Millimeter wave beam tracking method, computer readable storage medium and node equipment
By employing interval analysis and Bayesian filtering theory in millimeter-wave beam tracking, a beam pointing error interval is constructed and iteratively narrowed, thus solving the performance degradation problem caused by channel model mismatch and achieving robust beam tracking in nonlinear non-Gaussian scenarios.
Patent Information
- Application Number
- CN202511373484.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-25
- Publication Date
- 2026-01-20
AI Technical Summary
Existing millimeter-wave beam tracking methods suffer from performance degradation when the precise statistical model of the channel is unknown, especially in nonlinear non-Gaussian scenarios where they lack adaptability.
By employing the interval analysis approach, a beam pointing error interval is constructed. Through Bayesian filtering theory and iterative shrinking of the channel observation interval, a robust tracking of the beam pointing is achieved instead of the traditional error statistical model.
Without increasing computational load, it maintains tracking performance in nonlinear and non-Gaussian scenarios, and is suitable for uncertainties caused by factors such as multipath, analog-to-digital conversion quantization, and hardware nonlinearity errors, thus improving robustness and adaptability.
Smart Images
Figure CN121367521A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application belongs to the technical field of wireless communication, and particularly relates to a millimeter wave beam tracking method, a computer readable storage medium and a node device. BACKGROUND
[0002] In modern wireless communication systems, with the large-scale access of wireless devices, the communication demands of high density, high rate, high throughput and low delay promote the successful development and application of millimeter waves in wireless communication systems. By using the beam forming technology of large-scale array antenna, a high-gain narrow beam can be generated in a specific direction with low transmit power, which ensures the communication quality and improves the system anti-interference ability. At the same time, in this case, to improve the signal gain and maintain a high-quality communication link, the beam pointing of the transceiver node needs to be estimated and adjusted in real time in the time-varying scene. Therefore, the millimeter wave beam tracking technology has become an important research direction.
[0003] The current millimeter wave beam tracking technology mainly includes three methods of codebook training based method, Bayesian filtering based method and machine learning based method. The codebook training based method needs to search the predefined beam codebook for beam training before transmitting information, and select the codebook with the highest channel gain for beam pointing tracking. However, this method requires large training resource overhead, and is not suitable for millimeter wave networks with high-speed node motion. The beam tracking method based on Bayesian filtering is based on Bayesian filtering theory, and iteratively predicts and estimates the beam pointing by constructing a millimeter wave channel model. This method can achieve approximate optimal estimation of beam pointing and has good dynamic adaptability with low resource overhead without beam training process and sample data set through the time correlation of channel parameters and channel model. The existing Bayesian filtering based method mainly includes extended Kalman filter (EKF), unscented Kalman filter (UKF) and particle filter (PF) methods. The beam tracking method based on machine learning completes the tracking of beam pointing through neural network and other methods, which has the characteristics of low complexity and strong robustness. However, this method strongly depends on a large number of sample data sets collected, and the cost of collecting a large number of actual scene sample data is usually high. In addition, in the case where the sample data set cannot cover all the dynamic information of the millimeter wave channel, its performance will rapidly decline, and it is currently not suitable for millimeter wave communication networks in complex environments.
[0004] As described above, the beam tracking method based on Bayesian filtering better balances the relationship between resource overhead and tracking performance, but it needs to know the accurate channel model to realize the iterative correction of beam pointing. In actual application, due to the multipath phenomenon of millimeter wave channel, automatic gain control (AGC) of hardware receiver, analog-to-digital conversion quantization and non-linear error and other factors, the obtained channel observation does not conform to the constructed channel model, but fluctuates within a certain error interval and has a certain degree of "fuzzy" characteristics. In this case, due to the channel model mismatch, the performance of the existing beam tracking method based on Bayesian filtering is degraded. SUMMARY
[0005] The present application aims to solve the technical problem of performance degradation of beam tracking due to unknown accurate statistical model of channel in the prior art. To this end, the present application provides a millimeter wave beam tracking method and a computer readable storage medium and a node device, which can maintain the tracking performance in a nonlinear non-Gaussian scene without increasing the computational load.
[0006] In a first aspect, an embodiment of the present application provides a millimeter wave beam tracking method applied to a current node in a wireless communication system, which includes the following steps:
[0007] At a first time, an initial beam pointing error interval of the node is constructed to represent the error range of the beam pointing, wherein the first time is the time when the current node starts to perform the millimeter wave beam tracking method;
[0008] At a second time, the beam pointing error interval is propagated or predicted by a state transition equation of the beam pointing to obtain a predicted beam pointing error interval;
[0009] According to the channel model of the millimeter wave, the channel observation under the current beam pointing is obtained;
[0010] The predicted beam pointing error interval is contracted according to the channel observation;
[0011] The beam pointing angle is estimated according to the contracted beam pointing error interval, and the contracted beam pointing error interval is transmitted as prior information to the next time;
[0012] The steps after the second time are repeated to perform the operation at multiple times iteratively until a preset tracking time is reached.
[0013] In some embodiments, at the first time, according to the constructed initial beam pointing error interval of the node, the box particles are initialized to generate N box particles with equal weight of 1 / N and mutual disjoint, and each box particle has the same interval length.
[0014] In some embodiments, at the second time, each bin particle is predicted by a state transition equation when the beam pointing error interval is propagated or predicted.
[0015] In some embodiments, after the channel observation under the current beam pointing is obtained, a likelihood function between each bin particle and the channel observation is constructed according to a channel model of the millimeter wave.
[0016] In some embodiments, when the predicted beam pointing error interval is shrunk according to the channel observation, a likelihood function value of each bin particle is calculated according to the real channel observation and the likelihood function, and each bin particle is shrunk.
[0017] In some embodiments, after the shrink, the weight of the bin particle is updated and weight normalization processing is performed; the estimation result of the beam pointing angle is obtained according to the bin particle state and the weight value, and the bin particle state and the weight value are passed to the next time.
[0018] In some embodiments, when the channel observation under the current beam pointing is obtained:
[0019] For each predicted bin particle, a predicted virtual channel observation is generated according to the channel model of the millimeter wave;
[0020] The channel observation at the next time is obtained, the predicted virtual channel observation is corrected to obtain a corrected channel observation interval;
[0021] The corrected channel observation interval is used to calculate the likelihood function value of each bin particle.
[0022] In some embodiments, the shrink means that each bin particle is constrained:
[0023] If the corrected channel observation interval ≠ The bin particle is constrained in size by the corrected channel observation interval and the interval constraint algorithm to obtain a constrained bin particle.
[0024] If the corrected channel observation interval = The constrained bin particle is recorded as the bin particle.
[0025] In a second aspect, the embodiments of the present application provide a computer readable storage medium storing a computer program, which, when executed, implements the steps corresponding to the millimeter wave beam tracking method described above.
[0026] In a third aspect, an embodiment of the present application provides a node device, comprising: at least one processor; and at least one memory and a bus connected with the processor; wherein the processor, the memory and the bus complete mutual communication; the processor is used to call program instructions in the memory to execute steps corresponding to the millimeter wave beam tracking method.
[0027] From the above technical solution, the beneficial effects of the present application are:
[0028] 1. The method of the present application constructs the beam pointing state sample into a beam pointing error interval, which belongs to a controllable interval existing in the sample space, rather than a single sample point. Then, according to the error interval of the channel observation, the channel observation is modeled as an observation interval within the error interval, rather than a single observation point. Then, the channel observation under the current beam pointing is obtained and is contracted. The Bayesian filtering theory and the interval analysis idea are used to perform an iterative contraction operation on the beam pointing interval by the observation interval, to realize segmentation and elimination of the observation interval, and to obtain the current beam pointing by estimating the beam interval. The present application replaces the observation data point with the maximum error with interval data, replaces the traditional error statistical model, avoids the problem of beam tracking performance degradation caused by fuzzy channel observation, and can maintain the tracking performance in a nonlinear non-Gaussian scene without increasing the calculation complexity / load, which is suitable for uncertain and incomplete channel state information introduced by factors such as multipath, analog-to-digital conversion quantization, and hardware nonlinear error.
[0029] 2. The computer readable storage medium of the present application stores the millimeter wave beam tracking method in the form of a program, which facilitates the application of the method, so that it can be applied to various electronic devices, and thus realizes fast tracking of the millimeter wave beam of the communication node with fuzzy channel state information.
[0030] 3. The node device of the present application solidifies the millimeter wave beam tracking method into an executable computer program, and integrates it with the memory and the processor into hardware, which expands the application of the millimeter wave beam tracking method, reduces the system calculation amount while ensuring the tracking performance, and reduces the requirement for hardware. BRIEF DESCRIPTION OF DRAWINGS
[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments description will be briefly introduced one by one as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other embodiments and drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings. In the drawings, various schematic diagrams according to the embodiments of the present application are shown, which are not necessarily drawn to scale, some details are exaggerated for the purpose of clear expression, and some details can be omitted.
[0032] Figure 1 An embodiment process schematic diagram of the millimeter wave beam tracking method of the present application is shown.
[0033] Figure 2 A tracking error comparison diagram of the millimeter wave beam tracking method of the present application and the EKF method when the angle change rate is 0.25° is shown.
[0034] Figure 3 A tracking error comparison diagram of the millimeter wave beam tracking method of the present application and the EKF method when the angle change rate is 0.5° is shown. DETAILED DESCRIPTION
[0035] The technical solutions in the embodiments of the present application will be described clearly and completely in the following detailed description of the embodiments of the present application with reference to the drawings corresponding to the specific embodiments of the present application. The detailed description of the embodiments of the present application provided in the following drawings is not intended to limit the scope of the claimed present application, and the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, various different configurations can be arranged and designed, and all other embodiments obtained by those skilled in the art without any creative effort fall within the scope of the present application.
[0036] The present application will be described in detail below with reference to the drawings and specific embodiments:
[0037] In the first aspect of the present application, a millimeter wave beam tracking method is provided, which is applied to a current node in a wireless communication system. In the wireless communication system, there is also a neighbor node in communication with the current node. The millimeter wave beam tracking method comprises the following steps:
[0038] S1, at a first time, an initial beam pointing error interval of the node is constructed to represent an error range of the beam pointing, quantify the initial uncertainty of the beam pointing, and avoid tracking misalignment caused by point estimation; wherein the first time is the time when the current node starts to execute the millimeter wave beam tracking method; the estimated beam pointing state and channel observation are modeled as sample intervals subject to a certain error range, rather than sample points subject to a certain statistical model by using the idea of interval analysis.
[0039] S2, at the second time, the beam pointing error interval is propagated or predicted by a state transition equation of the beam pointing, to obtain a predicted beam pointing error interval; the beam pointing error interval at the next time is predicted by the state transition equation in combination with a Bayesian filtering method.
[0040] S3, according to a channel model of the millimeter wave, a channel observation under the current beam pointing is obtained; a pilot signal is sent in the current predicted pointing interval, and a channel observation value is extracted by a received signal strength ratio (RSSR) or a signal-to-noise ratio (SNR).
[0041] S4, the predicted beam pointing error interval is shrunk according to the channel observation; a matching degree (likelihood function) of each sub-interval in the predicted interval to the observation value is calculated, the low matching sub-interval is eliminated, and the high matching interval is retained and shrunk. Based on the Bayesian filtering, the beam pointing interval is iteratively constrained and shrunk by the channel observation interval, so that a robust beam tracking performance can be obtained, and the method is suitable for highly nonlinear and non-Gaussian scenarios.
[0042] S5, a beam pointing angle is estimated according to the shrunk beam pointing error interval, the estimated angle takes into account efficiency and robustness, avoids noise interference in single-point estimation, and the shrunk beam pointing error interval is transmitted as prior information to the next time, so that the prediction parameters can be dynamically adjusted.
[0043] S6, the steps (the above-mentioned prediction-observation-shrink-estimate process) after the second time are repeated, so that the operation at multiple times is iterated, the beam pointing error interval of uncertainty is compressed each time, until the preset tracking time is reached, and then the robust beam tracking under the condition of fuzzy channel observation information is completed.
[0044] Most of the existing technologies are only suitable for the case where the channel observation model is accurately known, so that better beam tracking performance can be obtained. In millimeter wave communication, due to factors such as multipath, analog-to-digital conversion quantization, AGC, and hardware nonlinear error, the channel observation received by the communication node does not conform to the constructed channel statistical model, but fluctuates within a certain error range. In this case, due to model mismatch, the traditional beam tracking method will have performance loss. However, the present application models the beam pointing state and the channel observation as sample intervals rather than sample points through the idea of interval analysis, avoids the influence of model mismatch and fuzzy channel observation through iteration of the sample interval, has higher robustness and tracking performance in actual application. Compared with the existing beam tracking method based on EKF, which can only be used in Gaussian nonlinear application scenarios, the present application can be applied to highly nonlinear and non-Gaussian scenarios, and has stronger adaptability.
[0045] In some embodiments, at the first time, according to the constructed node beam pointing error initial interval, the box particles are initialized, N box particles with the same length and with the weight of 1 / N are generated, each box particle is mutually exclusive. By constructing multiple box particles to fit the beam pointing state sample interval, the box particle interval is used to replace multiple statistical sample points in the traditional method, the same tracking effect can be achieved by using fewer particles, the tracking performance is ensured without increasing the calculation load, and the system calculation amount is reduced.
[0046] In some embodiments, at the second time, when the beam pointing error interval is propagated or predicted, each box particle is predicted by the state transition equation.
[0047] In some embodiments, after the channel observation under the current beam pointing is obtained, the likelihood function between each box particle and the channel observation is constructed according to the channel model of the millimeter wave.
[0048] In some embodiments, when the predicted beam pointing error interval is contracted according to the channel observation, the likelihood function value of each box particle is calculated according to the real channel observation and the likelihood function, and each box particle is contracted.
[0049] In some embodiments, after the contraction processing, the weight of the box particle is updated and the weight normalization processing is performed; the estimation result of the beam pointing angle is obtained according to the box particle state and the weight, and the box particle state and the weight are transmitted to the next time. The contraction processing of the beam pointing sample interval is realized through the propagation, constraint and weight update of the box particle.
[0050] In some embodiments, when the channel observation under the current beam pointing is obtained:
[0051] For each predicted box particle, a predicted virtual channel observation is generated according to the channel model of the millimeter wave;
[0052] The channel observation at the next time is obtained, the predicted virtual channel observation is corrected to obtain a corrected channel observation interval;
[0053] The corrected channel observation interval is used to calculate the likelihood function value of each box particle.
[0054] In some embodiments, the contraction means that each box particle is constrained:
[0055] If the corrected channel observation interval ≠ The size of the box particle is constrained by using the corrected channel observation interval and the interval constraint algorithm, and the constrained box particle is obtained;
[0056] If the corrected channel observation interval = The box particle after the constraint is recorded as a box particle.
[0057] Please refer to Figure 1 The method (hereinafter referred to as the method) is used to simulate and model the beam pointing error interval, and the steps are as follows:
[0058] S1, the node starts beam tracking, initializes the beam pointing box particle, at k=0 moment, generates N box particles with the same interval length and no intersection The weight of each box particle is 1 / N. Wherein, N is the number of box particles, which can be set to 4 or more, k is the ordinal number of the marking moment, and i is the ordinal number of the box particle.
[0059] S2, for each box particle, propagate or predict the box particle at k+1 moment through the beam pointing state transition function, and obtain the predicted box particle at k+1 moment as:
[0060]
[0061] Wherein, f(·) is the beam pointing state transition equation, u k is the transition process noise parameter, in this embodiment, the propagation or prediction noise of the box particle is set to zero mean Gaussian noise.
[0062] S31, for each predicted box particle, generate a predicted virtual channel observation according to the channel model:
[0063]
[0064] Wherein, g(·) represents the channel model, [g]([x]) represents the channel observation interval generated by the channel model, and the size is determined by the channel error interval. The channel observation interval is the possible value range of the channel parameter determined by the receiving end based on the actual measurement data (such as pilot signal), and the channel error interval represents the deviation range between the estimation result and the actual channel state, and the two have a close mathematical coupling relationship.
[0065] S32, obtain the channel observation [y k+1 ] at k+1 moment, correct the predicted virtual channel observation generated in the third step to obtain the corrected channel observation interval:
[0066]
[0067] S33, for each particle, calculate its likelihood function value Wherein
[0068]
[0069] Wherein, [x] represents the length of the interval [x], p is the dimension of the box particle, for example, if the channel observation is only the signal strength, then p = 1; j represents an integer in the range of [1, p], representing the number.
[0070] S41, for each box particle, the box particle constraint is performed: if then the interval constraint algorithm is used and the size of the box particle is constrained to obtain the constrained box particle If then In this embodiment, the constraint method adopts the forward and backward constraint method.
[0071] S42, for each box particle, update its particle weight:
[0072]
[0073] As above, p is the dimension of the box particle; j represents an integer in the range of [1, p], representing the number.
[0074] S43, perform particle weight normalization processing:
[0075]
[0076] S51, according to the interval distribution and weight of the box particle, the beam pointing estimation result is obtained:
[0077]
[0078] S52, the particle resampling is performed on the box particle, the particle weight is calculated according to formula (6), the particles are sorted according to their size, the box particles with smaller weight are removed, and the box particles with larger weight are selected for interval segmentation. There are many methods for box particle resampling. A simple method is to normalize the weight to the cumulative probability in the interval [0, 1], generate random numbers to select the box, for example, the probability of selecting the box with weight 0.3 is 30%, select the particle according to the probability and segment according to the minimum particle weight.
[0079] S6, let k = k + 1, return to step S2 to perform beam tracking at the next time, until the tracking time ends k = k end .
[0080] In this embodiment, the beam system includes 2 nodes, each node is equipped with an adjustable beam antenna, each beam antenna is a uniform linear array composed of 10 antenna elements, the antenna can transmit and receive signals, the frequency of the communication signal between the nodes is 35GHz, the spacing between the transmit and receive antenna elements is 0.5 times the beam width, the communication channel model is established as an S-V channel model containing only the direct path, and the observation error interval is the true observation plus 0.5 times the noise power. After tracking, the deviation value of the beam pointing estimation result at each tracking time from the true pointing is calculated, the algorithm is repeatedly run 100 times, the average value is calculated, and the results of the present method and the existing EKF-based beam tracking algorithm are compared as shown in Figure 2 and Figure 3 As shown in the figure, compared with the existing EKF method, the present method can effectively reduce the beam tracking error.
[0081] The second aspect embodiment of the present application provides a computer readable storage medium storing a computer program, which, when executed, implements the steps corresponding to the above-mentioned millimeter wave beam tracking method. If the millimeter wave beam tracking method is implemented in the form of a program function unit and sold or used as an independent product, it can be stored in a computer readable storage medium hardware. Based on such understanding, all or part of the processes of the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of each method in the above-mentioned embodiment can be implemented. When the computer program is executed by the processor, the specific implementation of each step and the generated technical effects are the same as those of the foregoing method embodiments. For brief description, the parts not mentioned in this embodiment can be referred to the corresponding contents in the foregoing method embodiments.
[0082] The third aspect embodiment of the present application provides a node device, which includes at least one processor and at least one memory connected with the processor through a bus; wherein the processor and the memory complete mutual communication through the bus; the processor is used to call the program instruction in the memory to execute the steps corresponding to the above-mentioned millimeter wave beam tracking method. When the processor executes the terminal program, the steps in the above-mentioned method embodiment are implemented, or when the processor executes the terminal program, the functions of each module / unit in the above-mentioned computer readable storage medium are implemented. It should be understood that the node device of each embodiment of the present application can be realized based on the memory and the processor. Each memory is used to store the terminal program for executing the above-mentioned method of the present application, and the processor executes the above-mentioned terminal program, so that the node device realizes the method of each embodiment.
[0083] In some embodiments, the node device described above can be a wireless communication device, a millimeter wave base station, a vehicle-mounted millimeter wave radar terminal, an underwater detector or other mobile terminal, and the node device is not limited to what operating system it carries. Those skilled in the art can understand that the above description does not constitute a limitation on the node device, and can include more or fewer components than the above, or combine certain components, or different components, for example, the node device can also include an input module, an output module, a network access module, a wiring, etc.
[0084] With regard to the specific implementations of the present application, it should be noted that:
[0085] In the description of the present application, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, device or computer readable storage medium including a series of elements not only includes those elements, but also includes other elements not explicitly listed that meet the concept of the present application, or further includes elements inherent to such process, method, device or computer readable storage medium. Without more limitations, the element defined by the statement "comprising a" does not exclude the existence of other elements in the process, method, device or computer readable storage medium including the element.
[0086] In the description of the present application, the description with reference to the terms "some embodiments", "optional embodiments", "examples", "specific examples", "optional examples" or "optional embodiments" means that the specific features, structures, materials or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application, but do not mean that all possible forms of the present application are described and described by these embodiments. In the present specification, the illustrative description of the above terms is not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in the specification.
[0087] The above detailed description together with examples will not be exhaustively described to limit the embodiments disclosed herein; rather, the above description together with examples will be understood in connection to explain the embodiments disclosed herein; the above description together with examples cannot be understood as a limitation on the embodiments disclosed herein. The technical solutions among the embodiments can be combined with each other, but it must be based on the implementation of the ordinary skilled in the art, when the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist, and is not within the protection scope required by the present application. Although the embodiments of the present application have been shown and described, various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and purposes of the present application. It can be understood by those skilled in the art that various other specific changes and combination embodiments made according to the technical inspirations disclosed in the present application without departing from the essence of the present application are still within the protection scope defined by the claims of the present application and equivalent technical solutions.
Claims
1. A millimeter wave beam tracking method applied to a current node in a wireless communication system, characterized in that, The method comprises the following steps: At a first time, an initial beam pointing error interval of a node is constructed to represent an error range of beam pointing, wherein the first time is a time when the current node starts to perform the millimeter wave beam tracking method; At a second time, the beam pointing error interval is propagated or predicted by a state transition equation of beam pointing to obtain a predicted beam pointing error interval; According to a channel model of the millimeter wave, a channel observation under the current beam pointing is obtained; The predicted beam pointing error interval is shrunk according to the channel observation; The beam pointing angle is estimated according to the shrunk beam pointing error interval, and the shrunk beam pointing error interval is transmitted as prior information to the next time; The steps after the second time are repeated to iteratively perform the operation at multiple times until a preset tracking time is reached.
2. The millimeter wave beam tracking method of claim 1, wherein, At the first time, the box particles are initialized according to the constructed initial beam pointing error interval of the node, and N box particles with equal length and mutually exclusive weights of 1 / N are generated.
3. The millimeter wave beam tracking method of claim 2, wherein, At the second time, each box particle is predicted by the state transition equation when the beam pointing error interval is propagated or predicted.
4. The millimeter wave beam tracking method of claim 3, wherein, After the channel observation under the current beam pointing is obtained, a likelihood function between each box particle and the channel observation is constructed according to the channel model of the millimeter wave.
5. The millimeter wave beam tracking method of claim 4, wherein, When the predicted beam pointing error interval is shrunk according to the channel observation, the likelihood function value of each box particle is calculated according to the real channel observation and the likelihood function, and each box particle is shrunk.
6. The millimeter wave beam tracking method of claim 5, wherein, After the shrinkage processing, the weights of the box particles are updated and weight normalization processing is performed; the estimation result of the beam pointing angle is obtained according to the box particle state and weight, and the box particle state and weight are transmitted to the next time.
7. The millimeter wave beam tracking method of claim 5, wherein, When the channel observation under the current beam pointing is obtained: For each predicted box particle, a predicted virtual channel observation is generated according to the channel model of the millimeter wave; The next time channel observation is obtained, the predicted virtual channel observation is corrected to obtain a corrected channel observation interval; The likelihood function value of each box particle is calculated using the corrected channel observation interval.
8. The millimeter wave beam tracking method of claim 7, wherein, The shrinkage means that each box particle is constrained: If the modified Then, the box particle is constrained by using the modified channel observation interval and interval constraint algorithm, and the size of the box particle is constrained to obtain a constrained box particle. If the modified The modified bin particle is then recorded as the bin particle.
9. A computer-readable storage medium, characterized in that, The computer program comprises a computer program which, when executed, implements the steps corresponding to the millimeter wave beam tracking method of any one of claims 1-8.
10. A node device, characterized by It comprises: At least one processor; And at least one memory connected with the processor, bus; wherein the processor, memory complete mutual communication through the bus; The processor is used to call the program instruction in the memory, so as to execute the steps corresponding to the millimeter wave beam tracking method of any one of claims 1-8.