RIS codebook determination method, device and equipment

Abstract

The invention discloses an RIS codebook determination method, apparatus and device. The method comprises the steps of obtaining an initial fingerprint database corresponding to an initial RIS codebook set; determining a candidate RIS codebook set in the initial RIS codebook set based on the non-uniformly quantized RSS information entropy of the RSS data of each reference node; based on a genetic simulated annealing algorithm, a target RIS codebook set is determined from the candidate RIS codebook set, and each RIS codebook in the target RIS codebook set is used for determining the position information of the user equipment in the positioning area. According to the method, the final target RIS codebook set is determined by introducing two-stage RIS codebook selection, and introduction of RIS configuration with small RSS difference is effectively avoided, so that when the target RIS codebook set is used for positioning, multiple times of information interaction and RSS measurement of the AP and the UE are not needed, and the positioning precision and the positioning time can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of wireless communication technology, and in particular to a method, device and equipment for determining a RIS codebook. Background Art

[0002] Wireless fingerprint positioning technology is an effective positioning technology. In order to improve the positioning accuracy of fingerprint positioning technology in complex scenarios, the emerging technology that can control the propagation path of radio waves, namely the reconfigurable intelligent surface (RIS), is used to enhance the strength and stability of the received signal through directional control of the reflected signal.

[0003] At present, in RIS-assisted fingerprint positioning technology, there are mainly two methods to improve positioning accuracy. One is to optimize the RIS reflection coefficient to make the difference in RSS values ​​in different areas more significant. In addition, some RIS codebook configurations are generated in advance during the offline fingerprint data collection stage. After the fingerprint data collection is completed, some search methods are applied to screen and record RIS codebooks with better positioning effects from multiple fixed RIS codebook configurations. In the online positioning process, only these better RIS codebooks need to be measured to complete the positioning.

[0004] However, the first method often requires multiple information exchanges and RSS measurements between the AP and the UE to obtain a better positioning effect, and the positioning takes longer time. For the second method, the codebook generated in advance is not specially designed for the environment, and some RIS configurations with very small RSS differences will inevitably be introduced, which will increase the overhead when they are included in the search range, thereby reducing the positioning efficiency and affecting the accuracy of actual positioning. Summary of the invention

[0005] The problem solved by this application is how to determine an effective RIS codebook set to improve positioning accuracy and efficiency.

[0006] To solve the above problems, in a first aspect, the present invention provides a method for determining a RIS codebook, the method comprising:

[0007] Acquire an initial fingerprint database corresponding to the initial RIS codebook set, wherein the initial fingerprint database includes RSS data corresponding to each RIS codebook in the initial RIS codebook set at each reference node in the positioning area;

[0008] Based on the non-uniformly quantized RSS information entropy of the RSS data of each of the reference nodes, a candidate RIS codebook set is determined in the initial RIS codebook set, and the difference between the RSS information entropies corresponding to the RIS codebooks in the candidate RIS codebook set meets the set threshold;

[0009] Based on the genetic simulated annealing algorithm, a target RIS codebook set is determined from the candidate RIS codebook set, and each RIS codebook in the target RIS codebook set is used to determine the location information of the user equipment in the positioning area.

[0010] Optionally, the RIS codebook determination method provided by the present application, after determining the target RIS codebook configuration set, the method further includes:

[0011] Obtain target RSS data corresponding to each RIS codebook in the target RIS codebook set of the user equipment to be located in the positioning area;

[0012] The target RSS data is compared with the RSS data in the initial fingerprint database to determine the location information of the user equipment.

[0013] Optionally, the RIS codebook determination method provided by the present application, wherein the determining of the candidate RIS codebook set in the initial RIS codebook set based on the non-uniformly quantized RSS information entropy of the RSS data of each of the reference nodes comprises:

[0014] Determine the non-uniformly quantized RSS information entropy of each RIS codebook in the initial RIS codebook set at each reference node to obtain a corresponding information entropy vector;

[0015] The information entropy vectors are sorted to determine the RIS codebooks corresponding to the RSS information entropy exceeding the set threshold, and the candidate RIS codebook set is obtained.

[0016] Optionally, the RIS codebook determination method provided by the present application, wherein determining the non-uniformly quantized RSS information entropy of each RIS codebook in the initial RIS codebook set at each reference node comprises:

[0017] Based on the maximum RSS data value and the minimum RSS data value in the initial fingerprint database, normalizing each RSS data in the initial fingerprint database to obtain a normalized result of each RSS data;

[0018] Based on the non-uniform quantization function, each normalization result is standardized to obtain the standardized RSS data;

[0019] Calculate the frequency value of each interval of the RSS data after standardization within the frequency range;

[0020] According to the frequency value, the information entropy of the non-uniform quantized RSS of each reference node is calculated by a set function.

[0021] Optionally, in the RIS codebook determination method provided by the present application, the non-uniform quantization function is expressed as follows:

[0022]

[0023] in, Indicates the normalized result of each RSS data. represents the RSS data of the i-th RIS codebook after normalization, and the variable μ represents the compression constant in the μ-law compression.

[0024] Optionally, in the RIS codebook determination method provided by the present application, the functional formula is:

[0025]

[0026] in, is the frequency value of the RSS data corresponding to the i-th codebook belonging to the t-th interval.

[0027] Optionally, in the RIS codebook determination method provided by the present application, each RSS data set corresponding to the RIS codebook in the candidate RIS codebook set is a new fingerprint database, and determining the target RIS codebook set from the candidate RIS codebook set based on the genetic simulated annealing algorithm includes:

[0028] Performing genetic algorithm chromosome encoding processing on each RSS data in the new fingerprint database to obtain an initialized population;

[0029] Performing intermediate processing on the initialized population data to obtain intermediate processed population data;

[0030] The population data after the intermediate processing is subjected to simulated annealing processing to obtain a target RSS database, and the RIS codebook corresponding to each RSS data in the RSS database is used as the target RIS codebook.

[0031] Optionally, in the RIS codebook determination method provided by the present application, the performing simulated annealing processing on the intermediate processed population data comprises:

[0032] Based on the initial temperature and the annealing rate, the intermediate processed population data is randomly flipped to generate a flipped RSS database;

[0033] Calculate the acceptance probability of each RSS data in the flip RSS database;

[0034] Based on the annealing relationship, the flipped RSS database is iteratively annealed until the set number of iterations or temperature threshold is reached.

[0035] In a second aspect, the present application provides a RIS codebook determination device, the device comprising:

[0036] An acquisition module is used to acquire an initial fingerprint database corresponding to an initial RIS codebook set, wherein the initial fingerprint database includes RSS data corresponding to each RIS codebook in the initial RIS codebook set at each reference node in the positioning area;

[0037] A first selection module is used to determine a candidate RIS codebook set in the initial RIS codebook set based on the non-uniformly quantized RSS information entropy of the RSS data of each reference node, wherein the difference between the RSS information entropies corresponding to the RIS codebooks in the candidate RIS codebook set meets a set threshold;

[0038] The second selection module is used to determine a target RIS codebook set from the candidate RIS codebook set based on a genetic simulated annealing algorithm, wherein each RIS codebook in the target RIS codebook set is used to determine the location information of the user equipment in the positioning area.

[0039] In a third aspect, the present application provides a computer device, comprising: 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 RIS codebook determination method as described in the first aspect.

[0040] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0041] The RIS codebook determination method, device and equipment provided in the present application introduce a two-stage RIS codebook selection method, that is, firstly, based on the non-uniformly quantized RSS information entropy, the RIS codebook corresponding to the RSS data with sufficient difference in information entropy is selected from the initial fingerprint database corresponding to the initial RIS codebook set to construct a new initial fingerprint database, and then in the second stage, the supervised learning feature selection method is used to search and simulate the RSS data in the initial fingerprint database, and finally select the best RIS codebook configuration subset, that is, the target RIS codebook set. Therefore, in the RIS codebook set determination process of the present application, the RIS codebook set determined by the two-stage selection can effectively avoid the introduction of some RIS configurations with very small RSS differences, thereby narrowing the search range and ultimately improving the positioning accuracy. Even if the target RIS codebook set is used to locate the user equipment in the positioning area, since the RIS codebook is selected in the offline fingerprint data collection stage, only the RSS data corresponding to the target RIS codebook set needs to be collected in the positioning stage, thereby reducing the positioning delay, and no multiple information interactions and RSS measurements between the AP and the UE are required, which ultimately improves the positioning efficiency and positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A flowchart of a RIS codebook determination method according to some embodiments of the present application;

[0044] Figure 2 A flowchart of a RIS codebook determination method according to some other embodiments of the present application;

[0045] Figure 3 A flowchart of a RIS codebook determination method according to some further embodiments of the present application;

[0046] Figure 4 A schematic diagram of the structure of a RIS codebook determination device in some embodiments of the present application;

[0047] Figure 5 A schematic diagram of the structure of a computer device provided for some embodiments of the present application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0049] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0050] It can be understood that for a positioning system based on cellular network technology, user positioning can be achieved using methods such as received signal strength (RSS), arrival time, and arrival angle. Among them, RSS fingerprint positioning technology is widely used because it is easy to obtain RSS information.

[0051] RSS fingerprint positioning is a positioning technology based on wireless signal characteristics, which is widely used in indoor environments. The main idea is to build a fingerprint database by measuring the wireless signal characteristics of a specific location (such as received signal strength RSS, channel state information CSI, etc.). When the device location needs to be determined, the real-time measured signal characteristics are matched with the fingerprint in the database to estimate the actual location of the device.

[0052] In practice, wireless fingerprint positioning is generally divided into two stages:

[0053] Offline phase (fingerprint construction phase): In this phase, multiple reference points are first divided in a predetermined area, and the signal characteristics of each wireless access point (such as Wi-Fi hotspots or base stations) are measured at each reference point to generate a fingerprint library. The position of each reference point and the corresponding signal characteristics form a pair of data. Since RSS signals are greatly affected by environmental changes (such as obstructions, reflections, etc.), the fingerprint library needs sufficient density and coverage to ensure positioning accuracy.

[0054] Online stage (position estimation stage): When positioning, the device measures the characteristics of the wireless signal it receives and matches these characteristics with the data in the fingerprint library. Common matching methods include K-Nearest Neighbor (KNN) and weighted KNN, which determine the reference point or area where the device is most likely to be located based on similarity.

[0055] It can also be understood that the intelligent metasurface (RIS) is an emerging electromagnetic wave control technology. It is a two-dimensional structure composed of a large number of small, adjustable units that can dynamically control the reflection, transmission and absorption behavior of incident electromagnetic waves by changing their electromagnetic properties. It can realize functions such as beam forming, beam deflection, and suppression of multipath effects.

[0056] Moreover, unlike traditional ultra-large-scale MIMO wireless communication systems, RIS can be independent of the surrounding environment. By controlling the reflection coefficient of each unit in real time, it can actively change the wireless channel and directionally reflect, refract or transmit the incident electromagnetic wave, thereby obtaining a strong coverage capability. RIS also has the advantages of passivity, extremely low thermal noise, software customization and simplicity, unique nanosecond-level low latency, and full-band response. Compared with the technology currently used in wireless networks, the significant feature of RIS is that the wireless transmission environment can be artificially controlled, and the electromagnetic response in the entire network can be fully controlled, which has a revolutionary impact on 6G wireless networks. Therefore, as a possible key technology for 6G, RIS provides new degrees of freedom to further enhance performance due to its advantages of low power consumption, low cost, wide coverage, wide access, and fast response.

[0057] It can be understood that in the present application, in order to solve the problem in the related art that multiple groups of UE measurements are required and RIS optimization increases the positioning delay, a two-stage RIS codebook determination method is introduced, and a RIS codebook configuration with sufficient signal difference is selected to build an initial fingerprint database, and a supervised learning feature selection method is used to search for the best RIS codebook configuration subset to improve the positioning accuracy. In addition, in a scenario where only a single base station exists, the fingerprint positioning process in this scenario becomes possible by introducing RIS.

[0058] The method may be executed by a processing device having a data processing function, and the user device may be an electronic device such as a computer, a mobile phone, a tablet computer, etc. within the positioning area.

[0059] In order to better understand the RIS codebook determination method provided in the embodiment of the present application, it is described in detail below with reference to the accompanying drawings.

[0060] Figure 1 FIG. 4 is a flow chart of a method for determining a RIS codebook in some embodiments. Figure 1 As shown, the method specifically includes:

[0061] S110, obtaining an initial fingerprint database corresponding to the initial RIS codebook set, wherein the initial fingerprint database includes RSS data corresponding to each RIS codebook in the initial RIS codebook set at each reference node in the positioning area.

[0062] S120, based on the non-uniformly quantized RSS information entropy of the RSS data of each reference node, determine a candidate RIS codebook set in the initial RIS codebook set, and the difference between the RSS information entropies corresponding to the RIS codebooks in the candidate RIS codebook set meets the set threshold.

[0063] S130: Determine a target RIS codebook set from the candidate RIS codebook set based on a genetic simulated annealing algorithm, wherein each RIS codebook in the target RIS codebook set is used to determine location information of a user equipment in the positioning area.

[0064] Specifically, in an embodiment of the present application, first, offline initial fingerprint data corresponding to the initial RIS codebook set can be collected, that is, a reference node (Reference Point, RP) with a known position can be set in a positioning area, such as an indoor positioning area, and then the RSS data corresponding to each different RIS codebook in the initial RIS codebook set is measured and obtained to obtain an initial fingerprint database corresponding to the initial RIS codebook set.

[0065] For example, an initial RIS codebook set including q different RIS codebooks may be constructed first, and then the RIS codebooks may be switched at L different points in the positioning area, ie, reference nodes, and RSS data may be collected and then stored in an L×q database.

[0066] Among them, the RSS of the i-th RIS codebook collected at grid point l can be defined as

[0067] Furthermore, non-uniformly quantized RSS information entropy can be introduced, and a candidate RIS codebook set can be determined in the initial RIS codebook set through the non-uniformly quantized RSS information entropy of the RSS data of each reference node, so that the difference between the RSS information entropies corresponding to the RIS codebooks in the candidate RIS codebook set meets the set threshold.

[0068] That is, by differentiating the RSS information entropy, a RIS codebook configuration with a sufficiently different signal is selected from the initial fingerprint database, so as to obtain a RIS codebook configuration that contributes more to the positioning capability from the initial RIS codebook set, thereby realizing the first stage selection of the RIS codebook in the initial RIS codebook set.

[0069] Among them, each RSS data set corresponding to the RIS codebook in the candidate RIS codebook set is a new fingerprint database; the set threshold represents the restriction on the difference of RSS information entropy corresponding to each RIS codebook during the screening process, so that only when the difference is large enough can a RIS codebook configuration that contributes more to the positioning capability be obtained.

[0070] The threshold can be flexibly set according to actual conditions, and this application does not impose any restrictions on this.

[0071] Finally, the genetic simulated annealing algorithm is used to further enhance the selection process using a data-driven training scheme. Specifically, the candidate RIS codebook set is further screened by annealing the new fingerprint database using the simulated annealing algorithm. That is, the main goal of this method is to determine a set of optimal RIS configurations and use fewer RIS codebook configurations to obtain sufficiently high positioning accuracy. Finally, the obtained set of optimal RIS codebook configurations, namely the target RIS codebook set, can be used.

[0072] It can be understood that the RIS codebook determination method provided in the embodiment of the present application introduces a two-stage RIS codebook selection method. First, based on the non-uniform quantized RSS information entropy, a RIS codebook with sufficiently different signals is selected from the initial fingerprint database corresponding to the initial RIS codebook set to construct a new initial fingerprint database. Then, in the second stage, a supervised learning feature selection method is used to search for the best RIS codebook configuration subset, that is, the target RIS codebook set. Finally, when positioning the user equipment in the positioning area, the target RIS codebook set can be used to obtain and match the corresponding RSS data, thereby effectively improving the positioning accuracy.

[0073] Moreover, the RIS codebook set determined by the two-stage selection can effectively avoid the introduction of some RIS configurations with small RSS differences, and ultimately improve the positioning efficiency and positioning accuracy.

[0074] In addition, in the scenario where there is only a single base station, the introduction of RIS makes the fingerprint positioning process possible in this scenario.

[0075] In addition, in terms of positioning delay, some studies in related technologies require UE to perform multiple groups of RSS measurements and perform RIS phase shift optimization to improve positioning accuracy, which increases positioning delay. This application is in the offline fingerprint data collection stage when selecting the RIS codebook. In the user equipment location estimation stage, only one set of RSS data of the screened RIS codebook needs to be collected, and the fitness function design also considers minimizing the number of measured RSS, that is, reducing positioning delay. Finally, a search algorithm with more robust local search capability is used to reduce the risk of the genetic algorithm falling into the local optimum during the optimization process and improve fingerprint positioning accuracy.

[0076] Optionally, in some embodiments of the present application, after the target RIS codebook configuration set is determined through the above steps, the target RIS codebook configuration set may be used to perform actual positioning, that is, the method may further include the following steps:

[0077] S140: Obtain target RSS data corresponding to each RIS codebook in the target RIS codebook set of the user equipment to be located in the positioning area.

[0078] S150: Compare the target RSS data with the RSS data in the initial fingerprint database to determine the location information of the user equipment.

[0079] Specifically, after the selection of the best RIS codebook configuration subset, ie, the target RIS codebook set, is completed through the above steps, the target RIS codebook set can be used to accurately locate the user equipment.

[0080] That is, in practice, the RIS codebook may be switched first, that is, the selected target RIS codebook set is executed, and then the target RSS data corresponding to the target RIS codebook set of the user equipment to be located in the positioning area is measured and acquired.

[0081] Furthermore, the acquired target RSS data is compared with the RSS data in the above-mentioned offline fingerprint database, ie, the initial fingerprint database, and then the weighted K nearest neighbor method can be used to estimate and determine the precise location of the user device, thereby locating the user device.

[0082] It can be understood that in the embodiment of the present application, the method for positioning the user equipment is implemented by utilizing an accurately selected target RIS codebook set to measure the RSS data corresponding to the RIS codebook therein, and then comparing it with the original offline RSS. Since the corresponding RSS data is collected by the selected target RIS codebook set, the positioning delay is effectively reduced; and in the above-mentioned user equipment positioning process, multiple information exchanges and RSS measurements between the AP and the UE are not required, thereby greatly reducing the positioning time.

[0083] Optionally, in S120, RSS data corresponding to RSS information entropy with sufficient difference in signals are selected from the initial fingerprint database, so as to obtain a RIS codebook configuration that contributes more to the positioning capability from the initial RIS codebook set.

[0084] The specific steps are as follows:

[0085] S121, determining the non-uniformly quantized RSS information entropy of each RIS codebook in the initial RIS codebook set at each reference node, and obtaining a corresponding information entropy vector.

[0086] S122, sorting the information entropy vectors, determining the RIS codebooks corresponding to the RSS information entropy exceeding the set threshold, and obtaining the candidate RIS codebook set.

[0087] Specifically, after obtaining the RSS data of each codebook in the initial RIS codebook set on each node, the information entropy value of each codebook on each node can be calculated.

[0088] The corresponding calculation process is as follows: Figure 2 As shown, the following steps may be specifically included:

[0089] S01, normalizing each RSS data in the initial fingerprint database based on the maximum RSS data value and the minimum RSS data value in the initial fingerprint database to obtain a normalized result of each RSS data.

[0090] S02, based on the non-uniform quantization function, performing normalization processing on each normalization result to obtain RSS data after the normalization processing.

[0091] S03, calculating the frequency value of each interval of the RSS data after the standardization processing within the frequency range.

[0092] S04: Calculate the information entropy of the non-uniform quantized RSS of each reference node through a set function according to the frequency value.

[0093] Specifically, the initial fingerprint database can be normalized by the maximum and minimum RSS values ​​in the initial fingerprint database, and then the normalized result can be processed by the non-uniform quantization function to determine the frequency value of each RSS data after processing in each interval within the set frequency range. Finally, according to the frequency value, the information entropy of the non-uniform quantized RSS of each reference node is calculated by the set function.

[0094] In order to better understand the calculation process, a certain RIS codebook, such as the i-th RIS codebook configuration, is taken as an example to explain in detail the calculation of the non-uniformly quantized RSS information entropy.

[0095] Assume that Max i is the maximum RSS value among all grid points, i.e., the initial fingerprint database l∈[1,L], Min i is the minimum RSS value among them, that is

[0096]

[0097] Based on the above assumptions, we can first use the maximum-minimum normalization method to normalize the RSS signal, that is:

[0098]

[0099] Next, Substituting into the non-uniform quantization function, we get:

[0100]

[0101] where the variable μ represents the compression constant in μ-law compression, represents the RSS data of the i-th RIS codebook after standardization and compression. Its range is 0-1, that is, as the frequency range. Then, we divide the frequency range into W uniform intervals, that is, form multiple intervals, and calculate the frequency of occurrence of each RSS data value in each interval, where the signal When it belongs to the tth interval, it can be expressed as

[0102]

[0103] Among them, is the number of grid points corresponding to the tth interval of the i-th RIS codebook, then It means that in the case of the i-th RIS codebook, the signal belongs to the t-th interval, that is, the frequency value of the interval, and L is the number of reference nodes in the positioning area.

[0104] After the above process, according to the concept of information entropy, the non-uniformly quantized RSS information entropy of the i-th RIS codebook configuration is given by the following formula:

[0105]

[0106] in, is the frequency value, W is the total number of intervals in the frequency interval, and t is the interval number.

[0107] Through the above formula, we can get the non-uniform quantized RSS information entropy vector of all RIS codebook sets, expressed as:

[0108] H=[H 1 ,H 2 ,...,H q ].

[0109] Furthermore, through the above steps, the non-uniform quantized RSS information entropy vectors of all RIS codebook sets are calculated, and the information entropy vectors can be sorted to determine the codebooks whose information entropy exceeds a set threshold to obtain a candidate codebook set.

[0110] For example, H can be sorted in descending order, the largest s RIS codebook configurations are selected, and the RSS data corresponding to the selected RIS codebooks are used to construct a new RSS fingerprint database with a size of L×s.

[0111] It can be understood that, through the above process, the basic first-stage screening of the initial codebook is completed, and a candidate RIS codebook set is obtained, and the RSS data corresponding to each RIS codebook in the candidate RIS codebook set reconstructs a new offline fingerprint data, i.e., a new fingerprint database, so that in the second stage, further selection is made based on the new fingerprint database.

[0112] Optionally, in S130, an annealing algorithm is used to apply a genetic simulated annealing algorithm to the candidate codebook set, and a data-driven training scheme is used to enhance the selection process. The main goal of this method is to determine a set of optimal RIS configurations and obtain sufficiently high positioning accuracy using fewer RIS codebook configurations.

[0113] like Figure 3 As shown, the following steps may be specifically included:

[0114] S131, performing genetic algorithm chromosome encoding processing on each RSS data in the new fingerprint database to obtain an initialized population.

[0115] S132, performing intermediate processing on the initialized population data to obtain intermediate processed population data.

[0116] S132, performing simulated annealing processing on the intermediately processed population data to obtain a target RSS database, and the RIS codebook corresponding to each RSS data in the RSS database is used as the target RIS codebook.

[0117] Specifically, first, for the new fingerprint database obtained in the above steps, that is, the fingerprint database corresponding to the candidate RIS codebook set, the data set with a total of M data points is divided into two parts, namely, there are M u The training set with M data points v A validation set of data points.

[0118] Furthermore, an s-bit binary vector is used to encode the RSS data of each individual, that is, each candidate codebook on each node in the genetic algorithm chromosome, where 1 indicates that this RIS codebook is used, and 0 indicates that it is not selected. In order to minimize the positioning error under the configuration of as few RIS codebooks as possible, the fitness function is designed as follows:

[0119] F(Ind)=E(Ind)+b×max(0,|Ind| 1 -M),

[0120] Among them, M is the maximum number of RIS codebooks available for use, |Ind| 1 is the number of RIS codebooks used by the current individual, b is a penalty value as large as possible, which is used to limit the number of RIS codebooks, and E(Ind) is the root mean square error of the positioning result under the current RIS configuration.

[0121] Furthermore, the initialized population is subjected to selection, crossover, mutation and simulated annealing operations. The detailed process is as follows:

[0122] Selection operation: In order to ensure the positioning accuracy of the results, we adopt an elite selection mechanism to select a group of individuals with the lowest fitness function from the population of the previous generation and use them as the next generation.

[0123] Crossover operation: Each bit of a child individual is inherited from the previous generation with equal probability.

[0124] Mutation operation: For each bit in each individual, it is possible to be reversed, and the probability value is set to 1 / s.

[0125] It is understandable that although the genetic algorithm has a strong grasp of the search process, it has problems such as low local search ability. The simulated annealing algorithm shows a robust local search ability, which can reduce the risk of the genetic algorithm falling into the local optimum during the optimization process. Therefore, the simulated annealing operation is performed afterwards.

[0126] The specific steps include:

[0127] S001, based on the initial temperature and annealing rate, randomly flip the intermediate processed population data to generate a flipped RSS database.

[0128] S002, calculating the acceptance probability of each RSS data in the flip RSS database.

[0129] S003, based on the annealing relationship, performing iterative annealing processing on the flipped RSS database until the set number of iterations or temperature threshold is reached.

[0130] Specifically, first set the initial temperature t 0 , annealing rate a, and then generate a new individual by randomly inverting a bit value of each individual. The new individual is recorded as Ind', and the difference in fitness function between the new individual and the original individual is calculated:

[0131] ΔC=F(Ind′)-F(Ind).

[0132] Then, the acceptance probability of the new individual is calculated:

[0133] P r =min[exp(-ΔC / t k ),1],

[0134] Among them, t k is the current temperature. If the fitness function of the new individual is lower than that of the original individual, the new individual is accepted with probability 1. If not, the new individual is accepted with probability exp(-ΔC / t k ) is accepted with a probability of k is the current temperature.

[0135] Then, use t k+1 =at k Annealing is performed with an annealing rate a in the interval [0,1]. The annealing operation is stopped when a predefined number of iterations is reached or the temperature drops below a specified threshold.

[0136] When the predetermined number of iterations is reached or when the fitness value has not improved within a given number of consecutive iterations, the entire genetic simulated annealing process ends, and a set of optimal RIS codebook configurations and corresponding RSS values ​​are obtained.

[0137] It can be understood that in the embodiment of the present application, by introducing a two-stage RIS codebook selection method, the RIS codebook configuration with sufficient signal difference is selected to build an initial fingerprint database, and the supervised learning feature selection method is used to search for the best RIS codebook configuration subset to improve the positioning accuracy.

[0138] That is, in the positioning system, commonly used performance indicators have multiple dimensions, such as positioning accuracy, coverage, positioning delay, robustness, etc. In terms of coverage, traditional multi-base station positioning often requires more than three base stations to perform, which limits the coverage of positioning and makes fingerprint positioning impossible in some extreme scenarios. Therefore, in the scenario where there is only a single base station, the present invention introduces RIS to make the fingerprint positioning process possible in this scenario.

[0139] In addition, in terms of positioning delay, some existing studies require UE to perform multiple groups of RSS measurements and perform RIS phase shift optimization to improve positioning accuracy, which increases positioning delay. The present invention is in the offline fingerprint data collection stage when selecting the RIS codebook. In the user position estimation stage, only one set of RSS data of the screened RIS codebook needs to be collected, and the fitness function design also considers minimizing the number of measured RSS, that is, reducing positioning delay. Finally, a search algorithm with more robust local search capability is used to reduce the risk of the genetic algorithm falling into the local optimum during the optimization process and improve fingerprint positioning accuracy.

[0140] On the other hand, Figure 4 As shown, the present application also provides a RIS codebook determination device, the device comprising:

[0141] A first acquisition module 210 is used to acquire an initial fingerprint database corresponding to an initial RIS codebook set, wherein the initial fingerprint database includes RSS data corresponding to each RIS codebook in the initial RIS codebook set at each reference node in the positioning area;

[0142] A first selection module 220 is configured to determine a candidate RIS codebook set in the initial RIS codebook set based on the non-uniformly quantized RSS information entropy of the RSS data of each reference node, wherein the difference between the RSS information entropies corresponding to the RIS codebooks in the candidate RIS codebook set satisfies a set threshold;

[0143] The second selection module 230 is used to determine a target RIS codebook set from the candidate RIS codebook set based on a genetic simulated annealing algorithm, wherein each RIS codebook in the target RIS codebook set is used to determine the location information of the user equipment in the positioning area.

[0144] Optionally, the RIS codebook determination device provided in the embodiment of the present application further includes:

[0145] The second acquisition module 240 is used to acquire target RSS data corresponding to each RIS codebook in the target RIS codebook set of the user equipment to be located in the positioning area;

[0146] The positioning module 250 is used to compare the target RSS data with the RSS data in the initial fingerprint database to determine the location information of the user equipment.

[0147] Optionally, in the RIS codebook determination device provided in an embodiment of the present application, the first selection module is specifically configured to:

[0148] Determine the non-uniformly quantized RSS information entropy of each RIS codebook in the initial RIS codebook set at each reference node to obtain a corresponding information entropy vector;

[0149] The information entropy vectors are sorted to determine the RIS codebooks corresponding to the RSS information entropy exceeding the set threshold, and the candidate RIS codebook set is obtained.

[0150] Optionally, in the RIS codebook determination device provided in an embodiment of the present application, the first selection module is specifically configured to:

[0151] Based on the maximum RSS data value and the minimum RSS data value in the initial fingerprint database, normalizing each RSS data in the initial fingerprint database to obtain a normalized result of each RSS data;

[0152] Based on the non-uniform quantization function, each normalization result is standardized to obtain the standardized RSS data;

[0153] Calculate the frequency value of each interval of the RSS data after standardization within the frequency range;

[0154] According to the frequency value, the information entropy of the non-uniform quantized RSS of each reference node is calculated by a set function.

[0155] Optionally, in the RIS codebook determination device provided in an embodiment of the present application, the non-uniform quantization function is expressed as follows:

[0156]

[0157] in, Indicates the normalized result of each RSS data. represents the RSS data of the i-th RIS codebook after normalization, and the variable μ represents the compression constant in the μ-law compression.

[0158] Optionally, in the RIS codebook determination device provided in an embodiment of the present application, the functional formula is:

[0159]

[0160] in, is the frequency value of the RSS data corresponding to the ith codebook belonging to the tth interval, and L is the number of reference nodes in the positioning area.

[0161] Optionally, in the RIS codebook determination device provided in an embodiment of the present application, the second selection module is specifically configured to:

[0162] Performing genetic algorithm chromosome encoding processing on each RSS data in the new fingerprint database to obtain an initialized population;

[0163] Performing intermediate processing on the initialized population data to obtain intermediate processed population data;

[0164] The population data after the intermediate processing is subjected to simulated annealing processing to obtain a target RSS database, and the RIS codebook corresponding to each RSS data in the RSS database is used as the target RIS codebook.

[0165] Optionally, in the RIS codebook determination device provided in an embodiment of the present application, the second selection module is specifically configured to:

[0166] Based on the initial temperature and the annealing rate, the intermediate processed population data is randomly flipped to generate a flipped RSS database;

[0167] Calculate the acceptance probability of each RSS data in the flip RSS database;

[0168] Based on the annealing relationship, the flipped RSS database is iteratively annealed until the set number of iterations or temperature threshold is reached.

[0169] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store video tag processing data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the RIS codebook determination method is implemented.

[0170] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0171] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0172] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0173] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0174] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0175] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0176] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a computer device, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0177] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0178] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for determining a RIS codebook, characterized in that: The method comprises: Acquire an initial fingerprint database corresponding to the initial RIS codebook set, wherein the initial fingerprint database includes RSS data corresponding to each RIS codebook in the initial RIS codebook set at each reference node in the positioning area; Based on the non-uniformly quantized RSS information entropy of the RSS data of each of the reference nodes, a candidate RIS codebook set is determined in the initial RIS codebook set, and the difference between the RSS information entropies corresponding to the RIS codebooks in the candidate RIS codebook set meets the set threshold; Based on the genetic simulated annealing algorithm, a target RIS codebook set is determined from the candidate RIS codebook set, and each RIS codebook in the target RIS codebook set is used to determine the location information of the user equipment in the positioning area.

2. The RIS codebook determination method according to claim 1, characterized in that: After determining the target RIS codebook configuration set, the method further includes: Obtain target RSS data corresponding to each RIS codebook in the target RIS codebook set of the user equipment to be located in the positioning area; The target RSS data is compared with the RSS data in the initial fingerprint database to determine the location information of the user equipment.

3. The RIS codebook determination method according to claim 1, characterized in that: Determining the candidate RIS codebook set in the initial RIS codebook set based on the non-uniformly quantized RSS information entropy of the RSS data of each reference node comprises: Determine the non-uniformly quantized RSS information entropy of each RIS codebook in the initial RIS codebook set at each reference node to obtain a corresponding information entropy vector; The information entropy vectors are sorted to determine the RIS codebooks corresponding to the RSS information entropy exceeding the set threshold, and the candidate RIS codebook set is obtained.

4. The RIS codebook determination method according to claim 3, characterized in that: The determining of the non-uniformly quantized RSS information entropy of each RIS codebook in the initial RIS codebook set at each reference node includes: Based on the maximum RSS data value and the minimum RSS data value in the initial fingerprint database, normalizing each RSS data in the initial fingerprint database to obtain a normalized result of each RSS data; Based on the non-uniform quantization function, each normalization result is standardized to obtain the standardized RSS data; Calculate the frequency value of each interval of the RSS data after standardization within the frequency range; According to the frequency value, the information entropy of the non-uniform quantized RSS of each reference node is calculated by a set function.

5. The RIS codebook determination method according to claim 4, characterized in that: The non-uniform quantization function is expressed as follows: in, Indicates the normalized result of each RSS data. represents the RSS data of the i-th RIS codebook after normalization, and the variable μ represents the compression constant in the μ-law compression.

6. The RIS codebook determination method according to claim 4, characterized in that: The functional formula is: in, is the frequency value of the RSS data corresponding to the ith codebook belonging to the tth interval, and L is the number of reference nodes in the positioning area.

7. The RIS codebook determination method according to claim 1, characterized in that: Each RSS data set corresponding to the RIS codebooks in the candidate RIS codebook set is a new fingerprint database, and determining the target RIS codebook set from the candidate RIS codebook set based on the genetic simulated annealing algorithm includes: Performing genetic algorithm chromosome encoding processing on each RSS data in the new fingerprint database to obtain an initialized population; Performing intermediate processing on the initialized population data to obtain intermediate processed population data; The population data after the intermediate processing is subjected to simulated annealing processing to obtain a target RSS database, and the RIS codebook corresponding to each RSS data in the RSS database is used as the target RIS codebook.

8. The RIS codebook determination method according to claim 5, characterized in that: The simulated annealing process of the intermediate processed population data comprises: Based on the initial temperature and the annealing rate, the population data after the intermediate processing is randomly flipped to generate a flipped RSS database; Calculate the acceptance probability of each RSS data in the flip RSS database; Based on the annealing relationship, the flipped RSS database is iteratively annealed until the set number of iterations or temperature threshold is reached.

9. A RIS codebook determination device, characterized in that: The device comprises: An acquisition module is used to acquire an initial fingerprint database corresponding to an initial RIS codebook set, wherein the initial fingerprint database includes RSS data corresponding to each RIS codebook in the initial RIS codebook set at each reference node in the positioning area; A first selection module is used to determine a candidate RIS codebook set in the initial RIS codebook set based on the non-uniformly quantized RSS information entropy of the RSS data of each reference node, wherein the difference between the RSS information entropies corresponding to the RIS codebooks in the candidate RIS codebook set meets a set threshold; The second selection module is used to determine a target RIS codebook set from the candidate RIS codebook set based on a genetic simulated annealing algorithm, wherein each RIS codebook in the target RIS codebook set is used to determine the location information of the user equipment in the positioning area.

10. A computer device, characterized in that: The computer device comprises: 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 RIS codebook determination method according to any one of claims 1 to 8.

Images