A handheld ground station anti-jamming method and device

Abstract

The application is suitable for the technical field of signal optimization, and provides a handheld ground station anti-interference method and device.The handheld ground station anti-interference method comprises the following steps: periodically performing wideband scanning on a pre-defined candidate channel set to obtain interference observation of each candidate channel in each scanning period; calculating predicted cleanliness and switching cost of the candidate channel according to the interference observation; determining a channel switching evaluation value of the candidate channel by using a difference value of the predicted cleanliness and the switching cost after normalization; and switching a current working channel to a candidate channel with the highest channel switching evaluation value in response to a decision result of spectrum agility. The application avoids blind frequency switching in a high switching cost scenario, suppresses reciprocating frequency switching, and reduces accumulation of link interruption time during frequency switching.

Description

Technical Field

[0001] This invention belongs to the technical field of signal optimization, and particularly relates to a handheld ground station anti-interference method and device. Background Technology

[0002] Handheld ground stations are portable terminals in UAV systems responsible for issuing remote control commands, receiving telemetry data, and transmitting and displaying images. The wireless data link between the handheld ground station and the airborne terminal typically operates in unlicensed frequency bands such as 2.4GHz and 5.8GHz. These bands simultaneously accommodate multiple services including Wi-Fi, Bluetooth, wireless image transmission, and microwave transmission, resulting in a complex electromagnetic environment prone to co-channel and adjacent-channel interference. If the data link is subjected to continuous interference, it can cause minor issues like image transmission lag and missing telemetry data, or even lead to the loss of remote control commands or complete link failure, directly threatening flight safety.

[0003] To address the aforementioned issues, existing technologies generally base frequency switching decisions solely on the current signal quality at the target frequency. In scenarios with rapidly changing interference, this can easily lead to frequent and ineffective frequency switching. The accumulated link interruption time during frequency switching exacerbates the risk of loss of control. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a handheld ground station anti-interference method and device to solve the technical problem that the existing technology roughly decides frequency switching based only on the current signal quality of the target frequency point, which easily leads to frequent and invalid frequency switching in scenarios with rapidly changing interference. During the frequency switching, the link interruption time continues to accumulate, which exacerbates the risk of loss of control.

[0005] A first aspect of the present invention provides a handheld ground station anti-interference method, the handheld ground station anti-interference method comprising: A predefined set of candidate channels is periodically scanned with wideband to obtain interference observations for each candidate channel in each scanning period; wherein, the interference observations include the energy value of the candidate channel, the interference type, the center frequency of the interference signal, and the occupied bandwidth. Based on the interference observations, the predicted cleanliness and handover cost of the candidate channel are calculated; wherein, the predicted cleanliness is the probability that the candidate channel will not be contaminated by any interference source in the next transmission time slot; and the handover cost is an estimate of the time required for the radio frequency parameters to be readjusted from the current working channel to the candidate channel. The difference between the predicted cleanliness and the normalized handover cost is determined as the channel handover evaluation value of the candidate channel. In response to the decision result of spectrum agility, the current working channel is switched to the candidate channel with the highest channel switching evaluation value.

[0006] Furthermore, the step of periodically scanning a predefined set of candidate channels to obtain interference observations for each candidate channel in each scanning period includes: A predefined set of candidate channels is periodically scanned with wideband, and IQ data is collected for candidate channels whose energy values ​​are higher than a preset energy threshold. The IQ data is subjected to time-domain and frequency-domain analysis to extract interference signal features; wherein, the interference signal features include the concentration of power spectral density in the frequency domain, the direction of change of the center frequency of the power spectrum with the scanning period, and the duty cycle of the pulse power in time; The interference signal features are input into a pre-trained neural network classifier, which outputs the type probability of the interference on the candidate channel belonging to each preset interference type; wherein, the preset interference types include continuous wave interference, frequency sweep interference, periodic pulse interference, and broadband noise interference; The preset interference type with the highest probability is determined as the interference type of the candidate channel in this scanning period; The combination of the candidate channel's energy value, interference type, center frequency of the interference signal, occupied bandwidth, and the probability of the aforementioned type is recorded as the interference observation of the candidate channel in that scanning period.

[0007] Furthermore, the step of calculating the predicted cleanliness and handover cost of the candidate channel based on the interference observation includes: Extract the center frequency, occupied bandwidth, and type probability from each interference observation in the current scanning cycle; For each interference observation, calculate the difference between the interference observation and the most recently recorded center frequency, occupied bandwidth, and type probability in each historical interference source profile; wherein, the historical interference source profile is updated based on historical interference source observations; When there is a historical interference source profile with a difference less than a preset association threshold, the interference observation is associated with the historical interference source profile, and the data in the historical interference source profile is updated with the interference observation to obtain the first current interference source profile. When there is no historical interference source profile with a difference less than the preset correlation threshold, the interference observation is recorded as a newly created second current interference source profile; wherein, the interference source profile records the interference type, center frequency, occupied bandwidth and time domain activity pattern of the interference source, and the time domain activity pattern includes the time sequence of the interference source observed in each scanning cycle and the duration of each observation; Based on the temporal activity patterns recorded in the profiles of each current interference source, predict the candidate channels that each interference source will pollute in the next transmission time slot; Calculate the predicted cleanliness of each candidate channel based on the candidate channels contaminated by each interference source in the next transmission time slot. The switching cost of each candidate channel is determined based on the frequency interval between the current working channel and each candidate channel.

[0008] Furthermore, the step of predicting the candidate channel contaminated by each interference source in the next transmission time slot based on the temporal activity patterns recorded in the profiles of each current interference source includes: Based on the time sequence in the time-domain activity pattern, calculate the time interval sequence between two consecutive observations of each interference source; When the difference between the maximum and minimum time intervals in the time interval sequence corresponding to the target interference source is less than a preset interval threshold, the mean of the time interval sequence is determined as the active period of the target interference source. Based on the active period and the time when the target interference source was last observed, extrapolate the predicted active time of the target interference source in the next transmission time slot; When the predicted active moment falls into the next transmission time slot, it is determined that the target interference source is in an active state in the next transmission time slot; The interference source in the active state is identified as the active interference source. The difference between the center frequency of the active interference source and half of the occupied bandwidth is identified as the lower boundary of the frequency band. The sum of the center frequency and half of the occupied bandwidth is identified as the upper boundary of the frequency band. Candidate channels whose frequency range intersects with the frequency band between the lower boundary and the upper boundary of the frequency band are predicted as candidate channels contaminated by the active interference source in the next transmission time slot.

[0009] Furthermore, the step of calculating the predicted cleanliness of each candidate channel based on the candidate channels contaminated by each interference source in the next transmission time slot includes: For each candidate channel, find the interference sources whose frequency range of occupied bandwidth overlaps with the frequency range of the candidate channel from each current interference source profile, and obtain the interference source set of the candidate channel. For each current interference source profile in the interference source set, extract the interval between adjacent times in the time sequence recorded in the current interference source profile to determine the duty cycle of the interference source. Based on the duration recorded in the current interference source profile, determine the duration of the interference source's occupation. Read the most recently observed time from the time sequence recorded in the current interference source profile of the interference source; The predicted cleanliness of the candidate channel is calculated based on the duty cycle, the occupancy duration, and the most recently observed time.

[0010] Further, the step of calculating the predicted cleanliness of the candidate channel based on the duty cycle, the occupancy duration, and the most recently observed time includes: Add an integer multiple of the duty cycle to the most recently observed time to obtain the expected occurrence time; wherein the integer multiple of the duty cycle increases successively from 1 duty cycle until the obtained time exceeds the end time of the next transmission time slot. For each expected occurrence time falling within the next transmission time slot, the time period from that expected occurrence time to the duration of the occupation is determined as the observed period of the interference source in the next transmission time slot; The ratio of the overlap duration between the observed time period and the next transmission time slot to the total duration of the next transmission time slot is determined as the probability that the interference source will contaminate the candidate channel in the next transmission time slot. The predicted cleanliness of the candidate channel is obtained by multiplying 1 by the difference between the contamination probabilities of each interference source in the interference source set.

[0011] Furthermore, the step of determining the handover cost of each candidate channel based on the frequency interval between the current working channel and each candidate channel includes: Read the center frequency of the current working channel and the center frequency of each candidate channel respectively; For each candidate channel, the absolute value of the difference between the center frequency of the candidate channel and the center frequency of the current working channel is determined as the frequency interval of the candidate channel. The frequency interval of the candidate channel is compared with the preset range boundary, and the range in which the frequency interval falls is determined as the frequency interval range of the candidate channel. The candidate channel's frequency interval is used to query a pre-established interval time lookup table to obtain the interval readjustment time. When the frequency interval of the candidate channel is the cross-band interval, the filter channel reconfiguration time is superimposed on the interval readjustment time of the candidate channel; wherein, the filter channel reconfiguration time is the time required to switch the filter channel of the RF front end from the frequency band of the current working channel to the frequency band of the candidate channel. The switching cost of the candidate channel is determined by superimposing the reconfiguration time of the filter channel or the reconfiguration time of the frequency interval when it is a non-cross-band interval, and then superimposing the convergence time of the automatic gain control. The total duration obtained by superposition is determined by superimposing the automatic gain control convergence time as the time required for the gain of the receiving link to re-converge to a stable state after the local oscillator frequency is locked.

[0012] Furthermore, the step of switching the current working channel to the candidate channel with the highest channel switching evaluation value in response to the decision result of spectrum agility includes: The predicted cleanliness of the current working channel is compared with a first classification threshold and a second classification threshold, wherein the first classification threshold is higher than the second classification threshold. When the predicted cleanliness is not lower than the first classification threshold, the decision result for channel maintenance is determined; When the predicted cleanliness is lower than the first classification threshold and not lower than the second classification threshold, the decision result of spectrum agility is determined, and the main target channel is cached as a channel to be executed; When the predicted cleanliness is lower than the second classification threshold, the decision result of spectrum agility is determined and spectrum agility is executed immediately; wherein, the classification threshold is set according to the historical switching success rate within a preset statistical period, and the historical switching success rate is the proportion of the number of successful spectrum agility changes within the preset statistical period to the total number of spectrum agility changes. When the decision result is spectrum agility, the candidate channel with the highest channel switching evaluation value is determined as the primary target channel, and the candidate channel with the second highest channel switching evaluation value and a frequency interval greater than the primary target channel is determined as the backup target channel. The agile command packet carrying the primary target channel, the backup target channel, and the switching timestamp is sent to the peer device via the reverse control link. The radio frequency parameters are switched to the primary target channel in sync with the switching timestamp and the peer device, and the switching of radio frequency parameters is completed within the inter-frame gap of the physical layer frame; If no response is received from the peer device on the primary target channel within the preset verification window after switching, the radio frequency parameters will be switched to the backup target channel.

[0013] A second aspect of the present invention provides a handheld ground station anti-interference device, comprising: A scanning unit is used to periodically scan a predefined set of candidate channels to obtain interference observations for each candidate channel in each scanning period; wherein, the interference observations include the energy value of the candidate channel, the interference type, the center frequency of the interference signal, and the occupied bandwidth. The calculation unit is used to calculate the predicted cleanliness and handover cost of the candidate channel based on the interference observation; wherein, the predicted cleanliness is the probability that the candidate channel will not be contaminated by any interference source in the next transmission time slot; and the handover cost is an estimate of the time required for the radio frequency parameters to be readjusted from the current working channel to the candidate channel. The determining unit is used to determine the difference between the predicted cleanliness and the normalized switching cost as the channel switching evaluation value of the candidate channel. The switching unit is used to switch the current working channel to the candidate channel with the highest channel switching evaluation value in response to the decision result of spectrum agility.

[0014] A third aspect of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the handheld ground station anti-interference method described in the first aspect.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the handheld ground station anti-interference method described in the first aspect.

[0016] The beneficial effects of this invention compared to existing technologies are as follows: By periodically scanning the spectrum and collecting multi-dimensional interference observations such as the energy value, interference type, center frequency of the interference signal, and occupied bandwidth of candidate channels, a fine-grained characterization of the electromagnetic environment is achieved. This overcomes the limitations of traditional schemes that rely solely on a single energy intensity index to perceive interference and cannot distinguish interference patterns, providing a sufficient information foundation for interference situation analysis. The predicted cleanliness calculated based on interference observation reflects the probability that a candidate channel will not be contaminated by any interference source in the next transmission time slot, transforming channel decision-making from reactive remediation to proactive prediction. This allows channel selection to be completed before interference actually causes link quality degradation, avoiding the problem of damage already occurring due to detection lag in traditional schemes. Incorporating the estimated time required to re-adjust RF parameters from the current working channel to the candidate channel as a handover cost into the decision overcomes the shortcomings of existing frequency switching mechanisms that only consider the target channel signal quality and ignore the communication interruption introduced by the frequency switching action itself. This avoids blind frequency switching in high handover cost scenarios, suppresses repetitive frequency switching, and reduces the accumulation of link interruption time during frequency switching. By normalizing the predicted cleanliness and the handover cost and then subtracting them, a unified channel handover assessment value is obtained. This integrates two different dimensions of factors, channel availability and handover cost, into a single decision criterion, taking into account both the long-term stability of link quality and the real-time control of handover costs. This achieves a balance between timeliness and accuracy in spectrum agility decision-making, ultimately improving the anti-interference capability, image transmission continuity, and remote control real-time performance of the handheld ground station data link in complex electromagnetic environments. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic flowchart of an anti-interference method for a handheld ground station provided by the present invention is shown; Figure 2 A schematic diagram of a handheld ground station anti-interference device according to an embodiment of the present invention is shown; Figure 3 A schematic diagram of a terminal device according to an embodiment of the present invention is shown; Figure 4 This invention provides a flowchart illustrating a specific implementation of step 101 in a handheld ground station anti-interference method. Figure 5 This invention provides a flowchart illustrating a specific implementation of step 102 in a handheld ground station anti-interference method. Figure 6 This invention provides a flowchart illustrating a specific implementation of step B5 in a handheld ground station anti-interference method. Figure 7 This invention provides a flowchart illustrating a specific implementation of step B6 in a handheld ground station anti-interference method. Figure 8 This invention provides a flowchart illustrating a specific implementation of step B65 in a handheld ground station anti-interference method. Figure 9 This invention provides a flowchart illustrating a specific implementation of step B7 in a handheld ground station anti-interference method. Figure 10 The flowchart illustrates a specific implementation of step 104 in a handheld ground station anti-interference method provided by the present invention. Detailed Implementation

[0019] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0020] This invention provides a handheld ground station anti-interference method and device to address the technical problem that existing technologies rely solely on the current signal quality of the target frequency for frequency switching decisions. In scenarios with rapidly changing interference, this can easily lead to frequent and ineffective frequency switching, with the link interruption time accumulating during frequency switching, which in turn exacerbates the risk of loss of control.

[0021] First, this invention provides a method for anti-interference of handheld ground stations. Please refer to [link / reference]. Figure 1 , Figure 1 A schematic flowchart of an anti-interference method for a handheld ground station provided by the present invention is shown. Figure 1 As shown, the anti-interference method for this handheld ground station may include the following steps: Step 101: Perform periodic broadband scanning on a predefined set of candidate channels to obtain interference observations for each candidate channel in each scanning period; wherein, the interference observations include the energy value of the candidate channel, the interference type, the center frequency of the interference signal, and the occupied bandwidth; The handheld ground station and the UAV exchange telemetry data and control commands via a wireless telemetry and control link. This link operates in an open frequency band of 2400MHz to 2483.5MHz, which also contains numerous other wireless devices, such as wireless LAN access points, Bluetooth devices, and wireless image transmission devices. In this embodiment, the frequency band is divided into 40 candidate channels with a bandwidth of 2MHz each, with a 2MHz center frequency interval between adjacent candidate channels. These 40 candidate channels are predefined as a candidate channel set. The handheld ground station performs periodic broadband scans of the candidate channel set during service transmission intervals, traversing all candidate channels once in each scan cycle to obtain the energy value of each candidate channel. For example, the scan cycle can be 50 milliseconds, matching a 20-millisecond transmission time slot, ensuring that the latest or near-latest spectrum status is obtained in each transmission time slot. The core idea of ​​this step is not to directly judge the quality of the channel based on the instantaneous energy value, but to combine the energy value, interference type, center frequency of the interference signal and occupied bandwidth into an interference observation record, so as to provide a data foundation for subsequent tracking and prediction of the behavior of the interference source from the time dimension.

[0022] Specifically, step 101 includes steps A1 to A5: Figure 4 The following is a flowchart illustrating a specific implementation of step 101 in a handheld ground station anti-interference method provided by the present invention: Step A1: Perform periodic broadband scanning on a predefined set of candidate channels, and collect IQ data for candidate channels whose energy values ​​are higher than a preset energy threshold. Wideband scanning is performed by the RF front-end of the handheld ground station. The scanning local oscillator locks onto the center frequency of each candidate channel sequentially with a fixed dwell time within the scanning period. The receiver measures the received power of the candidate channel during each dwell time, which is taken as the energy value of that candidate channel. A preset energy threshold is used to filter out clean channels that only have noise floor. For example, the preset energy threshold is set approximately 8 dB above the noise floor, and can be -98 dBm. Candidate channels with energy values ​​higher than the preset energy threshold are considered potentially interfered with. IQ data is collected from these candidate channels. For example, the collection duration can be 2 milliseconds, and the IQ data sampling rate can be 4 MSa / s, which is twice the bandwidth of the candidate channel, allowing for Nyquist sampling while providing a transition band margin. Collecting IQ data only for channels above the threshold avoids consuming acquisition and storage resources for clean channels, keeping the power consumption of the entire scanning process within the battery budget of the handheld device.

[0023] Step A2: Perform time-domain and frequency-domain analysis on the IQ data to extract interference signal features; wherein, the interference signal features include the concentration of power spectral density in the frequency domain, the direction of change of the center frequency of the power spectrum with the scanning period, and the duty cycle of the pulse power in time; The purpose of time-domain and frequency-domain analysis is to compress the original IQ data into a few interference signal features with clear physical meaning. Frequency-domain analysis performs a Fast Fourier Transform on the IQ data to obtain the power spectral density. The concentration of the power spectral density in the frequency domain is quantified by the ratio of the sum of the powers of the strongest spectral lines to the total power. Continuous wave interference has its power highly concentrated near a single spectral line, with this ratio close to 1. Broadband noise interference has its power evenly distributed across the entire bandwidth, with this ratio being very small. The direction of change of the power spectral center frequency with the scan period is obtained by comparing the spectral peak positions of the same candidate channel in two adjacent scan periods. The center frequency of swept-frequency interference exhibits a continuous unidirectional drift, while the center frequency of continuous wave interference remains essentially constant. The duty cycle of the pulse power in time is obtained from time-domain analysis. The instantaneous power of each sampling point of the IQ data is compared with the power level corresponding to a preset energy threshold. The ratio of the number of sampling points exceeding this threshold to the total number of sampling points is the duty cycle. The duty cycle of periodic pulse interference is between 0 and 1, while the duty cycles of continuous wave interference and broadband noise interference are close to 1.

[0024] Step A3: Input the interference signal features into a pre-trained neural network classifier and output the type probability of the interference on the candidate channel belonging to each preset interference type; wherein, the preset interference types include continuous wave interference, frequency sweep interference, periodic pulse interference and broadband noise interference; The neural network classifier is pre-trained offline. Its input is a 6-dimensional feature vector composed of interference signal features, including spectral concentration, center frequency drift rate, drift direction consistency, duty cycle, spectral flatness, and peak factor. The hidden layer is a two-layer fully connected structure with 16 and 12 nodes respectively. The output layer has 4 nodes, and after Softmax normalization, it outputs the probability of the interference on the candidate channel belonging to one of four preset interference types: continuous wave interference, frequency sweep interference, periodic pulse interference, and broadband noise interference. For example, the training samples can be 12,000 sets, obtained by feature extraction and labeling of various interference signals actually recorded in the RF laboratory. The classifier has low inference computation; a single inference time on the embedded digital signal processor of the handheld ground station is less than 0.2 milliseconds. A neural network classifier is used instead of manual rules because the boundaries of the four types of interference in the feature space are not non-linear. For example, the spectral concentration of weak frequency sweep interference and strong broadband noise interference may be similar. The neural network can characterize this non-linear boundary by learning from labeled samples, and the classification accuracy has been tested to reach over 97%. The network structure of the neural network classifier is a traditional technique and is not limited here.

[0025] As an optional embodiment of this application, the neural network classifier can also be replaced by rule-based decision based on feature thresholds. Specifically, when the spectral concentration is higher than a first threshold and the center frequency drift rate is lower than a second threshold, it is determined to be continuous wave interference; when the center frequency continuously drifts in one direction, it is determined to be frequency sweep interference; when the duty cycle is between 0.1 and 0.5, it is determined to be periodic pulse interference; and when the spectral flatness is higher than a third threshold, it is determined to be broadband noise interference. Rule-based decision does not require training samples and is simple to implement, but its classification accuracy is lower than that of the neural network classifier in scenarios with blurred feature boundaries. It is suitable for cost-sensitive devices or those lacking labeled samples.

[0026] Step A4: Determine the preset interference type with the highest type probability as the interference type of the candidate channel in this scanning period; The preset interference type with the highest type probability is the interference type that the candidate channel is most likely to suffer during that scanning period. For example, if the type probabilities of continuous wave interference, frequency sweep interference, periodic pulse interference, and broadband noise interference on a candidate channel are 0.86, 0.09, 0.03, and 0.02, respectively, then the interference type of the candidate channel during that scanning period is determined to be continuous wave interference. The type probability itself is also retained as part of the interference observation, rather than just the decision result. This is because subsequent interference source correlation steps need to use the similarity between type probabilities to determine whether two interference observations come from the same interference source; retaining only the decision result would lose this quantitative information.

[0027] Step A5: Record the combination of the energy value, interference type, center frequency of the interference signal, occupied bandwidth, and type probability of the candidate channel as the interference observation of the candidate channel in this scanning period.

[0028] Interference observations serve as the data source for all subsequent calculations. Their data structure is equivalent to a snapshot of the interference experienced by a candidate channel within a single scan cycle. In interference observations, the energy value comes from the received power measurement of the broadband scan, the interference type and type probability come from the output of the neural network classifier, and the center frequency and occupied bandwidth come from the power spectral density. The occupied bandwidth is determined by the frequency width between two points in the power spectral density that fall 3 dB below the peak. After each scan cycle, each candidate channel in the candidate channel set accumulates one interference observation. Multiple interference observations, arranged on the time axis, constitute the interference history of each candidate channel, which is used in step 102 to correlate and update the interference source profile.

[0029] In the embodiments corresponding to steps A1 to A5, through the cascaded processing of the above steps, broadband energy measurements are transformed into structured interference observation sequences carrying interference type semantics. Interference observation simultaneously includes quantitative parameters and type judgment, enabling subsequent processing to determine both the strength and type of interference received by the candidate channel, providing a complete data foundation for profiling and predicting the behavior of the interference source. The entire observation process only collects and processes IQ data for candidate channels exceeding the energy threshold. The computational load is proportional to the actual number of interferences within the frequency band and is independent of the size of the candidate channel set, making it suitable for real-time operation on low-power processors in handheld devices.

[0030] Step 102: Based on the interference observation, calculate the predicted cleanliness and handover cost of the candidate channel; wherein, the predicted cleanliness is the probability that the candidate channel will not be contaminated by any interference source in the next transmission time slot; the handover cost is an estimate of the time required for the radio frequency parameters to be readjusted from the current working channel to the candidate channel; This step is the core of the entire anti-interference method, taking interference observations of each candidate channel in each scanning period as input and outputting the predicted cleanliness and handover cost of each candidate channel. Predicted cleanliness is the probability that a candidate channel will not be contaminated by any interference source in the next transmission time slot, ranging from 0 to 1. Handover cost is an estimate of the time required for RF parameters to be retuned from the current working channel to the candidate channel. Selecting a target channel solely based on the energy value of the current scanning period is a reactive strategy. This strategy passively switches only after interference has already caused service impairment and cannot distinguish between departing and approaching interference. The core idea of ​​this step is to associate discrete interference observations with specific interference sources, creating an interference source profile for each source that includes temporal activity patterns. Then, the behavior of each interference source in the next transmission time slot is extrapolated from this profile, transforming the channel selection problem from a passive response to the current spectrum situation to an active prediction of the future spectrum situation.

[0031] Specifically, step 102 includes steps B1 to B7: Figure 5 The following is a flowchart illustrating a specific implementation of step 102 in a handheld ground station anti-interference method provided by the present invention: Step B1: Extract the center frequency, occupied bandwidth, and type probability from each interference observation in the current scanning period; Center frequency, occupied bandwidth, and type probability are three parameters related to the identity of the interference source in interference observations. Center frequency and occupied bandwidth characterize the location and range of the interference source in the frequency domain, while type probability characterizes the mechanistic category of the interference source. These three parameters are read one by one from the interference observations of each candidate channel in the current scanning period and used for subsequent correlation calculations.

[0032] Step B2: For each interference observation, calculate the difference between the interference observation and the most recently recorded center frequency, occupied bandwidth, and type probability in each historical interference source profile; wherein the historical interference source profile is updated based on historical interference source observations; Interference observations themselves do not carry interference source identifiers; their attribution needs to be determined by the similarity of parameters. For each interference observation, its center frequency, occupied bandwidth, and type probability are compared with the corresponding parameters recorded in the most recent historical interference source profile. The difference is the sum of the three normalized differences, calculated using the formula D=|fc-fc'| / B0+|B-B'| / B0+|p-p'|. Where fc and B are the center frequency and occupied bandwidth of the interference observation, fc' and B' are the center frequency and occupied bandwidth of the most recent record in the historical interference source profile, p and p' are the type probabilities of the two, and B0 is the normalization benchmark, which can be the bandwidth of the candidate channel, 2MHz. The smaller the difference, the greater the probability that the interference observation and the historical interference source profile are from the same source. For example, the preset association threshold can be 0.35. This value is based on the statistics of 3000 pairs of interference observations recorded in the field. The difference distribution of the same source observation pairs is mainly below 0.2, and the difference distribution of different source observation pairs is mainly above 0.5. 0.35 is located in the valley region of the intersection of the two distributions, which takes into account both the recall and precision of the association.

[0033] As an optional embodiment of this application, the item-by-item difference association can also be replaced by joint probability data association. That is, the normalized distance attribution probability is calculated for each interference observation and each historical interference source profile. Observations with probabilities exceeding a set threshold and which do not conflict with each other are paired with profiles. Observations that cannot be paired are then processed as new profiles. Joint probability data association can reduce erroneous associations in scenarios where multiple interference sources have similar frequency domain characteristics. The computational cost is higher than that of item-by-item difference association, making it suitable for complex electromagnetic environments with dense interference sources.

[0034] Step B3: When there is a historical interference source profile with a difference less than the preset association threshold, associate the interference observation with the historical interference source profile and update the data in the historical interference source profile with the interference observation to obtain the first current interference source profile; When a historical interference source profile exists with a difference less than a preset association threshold, it indicates that the interference observation likely originates from the interference source recorded in the profile. This interference observation is then associated with that historical interference source profile. After association, the data in the profile is updated using this interference observation. The center frequency and occupied bandwidth are updated using a moving average method, with new and old values ​​weighted by 0.3 and 0.7 respectively to suppress random fluctuations in single measurements. The type probability is also updated using a moving average. The time series is appended with the observed time, and the duration is recorded as the actual duration of this observation. The updated profile is called the first current interference source profile. The time series in the first current interference source profile extends continuously with the scanning cycle, constituting a record of the temporal behavior of the interference source.

[0035] Step B4: When there is no historical interference source profile with a difference less than the preset correlation threshold, the interference observation is recorded as a newly created second current interference source profile; wherein, the interference source profile records the interference type, center frequency, occupied bandwidth and time domain activity pattern of the interference source, and the time domain activity pattern includes the time sequence of the interference source observed in each scanning cycle and the duration of each observation; When no historical interference source profile exists with a difference less than a preset correlation threshold, it indicates that the interference observation does not belong to any known interference source, and it is recorded as a newly created second current interference source profile. The initial content of the new profile is directly taken from the interference observation. The interference type is taken from the interference type in the interference observation, and the center frequency, occupied bandwidth, and type probability are taken from the corresponding values ​​in the interference observation. The time sequence in the time-domain activity pattern is initialized to a single-element sequence containing only the observed time, and the duration is initialized to the duration of the current observation. The first and second current interference source profiles are completely identical in data structure and are collectively referred to as current interference source profiles. The only difference is the method of creation: one is an update of an existing profile, and the other is the creation of a new profile. The interference source profile as a whole records the interference type, center frequency, occupied bandwidth, and time-domain activity pattern of the interference source. The time-domain activity pattern includes the time sequence of the interference source observed in each scanning cycle and the duration of each observation.

[0036] Step B5: Based on the temporal activity patterns recorded in the profiles of each current interference source, predict the candidate channels that each interference source will pollute in the next transmission time slot; Interference sources are not continuous in time. Continuous wave interference usually persists for a long time, periodic pulse interference appears intermittently according to a fixed rhythm, and the center frequency of frequency sweep interference shifts in the frequency domain, causing intermittent energy rise on a candidate channel. Temporal activity patterns are the carriers that characterize these rhythms. Frequency domain parameters alone cannot determine which channels will be affected in the next transmission time slot; it is necessary to extract time parameters from the time series and then extrapolate the behavior of each interference source in the next transmission time slot.

[0037] Specifically, step B5 includes steps B51 to B56: Figure 6 The following is a flowchart illustrating a specific implementation of step B5 in a handheld ground station anti-interference method provided by the present invention: Step B51: Calculate the time interval sequence between two consecutive observations of each interference source based on the time sequence in the time domain activity pattern; The difference between two adjacent moments in a time series is the observation interval. Arranging all the differences between adjacent moments in chronological order yields the time interval sequence. Taking a periodic pulse interference source as an example, its time series is t0, t0+12 milliseconds, t0+24 milliseconds, and t0+36 milliseconds, then the time interval sequence is 12 milliseconds, 12 milliseconds, and 12 milliseconds. The uniformity of the time interval sequence directly reflects the regularity of the interference source's activity.

[0038] Step B52: When the difference between the maximum and minimum time intervals in the time interval sequence corresponding to the target interference source is less than a preset interval threshold, the mean of the time interval sequence is determined as the active period of the target interference source. If the difference between the maximum and minimum time intervals is less than a preset interval threshold, it indicates that the activity rhythm of the target interference source is highly stable, and the average of the time interval sequence is the active period. For example, the preset interval threshold can be 2 milliseconds. This value is based on the fact that the time measurement accuracy of the handheld ground station is approximately 0.5 milliseconds, the scan period is 50 milliseconds, and the quantization error of the observation time, after differential averaging, does not exceed 1 millisecond. The 2-millisecond threshold precisely covers the quantization error and leaves a margin. Taking the aforementioned interference source as an example again, all three time intervals are 12 milliseconds, and the difference between the maximum and minimum is 0, which is less than the threshold. Therefore, the active period is determined to be 12 milliseconds. For interference sources with unstable activity rhythms, this embodiment does not extrapolate their active time but conservatively assumes that they continue to exist in the next transmission time slot, directly marking the contamination according to their frequency domain range.

[0039] Step B53: Based on the active period and the time when the target interference source was last observed, extrapolate the predicted active time of the target interference source in the next transmission time slot; The extrapolation method involves successively adding the active period to the most recently observed time to obtain a series of predicted active times. Let the most recently observed time be t0, and the active period be T. Then the predicted active times are t0+T, t0+2T, t0+3T, and so on. Taking a time interval of T = 12 milliseconds, where the next transmission slot covers the period from t0+6 milliseconds to t0+26 milliseconds, as an example, the predicted active times t0+12 milliseconds and t0+24 milliseconds fall within this interval.

[0040] Step B54: When the predicted active moment falls into the next transmission time slot, it is determined that the target interference source is active in the next transmission time slot; The determination method involves comparing each predicted active time with the start and end times of the next transmission time slot. If any predicted active time falls within the time slot, the target interference source is determined to be active in the next transmission time slot. Being active means that the interference source will reappear at that time, and its frequency range needs to be included in the contamination prediction.

[0041] Step B55: Identify the interference source in the active state as an active interference source, determine the difference between the center frequency of the active interference source and half of the occupied bandwidth as the lower boundary of the frequency band, and determine the sum of the center frequency and half of the occupied bandwidth as the upper boundary of the frequency band. The lower and upper boundaries of a frequency band together define the pollution range of an active interference source in the frequency domain. For example, if the center frequency of an active interference source is 2430MHz and its occupied bandwidth is 4MHz, then the lower boundary of the frequency band is 2428MHz and the upper boundary is 2432MHz. The occupied bandwidth is taken from the most recently updated value in the interference source profile, reflecting the actual coverage of the interference signal rather than the nominal bandwidth of the candidate channel.

[0042] Step B56: The candidate channels whose frequency range intersects with the frequency band between the lower boundary and the upper boundary of the frequency band are predicted as candidate channels contaminated by the active interference source in the next transmission time slot.

[0043] The frequency range of candidate channels is calculated by adding or subtracting half of the channel bandwidth from its center frequency. Taking the 2428MHz to 2432MHz band as an example, the frequency ranges of three adjacent candidate channels with center frequencies of 2429MHz, 2430MHz, and 2431MHz all overlap with this band. These three candidate channels are predicted as candidate channels to be contaminated by the active interference source in the next transmission time slot. For frequency sweeping interference, its center frequency continuously drifts. The center frequency in the image is taken as the most recent recorded value and superimposed with the predicted offset in the drift direction to determine the frequency band boundary, ensuring that the contamination prediction covers the area ahead of its drift path.

[0044] In the embodiments corresponding to steps B51 to B56, through the cascaded processing of the above steps, the time series recorded in each current interference source profile is transformed into three types of outputs: active period, predicted active moment, and contaminated channel set. This achieves the extrapolation from the historical behavior of the interference source to its future frequency domain footprint. This contamination prediction considers both the rhythmicity of the time dimension and the coverage of the frequency domain dimension, ensuring that subsequent cleanliness calculations are based on the expected future behavior of the interference source, rather than on the current static snapshot.

[0045] Step B6: Calculate the predicted cleanliness of each candidate channel based on the candidate channels contaminated by each interference source in the next transmission time slot; Contamination prediction provides a qualitative set of data, indicating which candidate channels will be affected by which interference sources. Handover decisions also require a quantitative probabilistic measure: the likelihood that each candidate channel will remain clean in the next transmission time slot. This measure is provided by the predicted cleanliness, which is calculated by integrating the temporal behavior parameters of each interference source in the candidate channel's interference source set.

[0046] Specifically, step B6 includes steps B61 to B65: Figure 7 The following is a flowchart illustrating a specific implementation of step B6 in a handheld ground station anti-interference method provided by the present invention: Step B61: For each candidate channel, find the interference sources whose frequency range of occupied bandwidth overlaps with the frequency range of the candidate channel from each current interference source profile, and obtain the interference source set of the candidate channel. The search is based on frequency domain overlap. Unlike step B56, which predicts contamination based on active interference sources, this step searches all current interference source profiles, regardless of activity level. This is because interference sources with unstable activity rhythms that haven't been extrapolated, as well as long-term continuous wave interference, will also pose a threat to the candidate channel in the next transmission time slot and should be included in the interference source set. For example, if a candidate channel has a frequency range of 2434MHz to 2436MHz, and the frequency ranges of an interference source with a center frequency of 2435MHz and a bandwidth of 3MHz and an interference source with a center frequency of 2437MHz and a bandwidth of 4MHz overlap with it, then the interference source set for that candidate channel includes both of these interference sources.

[0047] Step B62: For each current interference source profile in the interference source set, extract the interval between adjacent times in the time sequence recorded in the current interference source profile to determine the duty cycle of the interference source. The meaning of duty cycle is consistent with that of active cycle, both referring to the time interval between two consecutive occurrences of an interference source, extracted here from the time series of the image. For the interference sources with the aforementioned time series t0, t0+12 ms, t0+24 ms, and t0+36 ms, each adjacent interval is 12 ms, and the duty cycle is 12 ms. Unlike step B52, the estimation of duty cycle does not require the uniformity of the intervals to meet a threshold condition, but directly takes the average of each adjacent interval for probabilistic calculation of cleanliness. The averaging process reduces the impact of a single missed observation on the cycle estimation.

[0048] Step B63: Determine the duration of the interference source's occupation based on the duration recorded in the current interference source profile; The duration of occupation is the length of time that the interference source remains in the frequency domain after each occurrence. The image records the duration of each observed event, and the duration of occupation is the average of the most recent durations to suppress random errors from single measurements. For example, if the durations of a certain periodic pulse interference source observed three times recently were 4.1 milliseconds, 3.9 milliseconds, and 4.0 milliseconds, then the duration of occupation is determined to be 4 milliseconds.

[0049] Step B64: Read the most recently observed time from the time sequence recorded in the current interference source profile of the interference source; The most recently observed moment is the reference point for time extrapolation; all calculations regarding future occurrences begin from this moment. This moment is read directly from the end of the time sequence without requiring additional calculations.

[0050] Step B65: Calculate the predicted cleanliness of the candidate channel based on the duty cycle, the occupancy duration, and the most recently observed time; Once the three time parameters are complete, the final step in cleanliness calculation is to perform an overlap analysis between the future occurrence time of each interference source and the next transmission time slot, convert the degree of overlap into a contamination probability, and then combine the contamination probabilities of each interference source to obtain the cleanliness level.

[0051] Specifically, step B65 includes steps B651 to B654: Figure 8 The following is a flowchart illustrating a specific implementation of step B65 in a handheld ground station anti-interference method provided by the present invention: Step B651: Add an integer multiple of the duty cycle to the most recently observed time to obtain each expected occurrence time; wherein, the integer multiple of the duty cycle increases sequentially from one duty cycle until the obtained time exceeds the end time of the next transmission time slot. The expected occurrence times are generated as integer multiples of the most recently observed time t0 plus the duty cycle T, i.e., t0+T, t0+2T, t0+3T, increasing sequentially. Accumulation stops once an expected occurrence time exceeds the end time of the next transmission time slot. For example, with T being 12 milliseconds and the next transmission time slot covering t0+6 milliseconds to t0+26 milliseconds, t0+12 milliseconds and t0+24 milliseconds do not exceed the end time and are retained; t0+36 milliseconds exceeds the end time and accumulation stops. There are a total of two expected occurrence times within this time slot.

[0052] Step B652: For each expected occurrence time falling within the next transmission time slot, the time period starting from that expected occurrence time and continuing for the duration of the occupation is determined as the observed time period of the interference source within the next transmission time slot; The observed time period is the predicted value of the energy coverage time window after the appearance of the interference source. For the expected occurrence time t0+12 milliseconds and the occupation duration of 4 milliseconds, the observed time period is from t0+12 milliseconds to t0+16 milliseconds; for the expected occurrence time t0+24 milliseconds, the observed time period is from t0+24 milliseconds to t0+28 milliseconds. Of the two observed time periods, the former falls entirely within the time slot, while the latter extends beyond the end of the time slot and is truncated to t0+24 milliseconds to t0+26 milliseconds.

[0053] Step B653: Determine the ratio of the overlap duration between the observed time period and the next transmission time slot to the total duration of the next transmission time slot as the probability of the interference source polluting the candidate channel in the next transmission time slot; The contamination probability measures the severity of interference by its time proportion. In the previous example, the total duration of the transmission time slot is 20 milliseconds. The overlap duration of the first observed time slot is 4 milliseconds, and the overlap duration of the second observed time slot after truncation is 2 milliseconds. The two observed time slots do not overlap, and the total overlap duration is 6 milliseconds. The contamination probability is 6 milliseconds divided by 20 milliseconds, which is 0.3. When multiple observed time slots of the same interference source are consecutive within the same time slot, the overlap duration is calculated as a union to avoid double counting.

[0054] Step B654: Multiply 1 by the difference in contamination probability of each interference source in the interference source set to obtain the predicted cleanliness of the candidate channel.

[0055] The predicted cleanliness level is obtained by multiplying the non-contamination probabilities of each interference source together, and the calculation formula is Pc=∏(1-pi).

[0056] Where pi is the probability that the i-th interference source in the interference source set of the candidate channel will contaminate the candidate channel in the next transmission time slot. The occurrence rhythm of each interference source is independent of each other, and whether one interference source contaminates or not does not affect the behavior of other interference sources. Therefore, the non-contamination probabilities of each interference source are synthesized by multiplication. Taking a candidate channel with two interference sources in the interference source set and contamination probabilities of 0.3 and 0.1 respectively as an example, the predicted cleanliness is 0.7 multiplied by 0.9, which is 0.63. The predicted cleanliness of a candidate channel with an empty interference source set is 1.

[0057] In the embodiments corresponding to steps B651 to B654, through the cascaded processing of the above steps, the temporal behavior parameters of each interference source are converted into contamination probabilities and synthesized into a single value. The predicted cleanliness thus obtains a clear probabilistic semantic, namely, the probability that the candidate channel will be completely uncontaminated in the next transmission time slot. This value is between 0 and 1 and can be directly used as input for channel ranking and threshold decision.

[0058] In the embodiments corresponding to steps B61 to B65, a complete quantitative process from the set of interference sources to the predicted cleanliness is implemented. The cleanliness of each candidate channel is jointly determined by the duty cycle, duration of occupation, and the time of most recent observation of each interference source in its set of interference sources. Interference sources with strong rhythms and large duty cycles have a significant effect on lowering the cleanliness, while interference sources with weak rhythms and short durations have a slight effect. The differences in the quality between channels are thus quantitatively characterized, providing the first input for the calculation of the evaluation value in step 103.

[0059] Step B7: Determine the switching cost of each candidate channel based on the frequency interval between the current working channel and each candidate channel.

[0060] The switching cost reflects the time overhead of hardware readjustment. The RF parameters of a handheld ground station consist of the local oscillator frequency, filter channels, and gain status. When switching from one channel to another, the local oscillator needs to be relocked, the filter channels need to be reconfigured when crossing frequency bands, and the gain of the receiver link needs to be reconverged after locking. The time consumption of these steps is related to the size of the frequency interval and needs to be estimated in increments according to the frequency interval.

[0061] Specifically, step B7 includes steps B71 to B76: Figure 9 The following is a flowchart illustrating a specific implementation of step B7 in a handheld ground station anti-interference method provided by the present invention: Step B71: Read the center frequency of the current working channel and the center frequency of each candidate channel respectively; The center frequency of the current working channel is stored in the radio frequency control register, and the center frequency of each candidate channel is stored in the configuration table of the candidate channel set. Both are static configuration data and can be read directly.

[0062] Step B72: For each candidate channel, the absolute value of the difference between the center frequency of the candidate channel and the center frequency of the current working channel is determined as the frequency interval of the candidate channel. Frequency spacing measures the frequency span between two tuning operations. For example, if the center frequency of the current operating channel is 2412MHz and the center frequency of the candidate channel is 2430MHz, the frequency spacing is 18MHz. The larger the frequency spacing, the greater the frequency division ratio that the phase-locked loop needs to change, and the longer the relocking time generally is.

[0063] Step B73: Compare the frequency interval of the candidate channel with the preset range boundary, and determine the range in which the frequency interval falls as the frequency interval range of the candidate channel. The tier boundaries divide the continuous frequency interval into several discrete intervals. For example, the preset tier boundary values ​​can be 10MHz and 30MHz. Adding the distinction between frequency bands, this results in four tiers: Tier 1 (less than 10MHz within the same frequency band), Tier 2 (10MHz to 30MHz within the same frequency band), Tier 3 (greater than 30MHz within the same frequency band), and a cross-band tier when the current working channel and the candidate channel belong to different frequency bands. The reason for tiered rather than continuous calculation is that the relationship between the relocking time of the phase-locked loop and the frequency interval is approximately linear in a small range but exhibits a step characteristic in a large range. Tiered estimation ensures accuracy while avoiding the overhead of point-by-point measurement.

[0064] Step B74: According to the frequency interval of the candidate channel, query the pre-established interval time lookup table to obtain the interval readjustment time of the candidate channel. The gear timing lookup table is calibrated by the production line before the equipment leaves the factory. The calibration method involves selecting several representative frequency points for each gear, measuring the time required for the local oscillator to relock from the center frequency of the current operating channel to each representative frequency point, and taking the envelope value of each gear as the gear readjustment time for that gear. For example, example values ​​for the gear timing lookup table are shown in Table 1 below: Table 1 Gear Time Comparison Table The lookup table is stored in non-volatile memory and can be read directly by the gear index during online operation, with negligible reading time.

[0065] As an optional embodiment of this application, the gear time lookup table can also be replaced by a fitting function. That is, a monotonically increasing piecewise linear function t=150+7.5·Δf microseconds is fitted to the three gears in the same frequency band with the frequency interval as the independent variable, and truncated at 300 microseconds, where Δf is the frequency interval in MHz. The fitting function eliminates the table lookup operation and has continuous transitions without jumps at the gear boundaries, but the accuracy depends on the sufficiency of the calibration data, and the error increases when the frequency interval exceeds the calibration range.

[0066] Step B75: When the frequency interval of the candidate channel is the cross-band interval, the filter channel reconfiguration time is superimposed on the interval readjustment time of the candidate channel; wherein, the filter channel reconfiguration time is the time required to switch the filter channel of the RF front end from the frequency band of the current working channel to the frequency band of the candidate channel. During cross-band switching, the filtering channel configured by the RF front-end to suppress out-of-band interference must be switched from the configuration of the current operating channel to the configuration of the candidate channel. This process involves the operation of the filter band selection switch and the stabilization wait after channel establishment. For example, the filter channel reconfiguration time can be 2600 microseconds, calibrated by production line testing and then fixed in the equipment. Switching within the same frequency band does not change the filter channel configuration, so this time does not need to be added.

[0067] Step B76: Add the automatic gain control convergence time to the position readjustment time after superimposing the filter channel reconfiguration time, or to the position readjustment time when the frequency interval position is a non-cross-band position, and determine the total duration obtained after superposition as the switching cost of the candidate channel; wherein, the automatic gain control convergence time is the time required for the gain of the receiving link to re-converge to stability after the local oscillator frequency is locked.

[0068] After the local oscillator frequency is locked, the signal level arriving at the receiving link changes due to channel switching. Automatic gain control (AGB) needs to readjust the gain to bring the input level of the analog-to-digital converter back to a suitable range. This convergence process can only begin after the phase-locked loop (PLL) is locked. For example, the convergence time of AGB can be 800 microseconds. The switching cost is the sum of the gear readjustment time, the filter channel reconfiguration time, and the convergence time of AGB. The switching cost for a candidate channel in gear 3 within the same frequency band is 300 microseconds plus 800 microseconds, totaling 1100 microseconds. The switching cost for a candidate channel in a cross-band gear is 3800 microseconds plus 2600 microseconds plus 800 microseconds, totaling 7200 microseconds. The maximum switching cost of 7200 microseconds is less than the 10-millisecond physical layer interframe gap, ensuring that any switching action can be completed within the interframe gap.

[0069] In the embodiments corresponding to steps B71 to B76, a switching cost estimation mechanism based on frequency interval levels and production line calibration data is implemented. The switching cost quantifies the time overhead of RF hardware retuning into a value comparable to the inter-frame gap. Cross-band candidate channels incur significantly higher costs than intra-band candidate channels due to filter channel reconfiguration. This difference is reflected in the evaluation value calculation in step 103, enabling the system to prioritize intra-band candidate channels with lower switching costs when cleanliness levels are similar.

[0070] In the embodiments corresponding to steps B1 to B7, through the cascaded processing of the above steps, the discrete interference observation sequence is organized into a profile system based on the interference source. The profile is then transformed into the predicted cleanliness of each candidate channel through time extrapolation and frequency domain matching. The readjustment characteristics of the RF hardware are then organized into a handover cost indexed by the level. The predicted cleanliness and handover cost are output together to step 103. The former answers whether the target channel will be clean in the future, and the latter answers how much time is required to reach the target channel. Together, they constitute the complete basis for the channel handover decision.

[0071] Step 103: Determine the difference between the predicted cleanliness and the normalized handover cost as the channel handover evaluation value of the candidate channel; This step combines two heterogeneous metrics into a single channel handover evaluation value. The predicted cleanliness itself ranges from 0 to 1 and does not require further normalization; what needs normalization is the handover cost. For example, the normalization benchmark can be the duration of the physical layer inter-frame interval, 10 milliseconds. That is, the normalized handover cost is the ratio of the handover cost to 10 milliseconds, reflecting the proportion of the inter-frame interval occupied by the handover action. The formula for calculating the channel handover evaluation value is E = Pc - Tsw / Tgap.

[0072] Where E is the channel handover evaluation value of the candidate channel, Pc is the predicted cleanliness of the candidate channel, Tsw is the handover cost of the candidate channel, and Tgap is the duration of the physical layer inter-frame interval. For example, candidate channel A has a predicted cleanliness of 0.94 and a handover cost of 950 microseconds, so its evaluation value is 0.94 minus 0.095, which is 0.845; candidate channel B has a predicted cleanliness of 0.97 and a handover cost of 7200 microseconds, so its evaluation value is 0.97 minus 0.72, which is 0.25. Although candidate channel B is cleaner, its evaluation value is significantly lower than that of candidate channel A, so the system will prioritize channel A. This synthesis method avoids the high cost of cross-band readjustment for a small improvement in cleanliness, and also avoids staying on a channel with poor cleanliness due to the pursuit of convenient handover.

[0073] Step 104: In response to the decision result of spectrum agility, switch the current working channel to the candidate channel with the highest channel switching evaluation value.

[0074] Spectrum agility refers to a transmission strategy in which both communicating parties synchronously switch from one channel to another at an agreed-upon time to avoid interference. The key to decision-making lies in when to trigger agility. Excessive sensitivity leads to frequent switching when interference is not severe, increasing the risk of loss of synchronization; excessive sluggishness causes continuous service degradation on the disrupted channel. This step employs a dual-threshold hierarchical decision-making process, comparing the predicted cleanliness of the current working channel with two hierarchical thresholds to differentiate between three actions: maintenance, opportunistic agility, and immediate agility. A reverse control link ensures synchronized actions between the two parties during the switch, making the entire spectrum agility process transparent to upper-layer services.

[0075] Specifically, step 104 includes steps D1 to D8: Figure 10 The following is a flowchart illustrating a specific implementation of step 104 in a handheld ground station anti-interference method provided by the present invention: Step D1: Compare the predicted cleanliness of the current working channel with the first classification threshold and the second classification threshold respectively; wherein the first classification threshold is higher than the second classification threshold; The predicted cleanliness of the current working channel is calculated in step 102 in the same way as the candidate channels, and the working channel is also included in the candidate channel set for unified evaluation. For example, the first level threshold can be 0.90, and the second level threshold can be 0.70. These two thresholds divide the cleanliness range into three intervals, each corresponding to one of three different treatment strategies.

[0076] Step D2: When the predicted cleanliness is not lower than the first classification threshold, determine the decision result for channel maintenance; A cleanliness level of 0.90 indicates that the current operating channel is highly likely to remain clean in the next transmission time slot. In this case, any switching is unnecessary overhead, RF parameters remain unchanged, and services continue to transmit as usual. Maintaining this cleanliness level is the system's normal practice, avoiding overreaction to minor interference disturbances.

[0077] Step D3: When the predicted cleanliness is lower than the first classification threshold and not lower than the second classification threshold, determine the decision result of spectrum agility and cache the main target channel as a channel to be executed; When the cleanliness level is between 0.70 and 0.90, interference threats exist but are not yet fatal. The system determines the decision result of spectrum agility, but does not execute it immediately. Instead, it buffers the primary target channel as a channel to be executed, postponing the handover to the next frame interval. The meaning of buffering is that the service data in the current time slot is still transmitted through the original channel, and the handover action does not interrupt the ongoing transmission. At the same time, the system reserves a fallback for possible deterioration. If the cleanliness level recovers to above the first level threshold in the next scan cycle, the buffered channel to be executed is discarded, avoiding an unnecessary handover.

[0078] Step D4: When the predicted cleanliness is lower than the second classification threshold, determine the decision result of spectrum agility and immediately execute spectrum agility; wherein, the classification threshold is set according to the historical switching success rate within a preset statistical period, and the historical switching success rate is the proportion of the number of successful spectrum agility changes within the preset statistical period to the total number of spectrum agility changes. A cleanliness level below 0.70 indicates a high probability of contamination in the next transmission time slot. If services remain on the original channel, they will suffer substantial damage. The system determines the spectrum agility decision and executes it immediately, completing the handover within the inter-frame gap after the current time slot ends. The tiered thresholds are not fixed but are adjusted based on the historical handover success rate within a preset statistical period. For example, the statistical period can be one hour, and the historical handover success rate is the proportion of successful spectrum agility handovers within that hour to the total number of spectrum agility handovers. Successful agility handover means that both parties resume communication on the new channel on time after the handover and receive a response within the verification window. When the historical handover success rate is higher than 95%, it indicates that the environmental prediction and synchronization mechanisms are reliable, and both tiered thresholds are lowered by 0.02, allowing the system to trigger agility earlier and avoid hesitation at the critical cleanliness level. When the historical handover success rate is lower than 85%, both tiered thresholds are raised by 0.02, reducing the number of handovers to avoid the risk of missing synchronization. The thresholds are adjusted in a closed-loop manner based on the historical success rate, ensuring that the decision sensitivity automatically matches the actual quality of the link.

[0079] Step D5: When the decision result is spectrum agility, the candidate channel with the highest channel switching evaluation value is determined as the primary target channel, and the candidate channel with the second highest channel switching evaluation value and a frequency interval greater than the primary target channel is determined as the backup target channel. The primary target channel is the candidate channel with the highest channel handover evaluation value. The backup target channel is not simply the one with the second highest evaluation value, but is constrained by a frequency spacing greater than a preset isolation interval. For example, the preset isolation interval can be 10MHz. The purpose of this constraint is to ensure that the primary and backup candidate channels are sufficiently far apart in the frequency domain, preventing the same interference source from simultaneously covering both with its occupied bandwidth. This ensures that if the primary target channel is contaminated by unforeseen sudden interference, the backup target channel still has a high probability of surviving.

[0080] Step D6: Send a fast-change instruction packet carrying the primary target channel, the backup target channel, and the switching timestamp to the peer device via the reverse control link; The reverse control link is a low-rate control channel sent from the handheld ground station to the peer device. It is independent of service data organization and has strong anti-fading redundancy. The agile command packet carries three items: the primary target channel, the backup target channel, and a handover timestamp. The handover timestamp is the handover execution time agreed upon by both parties, and its value can be 1 millisecond after the start of the inter-frame gap, allowing time for command propagation and preparation by both parties. The handover action begins at this time and can still be completed before the end of the inter-frame gap. The agile command packet is retransmitted twice consecutively at the link layer, with an interval of one control frame cycle between the two retransmissions, ensuring that the peer device receives it correctly with a high probability before the handover timestamp.

[0081] Step D7: The radio frequency parameters are switched to the primary target channel in sync with the switching timestamp and the peer device. The switching of radio frequency parameters is completed within the inter-frame gap of the physical layer frame. Both parties initiate RF parameter switching upon the arrival of the handover timestamp, including local oscillator frequency relocking, filter channel reconfiguration during cross-band switching, and gain reconvergence. The handover action is completed within the interframe gap of the physical layer frame, with an interframe gap duration of 10 milliseconds, exceeding the maximum handover cost of 7200 microseconds. The transmission and reception of service frames are naturally separated by the interframe gap during the handover, and upper-layer services do not experience any interruption. Both ends use the handover timestamp as a common starting point, with the time reference derived from periodic time synchronization signaling on the reverse control link. The synchronization accuracy is maintained within 0.2 milliseconds, ensuring that both parties arrive at the new channel almost simultaneously.

[0082] Step D8: If no response is received from the peer device on the primary target channel within the preset verification window after switching, switch the radio frequency parameters to the backup target channel.

[0083] The verification window is the time period during which the peer device is observed to follow the new channel after a switch. For example, a preset verification window value can be 50 milliseconds, approximately two and a half transmission time slots. During this window, the handheld ground station waits on the primary target channel for the peer device's heartbeat response. Receiving a response confirms successful agility, and communication continues on the primary target channel. If no response is received within the window, it indicates that the peer device likely did not receive the agility command packet, or that the primary target channel was covered by unforeseen sudden interference. In this case, the radio frequency parameters are switched to the backup target channel. The backup target channel is already carried in the agility command packet. If the peer device loses connection on the primary target channel and reaches its own timeout threshold, it will automatically switch to the backup target channel, and the two ends will reconnect on the backup target channel, restoring the link without further negotiation.

[0084] In the embodiments corresponding to steps D1 to D8, a complete spectrum agility mechanism is formed, consisting of dual-threshold hierarchical triggering, primary and backup dual-channel protection, timestamp synchronization, and a verification window as a fallback. Hierarchical triggering balances sensitivity and stability; the primary and backup dual-channel design uses 10MHz frequency isolation to mitigate the risk of single-point prediction errors; timestamp synchronization and inter-frame switching ensure seamless service continuity; and the verification window provides a final recovery measure. This entire mechanism, combined with the prediction capabilities of the aforementioned steps, elevates anti-interference from passive avoidance to active avoidance.

[0085] To verify the technical effectiveness of this embodiment, a comparative experiment was conducted on the telemetry and control link between the handheld ground station and the UAV in a real-world field environment containing three interference sources: continuous wave interference, frequency sweeping interference, and periodic pulse interference. The communication distance was approximately 1.5 kilometers, and the continuous test duration for each method was 2 hours. The experiment compared three methods: ① fixed channel method, which always operates on the initial channel without switching; ② reactive switching method, which selects the candidate channel with the lowest energy based only on the energy value of the current scanning cycle, without establishing an interference source profile; and ③ the spectrum agility method of this embodiment. The evaluation indicators were average uninterrupted duration, handover success rate, and link interruption time percentage. The average uninterrupted duration refers to the average of the continuous service availability time between two link interruptions, and a link interruption refers to an event where the bit error rate exceeds the threshold and lasts for more than one transmission time slot. For example, the experimental results are shown in Table 2 below: Table 2 Comparison of anti-interference performance of different methods The experimental results show that, compared with the reactive handover method, the spectrum agility method in this embodiment increases the average uninterrupted duration by approximately 3.7 times, reduces the link interruption time ratio from 9.7% to 2.2%, a decrease of 7.5 percentage points, and increases the handover success rate from 71.5% to 96.4%. Interruptions in the reactive handover method mainly arise from two aspects: first, the handover decision lags behind interference; second, it indiscriminately jumps to channels with momentarily lower energy, frequently falling into the drift path of frequency sweeping interference. This embodiment avoids contaminated channels in advance by extrapolating the interference source profile, avoids unnecessary cross-band readjustment by assessing handover costs, and mitigates the risk of disconnection due to prediction errors by employing a primary / backup dual-channel mechanism. These three mechanisms correspond to the three types of interruption causes mentioned above.

[0086] In the embodiments corresponding to steps 101 to 104, by periodically scanning the spectrum and collecting multi-dimensional interference observations such as the energy value, interference type, center frequency of the interference signal, and occupied bandwidth of the candidate channel, a fine-grained characterization of the electromagnetic environment is achieved. This overcomes the limitations of traditional schemes that rely solely on a single energy intensity index to perceive interference and cannot distinguish interference patterns, providing a sufficient information foundation for interference situation analysis. The predicted cleanliness calculated based on interference observation reflects the probability that the candidate channel will not be contaminated by any interference source in the next transmission time slot, transforming channel decision-making from reactive remediation to proactive prediction. This allows channel selection to be completed before the interference actually causes link quality degradation, avoiding the problem of damage already occurring due to detection lag in traditional schemes. Incorporating the estimated time required for RF parameters to be readjusted from the current working channel to the candidate channel as a handover cost into the decision overcomes the shortcomings of existing frequency switching mechanisms that only consider the target channel signal quality and ignore the communication interruption introduced by the frequency switching action itself. This avoids blind frequency switching in high handover cost scenarios, suppresses repetitive frequency switching phenomena, and reduces the accumulation of link interruption time during frequency switching. By normalizing the predicted cleanliness and the handover cost and then subtracting them, a unified channel handover assessment value is obtained. This integrates two different dimensions of factors, channel availability and handover cost, into a single decision criterion, taking into account both the long-term stability of link quality and the real-time control of handover costs. This achieves a balance between timeliness and accuracy in spectrum agility decision-making, ultimately improving the anti-interference capability, image transmission continuity, and remote control real-time performance of the handheld ground station data link in complex electromagnetic environments.

[0087] like Figure 2 This invention provides a handheld ground station anti-interference device; please refer to [link / reference]. Figure 2 , Figure 2 A schematic diagram of a handheld ground station anti-interference device provided by the present invention is shown, as follows: Figure 2 The handheld ground station anti-interference device shown includes: The scanning unit 21 is used to perform periodic broadband scanning on a predefined set of candidate channels to obtain interference observations of each candidate channel in each scanning period; wherein, the interference observations include the energy value of the candidate channel, the interference type, the center frequency of the interference signal, and the occupied bandwidth. The calculation unit 22 is used to calculate the predicted cleanliness and handover cost of the candidate channel based on the interference observation; wherein, the predicted cleanliness is the probability that the candidate channel will not be contaminated by any interference source in the next transmission time slot; and the handover cost is an estimate of the time required for the radio frequency parameters to be readjusted from the current working channel to the candidate channel. The determining unit 23 is used to determine the difference between the predicted cleanliness and the normalized switching cost as the channel switching evaluation value of the candidate channel. The switching unit 24 is used to switch the current working channel to the candidate channel with the highest channel switching evaluation value in response to the decision result of spectrum agility.

[0088] This invention provides a handheld ground station anti-interference device that periodically scans the spectrum and collects multi-dimensional interference observations, including the energy value of candidate channels, interference type, interference signal center frequency, and occupied bandwidth. This enables fine-grained characterization of the electromagnetic environment, overcoming the limitations of traditional schemes that rely solely on a single energy intensity index to perceive interference and cannot distinguish interference patterns. It provides a sufficient information foundation for interference situation analysis. The predicted cleanliness calculated based on interference observation reflects the probability that a candidate channel will not be contaminated by any interference source in the next transmission time slot, transforming channel decision-making from reactive remediation to proactive prediction. This allows channel selection to be completed before interference actually causes link quality degradation, avoiding the damage already occurring due to detection lag in traditional schemes. The estimated time required to re-adjust RF parameters from the current working channel to the candidate channel is incorporated as a handover cost into the decision-making process. This overcomes the shortcomings of existing frequency switching mechanisms that only consider the target channel signal quality and ignore the communication interruption introduced by the frequency switching action itself. It avoids blind frequency switching in high-handover-cost scenarios, suppresses repetitive frequency switching, and reduces the accumulation of link interruption time during frequency switching. By normalizing the predicted cleanliness and the handover cost and then subtracting them, a unified channel handover assessment value is obtained. This integrates two different dimensions of factors, channel availability and handover cost, into a single decision criterion, taking into account both the long-term stability of link quality and the real-time control of handover costs. This achieves a balance between timeliness and accuracy in spectrum agility decision-making, ultimately improving the anti-interference capability, image transmission continuity, and remote control real-time performance of the handheld ground station data link in complex electromagnetic environments.

[0089] Figure 3 This is a schematic diagram of a terminal device provided in an embodiment of the present invention. Figure 3As shown, a terminal device 3 in this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30, such as a handheld ground station anti-interference program. When the processor 30 executes the computer program 32, it implements the steps described in the various embodiments of the handheld ground station anti-interference method, for example... Figure 1 Steps 101 to 104 are shown. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each unit in the above-described device embodiments, for example... Figure 2 The function of the unit shown.

[0090] For example, the computer program 32 can be divided into one or more units, which are stored in the memory 31 and executed by the processor 30 to complete the present invention. The one or more units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 32 in the terminal device 3. For example, the specific functions of each unit of the computer program 32 can be divided as follows: A scanning unit is used to periodically scan a predefined set of candidate channels to obtain interference observations for each candidate channel in each scanning period; wherein, the interference observations include the energy value of the candidate channel, the interference type, the center frequency of the interference signal, and the occupied bandwidth. The calculation unit is used to calculate the predicted cleanliness and handover cost of the candidate channel based on the interference observation; wherein, the predicted cleanliness is the probability that the candidate channel will not be contaminated by any interference source in the next transmission time slot; and the handover cost is an estimate of the time required for the radio frequency parameters to be readjusted from the current working channel to the candidate channel. The determining unit is used to determine the difference between the predicted cleanliness and the normalized switching cost as the channel switching evaluation value of the candidate channel. The switching unit is used to switch the current working channel to the candidate channel with the highest channel switching evaluation value in response to the decision result of spectrum agility.

[0091] The terminal device includes, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of a terminal device 3 and does not constitute a limitation on a terminal device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, etc.

[0092] The processor 30 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0093] The memory 31 can be an internal storage unit of the terminal device 3, such as a hard disk or memory of the terminal device 3. The memory 31 can also be an external storage device of the terminal device 3, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device 3. Furthermore, the memory 31 can include both internal and external storage units of the terminal device 3. The memory 31 is used to store the computer program and other programs and data required by the roaming control device. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0094] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0095] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of the present invention. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0096] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0097] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0098] This invention provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0099] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0100] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0101] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0102] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0103] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units.

[0104] It should be understood that, when used in this specification and the appended claims, terms include indicating the presence of the described feature, integral, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0105] It should also be understood that the terms used in this specification and the appended claims refer to any combination of one or more of the associated listed items and all possible combinations, and include such combinations.

[0106] As used in this specification and the appended claims, the term "if" can be interpreted, depending on the context, as when, once, or in response to determination or in response to detection. Similarly, the phrase "if determined" or "if detected [the described condition or event]" can be interpreted, depending on the context, as once determined or in response to determination or detection [the described condition or event] or in response to detection [the described condition or event].

[0107] Furthermore, in the description of this invention and the appended claims, the terms first, second, third, etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0108] References to one or more embodiments described in this specification mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the invention. Therefore, phrases appearing in different parts of this specification as referring to one embodiment, some embodiments, some other embodiments, and others do not necessarily refer to the same embodiment, but rather mean one or more, but not all, embodiments, unless otherwise specifically emphasized. The terms include, encompass, have, and variations thereof mean including, but are not limited to, unless otherwise specifically emphasized.

[0109] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A handheld ground station anti-interference method, characterized in that, The anti-interference method for the handheld ground station includes: A predefined set of candidate channels is periodically scanned with wideband to obtain interference observations for each candidate channel in each scanning period; wherein, the interference observations include the energy value of the candidate channel, the interference type, the center frequency of the interference signal, and the occupied bandwidth. Based on the interference observations, the predicted cleanliness and handover cost of the candidate channel are calculated; wherein, the predicted cleanliness is the probability that the candidate channel will not be contaminated by any interference source in the next transmission time slot; and the handover cost is an estimate of the time required for the radio frequency parameters to be readjusted from the current working channel to the candidate channel. The difference between the predicted cleanliness and the normalized handover cost is determined as the channel handover evaluation value of the candidate channel. In response to the decision result of spectrum agility, the current working channel is switched to the candidate channel with the highest channel switching evaluation value.

2. The anti-interference method for handheld ground stations according to claim 1, characterized in that, The step of periodically scanning a predefined set of candidate channels to obtain interference observations for each candidate channel in each scanning period includes: A predefined set of candidate channels is periodically scanned with wideband, and IQ data is collected for candidate channels whose energy values ​​are higher than a preset energy threshold. The IQ data is subjected to time-domain and frequency-domain analysis to extract interference signal features; wherein, the interference signal features include the concentration of power spectral density in the frequency domain, the direction of change of the center frequency of the power spectrum with the scanning period, and the duty cycle of the pulse power in time; The interference signal features are input into a pre-trained neural network classifier, which outputs the type probability of the interference on the candidate channel belonging to each preset interference type; wherein, the preset interference types include continuous wave interference, frequency sweep interference, periodic pulse interference, and broadband noise interference; The preset interference type with the highest probability is determined as the interference type of the candidate channel in this scanning period; The combination of the candidate channel's energy value, interference type, center frequency of the interference signal, occupied bandwidth, and the probability of the aforementioned type is recorded as the interference observation of the candidate channel during the scanning period.

3. The anti-interference method for handheld ground stations according to claim 1, characterized in that, The step of calculating the predicted cleanliness and handover cost of the candidate channel based on the interference observation includes: Extract the center frequency, occupied bandwidth, and type probability from each interference observation in the current scanning cycle; For each interference observation, calculate the difference between the interference observation and the most recently recorded center frequency, occupied bandwidth, and type probability in each historical interference source profile; wherein, the historical interference source profile is updated based on historical interference source observations; When there is a historical interference source profile with a difference less than a preset association threshold, the interference observation is associated with the historical interference source profile, and the data in the historical interference source profile is updated with the interference observation to obtain the first current interference source profile. When there is no historical interference source profile with a difference less than the preset correlation threshold, the interference observation is recorded as a newly created second current interference source profile; wherein, the interference source profile records the interference type, center frequency, occupied bandwidth and time domain activity pattern of the interference source, and the time domain activity pattern includes the time sequence of the interference source observed in each scanning cycle and the duration of each observation; Based on the temporal activity patterns recorded in the profiles of each current interference source, predict the candidate channels that each interference source will pollute in the next transmission time slot; Calculate the predicted cleanliness of each candidate channel based on the candidate channels contaminated by each interference source in the next transmission time slot. The switching cost of each candidate channel is determined based on the frequency interval between the current working channel and each candidate channel.

4. The anti-interference method for handheld ground stations according to claim 3, characterized in that, The step of predicting the candidate channel contaminated by each interference source in the next transmission time slot based on the temporal activity patterns recorded in the profiles of each current interference source includes: Based on the time sequence in the time-domain activity pattern, calculate the time interval sequence between two consecutive observations of each interference source; When the difference between the maximum and minimum time intervals in the time interval sequence corresponding to the target interference source is less than a preset interval threshold, the mean of the time interval sequence is determined as the active period of the target interference source. Based on the active period and the time when the target interference source was last observed, extrapolate the predicted active time of the target interference source in the next transmission time slot; When the predicted active moment falls into the next transmission time slot, it is determined that the target interference source is in an active state in the next transmission time slot; The interference source in the active state is identified as the active interference source. The difference between the center frequency of the active interference source and half of the occupied bandwidth is identified as the lower boundary of the frequency band. The sum of the center frequency and half of the occupied bandwidth is identified as the upper boundary of the frequency band. Candidate channels whose frequency range intersects with the frequency band between the lower boundary and the upper boundary of the frequency band are predicted as candidate channels contaminated by the active interference source in the next transmission time slot.

5. The anti-interference method for handheld ground stations according to claim 3, characterized in that, The step of calculating the predicted cleanliness of each candidate channel based on the candidate channels contaminated by each interference source in the next transmission time slot includes: For each candidate channel, find the interference sources whose frequency range of occupied bandwidth overlaps with the frequency range of the candidate channel from each current interference source profile, and obtain the interference source set of the candidate channel. For each current interference source profile in the interference source set, extract the interval between adjacent times in the time sequence recorded in the current interference source profile to determine the duty cycle of the interference source. Based on the duration recorded in the current interference source profile, determine the duration of the interference source's occupation. Read the most recently observed time from the time sequence recorded in the current interference source profile of the interference source; The predicted cleanliness of the candidate channel is calculated based on the duty cycle, the occupancy duration, and the most recently observed time.

6. The anti-interference method for handheld ground stations according to claim 5, characterized in that, The step of calculating the predicted cleanliness of the candidate channel based on the duty cycle, the occupancy duration, and the most recently observed time includes: Add an integer multiple of the duty cycle to the most recently observed time to obtain the expected occurrence time; wherein the integer multiple of the duty cycle increases successively from 1 duty cycle until the obtained time exceeds the end time of the next transmission time slot. For each expected occurrence time falling within the next transmission time slot, the time period from that expected occurrence time to the duration of the occupation is determined as the observed period of the interference source in the next transmission time slot; The ratio of the overlap duration between the observed time period and the next transmission time slot to the total duration of the next transmission time slot is determined as the probability that the interference source will contaminate the candidate channel in the next transmission time slot. The predicted cleanliness of the candidate channel is obtained by multiplying 1 by the difference between the contamination probabilities of each interference source in the interference source set.

7. The anti-interference method for handheld ground stations according to claim 3, characterized in that, The step of determining the switching cost of each candidate channel based on the frequency interval between the current working channel and each candidate channel includes: Read the center frequency of the current working channel and the center frequency of each candidate channel respectively; For each candidate channel, the absolute value of the difference between the center frequency of the candidate channel and the center frequency of the current working channel is determined as the frequency interval of the candidate channel. The frequency interval of the candidate channel is compared with the preset range boundary, and the range in which the frequency interval falls is determined as the frequency interval range of the candidate channel. The candidate channel's frequency interval is used to query a pre-established interval time lookup table to obtain the interval readjustment time. When the frequency interval of the candidate channel is the cross-band interval, the filter channel reconfiguration time is superimposed on the interval readjustment time of the candidate channel; wherein, the filter channel reconfiguration time is the time required to switch the filter channel of the RF front end from the frequency band of the current working channel to the frequency band of the candidate channel. The switching cost of the candidate channel is determined by superimposing the reconfiguration time of the filter channel or the reconfiguration time of the frequency interval when it is a non-cross-band interval, and then superimposing the convergence time of the automatic gain control. The total duration obtained by superposition is determined by superimposing the automatic gain control convergence time as the time required for the gain of the receiving link to re-converge to a stable state after the local oscillator frequency is locked.

8. The anti-interference method for handheld ground stations according to claim 1, characterized in that, The step of switching the current working channel to the candidate channel with the highest channel switching evaluation value in response to the decision result of spectrum agility includes: The predicted cleanliness of the current working channel is compared with a first classification threshold and a second classification threshold, wherein the first classification threshold is higher than the second classification threshold. When the predicted cleanliness is not lower than the first classification threshold, the decision result for channel maintenance is determined; When the predicted cleanliness is lower than the first classification threshold and not lower than the second classification threshold, the decision result of spectrum agility is determined, and the main target channel is cached as a channel to be executed; When the predicted cleanliness is lower than the second classification threshold, the decision result of spectrum agility is determined and spectrum agility is executed immediately; wherein, the classification threshold is set according to the historical switching success rate within a preset statistical period, and the historical switching success rate is the proportion of the number of successful spectrum agility changes within the preset statistical period to the total number of spectrum agility changes. When the decision result is spectrum agility, the candidate channel with the highest channel switching evaluation value is determined as the primary target channel, and the candidate channel with the second highest channel switching evaluation value and a frequency interval greater than the primary target channel is determined as the backup target channel. The agile command packet carrying the primary target channel, the backup target channel, and the switching timestamp is sent to the peer device via the reverse control link. The radio frequency parameters are switched to the primary target channel in sync with the switching timestamp and the peer device, and the switching of radio frequency parameters is completed within the inter-frame gap of the physical layer frame; If no response is received from the peer device on the primary target channel within the preset verification window after switching, the radio frequency parameters will be switched to the backup target channel.

9. A handheld ground station anti-interference device, characterized in that, The handheld ground station anti-interference device includes: A scanning unit is used to periodically scan a predefined set of candidate channels to obtain interference observations for each candidate channel in each scanning period; wherein, the interference observations include the energy value of the candidate channel, the interference type, the center frequency of the interference signal, and the occupied bandwidth. The calculation unit is used to calculate the predicted cleanliness and handover cost of the candidate channel based on the interference observation; wherein, the predicted cleanliness is the probability that the candidate channel will not be contaminated by any interference source in the next transmission time slot; and the handover cost is an estimate of the time required for the radio frequency parameters to be readjusted from the current working channel to the candidate channel. The determining unit is used to determine the difference between the predicted cleanliness and the normalized switching cost as the channel switching evaluation value of the candidate channel. The switching unit is used to switch the current working channel to the candidate channel with the highest channel switching evaluation value in response to the decision result of spectrum agility.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps in the anti-interference method for handheld ground stations as described in any one of claims 1 to 8.

Images