TWI858999B - Beamforming-based positioning device and positioning method - Google Patents

Abstract

A beamforming-based positioning device and positioning method are provided. The positioning method includes: obtaining trajectory of a user equipment (UE) in a space and generating a dynamic model according to the trajectory; measuring a plurality of signal strengths of a first beam patten at plurality of locations in the space and generating a first radio power map (RPM) of the first beam pattern in the space according to the plurality of signal strengths; estimating a first probability distribution of the UE in the space according to the dynamic model; measuring a first current signal strength of the first beam patten through the UE and calculating a likelihood function between the first current signal strength and the first RPM to generate a second probability distribution of the UE in the space; and determining a current location of the UE according to the first probability distribution and the second probability distribution and outputting the current location.

Description

Positioning device and positioning method based on beamforming

The present invention relates to a positioning device and a positioning method, and in particular to a positioning device and a positioning method based on beamforming.

One of the important features of the technological evolution of 5G new radio (NR) is the introduction of beamforming technology. Different from the previous communication framework based on omnidirectional signal propagation, beamforming technology can provide signal gain and interference suppression based on direction and user position to improve the performance of the communication system. At the same time, beamforming technology also brings new challenges to the communication system, such as: directional signal gain brings interference in a specific direction or user-based beam configuration technology. In order to overcome the above difficulties, reduce interference between base stations, and thus improve the transmission quality of the system, one of the keys is to obtain the location information of users in the communication network and optimize the configuration of the beam accordingly. Therefore, how to perform user equipment (UE) positioning in a communication system under a beamforming communication architecture is one of the important topics in this field.

The present invention provides a positioning device and positioning method based on beamforming, which can accurately measure the current position of the user equipment.

The present invention discloses a positioning method based on beamforming, which is applicable to positioning a user equipment in space, and includes: obtaining a trajectory of the user equipment in space, and generating a dynamic model according to the trajectory; measuring multiple signal strengths of a first beam pattern at multiple positions in space, and generating a first signal strength spectrum of the first beam pattern in space according to the multiple signal strengths; predicting a first probability distribution of the user equipment in space according to the dynamic model; measuring a first current signal strength of the first beam pattern by the user equipment, and calculating a first likelihood function between the first current signal strength and the first signal strength spectrum to generate a second probability distribution of the user equipment in space; and determining a current position of the user equipment according to the first probability distribution and the second probability distribution, and outputting the current position.

In one embodiment of the present invention, the above-mentioned step of generating a first signal strength spectrum of a first beam pattern in space according to multiple signal strengths includes: obtaining multiple basis functions; updating parameters of at least one basis function among the multiple basis functions according to the multiple signal strengths, wherein the parameters include weights; selecting a subset of the multiple basis functions according to the parameters; and generating a first signal strength spectrum according to the subset.

In one embodiment of the present invention, the step of generating a first signal strength spectrum according to a subset includes: calculating a second likelihood function between a plurality of signal strengths and a subset, and determining whether the second likelihood function matches a convergence condition; and generating a first signal strength spectrum according to the subset in response to the second likelihood function matching the convergence condition.

In one embodiment of the present invention, the step of generating a first signal strength spectrum based on a subset further includes: in response to the second likelihood function not matching the convergence condition, updating the parameters of at least one basis function in the subset based on multiple signal strengths.

In one embodiment of the present invention, the above-mentioned multiple basis functions are related to the Laplace function.

In one embodiment of the present invention, the positioning method further includes: measuring the second beam pattern to generate a second signal strength spectrum of the second beam pattern in space; measuring the second current signal strength of the second beam pattern by the user equipment, and calculating the second likelihood function between the second current signal strength and the second signal strength spectrum to generate a third probability distribution of the user equipment in space; and determining the current position according to the first probability distribution, the second probability distribution and the third probability distribution.

In one embodiment of the present invention, the first beam pattern and the second beam pattern are provided by the same base station.

In one embodiment of the present invention, the first beam pattern and the second beam pattern are provided by different base stations.

In one embodiment of the present invention, the above-mentioned step of determining the current location of the user device based on the first probability distribution and the second probability distribution includes: using one of a particle filter and a machine learning model to determine the current location.

In one embodiment of the present invention, the above-mentioned dynamic model includes one of the following: autoregressive model, Newtonian mechanics model and machine learning model.

In one embodiment of the present invention, the above-mentioned multiple signal strengths are associated with the reference signal receiving power.

The present invention discloses a positioning device based on beamforming, which is applicable to user equipment in a positioning space and includes a processor and a transceiver. The processor is coupled to the transceiver and is configured to perform: obtaining the trajectory of the user equipment in space through the transceiver and generating a dynamic model according to the trajectory; measuring multiple signal strengths of the first beam pattern at multiple locations in space, and generating a first signal strength spectrum of the first beam pattern in space according to the multiple signal strengths; predicting a first probability distribution of the user equipment in space according to the dynamic model; measuring the first current signal strength of the first beam pattern through the user equipment, and calculating a first likelihood function between the first current signal strength and the first signal strength spectrum to generate a second probability distribution of the user equipment in space; and determining the current position of the user equipment according to the first probability distribution and the second probability distribution, and outputting the current position through the transceiver.

Based on the above, the signal strength spectrum reconstruction and positioning system based on the beam pattern aims to utilize the directional gain of the beam directivity and the signal strength measurement results of the user to reconstruct the signal strength spectrum (radio power map, RPM) based on the beam pattern, and provide user management functions in the application scenario. The positioning method of the present invention can be further divided into two parts, wherein the first part is the establishment of the signal strength spectrum. This method is based on a pre-known beam gain model and a Bayesian sparse learning algorithm, and reconstructs the signal strength spectrum in the environment by approximating the channel attenuation of the wireless signal, wherein the signal strength spectrum can be used to represent the variation of the signal strength with space. Based on the signal strength spectra corresponding to these different beam directivities, the positioning system can locate and track the user equipment in the target space. This tracking algorithm can utilize the directionality of the beam pattern to further improve the accuracy of positioning. These user location information can be used as reference information in the network to further optimize the settings in the network, thereby providing users with higher network speeds and service experience.

FIG1 shows a schematic diagram of a positioning device 100 based on beamforming according to an embodiment of the present invention, wherein the positioning device 100 can be communicatively connected to one or more base stations 200 (e.g., base stations 210 or 220) and a user equipment 300. The positioning device 100 is suitable for positioning a user equipment 300 in a specific space (e.g., an indoor space) through one or more base stations 200 with beamforming function.

FIG. 2 shows a block diagram of a positioning device 100 according to an embodiment of the present invention. The positioning device 100 may include a processor 110, a storage medium 120, and a transceiver 130.

The processor 110 is, for example, a central processing unit (CPU), or other programmable general-purpose or special-purpose micro control unit (MCU), microprocessor, digital signal processor (DSP), programmable controller, application specific integrated circuit (ASIC), graphics processing unit (GPU), image signal processor (ISP), image processing unit (IPU), arithmetic logic unit (ALU), complex programmable logic device (CPLD), field programmable gate array (FPGA), or other similar components or combinations of the above components. The processor 110 may be coupled to the storage medium 120 and the transceiver 130, and access and execute multiple modules and various applications stored in the storage medium 120.

The storage medium 120 is, for example, any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory, hard disk drive (HDD), solid state drive (SSD) or similar elements or a combination of the above elements, and is used to store multiple modules or various applications that can be executed by the processor 110.

The transceiver 130 transmits or receives signals wirelessly or wiredly. The transceiver 130 may also perform operations such as low noise amplification, impedance matching, mixing, up or down frequency conversion, filtering, amplification, and the like. The positioning device 100 may be connected to the base station 200 or the user equipment 300 through the transceiver 130.

FIG3 shows a flow chart of a positioning method based on beamforming according to an embodiment of the present invention, wherein the positioning method can be implemented by the positioning device 100 shown in FIG2 . The positioning method can include a process S300 for establishing a model and a process S400 for positioning the user equipment 300.

Referring to process S300, in step S310, the processor 110 of the positioning device 100 may obtain the trajectory of the user equipment 300 in a specific space, wherein the specific space is, for example, located within the service range of the base station 200. The above-mentioned trajectory is, for example, generated based on historical data associated with the past movement of the user equipment 300 in the specific space.

In step S320, the processor 110 may generate a dynamic model of the user equipment 300 in a specific space according to the trajectory. The dynamic model may include, for example, an autoregressive model, a Newtonian model, or a machine learning (ML) model.

On the other hand, in step S350, the processor 110 may obtain relevant information of the base station, including location information or beam pattern gain of the base station 200. In step S360, the processor 110 may measure multiple signal strengths of each of one or more beam patterns of the base station 200 at multiple locations in a specific space through the transceiver 130.

In one embodiment, the one or more beam patterns mentioned above may be provided by the same base station 200. For example, the processor 110 may measure multiple signal strengths of each of the two (or more) beam patterns provided by the base station 210 at multiple locations through the transceiver 130. In one embodiment, the one or more beam patterns mentioned above may be provided by different base stations 200. For example, the processor 110 may measure multiple signal strengths of the beam pattern of the base station 210 at multiple locations in a specific space, and may measure multiple signal strengths of the beam pattern of the base station 220 at multiple locations in a specific space. In one embodiment, the signal strength may include a reference signal received power (RSRP).

For example, the positioning device 100 can measure the signal strength of the base station 200 at every one meter in a specific space. The operating frequency of the base station 200 is, for example, 2.6GHz, and it can be equipped with three flat antennas to switch different field patterns. The beam width corresponding to each flat antenna is about 60 degrees. The processor 110 can communicate with the signal scanner (scanner) equipped with an omnidirectional antenna at a height of 1.5 meters through the transceiver 130, so that the signal scanner stays at each measurement location for 1 minute to measure the signal strength.

In step S370, the processor 110 may generate a signal strength spectrum of the beam pattern of the base station 200 in a specific space according to the measured multiple signal strengths. For example, the processor 110 may generate a signal strength spectrum of the beam pattern of the base station 210 in a specific space according to the multiple signal strengths corresponding to the base station 210, and may generate a signal strength spectrum of the beam pattern of the base station 220 in a specific space according to the multiple signal strengths corresponding to the base station 220.

Considering the multi-path channel effects such as reflection, penetration or refraction in the indoor space, the processor 110 can use multiple basis functions to generate a signal strength spectrum. FIG4 shows a flow chart of step S370 according to an embodiment of the present invention. In step S510, the processor 110 can perform pre-processing (e.g., normalization) on the measured multiple signal strengths. In step S520, the processor 110 can generate a signal strength spectrum of the beam pattern of the base station 200 in a specific space (e.g., a signal strength spectrum of the base station 210 and/or 220) based on the multiple signal strengths. In step S530, the processor 110 can store the signal strength spectrum. For example, the processor 110 may store the generated one or more signal strength spectra in the storage medium 120.

FIG5 is a flow chart of step S520 according to an embodiment of the present invention. The processor 110 may generate a signal intensity spectrum based on a plurality of signal intensities based on Bayesian sparse learning. Specifically, in step S610, a plurality of basis functions are obtained, and parameters of each basis function are initialized, wherein the parameters include, for example, weights associated with the position or intensity of the basis function. Each basis function may be used to represent a virtual signal source generated by a multipath effect. The basis function is, for example, a Laplace function, but the present invention is not limited thereto.

In step S620, the processor 110 may update the parameters of one or more basis functions among the multiple basis functions according to the multiple signal strengths.

In step S630, the processor 110 may select a subset of multiple basis functions according to the parameters. For example, the processor 110 may filter out basis functions with too small weights (e.g., weights less than a threshold) to retain one or more basis functions with larger weights (e.g., weights greater than or equal to the threshold), thereby obtaining a subset.

In step S640, the processor 110 may calculate the likelihood function between the multiple intensity signals and the subset, and determine whether the likelihood function matches the convergence condition. If the likelihood function matches the convergence condition (for example, the likelihood function is greater than the threshold), it means that the current subset and its parameters can faithfully reflect the characteristics of the multiple intensity signals. Accordingly, the processor 110 may execute step S650. If the likelihood function does not match the convergence condition, the processor 110 may re-execute step S620 to update the parameters of one or more basis functions in the multiple basis functions until a subset that can represent the multiple intensity signals is selected.

In step S650, the processor 110 may generate a signal strength spectrum based on a subset of multiple basis functions.

Returning to FIG. 3 , referring to process S400 , in step S410 , the processor 110 can obtain spatial information of a specific space through the transceiver 130 . The spatial information may include, for example, the layout of the space or the location of immovable objects (e.g., furniture) in the space.

In step S420, the processor 110 may predict the probability of the user device 300 at each position in the specific space according to the dynamic model to obtain a probability distribution. In one embodiment, the processor 110 may preset the probability distribution of the user device 300 in the specific space according to the dynamic model and the spatial information. For example, the processor 110 may set the probability of the user device 300 at a certain position to zero based on the spatial information indicating that there is an immovable object at the position.

On the other hand, in step S450, the processor 110 may be connected to the user equipment 300 through the transceiver 130, and measure the current signal strength of one or more beam patterns of the base station 200 through the user equipment 300. For example, assuming that the base station 210 provides two beam patterns, the user equipment 300 may measure two current signal strengths corresponding to the two beam patterns provided by the base station 210. For another example, assuming that the base station 210 provides one beam pattern and the base station 220 provides another beam pattern, the user equipment 300 may measure the current signal strength of the beam pattern provided by the base station 210 and the current signal strength of the beam pattern provided by the base station 220.

In step S460, the processor 110 may calculate a likelihood function between each of the one or more current signal strengths and the corresponding signal strength spectrum. Accordingly, the processor 110 may obtain one or more likelihood functions. The processor 110 may further generate a probability distribution of the user equipment 300 in a specific space based on each of the one or more likelihood functions. For example, it is assumed that the base station 210 and the base station 220 provide two beam patterns respectively. For the beam pattern provided by the base station 210, the processor 110 may calculate a likelihood function between the beam pattern and the signal strength spectrum corresponding to the beam pattern. For the beam pattern provided by the base station 220, the processor 110 may calculate the likelihood function between the beam pattern and the signal strength spectrum corresponding to the beam pattern. Accordingly, the processor 110 may obtain two likelihood functions. The processor 110 may further generate two probability distributions corresponding to the two likelihood functions respectively based on the two likelihood functions.

In step S470, the processor 110 may determine the current position of the user equipment 300 in a specific space according to the probability distribution obtained in step S420 and one or more probability distributions obtained in step S460, and output the current position of the user equipment 300 through the transceiver 130 for reference by the user of the positioning device 100. In one embodiment, the processor 110 may use a particle filter or a machine learning model to determine the current position of the user equipment 300 according to the probability distribution obtained in step S420 and one or more probability distributions obtained in step S460.

Through channel modeling and the introduction of beam patterns, the positioning method of the present invention can verify the positioning accuracy through simulated channels. For different beam scanning methods and omnidirectional and directional antenna pattern information, the corresponding positioning errors are shown in Table 1. Compared with the positioning results of omnidirectional antennas, the positioning method based on directional antennas of the present invention can improve the positioning error by 50%. In order to further improve the positioning error, users can perform the following steps: increase the number of base stations; or increase the number of directional antennas used by each base station (i.e., increase the number of beam patterns provided by the base station).

By introducing more precise beam patterns, users can verify the impact of various parameters on positioning accuracy in a simulated manner. Considering the complexity of beam scanning, for the application scenario of 24 directional antennas, this embodiment only considers using the beam pattern with the strongest intensity for positioning, and the results are shown in Table 2. From the results in Table 2, it can be found that whether by increasing the number of directional antennas or increasing the number of base stations, the positioning accuracy of the positioning method of the present invention can be effectively increased.

Table 2

FIG6 is a flow chart of a positioning method based on beamforming according to another embodiment of the present invention, wherein the positioning method can be implemented by the positioning device 100 shown in FIG2 . In step S710, the trajectory of the user equipment in space is obtained, and a dynamic model is generated based on the trajectory. In step S720, multiple signal strengths of a first beam pattern at multiple positions in space are measured, and a first signal strength spectrum of the first beam pattern in space is generated based on the multiple signal strengths. In step S730, a first probability distribution of the user equipment in space is predicted based on the dynamic model. In step S740, the first current signal strength of the first beam pattern is measured by the user equipment, and the first likelihood function between the first current signal strength and the first signal strength spectrum is calculated to generate a second probability distribution of the user equipment in space. In step S750, the current position of the user equipment is determined according to the first probability distribution and the second probability distribution, and the current position is output.

In summary, the purpose of the present invention is to provide signal strength spectrum reconstruction based on beam pattern and to provide user location management in beam communication networks, aiming to utilize the directional gain of beam directivity and the signal strength measurement in the environment to provide user management functions in application scenarios. This system can be further divided into two parts. The first part is the establishment of the signal strength spectrum. This method is based on the beam gain model and Bayesian sparse calculation. The technology is listed as follows: considering the multi-path channel effects such as reflection, penetration or refraction in the indoor space, the present invention introduces a basis selection mechanism to represent the signal reflection and refraction effects in the indoor environment through complex basis functions, that is, to learn the comprehensive effects of channel attenuation and reflection scattering of wireless signals in the indoor space; by removing the directional gain of the beam pattern, the modeling error caused by the inconsistency between the basis function and the beam pattern is overcome to reconstruct an accurate signal strength spectrum; and based on these signal strength spectra based on the beam pattern. In the second part, the positioning system can further provide user positioning and tracking in the target space through the signal strength received by the user. The technologies are listed as follows: Introducing indoor space planning, limiting the positioning range, and combining the user's dynamic model to form a comprehensive prior information distribution. The above tracking algorithm can use the directionality of the beam pattern to provide more accurate positioning accuracy compared to omnidirectional wireless communication systems; and selecting effective beams and base stations based on the signal strength to improve the accuracy of user tracking and reduce the complexity of calculations.

The positioning system of the present invention can be further combined with the computing architecture of the open radio access network (RAN) and provide the following advantages: by performing positioning calculations on edge computing servers, the transmission time of data in the core network is reduced. The positioning system architecture can reduce the response time of positioning and can be applied to time-sensitive networks such as smart factories. The current location information output by the positioning system can be used as reference information in the network, and further used to optimize the settings in the network to provide users with higher network speed and service experience.

100: Positioning device

110: Processor

120: Storage media

130: Transceiver

200, 210, 220: base stations

300: User equipment

S300, S400: Process

S310, S320, S350, S360, S370, S410, S420, S450, S460, S470, S510, S520, S530, S610, S620, S630, S640, S650, S710, S720, S730, S740, S750: Steps

FIG1 is a schematic diagram of a positioning device based on beamforming according to an embodiment of the present invention.

FIG. 2 shows a block diagram of a positioning device according to an embodiment of the present invention.

FIG3 shows a flow chart of a positioning method based on beamforming according to an embodiment of the present invention.

FIG. 4 shows a flow chart of step S370 according to an embodiment of the present invention.

FIG5 is a flow chart of step S520 according to an embodiment of the present invention.

FIG6 shows a flow chart of a positioning method based on beamforming according to another embodiment of the present invention.

S710, S720, S730, S740, S750: Steps

Claims (11)

A positioning method based on beamforming, applicable to positioning a user equipment in a space, comprising: obtaining a trajectory of the user equipment in the space, and generating a dynamic model according to the trajectory; measuring multiple signal strengths of a first beam pattern at multiple positions in the space, and generating a first signal strength spectrum of the first beam pattern in the space according to the multiple signal strengths, comprising: obtaining multiple basis functions; updating parameters of at least one basis function of the multiple basis functions according to the multiple signal strengths, wherein the parameters include weights; selecting the multiple basis functions according to the parameters; a subset of basis functions; and generating the first signal strength spectrum according to the subset; predicting the first probability distribution of the user equipment in the space according to the dynamic model; measuring the first current signal strength of the first beam pattern by the user equipment, and calculating the first likelihood function between the first current signal strength and the first signal strength spectrum to generate the second probability distribution of the user equipment in the space; and determining the current position of the user equipment according to the first probability distribution and the second probability distribution, and outputting the current position. The positioning method as described in claim 1, wherein the step of generating the first signal strength spectrum according to the subset includes: calculating the second likelihood function between the plurality of signal strengths and the subset, and determining whether the second likelihood function matches a convergence condition; and in response to the second likelihood function matching the convergence condition, generating the first signal strength spectrum according to the subset. The positioning method as described in claim 2, wherein the step of generating the first signal strength spectrum according to the subset further includes: in response to the second likelihood function not matching the convergence condition, updating the parameters of at least one basis function in the subset according to the multiple signal strengths. A positioning method as described in claim 1, wherein the multiple basis functions are related to the Laplace function. The positioning method as described in claim 1 further includes: measuring a second beam pattern to generate a second signal strength spectrum of the second beam pattern in the space; measuring a second current signal strength of the second beam pattern by the user equipment, and calculating a second likelihood function between the second current signal strength and the second signal strength spectrum to generate a third probability distribution of the user equipment in the space; and determining the current position according to the first probability distribution, the second probability distribution and the third probability distribution. A positioning method as described in claim 5, wherein the first beam pattern and the second beam pattern are provided by the same base station. A positioning method as described in claim 5, wherein the first beam pattern and the second beam pattern are provided by different base stations respectively. In the positioning method as described in claim 1, the step of determining the current position of the user equipment according to the first probability distribution and the second probability distribution includes: determining the current position using one of a particle filter and a machine learning model. A positioning method as described in claim 1, wherein the dynamic model includes one of the following: an autoregressive model, a Newtonian mechanics model, and a machine learning model. A positioning method as described in claim 1, wherein the multiple signal strengths are related to the reference signal receiving power. A positioning device based on beamforming, suitable for positioning a user equipment in space, comprising: a transceiver; and a processor, coupled to the transceiver and configured to execute: obtaining a trajectory of the user equipment in the space through the transceiver, and generating a dynamic model according to the trajectory; measuring multiple signal strengths of a first beam pattern at multiple positions in the space, and generating a first signal strength spectrum of the first beam pattern in the space according to the multiple signal strengths, comprising: obtaining multiple basis functions; updating parameters of at least one basis function of the multiple basis functions according to the multiple signal strengths, wherein the parameters include weights; re-selecting a subset of the plurality of basis functions according to the parameters; and generating the first signal strength spectrum according to the subset; predicting a first probability distribution of the user equipment in the space according to the dynamic model; measuring a first current signal strength of the first beam pattern by the user equipment, and calculating a first likelihood function between the first current signal strength and the first signal strength spectrum to generate a second probability distribution of the user equipment in the space; and determining a current position of the user equipment according to the first probability distribution and the second probability distribution, and outputting the current position through the transceiver.

Images