Self-adaptive spectrum management method based on intelligent gateway

Abstract

The invention provides a self-adaptive spectrum management method based on an intelligent gateway, and the intelligent gateway comprises an interference monitoring module, an equipment demand analysis module, a self-adaptive spectrum allocation module, a performance optimization and feedback module and the like. By combining a comprehensive solution of interference monitoring, equipment demand analysis and adaptive spectrum allocation, the problem of dynamic allocation of spectrum resources in an intelligent home environment is effectively solved, and the stability and communication performance of network coverage are improved. In the adaptive spectrum allocation, a dynamic weight adjustment strategy is adopted, the weight of each frequency band is dynamically adjusted according to the current network load and interference condition, and the frequency band with relatively small interference is preferentially allocated to high-priority equipment, so that the coverage and performance of the smart home network are optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless communications, and in particular to an adaptive spectrum management method based on an intelligent gateway. Background Art

[0002] With the popularity of smart home devices, the home network environment is becoming increasingly complex. Multiple devices such as set-top boxes, smart speakers, security cameras, thermostats, etc. rely on wireless signals to communicate.

[0003] In the prior art, the allocation of spectrum resources is usually static. This allocation method is based on a fixed frequency range and channel planning. Each device uses a specific frequency band for communication according to predetermined rules. Usually, this allocation method relies on the manual configuration of the network administrator, or is preset when the device leaves the factory. It often cannot effectively deal with the dynamic changes in wireless signal interference and device density in the home environment, resulting in a series of technical problems, such as low spectrum resource utilization. Static allocation may cause some frequency bands to be idle for a long time, while other frequency bands are overcrowded; the interference problem is serious. Since the frequency band is fixed, once a certain frequency band is interfered with, the affected device will not be able to communicate normally; and the scalability is poor. With the increase in smart home devices, the static allocation method is difficult to adapt to the dynamic changes in device density. In addition, it also leads to unstable network coverage. Due to unreasonable spectrum resource allocation, the network coverage in some areas may be insufficient and the communication quality may be reduced. Summary of the invention

[0004] In view of this, in order to solve the technical problems existing in the technical background such as low spectrum resource utilization, serious interference problems, poor scalability, unstable network coverage, etc., the present invention proposes an adaptive spectrum management method based on an intelligent gateway. The specific technical solution is as follows:

[0005] An adaptive spectrum management method based on an intelligent gateway comprises the following steps in sequence: an interference monitoring module, an equipment demand analysis module, an adaptive spectrum allocation module, and a performance optimization and feedback module;

[0006] The interference monitoring module is used to collect and analyze the signal interference in the home wireless environment in real time, and use the signal processing algorithm to perform spectrum analysis on the collected signals to identify the main interference sources and their frequency bands;

[0007] The device demand analysis module interacts with the communication protocols of each smart home device (such as Zigbee, Z-Wave, Wi-Fi, etc.) to collect the connection requirements of the devices, and then analyzes the communication priority, data transmission volume and real-time requirements of each device to provide a basis for spectrum allocation.

[0008] The adaptive spectrum allocation module runs an adaptive spectrum allocation algorithm in combination with the interference monitoring results and the device demand analysis structure; adopts a dynamic weight adjustment strategy to dynamically adjust the weight of each frequency band according to the current network load and interference situation, and preferentially allocates frequency bands with less interference to high-priority devices; at the same time, the algorithm adopts a spectrum sharing mechanism to allow low-priority devices to share the frequency band resources of high-priority devices while ensuring the communication quality of high-priority devices;

[0009] The performance optimization and feedback module continuously monitors network performance indicators (such as throughput, delay, packet loss rate, etc.) and evaluates the effectiveness of the spectrum allocation plan. Based on the performance monitoring results, the algorithm can self-learn and optimize and adjust parameters to adapt to the ever-changing network environment.

[0010] The interference monitoring module mainly adopts the following steps when identifying the main interference sources and their frequency bands:

[0011] Step 1.1 Signal collection: Use the built-in wireless signal receiver of the smart gateway to continuously collect signals in the home wireless environment at set time intervals.

[0012] The acquired signals are stored in a local buffer for subsequent processing.

[0013] Step 1.2 Spectrum analysis: Apply the fast Fourier transform (FFT) algorithm to the collected signal to convert the time domain signal into the frequency domain signal. The formula of FFT is:

[0014]

[0015] Where x[n] is the time domain signal sample, X[k] is the corresponding frequency domain signal sample, N is the total number of samples, and k is the frequency domain index.

[0016] Through FFT transformation, the energy distribution of the signal at different frequencies can be obtained.

[0017] Step 1.3: Interference source identification: Analyze the frequency domain signal after FFT transformation to find the frequency points or frequency bands with abnormally high energy. Set an energy threshold, and the frequency points or frequency bands exceeding this threshold are considered potential interference sources. Through further signal processing (such as filtering, demodulation, etc.), try to identify the specific type and source of the interference source.

[0018] Step 1.4: Interference frequency band determination: Based on the results of interference source identification, the interfered frequency bands are determined. These frequency bands will be temporarily excluded from the communication frequency bands of smart home devices to reduce interference.

[0019] In one embodiment of the present invention, the strategy of dynamically adjusting the weight of each frequency band by the adaptive spectrum allocation module mainly adopts the following steps:

[0020] Step 3.1: Weight initialization: In the algorithm initialization phase, an initial weight is set for each frequency band based on historical data and preset rules. Step 3.2. Real-time monitoring and data collection: The smart gateway continuously monitors the signal interference in the home wireless environment, including the interference intensity of each frequency band, channel utilization, etc. At the same time, it collects the connection status, communication requirements and priority information of each smart home device;

[0021] Step 3.3. Dynamic weight adjustment: Based on real-time monitoring data, a specific algorithm is used to dynamically adjust the weight of each frequency band. The weight adjustment takes into account factors such as current network load, interference level, device priority, and historical allocation.

[0022] Step 3.4. Spectrum allocation decision: Based on the adjusted weights, the smart gateway makes spectrum allocation decisions, giving priority to allocating frequency bands with less interference to high-priority devices. The algorithm for dynamically adjusting the weights of each frequency band is shown below:

[0023]

[0024] Where: W f represents the weight of frequency band f,

[0025] I f Indicates the interference level of frequency band f,

[0026] L f Indicates the current network load of frequency band f

[0027] P d Indicates the priority of device d.

[0028] and Represent the weights before and after adjustment respectively.

[0029] ∈ is a minimum value used to avoid the denominator being zero.

[0030] β, γ and δ are weight adjustment coefficients, which are set according to actual needs.

[0031] δ df is an indicator function that takes the value 1 when device d is assigned to frequency band f, and 0 otherwise.

[0032] In one embodiment of the present invention, the performance optimization and feedback module is responsible for continuously monitoring network performance indicators and evaluating the effectiveness of the spectrum allocation scheme based on these indicators. Through self-learning and optimization mechanisms, the module can adjust algorithm parameters to adapt to the ever-changing network environment. The specific steps in its operation are as follows:

[0033] Steps: 4.1. Performance indicator monitoring: The intelligent gateway regularly collects and analyzes network performance indicators (such as throughput, latency, packet loss rate, etc.), which can be obtained through communication protocols between devices or specialized network performance monitoring tools;

[0034] Step 4.2. Scheme effectiveness evaluation: Compare the collected performance indicators with the preset target values ​​or historical data to evaluate the effectiveness of the current spectrum allocation scheme. If the performance indicators deviate from the target values ​​or show a downward trend, the optimization process is triggered.

[0035] Step 4.3. Self-learning and optimization: The algorithm uses machine learning or reinforcement learning methods to analyze the deficiencies of the spectrum allocation scheme based on the performance monitoring results, and optimizes the spectrum allocation strategy by adjusting parameters such as the weight adjustment coefficient and the sharing condition threshold.

[0036] Step 4.4. Parameter adjustment and implementation: Based on the results of self-learning and optimization, the intelligent gateway dynamically adjusts the algorithm parameters, and the adjusted parameters are immediately applied to the spectrum allocation process to improve network performance.

[0037] Let P t represents the network performance index (such as throughput) at time t, P target represents the preset target performance indicator, α is the learning rate parameter, is the gradient of the performance index P with respect to the algorithm parameters. The optimization process can be simplified to an iterative update process based on gradient descent as follows:

[0038]

[0039] The specific steps are as follows:

[0040] Ⅰ. Initialize algorithm parameters.

[0041] II. Monitoring network performance index P t .

[0042] III. Calculate the difference ΔP between the performance index and the target value.

[0043] IV. According to the gradient And the learning rate α updates the algorithm parameters.

[0044] V. Repeat steps 2 to 4 until the performance index reaches a satisfactory level or the maximum number of iterations is reached.

[0045] The above technical solution has the following beneficial effects:

[0046] The intelligent gateway of the present invention includes steps such as an interference monitoring module, a device demand analysis module, an adaptive spectrum allocation module and a performance optimization and feedback module, which dynamically monitors the wireless signal interference in the home environment. The comprehensive solution combining interference monitoring, device demand analysis and adaptive spectrum allocation effectively solves the problem of dynamic allocation of spectrum resources in the smart home environment and improves the stability of network coverage and communication performance.

[0047] In adaptive spectrum allocation, a dynamic weight adjustment strategy is adopted to dynamically adjust the weight of each frequency band according to the current network load and interference situation, and give priority to allocating frequency bands with less interference to high-priority devices, thereby optimizing the coverage and performance of the smart home network;

[0048] The performance optimization and feedback module is responsible for continuously monitoring network performance indicators and evaluating the effectiveness of the spectrum allocation plan based on these indicators. At the same time, through self-learning and optimization mechanisms, the module can adjust the parameters in the spectrum allocation algorithm. Through such self-learning and optimization mechanisms, the smart gateway can continuously adapt to the changing network environment and ensure the stability and efficiency of the smart home network.

[0049] When identifying the main interference sources and their frequency bands, the interference monitoring module converts the time domain signal into a frequency domain signal through fast Fourier transform (FFT), so as to calculate the frequency points or frequency bands with abnormally high energy. By comparing with the preset energy threshold, the potential interference source can be determined. Through further signal processing (such as filtering, demodulation, etc.), the specific type and source of the interference source can be identified. According to the results of interference source identification, the interfered frequency band is determined. These frequency bands will be temporarily excluded from the communication frequency bands of smart home devices to reduce interference, thereby improving the stability and efficiency of spectrum allocation calculations. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a structural schematic diagram of a module of an adaptive spectrum management method based on an intelligent gateway of the present invention;

[0051] Figure 2 It is a control logic diagram of an adaptive spectrum allocation module of an adaptive spectrum management method based on an intelligent gateway of the present invention;

[0052] Figure 3 The present invention is a performance optimization and feedback module control logic diagram of an adaptive spectrum management method based on an intelligent gateway. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] An adaptive spectrum management method based on an intelligent gateway comprises the following steps in sequence: an interference monitoring module, an equipment demand analysis module, an adaptive spectrum allocation module, and a performance optimization and feedback module;

[0055] The interference monitoring module is used to collect and analyze the signal interference in the home wireless environment in real time, and use the signal processing algorithm to perform spectrum analysis on the collected signals to identify the main interference sources and their frequency bands;

[0056] The device demand analysis module interacts with the communication protocols of each smart home device (such as Zigbee, Z-Wave, Wi-Fi, etc.) to collect the connection requirements of the devices, and then analyzes the communication priority, data transmission volume and real-time requirements of each device to provide a basis for spectrum allocation.

[0057] The adaptive spectrum allocation module runs an adaptive spectrum allocation algorithm in combination with the interference monitoring results and the device demand analysis structure; adopts a dynamic weight adjustment strategy to dynamically adjust the weight of each frequency band according to the current network load and interference situation, and preferentially allocates frequency bands with less interference to high-priority devices; at the same time, the algorithm adopts a spectrum sharing mechanism to allow low-priority devices to share the frequency band resources of high-priority devices while ensuring the communication quality of high-priority devices;

[0058] The performance optimization and feedback module continuously monitors network performance indicators (such as throughput, delay, packet loss rate, etc.) and evaluates the effectiveness of the spectrum allocation plan. Based on the performance monitoring results, the algorithm can self-learn and optimize and adjust parameters to adapt to the ever-changing network environment.

[0059] The interference monitoring module mainly adopts the following steps when identifying the main interference sources and their frequency bands:

[0060] Step 1.1 Signal collection: Use the built-in wireless signal receiver of the smart gateway to continuously collect signals in the home wireless environment at set time intervals.

[0061] The acquired signals are stored in a local buffer for subsequent processing.

[0062] Step 1.2 Spectrum analysis: Apply the fast Fourier transform (FFT) algorithm to the collected signal to convert the time domain signal into the frequency domain signal. The formula of FFT is:

[0063]

[0064] Where x[n] is the time domain signal sample, X[k] is the corresponding frequency domain signal sample, N is the total number of samples, and k is the frequency domain index.

[0065] Through FFT transformation, the energy distribution of the signal at different frequencies can be obtained.

[0066] Step 1.3: Interference source identification: Analyze the frequency domain signal after FFT transformation to find the frequency points or frequency bands with abnormally high energy. Set an energy threshold, and the frequency points or frequency bands exceeding this threshold are considered potential interference sources. Through further signal processing (such as filtering, demodulation, etc.), try to identify the specific type and source of the interference source.

[0067] Step 1.4: Interference frequency band determination: Based on the results of interference source identification, the interfered frequency bands are determined. These frequency bands will be temporarily excluded from the communication frequency bands of smart home devices to reduce interference.

[0068] In one embodiment of the present invention, the strategy of dynamically adjusting the weight of each frequency band by the adaptive spectrum allocation module mainly adopts the following steps:

[0069] Step 3.1: Weight initialization: In the algorithm initialization phase, an initial weight is set for each frequency band based on historical data and preset rules. Step 3.2. Real-time monitoring and data collection: The smart gateway continuously monitors the signal interference in the home wireless environment, including the interference intensity of each frequency band, channel utilization, etc. At the same time, it collects the connection status, communication requirements and priority information of each smart home device;

[0070] Step 3.3. Dynamic weight adjustment: Based on real-time monitoring data, a specific algorithm is used to dynamically adjust the weight of each frequency band. The weight adjustment takes into account factors such as current network load, interference level, device priority, and historical allocation.

[0071] Step 3.4. Spectrum allocation decision: Based on the adjusted weights, the smart gateway makes spectrum allocation decisions, giving priority to allocating frequency bands with less interference to high-priority devices. The algorithm for dynamically adjusting the weights of each frequency band is shown below:

[0072]

[0073] Where: W f represents the weight of frequency band f,

[0074] I f Indicates the interference level of frequency band f,

[0075] L fIndicates the current network load of frequency band f

[0076] P d Indicates the priority of device d.

[0077] and Represent the weights before and after adjustment respectively.

[0078] ∈ is a minimum value used to avoid the denominator being zero.

[0079] β, γ and δ are weight adjustment coefficients, which are set according to actual needs.

[0080] δ df is an indicator function that takes the value 1 when device d is assigned to frequency band f, and 0 otherwise.

[0081] In one embodiment of the present invention, the performance optimization and feedback module is responsible for continuously monitoring network performance indicators and evaluating the effectiveness of the spectrum allocation scheme based on these indicators. Through self-learning and optimization mechanisms, the module can adjust algorithm parameters to adapt to the ever-changing network environment. The specific steps in its operation are as follows:

[0082] Steps: 4.1. Performance indicator monitoring: The intelligent gateway regularly collects and analyzes network performance indicators (such as throughput, latency, packet loss rate, etc.), which can be obtained through communication protocols between devices or specialized network performance monitoring tools;

[0083] Step 4.2. Scheme effectiveness evaluation: Compare the collected performance indicators with the preset target values ​​or historical data to evaluate the effectiveness of the current spectrum allocation scheme. If the performance indicators deviate from the target values ​​or show a downward trend, the optimization process is triggered.

[0084] Step 4.3. Self-learning and optimization: The algorithm uses machine learning or reinforcement learning methods to analyze the deficiencies of the spectrum allocation scheme based on the performance monitoring results, and optimizes the spectrum allocation strategy by adjusting parameters such as the weight adjustment coefficient and the sharing condition threshold.

[0085] Step 4.4. Parameter adjustment and implementation: Based on the results of self-learning and optimization, the intelligent gateway dynamically adjusts the algorithm parameters, and the adjusted parameters are immediately applied to the spectrum allocation process to improve network performance.

[0086] Let P t represents the network performance index (such as throughput) at time t, P target represents the preset target performance indicator, α is the learning rate parameter, is the gradient of the performance index P with respect to the algorithm parameters. The optimization process can be simplified to an iterative update process based on gradient descent as follows:

[0087]

[0088] The specific steps are as follows:

[0089] Ⅰ. Initialize algorithm parameters.

[0090] II. Monitoring network performance index P t .

[0091] III. Calculate the difference ΔP between the performance index and the target value.

[0092] IV. According to the gradient And the learning rate α updates the algorithm parameters.

[0093] V. Repeat steps 2 to 4 until the performance index reaches a satisfactory level or the maximum number of iterations is reached.

[0094] When the interference monitoring module identifies the main interference source and its frequency band, the specific operations are as follows:

[0095] Suppose we have a smart home system that uses the 2.4GHz frequency band for communication. In order to ensure the quality of communication, we need to collect and analyze the signals in the home wireless environment.

[0096] Signal collection: We use the built-in wireless signal receiver of the smart gateway to collect signals in the home wireless environment every 1 second. After 10 collections, we store these 10 signal samples in the local cache.

[0097] Next, we apply the Fast Fourier Transform (FFT) algorithm to these 10 signal samples. Assuming that each signal sample contains 1024 data points (N = 1024), we can calculate the energy distribution of each frequency point. Assuming that we calculate the energy distribution of the 512th frequency point (k = 512), substituting it into the FFT formula is:

[0098] The value of X

[512] is calculated to be 500.

[0099] Next, we compare the energy values ​​of all frequency points with the set threshold. Assuming the threshold is 400, the energy value of X

[512] , 500, exceeds the threshold, so we believe that there is potential interference at the 512th frequency point. Through further signal processing, such as filtering and demodulation, we find that the interference source is a microwave oven, which generates high-frequency noise when it is working.

[0100] In the adaptive spectrum management algorithm of the intelligent gateway, the adaptive spectrum allocation module dynamically adjusts the weight of each frequency band by adopting the following steps:

[0101] Assume that in a smart home environment, we have three frequency bands to choose from: 5GHz Band A, 5GHz Band B, and 2.4GHz Band C. At the same time, there are three types of devices: smart speakers (high priority, denoted as D1), security cameras (medium priority, denoted as D2), and thermostats (low priority, denoted as D3).

[0102] Initially, the weight WA of band A is 0.6, the weight WB of band B is 0.3, and the weight WC of band C is 0.1. Now, due to the addition of new equipment, the interference level IA of band A has risen to 0.7, while the channel utilization rate IC of band C has reached 0.8.

[0103] We set the weight adjustment coefficients (β=1.5), (γ=1), (δ=1), and take (∈=0.01).

[0104] According to the weight adjustment formula, we can calculate the new weight, specifically:

[0105] For frequency band A (the frequency band where smart speakers are located):

[0106]

[0107] Assume (δ D1A =1)(the smart speaker is currently in frequency band A), and the calculation results are About 0.5.

[0108] For Band C (where security cameras and thermostats reside):

[0109]

[0110] Assume (δ D2C =1) and (δ D3C =1)(the security camera and thermostat are currently in frequency band C), and the calculation results are About 0.2.

[0111] After the weight adjustment, the weight of band A has decreased, but it is still the highest, indicating that smart speakers should still have priority to use this band. However, considering the increase in interference, the smart gateway may decide to temporarily switch the smart speaker to band B, which has less interference.

[0112] For Band C, although its initial weight is low, its weight increases due to the presence of security cameras and thermostats and the high channel utilization. The smart gateway may look for a relatively high-quality sub-band within Band C to continue allocating to security cameras and thermostats.

[0113] Through such dynamic weight adjustment, the smart gateway can flexibly allocate spectrum resources according to real-time network conditions, ensuring the communication quality of high-priority devices while optimizing the performance of the entire smart home network.

[0114] In addition, in the adaptive spectrum management algorithm of the intelligent gateway of this application, the adaptive spectrum allocation module dynamically adjusts the weight of each frequency band, and the algorithm introduces a "spectrum sharing" mechanism. The so-called "spectrum sharing" mechanism is an important part of the adaptive spectrum allocation algorithm. It allows low-priority devices to share the frequency band resources occupied by high-priority devices while ensuring the communication quality of high-priority devices. This mechanism improves the utilization rate of spectrum resources while ensuring that the communication of key devices is not affected. Its specific implementation plan is:

[0115] Prioritization: During the algorithm initialization phase, a priority is assigned to each device based on its importance and communication needs.

[0116] Spectrum occupancy detection: The smart gateway continuously monitors the occupancy of each frequency band, including which devices are using which frequency bands and their communication quality.

[0117] Sharing condition determination: When a low-priority device needs to communicate and the frequency band resources occupied by a high-priority device are idle, the algorithm determines whether the sharing condition is met. The sharing condition can be set based on factors such as signal-to-interference ratio (SIR), channel utilization, and device communication requirements.

[0118] Spectrum sharing implementation: If the sharing conditions are met, the intelligent gateway will allocate the low-priority device to a sub-band or time slot of the corresponding high-priority device frequency band. Through power control, modulation and coding adjustment, etc., it ensures that the communication of the low-priority device will not interfere with the high-priority device.

[0119] The sharing conditions in the spectrum sharing mechanism can be defined as follows:

[0120] S f =1 and and

[0121] Where: S f Indicates the occupancy status of frequency band f (0 means idle, 1 means occupied)

[0122] Q d Indicates the communication quality indicator of device d (such as SIR)

[0123] P d Indicates the priority of device d

[0124] Indicates the communication quality indicator of high priority devices

[0125] Q th is the communication quality threshold set

[0126] and Indicates the priorities of low-priority and high-priority devices respectively.

[0127] Assume that in a smart home environment, the smart speaker (high priority) currently occupies most of the resources in 5GHz band A, while the security camera (medium priority) needs to communicate but the resources in 2.4GHz band C are tight. The smart gateway detects that there is an idle sub-band in band A that is not used by the smart speaker, and the communication quality index of the smart speaker is higher than the preset threshold Q th , and the priority of security cameras is lower than that of smart speakers. According to the "spectrum sharing" mechanism, the smart gateway determines that the security camera can share the idle sub-band in band A. Therefore, it allocates the security camera to this sub-band and ensures that its communication does not interfere with the smart speaker through power control and other means. Through such spectrum sharing, the smart gateway can improve the utilization rate of spectrum resources.

[0128] In the adaptive spectrum management algorithm of the macro intelligent gateway in this embodiment, the performance optimization and feedback module is responsible for continuously monitoring network performance indicators and evaluating the effectiveness of the spectrum allocation scheme based on these indicators. Through self-learning and optimization mechanisms, the module can adjust algorithm parameters to adapt to the ever-changing network environment.

[0129] Assume that in a smart home environment, the smart gateway detects that the network throughput has decreased recently and the latency has increased. Through the analysis of the performance optimization and feedback module, it is found that this is due to the failure of the spectrum allocation scheme to fully adapt to the newly added low-power devices. The smart gateway starts the self-learning and optimization process by adjusting the weight adjustment coefficients (β, γ, δ) and the sharing condition threshold (Q th ), so that low-power devices can share spectrum resources more effectively while ensuring the communication quality of high-priority devices. After several rounds of iterative updates, the smart gateway found that network throughput gradually recovered and latency was improved. Therefore, it applied the optimized parameters to the spectrum allocation algorithm to continuously improve the performance of the smart home network. Through such self-learning and optimization mechanisms, the smart gateway can continuously adapt to the changing network environment and ensure the stability and efficiency of the smart home network.

[0130] The basic principles and main features of the present invention are described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for illustrating the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which shall fall within the scope of the present invention to be protected. The scope of protection of the invention shall be defined by the attached claims and their equivalents.

Claims

1. An adaptive spectrum management method based on an intelligent gateway, characterized in that: The following steps are included in sequence: Interference monitoring module, equipment demand analysis module, adaptive spectrum allocation module, performance optimization and feedback module; The interference monitoring module is used to collect and analyze the signal interference in the home wireless environment in real time, and use the signal processing algorithm to perform spectrum analysis on the collected signals to identify the main interference sources and their frequency bands; The device demand analysis module collects the connection requirements of the devices by interacting with the communication protocols of each smart home device, and then analyzes the communication priority, data transmission volume and real-time requirements of each device to provide a basis for spectrum allocation; The adaptive spectrum allocation module runs an adaptive spectrum allocation algorithm in combination with the interference monitoring results and the device demand analysis structure; adopts a dynamic weight adjustment strategy to dynamically adjust the weight of each frequency band according to the current network load and interference situation, and preferentially allocates frequency bands with less interference to high-priority devices; at the same time, the algorithm adopts a spectrum sharing mechanism to allow low-priority devices to share the frequency band resources of high-priority devices while ensuring the communication quality of high-priority devices; The performance optimization and feedback module continuously monitors network performance indicators and evaluates the effectiveness of spectrum allocation schemes. Based on the performance monitoring results, the algorithm can self-learn and optimize and adjust parameters to adapt to the changing network environment; The interference monitoring module mainly adopts the following steps when identifying the main interference sources and their frequency bands: Step 1.1 Signal collection: Use the built-in wireless signal receiver of the smart gateway to continuously collect signals in the home wireless environment within a set time interval; The collected signals are stored in a local buffer for subsequent processing; Step 1.2 Spectrum analysis: Apply the fast Fourier transform (FFT) algorithm to the collected signal to convert the time domain signal into the frequency domain signal. The formula of FFT is: Where x[n] is a time domain signal sample, X[k] is the corresponding frequency domain signal sample, N is the total number of samples, and k is the frequency domain index; Through FFT transformation, the energy distribution of the signal at different frequencies can be obtained; Step 1.3: Interference source identification: Analyze the frequency domain signal after FFT transformation, find the frequency points or frequency bands with abnormally high energy, set an energy threshold, and the frequency points or frequency bands exceeding the threshold are considered potential interference sources. Through further signal processing, try to identify the specific type and source of the interference source; Step 1.4: Interference frequency band determination: Based on the results of interference source identification, the interfered frequency bands are determined, and these frequency bands will be temporarily excluded from the communication frequency bands of smart home devices to reduce interference.

2. The adaptive spectrum management method based on intelligent gateway according to claim 1, characterized in that: The strategy of dynamically adjusting the weight of each frequency band by the adaptive spectrum allocation module mainly adopts the following steps: Step 3.1: Weight initialization: In the algorithm initialization phase, an initial weight is set for each frequency band based on historical data and preset rules; Step 3.

2. Real-time monitoring and data collection: The smart gateway continuously monitors the signal interference in the home wireless environment, including the interference intensity of each frequency band, channel utilization, etc. At the same time, it collects the connection status, communication requirements and priority information of each smart home device; Step 3.

3. Dynamic weight adjustment: Based on real-time monitoring data, a specific algorithm is used to dynamically adjust the weight of each frequency band. The weight adjustment takes into account factors such as current network load, interference level, device priority, and historical allocation; Step 3.

4. Spectrum allocation decision: Based on the adjusted weights, the smart gateway makes spectrum allocation decisions, giving priority to allocating frequency bands with less interference to high-priority devices. The algorithm for dynamically adjusting the weights of each frequency band is shown below: Where: W f represents the weight of frequency band f, I f Indicates the interference level of frequency band f, L f Indicates the current network load of frequency band f P d Indicates the priority of device d. and Represent the weights before and after adjustment, ∈ is a minimum value used to avoid the denominator being zero. β, γ and δ are weight adjustment coefficients, which are set according to actual needs. δ df is an indicator function that takes the value 1 when device d is assigned to frequency band f, and 0 otherwise.

3. The adaptive spectrum management method based on intelligent gateway according to claim 1, characterized in that: The performance optimization and feedback module is responsible for continuously monitoring network performance indicators and evaluating the effectiveness of spectrum allocation schemes based on these indicators. Through self-learning and optimization mechanisms, this module is able to adjust algorithm parameters to adapt to the changing network environment. The specific steps in its work are as follows: Steps: 4.

1. Performance indicator monitoring: The intelligent gateway regularly collects and analyzes network performance indicators, which can be obtained through communication protocols between devices or dedicated network performance monitoring tools; Step 4.

2. Evaluation of scheme effectiveness: Compare the collected performance indicators with the preset target values ​​or historical data to evaluate the effectiveness of the current spectrum allocation scheme. If the performance indicators deviate from the target values ​​or show a downward trend, the optimization process is triggered; Step 4.

3. Self-learning and optimization: The algorithm uses machine learning or reinforcement learning methods to analyze the deficiencies of the spectrum allocation scheme based on the performance monitoring results, and optimizes the spectrum allocation strategy by adjusting parameters such as weight adjustment coefficients and sharing condition thresholds; Step 4.

4. Parameter adjustment and implementation: Based on the results of self-learning and optimization, the intelligent gateway dynamically adjusts the algorithm parameters, and the adjusted parameters are immediately applied to the spectrum allocation process to improve network performance; Let P t represents the network performance index (such as throughput) at time t, P target represents the preset target performance indicator, α is the learning rate parameter, is the gradient of the performance index P with respect to the algorithm parameters. The optimization process can be simplified to an iterative update process based on gradient descent as follows: The specific steps are as follows: Ⅰ. Initialize algorithm parameters. II. Monitoring network performance index P t . III. Calculate the difference ΔP between the performance index and the target value. IV. According to the gradient And the learning rate α updates the algorithm parameters. V. Repeat steps 2 to 4 until the performance index reaches a satisfactory level or the maximum number of iterations is reached.

Images