JP7846050B2 - Direction estimation device, direction estimation method, and program - Google Patents

Description

This invention relates to a direction estimation device, a direction estimation method, and a program.

In research on communication technologies for Beyond 5G (6G), virtualized terminals are being considered in which a user terminal device and multiple peripheral terminals, such as wearable devices, connected to the user terminal device communicate in a coordinated manner. In these virtualized terminals, the terahertz (THz) band is used for information communication between the user terminal device and the peripheral terminals. However, it is anticipated that the beam width will be narrow in the THz band. If the relative positions of the user terminal device and the peripheral terminals change based on the movements of a person wearing a wearable device, and information communication is being performed using radio waves with a narrow beam width such as the THz band, it is anticipated that the peripheral terminals will easily move outside the range of the beam transmitted from the user terminal device. To prevent peripheral terminals from moving outside the range of the beam transmitted from the user terminal device, it is conceivable that the user terminal device could predict the movement of the peripheral terminals and output a beam that tracks the predicted direction. Non-patent document 1 discloses a technology in which the user terminal device predicts the movement of peripheral terminals and limits the beam sweep range based on the prediction.

H. MATSUNO et al., "High-Accuracy and High-Speed Beam Tracking Algorithm for THz Short-Range Mobile Communication,"2023 17th European Conference on Antennas and Propagation (EuCAP), 2023

According to the technology disclosed in Non-Patent Document 1, the user terminal device outputs multiple directional beams with different output directions, and accurately estimates the direction in which the peripheral terminal is located based on the received power received by the peripheral terminal. Furthermore, the movement vector and dispersion of the peripheral terminal are calculated based on previously estimated directions, and the direction in which the user terminal device is moving is predicted. By outputting a beam in the predicted direction, the user terminal device can perform information communication using radio waves with a narrow beamwidth, such as in the THz band. Here, according to the technology disclosed in Non-Patent Document 1, if the received power is temporarily lost, the user terminal device initially outputs a beam in the position based on the movement prediction. If the peripheral terminal is not found, an omnidirectional search is performed to estimate the direction in which the peripheral terminal is located. However, in the THz band, which is being considered for use in Beyond 5G (6G), the beamwidth is narrow, posing a challenge in that omnidirectional search becomes time-consuming.

This invention was made in consideration of these circumstances, and its purpose is to easily estimate direction even when the received power is temporarily lost.

(1) One aspect of the present invention is a direction estimation device that receives radio waves output from a predetermined device in multiple directions and estimates a suitable direction of radio waves by measuring the radio wave intensity of the received radio waves for each direction of the output radio waves, comprising: a missing signal determination unit that determines whether or not there is a missing direction among the radio wave intensity of radio waves output from a predetermined device in multiple directions; and a missing signal cause determination unit that, if it is determined that there is a missing direction, determines the cause of the missing signal based on the time-dependent change in time-series data of radio wave intensity for each sweep direction.
(2) In one aspect of the present invention, in the direction estimation device described above, the missing data source determination unit comprises: a first determination unit that determines whether the missing data source is due to shielding based on the trend of changes over time in time-series data of radio wave intensity for each sweep direction; and, if the missing data source is not due to shielding, a second determination unit that determines whether the missing data source is due to moving out of the communication range based on the trend of changes over time in time-series data of radio wave intensity for each sweep direction.
(3) In one aspect of the present invention, in the direction estimation device described above, the second determination unit determines that if the time-series data of radio wave intensity for each sweep direction changes rapidly, the cause of the missing data is due to a movement prediction error.
(4) In one aspect of the present invention, the direction estimation device described above includes a pre-trained model that has been trained in advance using as training data the time-series change of time-series data of radio wave intensity for each sweep direction and information on whether or not the cause of the loss is due to shielding.
(5) In one aspect of the present invention, the direction estimation device described above includes a pre-trained model that has been trained in advance using as training data the time-series change of time-series data of radio wave intensity for each sweep direction and information on whether or not the cause of the loss is due to going outside the communication range.
(6) In one aspect of the present invention, in the direction estimation device described above, both the trained model included in the first determination unit and the trained model included in the second determination unit are trained using a classifier such as the K-nearest neighbors algorithm.
(7) In one aspect of the present invention, the direction estimation device described above further comprises a preprocessing unit which performs preprocessing to determine the range of radio wave intensity measured for each direction of radio waves that is input to the first determination unit.
(8) One aspect of the present invention is a direction estimation method for estimating a suitable direction of radio waves by receiving radio waves output from a predetermined device in multiple directions and measuring the radio wave intensity of the received radio waves for each direction of the output radio waves, comprising: a missing signal determination step for determining whether or not there is a missing direction among the radio wave intensity of radio waves output from a predetermined device in multiple directions; and a missing signal cause determination step for determining the cause of the missing signal based on the time-series change of time-series data of radio wave intensity for each sweep direction if it is determined that there is a missing direction.
(9) One aspect of the present invention is a program that causes a computer to receive radio waves output from a predetermined device in multiple directions and estimate a suitable direction of radio waves by measuring the radio wave intensity of the received radio waves for each direction of the output radio waves, the program to execute a missing signal determination step of determining whether or not there is a missing direction among the radio wave intensity of radio waves output from a predetermined device in multiple directions , and a missing signal cause determination step of determining the cause of the missing signal based on the time-dependent changes in time-series data of radio wave intensity for each sweep direction if it is determined that there is a missing direction.

According to the present invention, the effect of easily estimating direction can be obtained even when the received power is temporarily lost.

This figure illustrates an overview of a communication system to which a direction estimation device, direction estimation method, and program according to one embodiment are applied. This is a functional configuration diagram showing an example of the functional configuration of the transmitting device and receiving device according to this embodiment. This figure illustrates omnidirectional search and search with a limited search range in the beam search method according to this embodiment. This figure illustrates the types of defects according to this embodiment. This is a functional configuration diagram showing an example of the functional configuration of the direction estimation device according to this embodiment. This is a functional configuration diagram showing an example of the functional configuration of the defect cause determination unit according to this embodiment. This figure illustrates the change in received power when the cause of the loss according to this embodiment is due to a movement prediction error. This figure illustrates the change in received power when the cause of the power loss in this embodiment is due to the device going outside the communication range. This flowchart illustrates an example of a method for determining the beam sweep range according to this embodiment. This is a flowchart illustrating an example of a method for determining the cause of defects according to this embodiment. This is a block diagram showing an example of the internal configuration of the direction estimation device of this embodiment.

The orientation estimation device, orientation estimation method, and program according to aspects of the present invention will be described in detail below, with reference to the accompanying drawings, and will include preferred embodiments. Note that the embodiments of the present invention are not limited to these embodiments, and include various modifications and improvements. In other words, the components described below include those that are easily conceivable to those skilled in the art, and those that are substantially the same, and the components described below can be combined as appropriate. Furthermore, various omissions, substitutions, or modifications of components can be made without departing from the spirit of the present invention. Also, in the following drawings, the scale and numbers of each structure may differ from those of the actual structure in order to make each structure easier to understand.

[Embodiment]
Embodiments of the present invention will be described below with reference to the drawings. In the embodiments described below, as an example, the orientation estimation device, orientation estimation method, and program according to this embodiment will be described on the premise that they are applied to a communication system in Beyond 5G (6G). However, this embodiment is not limited to this example and can be applied to various communication systems other than Beyond 5G (6G).

Figure 1 is a diagram illustrating the outline of a communication system to which a direction estimation device, direction estimation method, and program according to one embodiment are applied. The outline of the communication system 1 will be described with reference to this figure. The communication system 1 comprises multiple communication devices. In the illustrated example, a first communication device D1, a second communication device D2, and an nth communication device Dn (where n is a natural number of 1 or greater) are shown. Hereinafter, when the multiple communication devices comprising the communication system 1 are not distinguished, they may simply be referred to as communication device D. Each communication device D communicates information with the others.

The communication device D may be a user terminal device such as a smartphone or tablet, or a wearable terminal device connected to the user terminal device. Examples of wearable terminal devices include smartwatches, smart glasses, and earphone-type wearable devices. The multiple communication devices D included in the communication system 1 may be held by the user or fixed in a predetermined location.

Furthermore, the multiple communication devices D included in communication system 1 constitute a virtualized terminal that communicates with each other in coordination. From an external perspective, communication system 1 can be perceived as behaving as a single virtualized terminal. For information communication between each communication device D, for example, the terahertz (THz) band is used. However, this embodiment is not limited to this example, and other frequency bands such as the gigahertz (GHz) band may also be used. In the following description, radio waves output from communication devices D that have narrow directivity may be referred to as a beam.

Here, in the THz frequency range, it is assumed that the beam width of the beam output from the communication device D becomes narrower. As the beam width narrows, the communication range between each communication device D is limited. In other words, since the communication devices D communicate information using beams with narrow directivity, if their relative positions change, they can easily move outside the communication range, making information communication impossible. On the other hand, some of the multiple communication devices D included in the communication system 1 may be wearable devices and may not be used in a fixed position. For example, in the case of information communication between a smartphone and a smartwatch, the relative positions of the devices can change due to the user swinging their arms while walking, easily causing them to move outside the communication range. Therefore, according to this embodiment, in order to prevent the communication devices D from moving outside the communication range, the movement of the communication devices D is predicted, and the beam is output in the predicted direction, thereby continuing information communication without interruption.

Figure 2 is a functional configuration diagram showing an example of the functional configuration of the transmitting and receiving devices according to this embodiment. Referring to this figure, an example of the functional configuration of the communication device D is shown. Multiple communication devices D communicate information by having one unit act as a transmitting device 2 and the other as a receiving device 3. At least one of the transmitting device 2 or receiving device 3 is equipped with a direction estimation device 10. Figure 2(A) is an example where the direction estimation device 10 is equipped in the receiving device 3, and Figure 2(B) is an example where the direction estimation device 10 is equipped in the transmitting device 2.

First, with reference to Figure 2(A), an example of the functional configuration of the transmitting device 2 and the receiving device 3 when the receiving device 3 is equipped with an azimuth estimation device 10 will be described. When the receiving device 3 is equipped with an azimuth estimation device 10, the transmitting device 2 will be referred to as the transmitting device 2A, and the receiving device 3 will be referred to as the receiving device 3A. The transmitting device 2A includes a beam sweeping unit 21. The beam sweeping unit 21 performs beam sweeping by outputting a narrowly directional beam in multiple directions by controlling an antenna array, for example, which is composed of multiple antennas. The receiving device 3A includes a beam receiving unit 31 and an azimuth estimation device 10. The beam receiving unit 31 receives radio waves output from the beam sweeping unit 21 and measures the radio wave strength (RSSI; Received Signal Strength Indicator) by measuring the received power of the received radio waves. The beam receiving unit 31 outputs information regarding the measured radio wave strength to the azimuth estimation device 10. The direction estimation device 10 acquires information regarding the radio wave intensity measured by the beam receiving unit 31 and estimates the direction in which the transmitting device 2A is located based on the acquired information. The direction estimation device 10 stores the results of the estimation of the direction in which the transmitting device 2A is located. Based on the stored direction estimation results (i.e., past direction estimation results), the direction estimation device 10 makes a prediction (i.e., movement prediction) about the change in the positional relationship between the transmitting device 2A and the receiving device 3A. Based on the movement prediction results, the direction estimation device 10 determines the beam sweep range and outputs information about the determined beam sweep range to the transmitting device 2A. By outputting a beam within the range based on the movement prediction, the transmitting device 2A can search for the direction in which the receiving device 3A is located without outputting beams in all directions, and can communicate information with the receiving device 3A.

Next, with reference to Figure 2(B), an example of the functional configuration of the transmitter 2 and receiver 3 when the transmitter 2 is equipped with an azimuth estimation device 10 will be described. When the receiver 3 is equipped with an azimuth estimation device 10, the transmitter 2 will be referred to as transmitter 2B and the receiver 3 as receiver 3B. Transmitter 2B differs from transmitter 2A in that it is equipped with an azimuth estimation device 10. Also, receiver 3B differs from receiver 3A in that it is not equipped with an azimuth estimation device 10. That is, in the example shown in Figure 2(B), the azimuth estimation processing is performed by transmitter 2B instead of receiver 3A, which is different from the example shown in Figure 2(A). In the example shown in Figure 2(A), information about the determined beam sweep range is output from receiver 3A to transmitter 2A, but in the example shown in Figure 2(B), information about radio wave intensity is output from receiver 3A to transmitter 2A. In the example shown in Figure 2(B), the functional configuration of the direction estimation device 10, beam sweep unit 21, and beam receiving unit 31 may be the same as in the example shown in Figure 2(A).

Figure 3 illustrates the beam search method according to this embodiment, specifically illustrating omnidirectional search and search with a limited search range. The beam search method according to this embodiment will be explained with reference to this figure.

Figure 3(A) shows an example of omnidirectional search. As shown in the figure, in omnidirectional search, the transmitter 2 emits radio waves in all directions from which it can emit radio waves. Therefore, with omnidirectional search, regardless of the location of the receiver 3, information communication between the transmitter 2 and the receiver 3 can be established as long as the transmitter 2 is located in a direction from which it can emit radio waves. However, as mentioned above, the beam width is narrow in the THz band, so omnidirectional search takes time and requires unnecessary power consumption for the search. Therefore, according to this embodiment, the transmitter 2 predicts the location of the receiver 3 and performs the search by limiting the search range.

Figure 3(B) shows an example of a search with a limited search range. As shown, the transmitter 2 outputs radio waves in the vicinity of the location where the receiver 3 is predicted to be located. By limiting the search range in this way, the search is not time-consuming and does not consume unnecessary power. However, with a search with a limited search range, if the receiver 3 is not present in the direction of the radio waves output from the transmitter 2 (i.e., if the transmitter 2 mispredicts the movement of the receiver 3), information communication between the transmitter 2 and the receiver 3 cannot be established, and it becomes necessary to perform an all-directional search again. In this case, the search will require even more time and power consumption. Therefore, it is necessary to accurately predict the movement of the receiver 3 and sweep the beam to a suitable range.

According to this embodiment, the transmitting device 2 predicts the direction in which the receiving device 3 is located based on previously estimated directions. The transmitting device 2 then limits the search range based on the predicted direction and performs a search. Here, if the receiving device 3 can continuously receive the radio waves output from the transmitting device 2, the movement of the receiving device 3 can be easily predicted. However, if the radio waves output from the transmitting device 2 are temporarily not received by the receiving device 3 (i.e., there is a loss of radio waves), the movement of the receiving device 3 is predicted based on past movement trajectories, and the search range is limited based on the prediction result to perform the search. Here, multiple factors can be considered as causes of radio wave loss. In this embodiment, by identifying the factors causing radio wave loss and predicting movement based on the identified loss factors, it becomes possible to sweep the beam to a suitable range.

Figure 4 is a diagram illustrating the types of loss factors according to this embodiment. The types of radio wave loss factors will be explained with reference to this figure. Referring to this figure, examples of cases where the radio wave loss factor is, firstly, due to shielding, secondly, due to movement prediction errors, and thirdly, due to movement prediction errors will be explained.

Figure 4(A) illustrates a case where the cause of signal loss is due to shielding. As shown in the figure, the presence of an obstruction between the transmitter 2 and the receiver 3 can cause signal loss from the transmitter 2. Examples of obstructions include human hands and feet. Other obstructions may include clothing or tables. For example, if the receiver 3 moves from a state where there is no signal loss to a position where an obstruction exists between the transmitter 2 and the receiver 3, the transmitter 2 determines the output range of the radio waves based on the receiver 3's past movement (for example, based on the movement vector and dispersion). However, the movement trajectory of the receiver 3 may have fluctuated during the period it was hidden by the obstruction, and in such cases, the accuracy of direction estimation may decrease.

Figure 4(B) illustrates the case where the cause of the missing data is due to an error in movement prediction. In this figure, the location where the receiver 3 is predicted to be located, as a result of movement prediction, is shown as receiver 3', and the actual location of receiver 3 is shown as receiver 3. The location of receiver 3' is predicted, for example, by machine learning, based on the past movement trajectory of receiver 3. Therefore, it is preferable that the location of receiver 3 and the location of receiver 3' are the same. However, if receiver 3 moves in a way that differs from its past movement trajectory, the location of receiver 3 and the location of receiver 3' may differ, as shown in the figure. In such cases where the movement prediction is inappropriate, it is preferable to perform an omnidirectional search to locate the actual receiver 3.

Figure 4(C) illustrates the case where the data loss is caused by the receiver 3 moving out of the communication range of the transmitter 2. In this figure, the position of the receiver 3 at the first time point is shown as receiver 3', and the position of the receiver 3 at the second time point (a later time point) is shown as receiver 3. When the receiver 3 is at the position shown by receiver 3', the transmitter 2 can communicate information with the receiver 3 normally. However, if the distance between the transmitter 2 and the receiver 3 increases, for example, as shown in the figure, the receiver 3 may move out of the communication range of the transmitter 2, making information communication impossible. In this case, the receiver 3 cannot be found no matter how much searching is performed. Therefore, when the receiver 3 is out of the communication range of the transmitter 2, it is preferable for the transmitter 2 not to search for the receiver 3 (i.e., not to output a beam to the receiver 3).

Figure 5 is a functional configuration diagram showing an example of the functional configuration of the direction estimation device according to this embodiment. An example of the functional configuration of the direction estimation device 10 will be described with reference to this figure. The direction estimation device 10 comprises a sweep result storage unit 11, a direction estimation unit 12, an estimation result storage unit 13, a missing data presence/absence determination unit 14, a missing data cause determination unit 15, and a sweep range determination unit 16. Each of these functional units is implemented, for example, using electronic circuits. Furthermore, each functional unit may internally incorporate storage means such as semiconductor memory or magnetic hard disk drives, as needed. Also, each function may be implemented by a computer and software.

The sweep result storage unit 11 acquires information regarding the beam sweep direction and received power from the beam receiving unit 31. The beam sweep direction is the direction of the narrowly directional beam output from the transmitting device 2. Received power is the power of the beam received by the receiving device 3, i.e., the radio wave intensity, corresponding to the beam direction. The sweep result storage unit 11 stores the acquired information. The information acquired by the sweep result storage unit 11 may include time information beforehand, or the sweep result storage unit 11 may store the acquired information in association with the time of acquisition.

The direction estimation unit 12 identifies the direction with the highest radio wave intensity at each time point, based on the information stored in the sweep result storage unit 11. The identified direction is the direction from which the receiving device 3 is located as viewed from the transmitting device 2 (or the direction from which the transmitting device 2 is located as viewed from the receiving device 3). The direction estimation unit 12 may also estimate the trajectory of the receiving device 3 as viewed from the transmitting device 2 (or the trajectory of the transmitting device 2 as viewed from the receiving device 3) based on the time-dependent changes in the identified direction. Based on the estimated trajectory, the direction estimation unit 12 estimates the direction of the receiving device 3 as viewed from the transmitting device 2 at a predetermined time. The direction estimation unit 12 stores the result of the direction estimation in the estimation result storage unit 13.

The estimation result storage unit 13 stores information about the direction estimated by the direction estimation unit 12. The estimation result storage unit 13 stores the information about the direction estimated by the direction estimation unit 12 in association with the time. It can also be said that the estimation result storage unit 13 stores past direction estimation results.

The loss detection unit 14 determines whether or not a loss has occurred in the received radio wave (i.e., whether or not a loss has occurred). Specifically, the loss detection unit 14 determines whether or not there is a direction with a loss among the multiple directions in which the radio wave was received. The loss detection unit 14 may determine that a loss has occurred if the radio wave intensity in at least one direction is below a predetermined value. The loss detection unit 14 outputs information regarding the determined presence or absence of a loss to the loss cause determination unit 15.

The Missing Signal Factor Determination Unit 15 acquires the direction estimation result stored in the Estimation Result Storage Unit 13 and obtains information regarding the presence or absence of missing signals from the Missing Signal Presence/Absence Determination Unit 14. If the Missing Signal Factor Determination Unit 15 determines that a direction with missing signals exists, it determines the cause of the missing signal based on the time-dependent change in the received radio wave intensity. Details regarding the determination of the missing signal factor will be described later with reference to Figure 6, etc. The Missing Signal Factor Determination Unit 15 outputs the information regarding the determined missing signal factor to the Sweep Range Determination Unit 16.

The sweep range determination unit 16 acquires the direction estimation result stored in the estimation result storage unit 13 and acquires the missing data factor determination result from the missing data factor determination unit 15. Based on the acquired information, the sweep range determination unit 16 determines the beam sweep range. The beam sweep range determined by the sweep range determination unit 16 is, in other words, the direction of the beam output from the transmitter 2. The direction of the beam output from the transmitter may be, firstly, all directions, secondly, a predetermined direction that is a narrower range than all directions, or thirdly, no output. The sweep range determination unit 16 outputs information regarding the determined sweep range to the beam sweep unit 21.

Figure 6 is a functional configuration diagram showing an example of the functional configuration of the defect cause determination unit according to this embodiment. An example of the functional configuration of the defect cause determination unit 15 will be described with reference to this figure. The defect cause determination unit 15 comprises a pre-processing unit 151, a first determination unit 152, a second determination unit 153, and an output unit 154.

The preprocessing unit 151 acquires the direction estimation results stored in the estimation result storage unit 13 and obtains information regarding the presence or absence of missing data from the missing data determination unit 14. If the missing data determination unit 14 determines that there is a direction with missing data, the preprocessing unit 151 performs preprocessing on the information to be input to the first determination unit 152. The information input to the first determination unit 152 is the time-series data of the direction estimation results. In this embodiment, the beam sweep range is variable. Therefore, the preprocessing unit 151 performs preprocessing to identify the range to be input to the first determination unit 152 and to adjust the format of the information to be input to the first determination unit 152. In other words, preprocessing can be described as the process of identifying the range to be input to the first determination unit 152 from the radio wave intensity measured for each direction of the radio waves. Furthermore, since the beam sweep range and sweep direction output by the transmitting device 2 fluctuate over time, preprocessing can also be described as the process of converting the data into a format that is easy for the first determination unit 152 to process.

The first determination unit 152 acquires pre-processed information from the pre-processing unit 151. Based on the acquired information, the first determination unit 152 performs a first determination. The first determination is whether or not the cause of the signal loss is due to shielding. The first determination is made based on the trend of changes in the received radio wave intensity over time. For example, the first determination unit 152 may determine that the cause of the signal loss is shielding if, among the radio wave intensity received in multiple directions, there is a direction with low radio wave intensity. The first determination can also be described as a determination to differentiate whether or not the cause of the signal loss is due to shielding. Furthermore, if the first determination unit 152 determines that the cause of the signal loss is due to shielding, it may perform estimation to identify the shielded section.

Furthermore, the first determination unit 152 may perform the first determination using machine learning. When the first determination is performed using machine learning, the first determination unit 152 includes a pre-trained model. This pre-trained model is pre-trained using supervised learning, with training data consisting of the time-dependent changes in the received radio wave intensity and information on whether the cause of the loss is due to shielding. It is preferable that this pre-trained model be trained using the K-nearest neighbors algorithm.

The second determination unit 153 obtains pre-processed information from the pre-processing unit 151 and obtains the result of the first determination from the first determination unit 152. Based on the obtained information, the second determination unit 153 performs a second determination. The second determination is performed when the cause of the signal loss is not due to shielding. The second determination determines whether the cause of the signal loss is due to moving outside the communication range. The second determination unit 153 performs the second determination based on the pre-processed information regarding the direction estimation result, i.e., the time-dependent trend of the received radio wave intensity. For example, if the received radio wave intensity changes rapidly, the second determination unit 153 may determine that the cause of the signal loss is due to a movement prediction error. In this case, if the cause of the signal loss is not due to moving outside the communication range, the second determination unit 153 may determine that a movement prediction error is the cause.

Furthermore, the second determination unit 153 may perform the second determination using machine learning. When the second determination is performed using machine learning, the second determination unit 153 includes a pre-trained model. This pre-trained model is pre-trained using supervised learning, with training data consisting of the time-dependent changes in the received radio wave intensity and information on whether the cause of the loss is due to the signal going outside the communication range. It is preferable that this pre-trained model be trained using the K-nearest neighbors algorithm.

The output unit 154 obtains a first determination result from the first determination unit 152 and a second determination result from the second determination unit 153. Based on the obtained results, the output unit 154 identifies the radio wave loss factors. The output unit 154 outputs the identified radio wave loss factors to the sweep range determination unit 16.

Next, referring to Figures 7 and 8, we will explain an example of how the received power changes when the relative positions of the transmitter 2 and the receiver 3 change. Furthermore, referring to Figures 7 and 8, we will explain why it is possible to determine the cause of signal loss from the trend of changes in received power.

Figure 7 illustrates the change in received power when the data loss is caused by a movement prediction error according to this embodiment. Figure 7(A) shows the beam sweep range and the positional relationship between the transmitter 2 and receiver 3 when the data loss is caused by a movement prediction error. Figure 7(B) shows the change in received power over time in the positional relationship shown in Figure 7(A). In this figure, the position where receiver 3 is predicted to be located is shown as receiver 3', and the actual location of receiver 3 is shown as receiver 3. In this case, the transmitter 2 outputs a beam towards receiver 3', which is the position where receiver 3 is predicted to be located. However, since receiver 3 is actually located at receiver 3, the beam is output in the wrong direction. In this case, the received power drops sharply.

Figure 8 illustrates the change in received power when the signal loss in this embodiment is caused by the receiver moving out of the communication range. Figure 8(A) shows the beam sweep range and the positional relationship between the receiver 2 and receiver 3 when the signal loss is caused by the receiver 3 moving out of the communication range of the transmitter 2. Figure 8(B) shows the change in received power over time in the positional relationship shown in Figure 8(A). In this figure, the position of receiver 3 at the first time point is shown as receiver 3', and the position of receiver 3 at the second time point (later than the first time point) is shown as receiver 3. In this case, the transmitter 2 continues to output a beam in the direction of receiver 3 at both the first and second time points. However, since receiver 3 gradually moves away from transmitter 2, the received power decreases slowly.

As explained with reference to Figures 7 and 8, the trend in the change in received power differs depending on whether the loss is due to an error in movement prediction or to moving out of the communication range. Therefore, it is possible to determine the cause of the radio wave loss based on the trend in the change in received power.

Figure 9 is a flowchart illustrating an example of a method for determining the beam sweep range according to this embodiment. The example of a method for determining the beam sweep range according to this embodiment will be explained with reference to this figure.

First, the transmitting device 2 performs a beam sweep in a predetermined direction (step S11). The receiving device 3 receives the beam output from the transmitting device 2 and estimates the direction in which the transmitting device 2 is located based on the radio wave intensity of the received beam (step S12). Specifically, the direction estimation may be performed based on previously estimated direction estimation results (for example, the direction estimation result from the previous step), i.e., time-series data of direction estimation results.

Next, the receiving device 3 estimates whether the received power of the beam is below a threshold (step S13). If the received power is below the threshold, the process proceeds to step S14. If the received power is not below the threshold, the process proceeds to step S15. If the received power is below the threshold (i.e., step S13; YES), the cause of the received power deficit is determined (step S14). Based on the determined cause of the received power deficit, the beam sweep range by the movement predictor is determined (step S16), or the beam sweep range is adaptively determined (step S17). If the received power is not below the threshold (i.e., step S13; NO), the beam sweep range by the movement predictor is determined (step S15).

Figure 10 is a flowchart illustrating an example of the method for determining the cause of missing data according to this embodiment. The example of the method for determining the cause of missing data according to this embodiment will be explained with reference to this figure. Note that the process described with reference to this figure is an example of the process performed by the missing data cause determination unit 15.

First, the missing data detection unit 15 performs preprocessing on the time-series data of the direction estimation results (step S21). Next, the missing data detection unit 15 determines the occlusion zone based on the preprocessed time-series data (step S22). The determination of the occlusion zone is performed using a pre-trained prediction model, which uses the fluctuation trend of the direction estimation results and the relationship with occlusion as training data. It is preferable to use the K-nearest neighbor method for determining the occlusion zone. Here, for example, at least one of the transmitting device 2 and the receiving device 3 may be a wearable device worn on the human body. Therefore, movement prediction can be performed by using a prediction model that has learned the relationship between human body movement and occlusion. Human body movement can be estimated, for example, based on the time-series data of the direction estimation results. Therefore, it becomes possible to determine the occlusion zone based on the time-series data of the direction estimation results.

If the loss cause determination unit 15 determines that the receiving device 3 is located in a shielded area (i.e., step S23; YES), it determines that the loss in the radio waves is due to shielding (step S26).

Furthermore, if the loss cause determination unit 15 does not determine that the receiving device 3 is located within the shielded section (i.e., step S23; NO), it performs a further determination (step S24). This further determination involves determining, based on the fluctuation trend of radio wave intensity for each sweep direction by the transmitting device 2, whether the loss is due to moving outside the communication range or due to a movement prediction error. A different prediction model may be used for this determination than the one used in step S23. It is preferable that this determination be performed using the K-nearest neighbor method.

The loss cause determination unit 15 determines that the cause is going outside the communication range if the fluctuation trend of the radio wave intensity for each sweep direction by the transmitting device 2 can be classified into a predetermined trend (first trend), and proceeds to step S27 (i.e., step S25; YES). The loss cause determination unit 15 determines that the cause of the loss is due to going outside the communication range.

Furthermore, the data loss cause determination unit 15 determines that the cause is not going outside the communication range if the fluctuation trend of radio wave intensity for each sweep direction by the transmitting device 2 can be classified into a predetermined trend (second trend) (or if it cannot be classified), and proceeds to step S28 (i.e., step S25; NO). The data loss cause determination unit 15 determines that the cause of the data loss is due to a movement prediction error.

Figure 11 is a block diagram showing an example of the internal configuration of the direction estimation device of this embodiment. At least some of the functions of the direction estimation device 10 can be realized using a computer. As shown in the figure, the computer is composed of a central processing unit 901, RAM 902, input/output ports 903, input/output devices 904 and 905, etc., and a bus 906. The computer itself can be realized using existing technology. The central processing unit 901 executes instructions contained in programs read from RAM 902, etc. The central processing unit 901 writes data to RAM 902, reads data from RAM 902, and performs arithmetic and logical operations according to each instruction. RAM 902 stores data and programs. Each element contained in RAM 902 has an address and can be accessed using that address. RAM stands for "Random Access Memory". Input/output ports 903 are ports for the central processing unit 901 to exchange data with external input/output devices, etc. Input/output devices 904 and 905 are input/output devices. Input/output devices 904 and 905 exchange data with the central processing unit 901 via input/output port 903. Bus 906 is a common communication channel used within the computer. For example, the central processing unit 901 reads and writes data to RAM 902 via bus 906. Also, for example, the central processing unit 901 accesses input/output ports via bus 906.

[Summary of Embodiments]
According to this embodiment, the direction estimation device 10 receives radio waves output in multiple directions from a predetermined device such as a transmitting device 2, and estimates the radio waves in a suitable direction by measuring the radio wave intensity of the received radio waves for each direction of the output radio waves. Furthermore, the direction estimation device 10 includes a loss detection unit 14 to determine whether or not there is a direction with a loss among the multiple directions in which radio waves are received, and includes a loss cause determination unit 15 to determine the cause of the loss based on the time-dependent change in the radio wave intensity of the received radio waves if it is determined that there is a direction with a loss. In other words, the direction estimation device 10 determines the presence or absence of radio wave loss and then determines the cause of the loss. Therefore, because the direction estimation device 10 can determine the cause of the loss, it is possible to easily estimate the direction of the receiving device 3 as seen from the transmitting device 2 even if the received power is temporarily lost. Because the direction estimation device 10 can determine the sweep range based on the estimated direction, it is not necessary to perform an all-directional search when a loss occurs, and the device to be communicated can be searched in a short time.

Furthermore, according to the above-described embodiment, the loss factor determination unit 15 includes a first determination unit 152 that determines whether the loss is due to shielding based on the trend of changes in the received radio wave intensity over time, and a second determination unit 153 that, if the loss is not due to shielding, determines whether the loss is due to moving out of the communication range based on the trend of changes in the received radio wave intensity over time. That is, the loss factor determination unit 15 performs the first determination to determine whether the loss is due to shielding, and then performs the second determination to determine whether the loss is due to moving out of the communication range. Therefore, according to this embodiment, the loss factor can be determined in detail. According to this embodiment, since the sweep range can be determined according to the loss factor determined in detail, beam sweeping can be performed within a suitable range, and the device to be communicated can be searched in a short time.

Furthermore, according to the above-described embodiment, the second determination unit 153 determines that if the signal strength of the received radio waves changes rapidly, the cause of the loss is due to a movement prediction error. Therefore, according to this embodiment, the cause of the loss can be determined in more detail. According to this embodiment, since the sweep range can be determined according to the more detailed determined cause of the loss, beam sweeping can be performed within a more suitable range, and the device to be communicated can be searched for in a shorter time.

Furthermore, according to the embodiment described above, the first determination unit 152 includes a pre-trained model that has been learned in advance using the time-dependent change in the received radio wave intensity and information on whether or not the cause of the loss is due to shielding as training data. That is, the first determination unit 152 performs the determination using machine learning. Therefore, according to this embodiment, the cause of the loss can be determined with greater accuracy. According to this embodiment, since the sweep range can be determined according to the more accurately determined cause of the loss, beam sweeping can be performed within a suitable range, and the device to be communicated can be searched for in a short time.

Furthermore, according to the embodiment described above, the second determination unit 153 includes a pre-trained model that has been learned in advance using training data, which includes the time-dependent change in the received radio wave intensity and information on whether or not the cause of the loss is due to the signal going outside the communication range. In other words, the second determination unit 153 performs the determination using machine learning. Therefore, according to this embodiment, the cause of the loss can be determined with greater accuracy. According to this embodiment, since the sweep range can be determined according to the more accurately determined cause of the loss, beam sweeping can be performed within a suitable range, and the device to be communicated can be searched for in a short time.

Furthermore, according to the embodiment described above, both the trained model included in the first determination unit 152 and the trained model included in the second determination unit 153 are trained using the K-nearest neighbors algorithm. Therefore, according to this embodiment, the cause of missing data can be accurately determined using a simple algorithm.

Furthermore, according to the embodiment described above, the data loss cause determination unit 15 includes a preprocessing unit 151 that performs preprocessing to identify the range of radio wave intensity measured for each direction of radio waves that will be input to the first determination unit 152. Preprocessing is the process of identifying the range of radio wave intensity measured for each direction of radio waves that will be input to the first determination unit 152, and converting it into a format easily processed by the first determination unit 152 and the second determination unit 153. Therefore, according to this embodiment, the cause of the data loss can be inferred using a pre-trained machine learning model.

Furthermore, this will enable an overall improvement in service quality in mobile communication systems, such as 5G systems, and thus contribute to Goal 9 of the United Nations-led Sustainable Development Goals (SDGs): "Build resilient infrastructure, promote sustainable industrialization and foster innovation."

The embodiments of the present invention have been described in detail above with reference to the drawings. However, the specific configuration is not limited to these embodiments, and design modifications and other changes that do not depart from the spirit of the present invention are also included.

Alternatively, computer programs for realizing the functions of each of the above-mentioned devices may be recorded on a computer-readable recording medium, and the programs recorded on this recording medium may be loaded into a computer system and executed. Note that the term "computer system" here may include hardware such as an operating system and peripheral devices.
Furthermore, "computer-readable recording media" refers to writable non-volatile memory such as flexible disks, magneto-optical disks, ROMs, and flash memory, portable media such as DVDs (Digital Versatile Discs), and storage devices such as hard disks built into computer systems.

Furthermore, "computer-readable recording media" includes volatile memory (e.g., DRAM (Dynamic Random Access Memory)) within computer systems that act as servers or clients when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, which retains the program for a certain period of time. In addition, the above program may be transmitted from the computer system that stores the program in a memory device, etc., to another computer system via a transmission medium or by transmission waves within the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium that has the function of transmitting information, such as a network such as the Internet (communication network) or a communication line such as a telephone line (communication line).
Furthermore, the above program may be intended to implement some of the functions described above. It may also be a so-called differential file (differential program) that can implement the aforementioned functions in combination with programs already recorded in the computer system.

1…Communication system, D…Communication device, 2…Transmitter, 3…Receiver, 21…Beam sweep unit, 31…Beam receiving unit, 10…Directional estimation device, 11…Sweep result storage unit, 12…Directional estimation unit, 13…Estimation result storage unit, 14…Defect detection unit, 15…Defect factor detection unit, 16…Sweep range determination unit, 151…Preprocessing unit, 152…First determination unit, 153…Second determination unit, 154…Output unit

Claims (9)

A direction estimation device that receives radio waves emitted in multiple directions from a predetermined device, and estimates the radio waves of a suitable direction by measuring the radio wave intensity of the received radio waves for each direction of the emitted radio waves,
A missing signal detection unit determines whether or not there is a missing direction among the radio wave intensity received from multiple directions of radio waves output from a predetermined device ,
If it is determined that there is a missing direction, a missing data cause determination unit determines the cause of the missing data based on the time-dependent changes in the time-series data of radio wave intensity for each sweep direction.
A direction estimation device equipped with the following features. The missing factor determination unit is,
A first determination unit determines whether the cause of the loss is due to shielding, based on the trend of changes over time in time-series data of radio wave intensity for each sweep direction .
A second determination unit, when the cause of the loss is not due to shielding, determines whether the cause of the loss is due to moving out of the communication range, based on the trend of changes over time in the time-series data of radio wave intensity for each sweep direction.
The direction estimation device according to claim 1, comprising: The direction estimation device according to claim 2, wherein the second determination unit determines that if the time-series data of radio wave intensity for each sweep direction changes abruptly, the cause of the missing data is due to a movement prediction error. The orientation estimation device according to claim 2, wherein the first determination unit includes a trained model that has been pre-trained using as training data the time-dependent changes in time-series data of radio wave intensity for each sweep direction and information on whether or not the cause of the loss is due to shielding. The orientation estimation device according to claim 2, wherein the second determination unit includes a trained model that has been pre-trained using as training data the time-dependent changes in time-series data of radio wave intensity for each sweep direction and information on whether or not the cause of the loss is due to going outside the communication range. The orientation estimation device according to claim 4 or 5, wherein both the trained model included in the first determination unit and the trained model included in the second determination unit are trained using a classifier such as the K-nearest neighbors algorithm. The direction estimation device according to any one of claims 2 to 5, wherein the missing factor determination unit further comprises a preprocessing unit that performs preprocessing to determine the range of radio wave intensities measured for each direction of radio waves that are input to the first determination unit. A direction estimation method that estimates the direction of radio waves by receiving radio waves emitted in multiple directions from a predetermined device and measuring the radio wave intensity of the received radio waves for each direction of the emitted radio waves,
A defect detection step that determines whether or not there is a direction with missing radio wave intensity among the radio wave intensity received from multiple directions output from a predetermined device ,
If it is determined that there is a missing direction, a missing data cause determination step is performed to determine the cause of the missing data based on the time-dependent changes in the time-series data of radio wave intensity for each sweep direction.
A method for estimating direction, comprising the following characteristics. On the computer,
A program that receives radio waves emitted in multiple directions from a predetermined device, and estimates the radio waves in a suitable direction by measuring the radio wave intensity of the received radio waves for each direction of the emitted radio waves,
A step to determine whether or not there is a missing direction among the radio wave intensity received from multiple directions of radio waves output from a predetermined device ,
If it is determined that there is a missing direction, a missing data cause determination step is performed to determine the cause of the missing data based on the time-dependent changes in the time-series data of radio wave intensity for each sweep direction .
A program that executes the command.