned in the above background technology that 5G anti-interference lacks dynamic adaptive adjustment, single anti-interference means, and lack of cross-base station collaboration, the present application provides a 5G network intelligent optimization method. In this method, the time domain signal is converted into a frequency domain spectrum by FFT, and combined with the 5G standard spectrum template comparison, the abnormal power area can be accurately located to improve the accuracy of interference separation; the random forest algorithm is used to fuse the time-frequency and spatial features of the interference signal, overcome the limitations of a single feature, and significantly improve the accuracy of interference type identification; different anti-interference means are adopted according to the interference type. For narrowband interference, a dual security mechanism of chaotic spread spectrum-quantum key is adopted. For broadband interference, a global avoidance of dynamic frequency hopping-blockchain collaboration is adopted. For directional interference, a spatial nulling suppression of beam space filtering is adopted, which comprehensively enhances the adaptability and collaboration of the anti-interference means; a local optimal strategy is formulated by a swarm algorithm to achieve intelligent optimization of resource scheduling. Thus, the existing technology solves the problems of lack of dynamic adaptive adjustment, single anti-interference means, and lack of cross-base station collaboration in 5G anti-interference.

[0020] Specifically, Figure 1 A flowchart of a 5G network intelligent optimization method provided in an embodiment of the present application.

[0021] like Figure 1 As shown, the 5G network intelligent optimization method includes the following steps: In step A1, 5G network signal data is obtained, and the time domain signal is converted into a frequency domain spectrum using the fast Fourier transform (FFT). The spectrum template of the 5G standard signal is compared to identify abnormal power spectrum density areas. The abnormal power spectrum density areas are filtered and extracted to separate the interference signal.

[0022] It is understandable that the embodiment of the present application utilizes fast Fourier transform (FFT) combined with spectrum template comparison to separate 5G network interference signals, which can accurately and efficiently locate the interference source. FFT converts the time domain signal into a frequency domain spectrum, which can intuitively display the frequency component distribution of the signal, and the 5G standard signal spectrum template is used as a reference benchmark to quickly lock areas that do not meet the normal signal characteristics. By targeted filtering of abnormal power spectrum density areas, the interference signal can be separated from the complex received signal. This method avoids blind processing of the entire signal, greatly improves the efficiency of interference separation, and reduces the waste of computing resources, ensuring that interference can still be accurately identified in complex electromagnetic environments, laying the foundation for subsequent interference suppression and network optimization.

[0023] In the embodiment of the present application, 5G network signal data is obtained, and the time domain signal is converted into a frequency domain spectrum using a fast Fourier transform (FFT). The spectrum template of the 5G standard signal is compared to identify abnormal power spectrum density areas, and the abnormal power spectrum density areas are filtered and extracted to separate the interference signal. The specific steps are as follows: B1. The RF front-end of the 5G base station collects the time domain signal sequence, pre-processes the time domain signal sequence, limits the bandwidth to the 5G operating frequency band, and obtains the target signal data x(f); B2. Convert the target signal data into frequency domain representation through fast Fourier transform (FFT). Calculate the power spectrum density based on the transformed result to obtain the power spectrum estimation. The specific formula of Fourier transform is: , , where N is the number of FFT points and k is the index of the discrete signal in the frequency domain. is a complex exponential function that weights the time domain signal; the power spectrum estimation formula is: ; B3. Construct a 5G standard spectrum template and calculate the deviation between the power spectrum estimate and the template spectrum ,in, is the template spectrum. If the deviation is greater than the target threshold, the region is judged to be abnormal; B4. Design of notch filter , filter and extract the abnormal area, then convert the filtered frequency domain signal back to the time domain according to the inverse FFT method, reconstruct the interference signal, and obtain the separated interference signal, where the filter extraction formula is: , the inverse FFT formula is .

[0024] Specifically, in a certain city's commercial district, 5G base stations simultaneously carry the communication needs of a large number of users, and the signal environment is complex. One day, the signal received by the base station becomes abnormal, the data transmission rate decreases, and the packet loss rate increases. Figure 2 As shown in the figure, the base station's RF front-end collects time-domain signals. After preprocessing to lock the 5G operating frequency band, the acquired time-domain signals are transformed using FFT to obtain the frequency-domain spectrum. Comparing this with the spectrum template of the 5G standard signal reveals that the power spectral density within a specific frequency range is significantly higher than normal, indicating interference in this area. A notch filter is then used to filter the abnormal frequency band, and an inverse FFT transform is then performed to successfully isolate the interfering signal. Based on the interference signal, the interference type is determined and appropriate anti-interference measures are implemented.

[0025] It is understandable that the embodiment of the present application first limits the signal bandwidth to the 5G operating frequency band, focusing on the effective signal range and reducing irrelevant data interference; uses FFT and power spectrum estimation to deeply analyze the signal frequency characteristics, combined with standard spectrum template comparison, it can keenly capture subtle anomalies; the notch filter accurately "cuts off" the interference frequency band, and restores the time domain signal through inverse FFT conversion to ensure that the interference signal is completely removed. It not only ensures the accuracy of separation, but also takes into account computational efficiency. It can quickly locate and separate various types of interference, provide reliable data support for subsequent interference suppression and network optimization, and significantly improve the 5G network's anti-interference ability and operational stability.

[0026] In step A2, features are extracted from the obtained interference signal, and the features are used to identify interference types according to the random forest algorithm. The interference types include narrowband interference, broadband interference, and directional interference.

[0027] It is understandable that the embodiment of the present application uses a random forest algorithm to identify the characteristics of the interference signal. By constructing multiple decision trees and integrating the decision results, it can fully explore the complex relationship between the multi-dimensional characteristics of the interference signal in the time domain, frequency domain, spatial domain, etc., and avoid the overfitting problem that may occur in a single decision tree. When faced with the complex and changeable interference environment in the 5G network, the characteristic patterns of different types of interference can be adaptively learned, and a high recognition accuracy can be maintained even in the presence of noise or missing feature data. By accurately extracting and classifying the characteristics of the interference signal, the type of interference can be quickly determined, providing a reliable basis for the subsequent targeted anti-interference measures.

[0028] In the embodiment of the present application, features are extracted from the obtained interference signal, and the features are used to identify the interference type according to the random forest algorithm. The interference types include narrowband interference, broadband interference, and directional interference. The specific steps are as follows: C1, the interference signal After preprocessing, the time domain, frequency domain, time-frequency domain, and spatial domain features of the signal are extracted respectively. The extracted feature vectors are spliced into a multidimensional feature vector. Through feature importance analysis, the M most discriminative features are selected to form a comprehensive feature vector x. Among them, the time domain features include signal variance, duty cycle, and crest factor; the frequency domain features include center frequency, spectral entropy, and normalized bandwidth; the time-frequency domain includes a time-frequency graph; and the spatial domain feature is the angle of arrival (AoA) of the interference signal estimated by the MUSIC algorithm using the base station's multi-antenna array.

[0029] C2. Input the comprehensive feature vector x into the trained random forest classification model and output the classification results. The classification results include narrowband interference, broadband interference, and directional interference. The classification formula is , where c is the value of the interference type, T is the number of decision trees, is the classification result of the t-th tree for sample x, .

[0030] Specifically, in a certain urban transportation hub area, 5G base stations were affected by multiple interferences at the same time, resulting in network signal quality degradation and unstable data transmission. Figure 3 As shown in the figure, after the base station detects the interference signal, it first extracts multi-dimensional features such as the signal's time-domain variance, frequency-domain spectral entropy, time-frequency graph texture characteristics, and spatial angle of arrival. Through feature importance analysis, the most discriminative features are selected to form a comprehensive feature vector. This vector is then input into a trained random forest classification model, and the model output indicates that the interference is broadband. Based on accurate interference type identification, an adaptive frequency hopping strategy is implemented to quickly suppress the interference, restoring normal 5G network service in the area, ensuring passengers' mobile network experience and the stable operation of the intelligent transportation system.

[0031] It is understandable that the embodiment of the present application significantly improves the accuracy and reliability of 5G network interference type identification by combining multi-dimensional feature extraction with the random forest algorithm. On the one hand, the interference signal characteristics are comprehensively captured from four dimensions: time domain, frequency domain, time-frequency domain, and spatial domain. The time domain features reflect the signal strength fluctuations, the frequency domain features reveal the frequency distribution law, the time-frequency domain features show the frequency changes of the signal over time, and the spatial domain features locate the direction of the interference source. Multi-dimensional information fusion avoids the one-sidedness of a single feature. On the other hand, the random forest algorithm is based on an ensemble learning strategy. By voting through multiple decision trees, it effectively reduces the risk of model overfitting and enhances adaptability to complex interference scenarios. Through feature importance analysis, the core features are further screened, redundant information is reduced, and computational efficiency is improved. It can quickly and accurately identify different types of interference in complex electromagnetic environments, providing a key basis for subsequent targeted anti-interference measures.

[0032] In step A3, the corresponding anti-interference measures are activated according to the interference type, and the operating status data is generated in real time. If the interference type is narrowband interference, a chaotic coding sequence is generated according to the improved Logistic mapping to increase the spread spectrum gain, and the chaotic system parameters are updated in real time through quantum key distribution technology; if the interference type is broadband interference, the adaptive frequency hopping mode is activated, the frequency hopping period is dynamically adjusted according to the interference situation, and a frequency hopping coordination mechanism is established with adjacent base stations, and the trusted interaction of frequency hopping information is ensured through blockchain technology; if the interference type is directional interference, it is suppressed according to beam space filtering.

[0033] It is understandable that the embodiments of the present application implement differentiated and intelligent anti-interference strategies according to different interference types, thereby improving the adaptability and reliability of 5G networks in complex environments. For narrowband interference, the improved Logistic mapping combined with quantum key distribution not only enhances the randomness and spread spectrum gain of the chaotic sequence, but also ensures the security of parameter updates; in the face of broadband interference, adaptive frequency hopping and blockchain collaborative mechanisms realize dynamic optimization of frequency resources and trusted interaction across base stations, effectively avoiding interference frequency bands; and directional interference uses beam space filtering to accurately suppress the interference direction using the spatial characteristics of the antenna array, avoiding the limitations of a single strategy for dealing with multiple interferences. Real-time generation of operating status data provides a dynamic basis for network optimization.

[0034] In the embodiment of the present application, if the interference type is narrowband interference, a chaotic coding sequence is generated according to the improved Logistic map to increase the spread spectrum gain, and the chaotic system parameters are updated in real time through quantum key distribution technology. The specific steps are as follows: D1. Transmit polarized photons through quantum channels to generate the original key , the original key Divided into three parts, mapped to the initial value of the chaotic system , control parameter r and disturbance coefficient k, each Re-execute QKD in milliseconds and update parameters synchronously to ensure that the key space complexity reaches above.

[0035] D2. Use improved Logistic mapping to iteratively generate chaotic sequences , where the generation formula is: ,in, is the chaos control parameter, is the perturbation coefficient, the initial value Randomly generated by quantum key distribution, the chaotic sequence Take the first N values and perform binarization processing, and map the continuous values into bipolar code elements c(n).

[0036] D3, multiply the data signal d(t) by the chaotic sequence c(n) to realize direct sequence spread spectrum. The specific formula is: , where t is the time variable, representing the state of the signal at different times, is the pulse shaping function, is the chip period.

[0037] Specifically, in a certain area's 5G private network, base stations are frequently interfered with by unknown sources. After judging the interference type, the system determines that the interference type is narrowband interference, and immediately starts the quantum key distribution and improved Logistic mapping mechanism. The chaotic system parameters are updated through the quantum channel every millisecond to generate a random chaotic sequence to spread the data signal. In one interference event, the interference source continuously emitted single-frequency interference in the 1.8GHz frequency band with a power intensity of -30dBm. However, since the chaotic spread spectrum gain used by the system is 18dB, and the sequence changes randomly every millisecond, the energy of the interference signal is effectively diluted, and the bit error rate at the receiving end is always maintained at . The following ensures reliable data transmission.

[0038] It can be understood that the embodiments of the present application provide dual guarantees of high security and strong robustness for 5G network narrowband interference suppression through the deep integration of quantum key distribution and improved Logistic mapping. The unconditional security of quantum keys ensures that the parameters of the chaotic system are updated in real time and are unpredictable. The update mechanism every millisecond makes the complexity of the key space far exceed the cracking ability of traditional encryption algorithms. The chaotic sequence generated by the improved Logistic mapping has better randomness and ergodicity, and the spread spectrum gain is improved by 3-5dB compared with the traditional PN sequence, which can effectively resist single-frequency and multi-frequency interference. Bipolar symbol conversion makes the sequence more suitable for digital communication systems, and direct sequence spread spectrum disperses the signal energy into a wider frequency band, significantly reducing the impact of narrowband interference on demodulation performance.

[0039] In the embodiment of the present application, if the interference type is broadband interference, the adaptive frequency hopping mode is activated, the frequency hopping period is dynamically adjusted according to the interference situation, and a frequency hopping coordination mechanism is established with adjacent base stations. The blockchain technology is used to ensure the trusted interaction of frequency hopping information. The specific steps are as follows: E1. Calculate the frequency modulation period: Scan the entire 5G operating frequency band and generate an interference heat map , filter the frequency bands whose interference power does not exceed the threshold, obtain the available frequency band A, and dynamically adjust the initial frequency hopping period according to the interference intensity , ,in, is the base period, is the adjustment factor and B is the total bandwidth.

[0040] E2. Generate frequency hopping sequence and frequency selection: Generate frequency hopping sequence using improved Tent mapping , the chaotic sequence Mapped to the available frequency band A, the mapping function is ,in is the lower limit of the available frequency band, is the frequency interval.

[0041] E3. Blockchain collaboration: Adjacent base stations form a consortium chain network. Each base station acts as a node and runs a blockchain client. Each base station encapsulates the current frequency hopping parameters (frequency hopping sequence, frequency hopping period, available frequency band) as a transaction and writes it into the blockchain through the PBFT consensus algorithm. Each base station queries the frequency hopping information of the neighboring station through the smart contract and executes the interference avoidance algorithm. The specific formula is: ,in, is the weight between base stations, and Overlap is the frequency band overlap.

[0042] Specifically, during a technology exhibition at a large convention center, a large number of wireless devices were deployed, generating severe broadband interference and causing widespread signal outages on the 5G network. Upon detecting the broadband interference, the base station immediately activated adaptive frequency hopping mode. First, the base station scanned the 5G frequency band to generate an interference heat map. High-intensity interference was detected in multiple frequency bands. Based on the interference intensity, the initial frequency hopping period was shortened from 10ms to 5ms. Next, a modified Tent mapping algorithm was used to generate a frequency hopping sequence, dynamically switching frequencies within the remaining available frequency band. Simultaneously, neighboring base stations formed a consortium network using blockchain and recorded their frequency hopping parameters on the blockchain. For example, if base station A discovered that its frequency band partially overlapped with base station B, it would automatically adjust the frequency hopping sequence through a smart contract to avoid frequency conflict. Ultimately, the convention center's 5G network quickly regained stability, supporting tens of thousands of people simultaneously enjoying high-traffic services such as HD live streaming and VR experiences.

[0043] It can be understood that the embodiment of the present application provides an efficient, secure and collaborative solution for 5G network broadband interference suppression through the deep integration of dynamic frequency hopping, chaotic sequence generation and blockchain collaboration. Adaptive frequency hopping dynamically adjusts the period according to the interference heat map, which can flexibly avoid high-intensity interference frequency bands. Compared with fixed-period frequency hopping, the interference avoidance success rate is increased by more than 40%; the frequency hopping sequence generated by the improved Tent mapping has good randomness and ergodicity, can make full use of available frequency band resources, and reduce the risk of being tracked by interference. The introduction of blockchain technology realizes the decentralized and trusted interaction of frequency hopping information between base stations. Smart contracts automatically execute interference avoidance strategies, reduce frequency conflicts between base stations, and reduce signaling overhead by 70% compared with traditional centralized collaborative mechanisms. At the same time, it improves the security and anti-attack capabilities of the network.

[0044] In the embodiment of the present application, if the interference type is directional interference, it is suppressed according to beam space filtering. The specific steps are as follows: F1. Precisely locate the interference direction: The MUSIC algorithm is used again to estimate the arrival angle of the interference signal with higher resolution, with an accuracy of less than 1°, to determine the specific incident direction of the interference.

[0045] F2. Beamforming matrix design: Design the beamforming matrix based on the base station antenna array structure and the specific incident direction of the interference. , so that the matrix forms a null in the interference direction and maintains gain in the target signal direction. The null constraint optimization formula is: , where w is the beamforming vector, is the beam space steering vector.

[0046] F3, signal weighting processing: Multiply the received signal with the designed beamforming matrix W, perform weighted summation on the signals received by different antennas, enhance the target signal, suppress interference signals from specific directions, and effectively suppress directional interference.

[0047] Specifically, a 5G base station deployed in an urban area with dense high-rise buildings was frequently subjected to interference from the southwest for a period of time, causing the uplink rate for users in that sector to drop to 30% of the normal level. Using the MUSIC algorithm to estimate the angle of arrival of the interference signal, the interference direction was determined to be 15° west-southwest with an accuracy of 0.8°. The base station designed a beamforming matrix based on the antenna array (a 64-element uniform linear array), creating a -35dB null at 15°. This also enhanced the signal gain in the northeast direction (where users are primarily concentrated). As a result, the average uplink rate in that sector returned to 85Mbps (close to the normal level of 90Mbps), and the bit error rate caused by interference dropped from 0.12 to 0.005.

[0048] It can be understood that the embodiment of the present application uses beam spatial filtering to achieve precise positioning and directional suppression of the interference direction, and achieves arrival angle estimation accuracy within 1° through high-resolution direction finding of the MUSIC algorithm, ensuring accurate capture of the interference source direction; the beamforming matrix forms a deep null in the interference direction while retaining the target signal gain, avoiding attenuation of useful signals.

[0049] In step A4, a bee swarm algorithm is used to receive the operating status data, and the objective function of the bee swarm algorithm is set according to the current network demand and anti-interference target. The bee behavior is simulated to iteratively search and determine the local optimal resource scheduling strategy.

[0050] It is understandable that the embodiments of the present application use a swarm algorithm for resource scheduling, which can efficiently process the complex and changeable operating status data in the 5G network and dynamically balance anti-interference and network performance requirements. By simulating the swarm intelligence behavior of bees foraging, it is possible to quickly find the local optimal solution in the vast resource scheduling solution space and avoid falling into the local minimum. By flexibly setting the objective function, multi-dimensional indicators such as throughput, latency, and interference suppression ratio can be included in the optimization range to achieve refined resource allocation.

[0051] In the embodiment of the present application, a bee swarm algorithm is used to receive operating status data. The objective function of the bee swarm algorithm is set according to the current network requirements and anti-interference goals. The bee behavior is simulated to iteratively search and determine the local optimal resource scheduling strategy. The specific steps are as follows: G1. Objective function definition: The objective function of the swarm algorithm is set according to the current network requirements and anti-interference goals. Let the throughput be T, the delay be D, and the interference intensity be I. The objective function is ,in, 、 、 is the weight coefficient, and .

[0052] G2. Initialize the bee colony: Randomly generate N individual worker bees. Each individual represents a resource scheduling strategy, forming the initial solution space. The resource scheduling strategy includes parameters such as channel allocation, power control, and beamforming.

[0053] G3, Iterative Search and Local Optimization: This method simulates the foraging behavior of bees, where leader, follower, and scout bees collaborate to perform iterative search. After multiple rounds of iterations, when the objective function value no longer increases significantly or reaches the preset number of iterations, the currently optimal resource scheduling strategy is output as the local optimal solution.

[0054] Specifically, in a certain port's intelligent operation area, a large number of automated equipment exchange data via the 5G network. At the same time, directional interference generated by equipment such as port radars severely impacts communication quality. After the base station collects network operation data, it initiates a swarm optimization algorithm. Given the latency sensitivity of automated equipment, the latency weight is set to 0.6, the throughput weight to 0.2, and the interference intensity weight to 0.2. 40 resource scheduling strategy solutions were initially generated. After 25 iterations of the swarm algorithm, the final strategy was determined: reallocate anti-interference channels to equipment in the interfered area, reduce base station transmit power to minimize signal reflection interference, and adjust beamforming direction to avoid interference sources. After implementation, communication latency for equipment in the operation area decreased from 60ms to 20ms, and the data transmission success rate increased from 75% to 98%, ensuring the efficient and stable operation of the port's automated operations.

[0055] It is understandable that the embodiment of the present application uses a swarm algorithm to simulate the collaborative mechanism of a bee colony, quantifies the operating status of the 5G network into an optimizable objective function, and realizes intelligent dynamic optimization of resource scheduling strategies. By customizing weight coefficients, key indicators such as throughput, latency, and interference intensity can be flexibly balanced to meet network requirements in different scenarios. The randomly initialized swarm provides a rich set of strategy samples, and the collaborative search of leading bees, following bees, and scout bees can effectively avoid falling into local optimal solutions and quickly find efficient solutions for resource allocation under complex interference and variable traffic loads.

[0056] In step A5, the resource scheduling strategy is executed and the execution effect is fed back. If the execution effect does not meet the set expectations, the bee swarm algorithm is restarted to determine a new resource scheduling strategy.

[0057] It is understood that the embodiments of this application can quickly identify policy deviations and trigger algorithm re-optimization by real-time monitoring of execution results, thus avoiding performance degradation caused by environmental changes. The combination of the feedback mechanism and the swarm algorithm forms a closed loop of "monitoring-optimization-execution-re-optimization", enabling the network to continuously approach the optimal state in complex scenarios, improving resource utilization while enhancing the timeliness and reliability of the anti-interference strategy.

[0058] In the embodiment of the present application, the resource scheduling strategy is executed and the execution effect is fed back. If the execution effect does not meet the set expectations, the bee swarm algorithm is restarted to determine a new resource scheduling strategy, including: The network control center encapsulates the optimal strategy parameters (channel allocation scheme, power control value, beamforming weight) determined by the swarm algorithm into control signaling and sends it to each device through the downlink control channel. The device adjusts the hardware configuration in real time based on the received parameters. For example, the terminal adjusts the uplink transmit power and modulation coding method, the base station adjusts the phase and amplitude weights of the antenna array, and dynamically allocates physical resource blocks and time-frequency resources.

[0059] The device collects key performance indicators in real time and reports the monitoring results every 20ms. The deviation between the actual performance and the expected target is calculated based on the evaluation function. If the deviation is greater than the target threshold, it is determined that the performance has not met the expectations and the swarm algorithm is restarted. ,in, is the weight of the i-th indicator, is the actual value, is the target value.

[0060] The bee swarm algorithm adjusts the parameters of the bee swarm algorithm according to the last execution result, initializes the bee swarm with the historical optimal solution, and accelerates convergence.

[0061] Specifically, during peak hours at a certain transportation hub, a dramatic increase in passenger traffic led to a surge in 5G network load. The network control center used a swarm algorithm to determine the optimal strategy parameters, such as channel allocation and power control, and encapsulated and distributed them. Base stations adjusted the antenna array phase and amplitude weightings accordingly to enhance coverage, and terminal devices adjusted transmit power and modulation coding to ensure data transmission. Devices reported latency, throughput, and other metrics every 20ms. An evaluation revealed that the actual average latency in the waiting hall area reached 80ms, exceeding the target of 50ms, exceeding the threshold. The network control center immediately restarted the swarm algorithm, reducing the search step size, increasing the number of iterations, and using the historical optimal solution for initialization to quickly generate a new strategy. After implementation, the average latency in the waiting hall area dropped to 45ms, effectively resolving network lag issues and ensuring smooth access to services like mobile payment and real-time navigation for passengers.

[0062] It is understandable that the embodiment of the present application realizes intelligent closed-loop management through policy issuance, intelligent device response, real-time data feedback, and algorithm iterative optimization, which greatly enhances the dynamic adaptive capability of the 5G network. The network control center uniformly encapsulates the optimal policy parameters obtained by the swarm algorithm into control signaling and issues it, realizing efficient coordination of resource scheduling of equipment across the entire network; the equipment reports key performance indicators at a high frequency of 20ms, and cooperates with the weighted evaluation function to accurately capture network operation deviations, making anomaly detection more sensitive and accurate. When the deviation exceeds the threshold, the swarm algorithm quickly restarts, and dynamically adjusts parameters based on historical execution effects, and initializes with the help of historical optimal solutions, so that the algorithm convergence speed is increased by more than 50%, and can quickly generate new strategies to respond to network changes, significantly improving network stability and resource utilization efficiency, and ensuring that the 5G network always maintains high-performance operation.

[0063] The following will describe a 5G network intelligent optimization method in detail through a specific embodiment. Figure 4 As shown, the content is as follows: A technology expo was held in a convention and exhibition center, attracting tens of thousands of visitors. At the same time, a large number of IoT devices, live broadcast equipment and smart service terminals were deployed, putting tremendous pressure on the 5G network.

[0064] After the exhibition opened, the 5G base station's RF front-end continuously collected time-domain signal sequences, capturing hundreds of megabytes of data per second. The system converted the collected signals into frequency-domain spectra using a Fast Fourier Transform (FFT). Comparing these spectra with a pre-set 5G standard signal spectrum template, the system discovered significant abnormalities in the power spectral density of the 3.55-3.6 GHz band, far exceeding the normal range. The system immediately filtered and extracted this region, successfully isolating the interfering signal and confirming the presence of high-intensity interference in this frequency band.

[0065] For the separated interference signal, the system extracts multi-dimensional features in the time domain, frequency domain, and spatial domain, splices these features into a comprehensive feature vector, and inputs it into the trained random forest algorithm model. After analysis, it is determined that the interference is broadband interference.

[0066] To combat broadband interference, the system implements anti-interference measures, activates adaptive frequency hopping mode, scans the entire 5G operating frequency band to generate an interference heat map, shortens the initial frequency hopping period from 10ms to 5ms, and uses improved Tent mapping to generate the frequency hopping sequence, dynamically switching frequencies within the remaining available frequency band. Furthermore, neighboring base stations form a consortium chain network through blockchain, write their respective frequency hopping parameters to the blockchain, and automatically adjust the frequency hopping sequence through smart contracts to avoid frequency conflicts between base stations.

[0067] After anti-interference measures are activated, real-time operational status data is collected, including metrics such as throughput, latency, and interference intensity in each area. A bee colony algorithm is used to optimize resource scheduling. Fifty artificial bees are randomly generated, each representing a resource scheduling strategy. Simulating bee foraging behavior, 30 rounds of iterative search determine the locally optimal resource scheduling strategy. For example, this strategy dynamically allocates more physical resource blocks to high-traffic areas and adjusts base station transmit power to balance coverage and interference mitigation.

[0068] The network control center encapsulates the optimal strategy parameters as control signaling and sends it to each device via the downlink control channel. The terminal devices adjust the uplink transmit power and modulation coding method to improve data transmission efficiency. Every 20ms, the devices report key performance indicators, and the system calculates the deviation between actual performance and the expected target based on an evaluation function. After 10 minutes of operation, it was found that the actual average latency in some live broadcast areas was 80ms, while the target was 50ms, exceeding the threshold. The system immediately restarted the swarm algorithm, combined the newly collected data with the previous execution results, adjusted the swarm algorithm parameters, initialized the swarm using the historical optimal solution, and quickly generated a new resource scheduling strategy. After the new strategy was implemented, the average latency in the live broadcast area dropped to 45ms, effectively resolving network lag issues and ensuring the smooth operation of high-definition live broadcasts, real-time interactive services, and other services during the exhibition.

[0069] In summary, the embodiment of the present application realizes automation and intelligent collaboration in the entire process, from interference detection to type identification, and then to targeted anti-interference and resource scheduling. The interference frequency band is quickly locked based on FFT and spectrum template comparison, and the interference type is accurately classified through the random forest algorithm; for broadband interference, adaptive frequency hopping is used to effectively suppress interference; the swarm algorithm is used to dynamically optimize resource scheduling, combined with a real-time feedback mechanism, and the strategy is flexibly adjusted according to network needs. The entire solution ensures the stable operation of the 5G network in a high-concurrency and strong interference environment during the exhibition, and ensures the smooth development of key businesses such as high-definition live broadcast and mobile payment.

[0070] Next, a 5G network intelligent optimization system proposed according to an embodiment of the present application is described with reference to the accompanying drawings.

[0071] Figure 5 This is a schematic diagram of the composition of a 5G network intelligent optimization system according to an embodiment of the present application.

[0072] like Figure 5 As shown, the 5G network intelligent optimization system 10 includes: a 5G interference data separation module 100, an interference type identification module 200, an anti-interference module 300, and a data scheduling module 400.

[0073] Among them, the 5G interference data separation module 100 is used to obtain 5G network signal data, use fast Fourier transform (FFT) to convert the time domain signal into a frequency domain spectrum, compare it with the spectrum template of the 5G standard signal, identify the abnormal power spectrum density area, filter and extract the abnormal power spectrum density area, and separate the interference signal.

[0074] The interference type identification module 200 is used to extract features from the obtained interference signal and identify the interference type based on the random forest algorithm. The interference types include narrowband interference, broadband interference, and directional interference. The anti-interference module 300 is used to activate corresponding anti-interference measures according to the interference type and generate operating status data in real time. If the interference type is narrowband interference, a chaotic coding sequence is generated based on the improved logistic map to increase the spread spectrum gain, and the chaotic system parameters are updated in real time through quantum key distribution technology. If the interference type is broadband interference, an adaptive frequency hopping mode is activated, the frequency hopping period is dynamically adjusted according to the interference situation, and a frequency hopping coordination mechanism is established with adjacent base stations. The blockchain technology is used to ensure the trustworthy interaction of frequency hopping information. If the interference type is directional interference, it is suppressed based on beam space filtering. The data scheduling module 400 is used to receive operating status data using a swarm algorithm, set the objective function of the swarm algorithm according to current network requirements and anti-interference goals, simulate the iterative search of bee behavior, and determine the local optimal resource scheduling strategy; the device executes the resource scheduling strategy and provides feedback on the execution effect. If the execution effect does not meet the set expectations, the swarm algorithm is restarted to determine a new resource scheduling strategy.

[0075] A 5G network intelligent optimization system will be specifically described below through a specific embodiment, as follows: In a modern smart industrial park, numerous industrial automation equipment, smart sensors, and high-definition surveillance cameras rely on 5G networks for data transmission and interaction. One weekday morning, the park's 5G network suddenly experienced severe packet loss and device communication interruptions.

[0076] The 5G base station radio frequency front-end within the campus collected time-domain signal data thousands of times per second. Using a Fast Fourier Transform (FFT), the signal was converted to a frequency-domain spectrum and compared with the standard 5G signal spectrum template. This revealed an abnormally high power spectral density region in the 28 GHz band. The system immediately filtered and extracted this region, successfully isolating the interference signal.

[0077] For the separated interference signal, the module extracts features from multiple aspects such as time domain, frequency domain and spatial domain, splices the features into a comprehensive feature vector and inputs it into the random forest algorithm model. After analysis, it is determined that the interference is directional interference.

[0078] After determining that the interference was directional, anti-interference measures were implemented. Using beam spatial filtering, a beamforming matrix was designed for the interference incident direction (the southeast corner of the campus), creating a -35dB null in that direction to ensure that the gain in the target signal direction was not affected.

[0079] After anti-interference measures were activated, operating status data was collected in real time. The swarm algorithm's objective function was set based on the industrial park's requirements for stable equipment communication and low data transmission latency. Thirty artificial bees were randomly generated, each representing a resource scheduling strategy encompassing parameters such as channel allocation, power control, and beamforming. After 20 rounds of iterative search, simulating bee foraging behavior, a local optimal strategy was identified: dedicated channels were allocated to key industrial control equipment, base station transmit power was adjusted to balance coverage and interference suppression, and beamforming parameters were optimized to enhance signal directivity. After implementing this strategy, the system continuously monitored and confirmed that communication latency met the target. Communication stability within the industrial park was restored, with the packet loss rate dropping from 15% to below 1% and the average latency dropping from 80ms to 20ms, ensuring the normal operation of automated production within the industrial park.

[0080] In summary, the embodiments of this application use FFT to convert the signal into a frequency-domain spectrum and compare it with a spectrum template, accurately locking the interference frequency band and laying a solid foundation for subsequent processing. Based on multi-dimensional features and the random forest algorithm, it can quickly and accurately determine the type of interference. For directional interference, beam space filtering is used for efficient suppression. With the help of the bee swarm algorithm, resource scheduling strategies are dynamically optimized and continuously iterated through a feedback mechanism. In industrial park scenarios, the entire system not only quickly resolves network anomalies, reduces data packet loss rates, and reduces latency, but also ensures the stable operation of automated production.

[0081] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include: A memory 601 , a processor 602 , and a computer program stored in the memory 601 and executable on the processor 602 .

[0082] When the processor 602 executes the program, a 5G network intelligent optimization method provided in the above embodiment is implemented.

[0083] Furthermore, the electronic device further includes: The communication interface 603 is used for communication between the memory 601 and the processor 602 .

[0084] The memory 601 is used to store computer programs that can be run on the processor 602 .

[0085] The memory 601 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0086] If the memory 601, processor 602, and communication interface 603 are implemented independently, the communication interface 603, memory 601, and processor 602 can be connected to each other via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0087] Optionally, in a specific implementation, if the memory 601, the processor 602 and the communication interface 603 are integrated on a chip, the memory 601, the processor 602 and the communication interface 603 can communicate with each other through an internal interface.

[0088] The processor 602 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0089] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0090] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Thus, a feature specified as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of this application, "N" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0091] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0092] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiment, the N steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logical functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0093] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

Claims

1. A 5G network intelligent optimization method, characterized in that: The following steps are involved: A1: Acquire 5G network signal data, convert the time-domain signal into a frequency-domain spectrum using a fast Fourier transform (FFT), compare the spectrum with the 5G standard signal's spectrum template, identify areas of abnormal power spectrum density, and perform filtering and extraction on these areas to isolate the interference signal. A2: Extracting features from the obtained interference signal and using the features to identify interference types using a random forest algorithm. The interference types include narrowband interference, broadband interference, and directional interference. A3: Activate corresponding anti-interference measures based on the interference type and generate operating status data in real time. If the interference type is narrowband interference, generate a chaotic coding sequence based on the improved logistic map to increase the spread spectrum gain, and update the chaotic system parameters in real time through quantum key distribution technology. If the interference type is broadband interference, activate the adaptive frequency hopping mode, dynamically adjust the frequency hopping period according to the interference situation, establish a frequency hopping coordination mechanism with adjacent base stations, and ensure the trusted interaction of frequency hopping information through blockchain technology. If the interference type is directional interference, suppress it based on beam space filtering. A4: Using a bee swarm algorithm to receive the operating status data, setting an objective function of the bee swarm algorithm based on current network requirements and anti-interference goals, simulating bee behavior to iteratively search and determine a locally optimal resource scheduling strategy; A5: Execute the resource scheduling strategy and provide feedback on the execution effect. If the execution effect does not meet the set expectations, restart the bee swarm algorithm to determine a new resource scheduling strategy.

2. A 5G network intelligent optimization method according to claim 1, characterized in that: Obtain 5G network signal data, use Fast Fourier Transform (FFT) to convert the time domain signal into a frequency domain spectrum, compare it with the spectrum template of the 5G standard signal, identify abnormal power spectrum density areas, filter and extract the abnormal power spectrum density areas, and separate the interference signal. The specific steps are as follows: B1. Collect a time domain signal sequence through the RF front end of the 5G base station, preprocess the time domain signal sequence, limit the bandwidth to the 5G operating frequency band, and obtain target signal data x(f); B2. Convert the target signal data into a frequency domain representation by fast Fourier transform (FFT), calculate the power spectrum density based on the transformed result, and obtain a power spectrum estimate. The specific formula of Fourier transform is: , , where N is the number of FFT points and k is the index of the discrete signal in the frequency domain. is a complex exponential function that weights the time domain signal; the power spectrum estimation formula is: ; B3. Construct a 5G standard spectrum template and calculate the deviation between the power spectrum estimate and the template spectrum ,in, is a template spectrum, and if the deviation is greater than the target threshold, the region is determined to be abnormal; B4. Design of notch filter , filter and extract the abnormal area, then convert the filtered frequency domain signal back to the time domain according to the inverse FFT method, reconstruct the interference signal, and obtain the separated interference signal, where the filter extraction formula is: , the inverse FFT formula is .

3. A 5G network intelligent optimization method according to claim 1, characterized in that: Extract features from the obtained interference signal and use the features to identify interference types using a random forest algorithm. The interference types include narrowband interference, broadband interference, and directional interference. The specific steps are as follows: C1, the interference signal After preprocessing, extract the signal's time domain, frequency domain, time-frequency domain, and spatial domain features respectively. Concatenate the extracted feature vectors into a multidimensional feature vector. Through feature importance analysis, select the M most discriminative features to form a comprehensive feature vector x. The time domain features include signal variance, duty cycle, and crest factor; the frequency domain features include center frequency, spectral entropy, and normalized bandwidth; the time-frequency domain includes a time-frequency graph; and the spatial domain features are the angle of arrival (AoA) of the interference signal estimated using the MUSIC algorithm using the base station's multi-antenna array. C2. Input the comprehensive feature vector x into the trained random forest classification model and output the classification results, which include narrowband interference, broadband interference, and directional interference. The classification formula is: , where c is the value of the interference type, T is the number of decision trees, is the classification result of the t-th tree for sample x, .

4. A 5G network intelligent optimization method according to claim 1, characterized in that: If the interference type is narrowband interference, a chaotic coding sequence is generated according to the improved Logistic map to increase the spread spectrum gain, and the chaotic system parameters are updated in real time through quantum key distribution technology. The specific steps are as follows: D1. Transmit polarized photons through quantum channels to generate the original key , the original key Divided into three parts, mapped to the initial value of the chaotic system , control parameter r and disturbance coefficient k, each Re-execute QKD in milliseconds and update parameters synchronously to ensure that the key space complexity reaches above; D2. Use improved Logistic mapping to iteratively generate chaotic sequences , where the generation formula is: ,in, is the chaos control parameter, is the perturbation coefficient, the initial value Randomly generated by quantum key distribution, the chaotic sequence Take the first N values and perform binarization processing, mapping the continuous values into bipolar code elements c(n); D3, multiply the data signal d(t) by the chaotic sequence c(n) to realize direct sequence spread spectrum. The specific formula is: , where t is the time variable, representing the state of the signal at different times, is the pulse shaping function, is the chip period.

5. A 5G network intelligent optimization method according to claim 1, characterized in that: If the interference type is broadband interference, the adaptive frequency hopping mode is activated, the frequency hopping period is dynamically adjusted according to the interference situation, and a frequency hopping coordination mechanism is established with adjacent base stations. The blockchain technology is used to ensure the trustworthy interaction of frequency hopping information. The specific steps are as follows: E1. Calculate the frequency modulation period: Scan the entire 5G operating frequency band and generate an interference heat map , filter the frequency bands whose interference power does not exceed the threshold, obtain the available frequency band A, and dynamically adjust the initial frequency hopping period according to the interference intensity , ,in, is the base period, is the adjustment factor, B is the total bandwidth; E2. Generate frequency hopping sequence and frequency selection: Generate frequency hopping sequence using improved Tent mapping , the chaotic sequence Mapped to the available frequency band A, the mapping function is ,in is the lower limit of the available frequency band, is the frequency interval; E3. Blockchain collaboration: Adjacent base stations form a consortium chain network. Each base station acts as a node and runs a blockchain client. Each base station encapsulates the current frequency hopping parameters (frequency hopping sequence, frequency hopping period, available frequency band) as a transaction and writes it into the blockchain through the PBFT consensus algorithm. Each base station queries the frequency hopping information of the neighboring station through the smart contract and executes the interference avoidance algorithm. The specific formula is: ,in, is the weight between base stations, and Overlap is the frequency band overlap.

6. A 5G network intelligent optimization method according to claim 1, characterized in that: If the interference type is directional interference, it is suppressed according to beam space filtering. The specific steps are as follows: F1. Accurately locate the interference direction: The MUSIC algorithm is used again to estimate the arrival angle of the interference signal with higher resolution, with an accuracy of less than 1°, to determine the specific incident direction of the interference. F2. Beamforming matrix design: Design the beamforming matrix based on the base station antenna array structure and the specific incident direction of the interference. , so that the matrix forms a null in the interference direction and maintains gain in the target signal direction. The null constraint optimization formula is: , where w is the beamforming vector, is the beam space steering vector; F3, signal weighting processing: Multiply the received signal with the designed beamforming matrix W, perform weighted summation on the signals received by different antennas, enhance the target signal, suppress interference signals from specific directions, and effectively suppress directional interference.

7. A 5G network intelligent optimization method according to claim 1, characterized in that: The bee swarm algorithm is used to receive the operating status data, and the objective function of the bee swarm algorithm is set according to the current network demand and anti-interference target. The bee behavior is simulated to iteratively search and determine the local optimal resource scheduling strategy. The specific steps are as follows: G1. Objective function definition: The objective function of the swarm algorithm is set according to the current network requirements and anti-interference goals. Let the throughput be T, the delay be D, and the interference intensity be I. The objective function is ,in, 、 、 is the weight coefficient, and ; G2. Initialize the bee colony: Randomly generate N individual worker bees, each of which represents a resource scheduling strategy, forming the initial solution space. The resource scheduling strategy includes parameters such as channel allocation, power control, and beamforming. G3, Iterative Search and Local Optimization: This method simulates the foraging behavior of bees, where leader, follower, and scout bees collaborate to perform iterative search. After multiple rounds of iterations, when the objective function value no longer increases significantly or reaches the preset number of iterations, the currently optimal resource scheduling strategy is output as the local optimal solution.

8. A 5G network intelligent optimization method according to claim 1, characterized in that: Execute the resource scheduling strategy and provide feedback on the execution effect. If the execution effect does not meet the set expectations, restart the bee swarm algorithm to determine a new resource scheduling strategy, including: The network control center encapsulates the optimal strategy parameters (channel allocation scheme, power control value, beamforming weight) determined by the swarm algorithm into control signaling and sends it to each device via the downlink control channel. The device then adjusts its hardware configuration in real time based on the received parameters. For example, the terminal adjusts the uplink transmit power and modulation coding scheme, the base station adjusts the phase and amplitude weights of the antenna array, and dynamically allocates physical resource blocks and time-frequency resources. The device collects key performance indicators in real time and reports the monitoring results every 20ms. The deviation between the actual performance and the expected target is calculated based on the evaluation function. If the deviation is greater than the target threshold, it is determined that the performance has not met the expectations and the swarm algorithm is restarted. ,in, is the weight of the i-th indicator, is the actual value, is the target value; The bee swarm algorithm adjusts the parameters of the bee swarm algorithm according to the last execution result, initializes the bee swarm with the historical optimal solution, and accelerates convergence.

9. A 5G network intelligent optimization system, characterized in that: Including: 5G interference data separation module, interference type identification module, anti-interference module, data scheduling module, among which, The 5G interference data separation module is used to obtain 5G network signal data, convert the time domain signal into a frequency domain spectrum using a fast Fourier transform (FFT), compare the spectrum template of the 5G standard signal, identify abnormal power spectrum density areas, filter and extract the abnormal power spectrum density areas, and separate the interference signal; The interference type identification module is used to extract features from the obtained interference signal and identify the interference type based on the features according to the random forest algorithm. The interference types include narrowband interference, broadband interference, and directional interference; The anti-interference module is used to activate corresponding anti-interference measures according to the interference type and generate operating status data in real time. If the interference type is narrowband interference, a chaotic coding sequence is generated according to the improved Logistic map to increase the spread spectrum gain, and the chaotic system parameters are updated in real time through quantum key distribution technology; if the interference type is broadband interference, an adaptive frequency hopping mode is activated, the frequency hopping period is dynamically adjusted according to the interference situation, and a frequency hopping coordination mechanism is established with adjacent base stations to ensure the trusted interaction of frequency hopping information through blockchain technology; if the interference type is directional interference, it is suppressed according to beam space filtering; The data scheduling module is used to receive the operating status data using a bee swarm algorithm, set the objective function of the bee swarm algorithm according to the current network demand and anti-interference target, simulate the bee behavior iterative search, and determine the local optimal resource scheduling strategy; the device executes the resource scheduling strategy and provides feedback on the execution effect. If the execution effect does not meet the set expectations, the bee swarm algorithm is restarted to determine a new resource scheduling strategy.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a 5G network intelligent optimization method as described in any one of claims 1 to 8.

Images