Self-adaptive frequency hopping method and device and readable storage medium

Abstract

The invention provides a self-adaptive frequency hopping method and device and a readable storage medium, and the method comprises the steps: collecting the time sequence state information of a current period of a plurality of channels in a 2.4 GHz frequency band, the time sequence state information comprising a multi-dimensional observation feature corresponding to each channel in the plurality of channels; inputting the time sequence state information of the first n periods stored in the memory into a pre-trained neural network model to obtain interference prediction results of the plurality of channels in m periods including the current period, the neural network model being a one-way time sequence prediction network; generating a frequency hopping sequence according to the time sequence state information of the current period and the interference prediction result; controlling the communicator to perform frequency hopping communication according to the frequency hopping sequence; and storing the time sequence state information of the current period into the memory. According to the method provided by the invention, on the premise of considering the real-time information of the channel, the prediction of multiple steps in the future is output by integrating historical information, and a smoother and more efficient frequency hopping track is realized.

Description

Technical Field

[0001] This application belongs to the field of wireless communication technology, specifically relating to an adaptive frequency hopping method, apparatus, and readable storage medium. Background Technology

[0002] In the 2.4GHz band, multiple wireless communication standards such as Bluetooth, Wi-Fi, and Zigbee coexist, leading to channel congestion and severe interference. To improve anti-interference capabilities, frequency hopping spread spectrum technology is widely used. However, existing adaptive frequency hopping technology cannot predict the changing trends of interference. By the time interference is detected and frequency hopping is triggered, communication has already been lost. In scenarios with rapidly changing interference (such as bursty Wi-Fi data transmission), the system may fall into an oscillating state of "frequent detection-frequent frequency hopping," resulting in low efficiency.

[0003] Therefore, how to provide a more efficient adaptive frequency hopping method is an urgent problem to be solved. Summary of the Invention

[0004] To address the problems existing in the prior art, an adaptive frequency hopping method, apparatus, and computer-readable storage medium are proposed. Using this method, apparatus, and computer-readable storage medium, the aforementioned problems can be solved.

[0005] This application provides the following solutions.

[0006] In a first aspect, this application provides an adaptive frequency hopping method, which is applied to a terminal device, the terminal device being equipped with a memory and a communicator, and the method includes:

[0007] The timing status information of multiple channels in the 2.4 GHz band during the current period is collected. The timing status information includes the multi-dimensional observation features corresponding to each channel in the multiple channels. The timing state information of the first n cycles stored in the memory is input into the pre-trained neural network model to obtain the interference prediction results of multiple channels in m cycles including the current cycle. The neural network model is a unidirectional timing prediction network, where m and n are both integers greater than 0. Based on the timing information of the current cycle and the interference prediction results, a frequency hopping sequence is generated; The controller performs frequency hopping communication according to the frequency hopping sequence; Store the timing status information of the current cycle into the memory.

[0008] In some possible embodiments, the timing state information of the first n cycles stored in the memory is input into a pre-trained neural network model to obtain interference prediction results for multiple channels in m cycles including the current cycle, including: Stack the temporal state information of each of the first n cycles stored in the memory to generate a three-dimensional tensor; By inputting a three-dimensional tensor into a pre-trained neural network model, interference prediction results for multiple channels in m cycles, including the current cycle, are obtained.

[0009] In some possible embodiments, the interference prediction results include predicted interference risk values ​​for multiple channels and interference variation trends for multiple channels. Based on the timing state information of the current period and the interference prediction results, a frequency hopping sequence is generated, including: Eliminate channels with predicted interference risk values ​​greater than preset values ​​from multiple channels to obtain candidate channels; The next-hop channel is determined from the candidate channels based on the timing status information of the current period; Frequency hopping sequences are generated based on the next-hop channel and interference variation trends.

[0010] In some possible embodiments, the frequency hopping sequence includes a sequence of k-bit channels, and the frequency hopping sequence is determined based on the next-hop channel and interference variation trends, including: The first channel in the frequency hopping sequence is determined based on the next hop channel; Based on the interference change trend and the historical frequency hopping sequence in the memory, the subsequent channel in the frequency hopping sequence is generated. The historical frequency hopping sequence includes the frequency hopping sequence of the first k-1 cycles, where k is an integer greater than 0.

[0011] In some possible embodiments, the method further includes: After the communication in the current cycle is completed, the communication data of the current cycle is stored in the memory. The communication data of the current cycle is used to update the neural network model.

[0012] In some possible embodiments, the multidimensional observation features include at least one of the following: Received Signal Strength Indication (RSSI), Packet Error Rate (PER), Retransmission Rate, Channel Occupancy Time Ratio, Recent Frequency Hopping History, and Interference Energy Indicator.

[0013] In some possible embodiments, the neural network model includes a gated recurrent unit neural network model.

[0014] In some possible embodiments, the terminal device includes a Bluetooth device or a Wi-Fi device.

[0015] Secondly, this application provides an adaptive frequency hopping device, which is applied to a terminal device. The terminal device includes a memory and a communicator. The adaptive frequency hopping device includes: The acquisition module is used to acquire the timing status information of the current period of multiple channels in the 2.4 GHz band. The timing status information includes the multi-dimensional observation features corresponding to each of the multiple channels. The prediction module is used to input the timing state information of the previous n cycles stored in the memory into the pre-trained neural network model to obtain the interference prediction results of multiple channels in the m cycles including the current cycle. The neural network model is a unidirectional timing prediction network, where m and n are both integers greater than 0. The sequence generation module is used to generate frequency hopping sequences based on the timing state information of the current period and the interference prediction results; The communication module is used to control the communicator to perform frequency hopping communication according to the frequency hopping sequence; The storage module is used to store the timing status information of the current cycle into the memory.

[0016] Thirdly, this application provides a computer-readable storage medium storing a program that, when executed by a multi-core processor, causes the multi-core processor to perform the aforementioned adaptive frequency hopping method.

[0017] The adaptive frequency hopping method provided in this application generates a frequency hopping sequence by inputting the timing state information of the previous n cycles stored in the memory into the interference prediction result obtained by the pre-trained neural network model, and the timing state information of the current cycle. Under the premise of taking into account the real-time information of the channel, it can output predictions for multiple future steps by comprehensively considering historical information, enabling the decision module to perform sequence optimization with a finite step length, avoiding getting trapped in local optima, and achieving a smoother and more efficient frequency hopping trajectory.

[0018] Other advantages of this application will be explained in more detail with reference to the following description and figures.

[0019] It should be understood that the above description is merely an overview of the technical solution of this application, so as to enable a clearer understanding of the technical means of this application and thus allow for its implementation in accordance with the contents of the specification. To make the above and other objects, features, and advantages of this application more apparent and understandable, specific embodiments of this application are illustrated below. Attached Figure Description

[0020] By reading the detailed description of the exemplary embodiments below, those skilled in the art will understand the advantages and benefits described herein, as well as other advantages and benefits. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. In the drawings: Figure 1 A flowchart illustrating an adaptive frequency hopping method provided in an embodiment of this application; Figure 2 This is a schematic diagram of an adaptive frequency hopping device provided in an embodiment of this application.

[0021] In the accompanying drawings, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Implementation

[0022] Exemplary embodiments of this application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of this application and to fully convey the scope of this application to those skilled in the art.

[0023] In the description of embodiments of this application, it should be understood that terms such as “comprising” or “having” are intended to indicate the presence of the disclosed features, figures, steps, behaviors, components, portions or combinations thereof in this specification, and do not exclude the possibility of the presence of one or more other features, figures, steps, behaviors, components, portions or combinations thereof.

[0024] Unless otherwise stated, " / " means "or". For example, A / B can mean A or B. In this article, "and / or" is merely a way of describing the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A alone, A and B at the same time, and B alone.

[0025] The terms "first," "second," etc., are used only for ease of description to distinguish identical or similar technical features and should not be construed as indicating or implying the relative importance or number of these technical features. Therefore, a feature defined by "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of embodiments of this application, unless otherwise stated, the term "multiple" means two or more.

[0026] It should also be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0027] The adaptive frequency hopping method provided in this application is applied to a terminal device, which includes a memory and a communicator. For example... Figure 1 As shown in the embodiments of this application, an adaptive frequency hopping method includes: S101: Collects the timing status information of multiple channels in the 2.4GHz band during the current period. The timing status information includes the multi-dimensional observation features corresponding to each of the multiple channels.

[0028] It should be noted that, in the embodiments of this application, the terminal device includes Bluetooth devices or WIFI devices, and may also include other communication devices using the 2.4GHz frequency band; this application embodiment does not limit this. The multi-dimensional observation features in the embodiments of this application include one or more of the following: Received Signal Strength Indicator (RSSI), Packet Error Rate (PER), Retransmission Rate, Channel Occupancy Time Ratio, Recent Frequency Hopping History Information, and Interference Energy Indicator; this application embodiment does not limit this. The period in the embodiments of this application can be the scheduling period of the terminal device; this application embodiment does not limit this.

[0029] S102: Input the timing state information of the first n cycles stored in the memory into the pre-trained neural network model to obtain the interference prediction results of multiple channels in m cycles including the current cycle.

[0030] It should be noted that the order of steps S102 and S101 is not limited in this application embodiment. They can be executed sequentially or simultaneously, and this application embodiment does not impose any restrictions on this. In this application embodiment, the first n cycles refer to the n cycles before the current cycle, and the m cycles include the current cycle and the m-1 cycles after the current cycle. In this application embodiment, the neural network model can be a unidirectional time series prediction network, where m and n are both integers greater than 0. The neural network model in this application embodiment can be a neural network model with a Gated Recurrent Unit (GRU) neural network as its core. The GRU neural network is a unidirectional time series prediction network, consisting of a GRU network layer and a fully connected output layer. The GRU network is used to extract the correlation of channel interference states in the time dimension, and the fully connected layer is used to map the time series features to the predicted interference risk values ​​of each channel in future communication cycles. Compared with Long Short-Term Memory (LSTM) networks, the GRU network used in this application embodiment significantly reduces the model parameter size and computational complexity while maintaining the ability to model time series dependencies, making it more suitable for deployment on resource-constrained wireless terminals or embedded platforms.

[0031] In this embodiment, the training data for the neural network model can be derived from the wireless spectrum energy distribution collected in real time or offline, and the WiFi interference features and Bluetooth interference features obtained by scanning the target frequency band at a fixed frequency resolution.

[0032] In its research, the applicant discovered that interference in the 2.4GHz channel exhibits a certain distribution pattern. Specifically, WiFi interference is concentrated in the 2.4GHz ISM band, such as 2420–2460MHz, with a spectrum characterized by a continuous wideband high-energy range and significantly lower energy at non-interference frequencies. Bluetooth interference often employs frequency hopping mechanisms, exhibiting short-duration, low- to medium-energy pulses at multiple discrete frequency points, with both time and frequency distributions being relatively dispersed. The neural network model in the embodiments of this application can learn these characteristics, thereby avoiding different types of interference and achieving a smoother and more efficient frequency hopping trajectory.

[0033] In practical applications, in order to accelerate the computation speed of the neural network model, the embodiments of this application can stack the time-series state information of each of the first n cycles stored in the memory to generate a three-dimensional tensor, and then input the three-dimensional tensor into the pre-trained neural network model to obtain the interference prediction results of multiple channels in m cycles including the current cycle.

[0034] As an example, in each embodiment of this application, multi-dimensional observation features can be collected for each of the multiple channels in each period to form a state matrix at time t, resulting in:

[0035] Where A is the number of channels, and F is the number of observed features. Fold the timing state information for the t-th cycle.

[0036] To capture the temporal correlation of interference, this embodiment of the application can stack the states of the first n cycles to form the model input:

[0037] That is, a three-dimensional tensor

[0038] In practical applications, the selection of the time length for n periods can follow these principles: a. Sufficient to cover the duty cycle of typical interference sources (such as Wi-Fi / Bluetooth beacon / zigbee cycles). b. Control the complexity of the neural network model to avoid the computational and storage pressure caused by excessively long windows; c. Modeling that takes into account both short-term sudden disturbances and medium-term disturbance trends.

[0039] In practice, n can be 20 to 50 time steps.

[0040] S103: Generate a frequency hopping sequence based on the timing status information of the current cycle and the interference prediction results.

[0041] As one possible implementation, the interference prediction results in this application embodiment include interference risk prediction values ​​for multiple channels and interference change trends for multiple channels. The interference risk prediction value indicates the risk of interference to a channel. The interference risk prediction value can be a one-dimensional vector, where each element corresponds to an interference risk score or probability value for a channel; a larger value indicates a higher probability that the channel will be interfered with in future periods. The interference change trend indicates the interference change trend of a channel over m periods. The method provided in this application embodiment can first exclude channels with interference risk prediction values ​​greater than a preset value from multiple channels to obtain candidate channels. Then, based on the timing status information of the current period, the next-hop channel is determined from the candidate channels. Thus, this application can generate a complete frequency hopping sequence based on the next-hop channel and the interference change trend.

[0042] In practical applications, the frequency hopping sequence in this embodiment may include a sequence of k channels. This embodiment determines the first channel in the frequency hopping sequence based on the next hop channel, and generates subsequent channels in the frequency hopping sequence based on interference variation trends and historical frequency hopping sequences in memory. The historical frequency hopping sequence includes the frequency hopping sequences of the previous k-1 periods, where k is an integer greater than 0. It should be noted that using historical frequency hopping sequences in this embodiment can avoid a significant impact of a single frequency hopping sequence on the client when the frequency hopping sequence is updated frequently, such as every period, thus maintaining a smoother frequency hopping trajectory for the client.

[0043] S104: Control the communicator to perform frequency hopping communication according to the frequency hopping sequence.

[0044] It should be noted that the frequency hopping sequence in this embodiment can be updated in every cycle, or once every two or more cycles; this embodiment does not limit this. In this embodiment, the communicator is controlled to perform frequency hopping communication according to the first channel of the frequency hopping sequence within the current cycle. In the next cycle, if the frequency hopping sequence has not been updated, then frequency hopping communication can be performed according to the second channel of the frequency hopping sequence. However, if the frequency hopping sequence has been updated in the next cycle, then the communicator will perform frequency hopping communication according to the first channel of the updated frequency hopping sequence.

[0045] S105: Store the timing status information of the current cycle into the memory.

[0046] It should be noted that the embodiments of this application do not limit the order of steps S105 and S102-S104. Figure 1The order relationship shown is merely illustrative. In practical applications, the frequency hopping sequence corresponding to the current cycle can also be stored in the memory. The frequency hopping sequence corresponding to the current cycle can be used as a reference for generating subsequent frequency hopping sequences, as detailed in the embodiment of step S103, which will not be repeated here. As a possible implementation, after the communication of the current cycle is completed, the communication data of the current cycle can also be stored in the memory, and the communication data of the current cycle can be used for subsequent updates to the neural network model.

[0047] The adaptive frequency hopping method provided in this application generates a frequency hopping sequence by inputting the timing state information of the previous n cycles stored in the memory into the interference prediction result obtained by the pre-trained neural network model, and the timing state information of the current cycle. The historical frequency hopping sequence can be combined with the historical information to output predictions for the next multiple steps while taking into account the real-time channel information. This enables the decision module to perform sequence optimization with a finite step size, avoid getting trapped in local optima, and achieve a smoother and more efficient frequency hopping trajectory.

[0048] In the description of this specification, references to terms such as "some possible implementations," "some implementations," "example," "specific example," or "some examples" indicate that a specific feature, structure, material, or characteristic described in connection with that implementation or example is included in at least one implementation or example of this application, and the aforementioned terms do not necessarily refer to the same implementation or example. Furthermore, the described specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more implementations or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different implementations or examples described in this specification, as well as the features of different implementations or examples.

[0049] The method flowcharts for embodiments of this application describe certain operations as different steps performed in a certain order. Such flowcharts are illustrative and not restrictive. Some steps described herein may be grouped together and performed in a single operation, or some steps may be divided into multiple sub-steps, and some steps may be performed in an order different from that shown herein. The various steps shown in the flowcharts may be implemented in any way by any circuit structure and / or tangible mechanism (e.g., by software running on a computer device, hardware (e.g., logic functions implemented by a processor or chip), and / or any combination thereof).

[0050] Those skilled in the art will understand that in the methods described in the above specific embodiments, the order in which the steps are written does not imply a strict execution order, and the specific execution order of each step should be determined by its function and possible internal logic.

[0051] Based on the adaptive frequency hopping method provided in the above embodiments, this application also provides an adaptive frequency hopping device.

[0052] The adaptive frequency hopping device provided in this application embodiment is applied to a terminal device, which includes a memory and a communicator. For example... Figure 2 As shown, the adaptive frequency hopping device includes: The acquisition module 100 is used to acquire the timing status information of the current period of multiple channels in the 2.4 GHz band. The timing status information includes the multi-dimensional observation features corresponding to each of the multiple channels.

[0053] The prediction module 200 is used to input the timing state information of the first n cycles stored in the memory into the pre-trained neural network model to obtain the interference prediction results of multiple channels in the m cycles including the current cycle. The neural network model is a unidirectional timing prediction network, where m and n are both integers greater than 0.

[0054] The sequence generation module 300 is used to generate a frequency hopping sequence based on the timing state information of the current period and the interference prediction results.

[0055] The communication module 400 is used to control the communicator to perform frequency hopping communication according to the frequency hopping sequence.

[0056] The first storage module 500 is used to store the timing status information of the current cycle into the memory.

[0057] It should be noted that the multidimensional observation features in this embodiment include at least one of the following: Received Signal Strength Indication (RSSI), Packet Error Rate (PER), Retransmission Rate, Channel Occupancy Time Ratio, Recent Frequency Hopping History, and Interference Energy Indicator. The neural network model includes a gated recurrent unit neural network model. The terminal device includes a Bluetooth device or a Wi-Fi device.

[0058] As one possible implementation, the prediction module 200 is used to stack the timing state information of each of the previous n cycles stored in the memory to generate a three-dimensional tensor; input the three-dimensional tensor into a pre-trained neural network model to obtain the interference prediction results of multiple channels in m cycles including the current cycle.

[0059] As one possible implementation, the interference prediction results include the interference risk prediction values ​​of multiple channels and the interference change trends of multiple channels. The sequence generation module 300 includes: an exclusion module, used to exclude channels with interference risk prediction values ​​greater than preset values ​​from multiple channels to obtain candidate channels; a determination module, used to determine the next-hop channel from the candidate channels based on the timing status information of the current period; and a generation module, used to generate a frequency hopping sequence based on the next-hop channel and the interference change trend.

[0060] As one possible implementation, the generation module is used to determine the first channel in the frequency hopping sequence based on the next hop channel; and to generate subsequent channels in the frequency hopping sequence based on the interference change trend and the historical frequency hopping sequence in the memory, wherein the historical frequency hopping sequence includes the frequency hopping sequence of the first k-1 cycles, where k is an integer greater than 0.

[0061] As one possible implementation, the adaptive frequency hopping device provided in this application embodiment further includes a second storage module, which is used to store the communication data of the current cycle into the memory after the communication of the current cycle is completed, and the communication data of the current cycle is used to update the neural network model.

[0062] It should be noted that the adaptive frequency hopping device in the embodiments of this application can implement the various processes of the aforementioned adaptive frequency hopping method embodiments and achieve the same effect and function, which will not be repeated here.

[0063] According to some embodiments of this application, an adaptive frequency hopping device according to an embodiment of this application is provided for performing... Figure 1 The adaptive frequency hopping method shown includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method described in the above embodiments.

[0064] According to some embodiments of this application, a non-volatile computer storage medium for an adaptive frequency hopping method is provided, having stored computer-executable instructions configured to execute, when run by a processor, the method described in the above embodiments.

[0065] Computer-readable media include permanent and non-permanent, removable and non-removable media, which can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory, read-only memory, electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device. Furthermore, although the operations of the methods of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally, certain steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple sub-steps.

[0066] While the spirit and principles of this application have been described above with reference to several specific embodiments, it should be understood that this application is not limited to the disclosed specific embodiments, and the division of aspects does not imply that features in these aspects cannot be combined. This application is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. An adaptive frequency hopping method, characterized in that, The method is applied to a terminal device, the terminal device being provided with a memory and a communicator, and the method includes: The timing status information of multiple channels in the 2.4 GHz band during the current period is collected. The timing status information includes the multi-dimensional observation features corresponding to each of the multiple channels. The timing state information of the first n cycles stored in the memory is input into the pre-trained neural network model to obtain the interference prediction results of the multiple channels in the m cycles including the current cycle. The neural network model is a one-way timing prediction network, and m and n are both integers greater than 0. Based on the timing status information of the current period and the interference prediction results, a frequency hopping sequence is generated; Control the communicator to perform frequency hopping communication according to the frequency hopping sequence; The timing status information of the current cycle is stored in the memory.

2. The method according to claim 1, characterized in that, The step of inputting the timing state information of the previous n cycles stored in the memory into a pre-trained neural network model to obtain interference prediction results for the multiple channels in m cycles including the current cycle includes: Stack the temporal state information of each of the first n cycles stored in the memory to generate a three-dimensional tensor; The three-dimensional tensor is input into a pre-trained neural network model to obtain interference prediction results for the multiple channels in m cycles, including the current cycle.

3. The method according to claim 1, characterized in that, The interference prediction results include the predicted interference risk values ​​for the multiple channels and the interference change trends for the multiple channels. The step of generating a frequency hopping sequence based on the timing information of the current period and the interference prediction results includes: Channels with predicted interference risk values ​​greater than a preset value are excluded from the plurality of channels to obtain candidate channels; The next-hop channel is determined from the candidate channels based on the timing status information of the current period; A frequency hopping sequence is generated based on the next-hop channel and the interference change trend.

4. The method according to claim 3, characterized in that, The frequency hopping sequence includes a sequence of k-bit channels. Determining the frequency hopping sequence based on the next-hop channel and the interference variation trend includes: The first channel in the frequency hopping sequence is determined based on the next hop channel; Based on the interference change trend and the historical frequency hopping sequence in the memory, the subsequent channel in the frequency hopping sequence is generated. The historical frequency hopping sequence includes the frequency hopping sequence of the first k-1 cycles, where k is an integer greater than 0.

5. The method according to claim 1, characterized in that, The method further includes: After the communication of the current cycle is completed, the communication data of the current cycle is stored in the memory, and the communication data of the current cycle is used to update the neural network model.

6. The method according to claim 1, characterized in that, The multidimensional observation features include at least one of the following: Received Signal Strength Indicator (RSSI), Packet Error Rate (PER), Retransmission Rate, Channel Occupancy Time Ratio, Recent Frequency Hopping History, and Interference Energy Indicator.

7. The method according to claim 1, characterized in that, The neural network model includes a gated recurrent unit neural network model.

8. The method according to any one of claims 1-7, characterized in that, The terminal device includes a Bluetooth device or a WIFI device.

9. An adaptive frequency hopping device, characterized in that, The adaptive frequency hopping device is applied to a terminal device, which includes a memory and a communicator. The adaptive frequency hopping device comprises: The acquisition module is used to acquire the timing status information of the current period of multiple channels in the 2.4 GHz band. The timing status information includes the multi-dimensional observation features corresponding to each of the multiple channels. The prediction module is used to input the timing state information of the previous n cycles stored in the memory into a pre-trained neural network model to obtain the interference prediction results of the multiple channels in the m cycles including the current cycle. The neural network model is a one-way timing prediction network, and m and n are both integers greater than 0. The sequence generation module is used to generate a frequency hopping sequence based on the timing state information of the current period and the interference prediction result; A communication module is used to control the communicator to perform frequency hopping communication according to the frequency hopping sequence; The first storage module is used to store the timing state information of the current period into the memory.

10. A computer-readable storage medium storing a program that, when executed by a multi-core processor, causes the multi-core processor to perform the method as described in any one of claims 1-8.

Images