A channel measurement based on-site metrology method and system for peripheral base stations

Abstract

The application relates to the technical field of image communication, in particular to a presence measurement method and system of a peripheral base station based on channel measurement, which comprises collecting multimedia interaction signals sent by a user interaction terminal, collecting interaction characteristic sampling data containing phase information and perception parameters; calculating an angle of arrival (AoA) based on the interaction characteristic sampling data, obtaining distance information based on the perception parameters, fusing signal strength indication, and generating space parameters; extracting multipath characteristics based on the interaction characteristic sampling data, judging a viewing direction state of a user; matching the space parameters with a multimedia visual area model, performing time sequence fusion on the space parameters within a preset time window, and calculating a confidence degree in combination with the viewing direction state of the user; when it is detected that multiple user interaction terminals are simultaneously in a presence state, generating a viewer presence event, mapping the viewer presence event and an interaction event of the user interaction terminal into a multimedia control instruction, and transmitting the multimedia control instruction to the smart television.

Description

Technical Field

[0001] This invention relates to the field of image communication technology, specifically to a method and system for measuring the presence of peripheral base stations based on channel measurement. Background Technology

[0002] With the increasing popularity of smart homes and wearable devices, the interaction between smart TVs and users is gradually evolving from traditional remote controls to intelligent sensing. User presence detection, as the foundation for functions such as automatic playback control, personalized recommendations, and ad exposure measurement, has become an important research direction in the field of smart TVs.

[0003] Currently, the main technical approaches for detecting the presence of smart TV users are as follows: The first approach is a TV-integrated solution, which achieves presence detection by deploying a positioning algorithm or installing a system-level SDK within the TV system. This solution requires obtaining system permissions or HAL layer adaptation. Significant differences in system customization among different TV manufacturers result in poor portability, high adaptation costs, and high consumption of computing resources on the TV.

[0004] The second approach is based on on-device computing, which involves deploying presence detection algorithms on wearable devices such as smart bracelets or smart rings. However, due to the limited battery capacity and processing power of wearable devices, complex signal processing and positioning algorithms can significantly increase power consumption, shorten battery life, and increase product size.

[0005] The third approach is a simple bridging scheme based on Received Signal Strength Indication (RSSI), which determines the relative distance between the user and the television by measuring the RSSI value. However, RSSI, as a single scalar parameter, cannot accurately characterize the multipath propagation characteristics of the wireless channel. In a living room environment, it is easily affected by factors such as people walking around, furniture obstruction, and wall reflections, causing the detection results to fluctuate frequently near the area boundary, making it difficult to achieve stable and reliable presence determination.

[0006] In summary, existing technical solutions have shortcomings in terms of system invasiveness, computing resource allocation, detection accuracy, and state stability, making it difficult to meet the actual needs of high-precision detection and measurement of user presence in smart TV scenarios.

[0007] To address this, a method and system for measuring the presence of peripheral base stations based on channel measurements is proposed. Summary of the Invention

[0008] The purpose of this invention is to provide a peripheral base station presence measurement method and system based on channel measurement. By collecting interactive terminal signals to obtain interactive feature sampling data, the AoA and distance are calculated to generate spatial parameters. Utilizing the multipath characteristics of the data, and based on the interactive link quality and phase consistency indicators, the method accurately determines whether the user is focused on watching a movie or has their gaze deviated. Combined with a multimedia visual area model with a hysteresis mechanism, the method performs time-series fusion to map multi-terminal presence events into multimedia control commands and transmit them to the smart TV.

[0009] To achieve the above objectives, the present invention provides the following technical solution: A peripheral base station presence measurement method based on channel measurement includes: Collect multimedia interaction signals sent by the user interaction terminal, and collect interaction feature sampling data and perception parameters containing phase information; The angle of arrival (AoA) is calculated based on interactive feature sampling data, distance information is obtained based on sensing parameters, and signal strength indication is fused to generate spatial parameters including angle information and distance partition information. Multipath features are extracted based on interactive feature sampling data to determine the user's viewing orientation. Specifically, when the interactive link quality index is higher than the first feature threshold and the phase consistency index is higher than the second feature threshold, it is determined to be a face-to-face state. When the interactive link quality index is lower than the third feature threshold and / or the phase consistency index is lower than the fourth feature threshold, it is determined to be a divergence state. The spatial parameters are matched with the multimedia visual area model, and the spatial parameters are temporally fused within a preset time window. The confidence level is calculated in combination with the user's viewing orientation. The multimedia visual area model is a spatial area centered on the smart TV display terminal. An entry threshold and an exit threshold are set to form a hysteresis mechanism. When multiple user interaction terminals are detected to be present at the same time, a viewer presence event is generated. The viewer presence event and the interaction events of the user interaction terminals are mapped into multimedia control commands and transmitted to the smart TV.

[0010] Preferably, the specific process for calculating the angle of arrival (AoA) includes: receiving signals from the same device through at least two antenna elements in a multi-antenna array, acquiring interactive feature sampling data corresponding to each element, wherein the interactive feature sampling data includes in-phase component I and quadrature component Q; calculating the phase difference between adjacent elements based on the interactive feature sampling data received by different elements, wherein the phase difference is obtained by comparing the phase information of the interactive feature sampling data of different elements; determining the incident angle of the signal according to the phase difference, the physical spacing between antenna elements and the wavelength of the device signal, and obtaining the angle of arrival (AoA) value relative to the front direction of the external base station; and repeating the calculation within a preset sampling period to obtain time series data of the angle of arrival (AoA).

[0011] Preferably, the specific process for generating distance interval information includes: initiating a sensing parameter request during device connection, receiving sensing parameter response data returned by the user interaction terminal, wherein the sensing parameter response data includes round-trip time information and signal quality parameters; extracting the round-trip time information and the received signal strength indication value at the current moment to form a distance feature vector; comparing the distance feature vector with a preset distance interval threshold table, wherein the distance interval threshold table includes the judgment boundaries of near distance interval, medium distance interval, and far distance interval; classifying the current distance into one of the near distance interval, medium distance interval, and far distance interval levels according to the comparison result, and outputting a distance interval label.

[0012] Preferably, the method for obtaining the spatial parameters includes: establishing a multimedia visible area model with the center of the antenna array as the reference origin and the frontal orientation as the reference direction; using the angle of arrival (AoA) as the azimuth component in the multimedia visible area model to characterize the angular orientation of the user interaction terminal relative to the peripheral base station; using the distance interval label obtained based on the sensing parameters as the distance level component in the multimedia visible area model to characterize the relative distance range between the user interaction terminal and the peripheral base station; combining the azimuth component and the distance level component to form spatial parameters, wherein the spatial parameters also include corresponding sampling timestamps and device identifiers; and determining the spatial region where the user interaction terminal is located based on the azimuth component and the distance level component in the spatial parameters, wherein the spatial region is a sector-shaped interval centered on the peripheral base station and jointly defined by the angle range and the distance level.

[0013] Preferably, the specific process for determining the viewing orientation status includes: collecting interactive feature sampling data sequences within multiple consecutive sampling periods; calculating the signal amplitude value for the interactive feature sampling data in each sampling period; statistically analyzing the variance of the signal amplitude value within a time window; and using the reciprocal of the variance as a signal amplitude stability index; calculating the phase value sequence for the interactive feature sampling data in consecutive sampling periods; statistically analyzing the mean absolute value difference of the phase values ​​between adjacent sampling periods; and using the reciprocal of the mean absolute value difference as a phase consistency index; establishing an orientation determination feature space, with the signal amplitude stability index as the first dimension and the phase consistency index as the second dimension; setting determination boundaries for facing and departing regions in the feature space; mapping the current signal amplitude stability index and phase consistency index to the feature space; and outputting a viewing orientation status label based on the region it falls into; performing temporal filtering on the viewing orientation status labels at multiple consecutive times; and confirming the user's orientation as facing when the proportion of facing status labels exceeds a preset threshold, otherwise confirming it as departing.

[0014] Preferably, the specific implementation of the hysteresis mechanism includes: a spatial area centered on the front of the smart TV; setting entry conditions, including a first angle threshold and a first distance threshold, whereby the absolute value of the arrival angle AoA is less than the first angle threshold and the distance interval label does not exceed the first distance threshold, thus indicating that the entry condition is met; setting exit conditions, including a second angle threshold and a second distance threshold, whereby the absolute value of the arrival angle AoA is greater than the second angle threshold or the distance interval label exceeds the second distance threshold, thus indicating that the exit condition is met, wherein the second angle threshold is greater than the first angle threshold and the second distance threshold is less stringent than the first distance threshold; maintaining a current presence status identifier, whereby the current status is an exit status and the entry condition is met continuously for more than a first preset duration, the status is switched to an entry status; and the current status is an entry status and the exit condition is met continuously for more than a second preset duration, thus switching the status to an exit status, avoiding frequent status switching through the threshold difference between the entry and exit conditions and the time constraint of the orientation status; and confirming a valid presence status only when the spatial parameters match the multimedia visible area model and the viewing orientation status is a face-viewing status.

[0015] Preferably, the specific process of confidence calculation includes: collecting spatial parameter samples at multiple consecutive moments within a preset time window, each sample including the angle of arrival (AoA) value, distance interval label, viewing orientation status label, and sampling timestamp; performing consistency analysis on the AoA values ​​within the time window, and counting the number of samples falling within the viewing area angle range as a first proportion of the total number of samples; performing stability analysis on the distance interval labels within the time window, and counting the number of samples remaining within the acceptable distance range as a second proportion of the total number of samples; performing statistical analysis on the viewing orientation status labels within the time window, and counting the number of samples marked as facing the view as a third proportion of the total number of samples; extracting motion state data reported by the user interaction terminal, determining it as a stationary state or a moving state, assigning a first weight coefficient to the stationary state, and assigning a second weight coefficient to the moving state, wherein the first weight coefficient is greater than the second weight coefficient.

[0016] Preferably, the confidence calculation process further includes: determining an orientation weight coefficient based on the viewing orientation status label at the current moment; assigning a third weight coefficient when the viewing orientation status is facing the viewer; assigning a fourth weight coefficient when the viewing orientation status is facing away from the viewer; the third weight coefficient being greater than the fourth weight coefficient; calculating the presence confidence value at the current moment by combining the first ratio, the second ratio, the third ratio, the motion state weight coefficient, and the orientation weight coefficient; confirming the presence status is valid when the confidence value is higher than a preset threshold and the current viewing orientation status is facing the viewer; and outputting a judgment result including the presence status identifier and the confidence value.

[0017] A peripheral base station presence metering system based on channel measurement includes: The data analysis module collects multimedia interactive signals sent by the user's interactive terminal, and collects interactive feature sampling data and sensing parameters containing phase information; it calculates the angle of arrival (AoA) based on the interactive feature sampling data, obtains distance information based on the sensing parameters, fuses signal strength indicators, and generates spatial parameters including angle information and distance partition information. The orientation state analysis module extracts multipath features based on interactive feature sampling data to determine the user's viewing orientation state. The hysteresis mechanism analysis module matches spatial parameters with the multimedia visual area model, performs temporal fusion of spatial parameters within a preset time window, and calculates confidence based on the user's viewing orientation. The multimedia visual area model is a spatial area centered on the smart TV display terminal, and entry and exit thresholds are set to form a hysteresis mechanism. The data transmission module generates a viewer presence event when it detects that multiple user interaction terminals are present at the same time. It then maps the viewer presence event and the interaction events of the user interaction terminals into multimedia control commands and transmits them to the smart TV.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention integrates multi-dimensional channel measurement information, including Angle of Arrival (AoA), distance range, RSSI, phase consistency, and multipath characteristics, to construct a unified spatial parameter and viewing orientation determination mechanism. Compared to existing technologies that rely solely on infrared, cameras, or single signal strength, this invention effectively overcomes misjudgment problems caused by obstruction, lighting variations, multi-user interference, and non-directly facing scenarios. In particular, through phase difference calculation and time series consistency analysis using multi-antenna arrays, it enables presence recognition to operate independently of visual perception devices, significantly improving stability and accuracy in complex indoor environments while ensuring privacy. This makes it suitable for real-world application scenarios such as living rooms and multi-user environments.

[0019] 2. This invention not only determines whether a user is within a spatial range, but also constructs a viewing orientation determination feature space through signal amplitude stability and phase consistency, achieving accurate differentiation between "face-to-face viewing" and "viewing direction deviation" states. Furthermore, it introduces a hysteresis mechanism with different entry and exit thresholds. This design avoids frequent switching of presence status caused by slight head turns, brief standing, or instantaneous channel fluctuations. Compared to existing technologies that rely on instantaneous condition triggers for presence determination, this approach is smoother and more reliable in the time dimension, more accurately reflecting the user's continuous viewing behavior.

[0020] 3. This invention performs statistical analysis on angle consistency, distance stability, and orientation state proportion within a preset time window, and combines this with terminal motion state and orientation weights to calculate presence confidence, thereby achieving a quantitative assessment of presence status. Compared to the traditional binary "present / absent" judgment method, this invention can output a more reliable presence result and generate a unified audience presence event when multiple users are present simultaneously. This event is used to drive multimedia control commands on smart TVs, thereby supporting more refined and intelligent content playback, pause, and interactive control, significantly improving the user experience. Attached Figure Description

[0021] Figure 1 A flowchart of a peripheral base station presence measurement method based on channel measurement provided by the present invention; Figure 2 This invention provides a structural diagram of a peripheral base station on-site metering system based on channel measurement. Figure 3 The flowchart for determining the viewing orientation state provided by the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0023] like Figure 1 and Figure 2 As shown, the peripheral base station presence metering system based on channel measurement provided by the present invention includes three parts: a peripheral base station, a user interaction terminal, and a smart TV.

[0024] The external base station is positioned at the top or bottom edge of the smart TV, with its front facing the same direction as the TV screen. The external base station includes a multi-antenna array module, a signal processing module, and a USB interface module. The multi-antenna array module contains at least two antenna elements arranged linearly in a horizontal direction, with the physical spacing between adjacent antenna elements being half the Bluetooth signal wavelength, approximately 6.25 cm in the 2.4 GHz band. The signal processing module, integrated within the external base station, performs data processing tasks such as signal acquisition, feature extraction, spatial parameter calculation, orientation determination, and presence status assessment. The USB interface module connects to the smart TV's USB Host port via a USB Type-A or Type-C interface, and the external base station is recognized as a standard USB HID keyboard device on the smart TV.

[0025] The user interaction terminal is a smartwatch or smart bracelet worn on the user's wrist, with a built-in Bluetooth Low Energy communication module, supporting Bluetooth 5.1 and above for angle-of-arrival positioning and channel detection. The user interaction terminal periodically sends broadcast signals containing device identification, a side tag, and status information, with a broadcast period of 100 milliseconds. The side tag identifies whether the user interaction terminal is worn on the user's left or right wrist and is set by the user during initial pairing.

[0026] A smart TV is a TV terminal device with a USB Host interface, which can receive and respond to USB HID keyboard events and perform corresponding media control operations, including play, pause, volume adjustment, and navigation functions.

[0027] This invention provides a method and system for measuring the presence of peripheral base stations based on channel measurement. The technical solution is as follows: Multimedia interactive signals sent by user interactive terminals are collected, along with interactive feature sampling data and sensing parameters containing phase information; the angle of arrival (AoA) is calculated based on the interactive feature sampling data, distance information is obtained based on the sensing parameters, and signal strength indicators are fused to generate spatial parameters including angle information and distance partitioning information; multipath features are extracted based on the interactive feature sampling data to determine the user's viewing orientation; wherein, when the interactive link quality index is higher than a first feature threshold and the phase consistency index is higher than a second feature threshold, it is determined to be a face-to-face state; when... When the interaction link quality index is lower than the third feature threshold and / or the phase consistency index is lower than the fourth feature threshold, it is determined to be in a divergent state. Based on the spatial parameters and the multimedia visual area model, the spatial parameters are temporally fused within a preset time window, and the confidence level is calculated in combination with the user's viewing orientation. The multimedia visual area model is a spatial area centered on the smart TV display terminal, and an entry threshold and an exit threshold are set to form a hysteresis mechanism. When multiple user interaction terminals are detected to be present at the same time, a viewer presence event is generated, and the viewer presence event and the interaction event of the user interaction terminal are mapped into multimedia control commands and transmitted to the smart TV.

[0028] Preferably, the specific process for calculating the angle of arrival (AoA) includes: receiving signals from the same device through at least two antenna elements in a multi-antenna array, acquiring interactive feature sampling data corresponding to each element, wherein the interactive feature sampling data includes in-phase component I and quadrature component Q; calculating the phase difference between adjacent elements based on the interactive feature sampling data received by different elements, wherein the phase difference is obtained by comparing the phase information of the interactive feature sampling data of different elements; determining the incident angle of the signal according to the phase difference, the physical spacing between antenna elements and the wavelength of the device signal, and obtaining the angle of arrival (AoA) value relative to the front direction of the external base station; and repeating the calculation within a preset sampling period to obtain time series data of the angle of arrival (AoA).

[0029] Specifically, the peripheral base station receives multimedia interactive signals sent by the user interaction terminal through a multi-antenna array and collects interactive feature sampling data containing phase information.

[0030] The interactive feature sampling data consists of in-phase component I and quadrature component Q sequences obtained by the antenna elements of the multi-antenna array after receiving Bluetooth signals from the user's interactive terminal, through low-noise amplification, mixing, filtering, and analog-to-digital conversion. The interactive feature sampling data includes in-phase component I and quadrature component Q, which together characterize the amplitude and phase information of the received signal at the current sampling moment. The signal amplitude is obtained by calculating the square root of the sum of the squares of in-phase component I and quadrature component Q, and the signal phase is obtained by calculating the arctangent of the ratio of quadrature component Q to in-phase component I.

[0031] The calculation process for the Angle of Arrival (AoA) is as follows: The external base station acquires the interactive feature sampling data received by two adjacent antenna elements at the same time, calculates the signal phase value corresponding to each of the two antenna elements, and then calculates the difference between the two phase values ​​to obtain the phase difference. Since the Bluetooth signal arrives at the antenna array in the form of a plane wave, there is a fixed path difference between the signals received by adjacent antenna elements, which is related to the incident angle of the signal. Based on the phase difference, the physical distance between the antenna elements, and the wavelength of the Bluetooth signal, the incident angle of the signal relative to the normal direction of the antenna array is determined by arcsine operation. This angle is the Angle of Arrival (AoA) value. The range of the Angle of Arrival (AoA) value is from -90 degrees to +90 degrees, where 0 degrees indicates that the signal arrives from directly in front of the external base station, a positive value indicates that the signal arrives from the right side of the external base station, and a negative value indicates that the signal arrives from the left side of the external base station.

[0032] The peripheral base station repeats the above calculation process in each sampling period to obtain the time-series data of the angle of arrival (AoA). The sampling period is consistent with the broadcast period of the user interaction terminal, which is 100 milliseconds. The time-series data is used for subsequent time-series fusion processing to improve the stability and accuracy of the angle of arrival estimation.

[0033] Preferably, the specific process for generating distance interval information includes: initiating a sensing parameter request during device connection, receiving sensing parameter response data returned by the user interaction terminal, wherein the sensing parameter response data includes round-trip time information and signal quality parameters; extracting the round-trip time information and the received signal strength indication value at the current moment to form a distance feature vector; comparing the distance feature vector with a preset distance interval threshold table, wherein the distance interval threshold table includes the judgment boundaries of near distance interval, medium distance interval, and far distance interval; classifying the current distance into one of the near distance interval, medium distance interval, and far distance interval levels according to the comparison result, and outputting a distance interval label.

[0034] Specifically, the peripheral base station obtains distance estimation information based on sensing parameters and converts the distance estimation results into distance interval labels.

[0035] After the Bluetooth connection is established, the peripheral base station sends a sensing parameter request to the user interface terminal. Upon receiving the request, the user interface terminal immediately returns sensing parameter response data. The peripheral base station records the time interval from sending the request to receiving the response, and after deducting the processing delay of the user interface terminal, obtains the round-trip time information of the signal. Simultaneously, the peripheral base station extracts the signal strength indicator value of the received signal at the current moment.

[0036] The external base station combines round-trip time (RTT) information and RSSI values ​​to form a distance feature vector. RTT information reflects the actual distance the signal travels in space, while RSSI values ​​reflect the signal strength attenuation after path loss. Due to multipath effects and obstructions in indoor environments, the accuracy of distance estimation using either RTT or RSSI alone is limited. Therefore, this invention employs a fusion of both methods to improve the reliability of distance estimation. The RTT is obtained through Bluetooth LE channel model readability or proprietary Bluetooth extension functionality.

[0037] The peripheral base station pre-stores a distance interval threshold table, which defines the boundaries for three distance intervals: the near distance interval corresponds to the range within 2 meters directly in front of the smart TV, the medium distance interval corresponds to the range between 2 and 4 meters, and the far distance interval corresponds to the range beyond 4 meters. The peripheral base station compares the current distance feature vector with the threshold table, and based on the comparison result, classifies the current distance into one of the three levels: near, medium, or far, and outputs the corresponding distance interval label.

[0038] Preferably, the method for obtaining the spatial parameters includes: establishing a multimedia visible area model with the center of the antenna array as the reference origin and the frontal orientation as the reference direction; using the angle of arrival (AoA) as the azimuth component in the multimedia visible area model to characterize the angular orientation of the user interaction terminal relative to the peripheral base station; using the distance interval label obtained based on the sensing parameters as the distance level component in the multimedia visible area model to characterize the relative distance range between the user interaction terminal and the peripheral base station; combining the azimuth component and the distance level component to form spatial parameters, wherein the spatial parameters also include corresponding sampling timestamps and device identifiers; and determining the spatial region where the user interaction terminal is located based on the azimuth component and the distance level component in the spatial parameters, wherein the spatial region is a sector-shaped interval centered on the peripheral base station and jointly defined by the angle range and the distance level.

[0039] Specifically, the peripheral base station generates spatial parameters based on the angle of arrival (AoA) and distance interval labels, and determines the spatial area where the user interaction terminal is located.

[0040] The peripheral base station uses the geometric center of the multi-antenna array as the reference origin and the front orientation of the peripheral base station as the reference direction to establish a multimedia viewing area model. Since the peripheral base station is installed on the smart TV and its front orientation is consistent with the orientation of the TV screen, the reference direction points to the viewing area directly in front of the smart TV.

[0041] In the multimedia visible area model, the angle of arrival (AoA) serves as the azimuth component, representing the angular orientation of the user terminal relative to the external base station. The distance interval label, as the distance level component, represents the relative distance range between the user terminal and the external base station. The external base station combines the azimuth and distance level components to form spatial parameters. These spatial parameters also include the corresponding sampling timestamp and the user terminal's device identifier, used to distinguish measurement data from different times and different devices.

[0042] The method for constructing the multimedia visible area model includes: establishing a polar coordinate system with the geometric center of the antenna array of the peripheral base station as the reference origin and the frontal orientation of the peripheral base station as the reference direction; using the angle of arrival (AoA) calculated based on interactive feature sampling data as the azimuth component to represent the angular orientation of the user interaction terminal relative to the peripheral base station; using the distance interval label obtained based on sensing parameters as the distance level component to represent the relative distance range between the user interaction terminal and the peripheral base station; combining the azimuth component and the distance level component, and adding a sampling timestamp and device identifier to form complete spatial parameters. Based on the azimuth component and the distance level component in the spatial parameters, the space centered on the peripheral base station is divided into multiple sector intervals defined by both the angle range and the distance level, achieving a discretized representation of the spatial region. The multimedia visible area model is a deterministic model based on a geometric coordinate system, and its parameters are calibrated using preset angle range thresholds and distance binning threshold tables, eliminating the need for iterative training based on sample data.

[0043] The external base station determines the spatial region where the user terminal is located based on the azimuth and distance class components in the spatial parameters. The spatial region is a sector-shaped area centered on the external base station, defined by both the azimuth range and the distance class. In this embodiment, the external base station divides the entire space into nine regions, each corresponding to a combination of three azimuth ranges and three distance classes. The three azimuth ranges are: the left region corresponds to an angle of arrival (AoA) less than -30 degrees; the region directly in front corresponds to an angle of arrival (AoA) between -30 and +30 degrees; and the right region corresponds to an angle of arrival (AoA) greater than +30 degrees. The three distance classes are near range, medium range, and long range.

[0044] Preferably, the specific process for determining the viewing orientation is as follows: Figure 3The process includes: collecting interactive feature sampling data sequences over multiple consecutive sampling periods; calculating the signal amplitude value for the interactive feature sampling data in each sampling period; calculating the variance of the signal amplitude value within a time window; and using the reciprocal of the variance as a signal amplitude stability index; calculating the phase value sequence for the interactive feature sampling data in consecutive sampling periods; calculating the mean absolute difference of the phase values ​​between adjacent sampling periods; and using the reciprocal of the mean absolute difference as a phase consistency index; establishing an orientation determination feature space, with the signal amplitude stability index as the first dimension and the phase consistency index as the second dimension; setting determination boundaries for facing and departing regions in the feature space; mapping the current signal amplitude stability index and phase consistency index to the feature space; and outputting a viewing orientation status label based on the region it falls into; performing temporal filtering on the viewing orientation status labels over multiple consecutive times; confirming the user's orientation as facing when the proportion of facing status labels exceeds a preset threshold, otherwise confirming it as departing.

[0045] Specifically, the peripheral base station extracts multipath features based on the interaction feature sampling data and determines the viewing orientation state based on the multipath features.

[0046] The physical basis for judging viewing orientation is the blocking effect of the human body on Bluetooth signals. In the 2.4GHz band, human tissue has a strong absorption and scattering effect on electromagnetic waves. When the user faces the smart TV, the signal propagation path between the user interaction terminal worn on the wrist and the external base station is less blocked by the human torso, the direct component dominates, and the amplitude and phase of the received signal are relatively stable. When the user turns away from the smart TV, the human torso is located between the user interaction terminal and the external base station, the direct path is blocked, and the signal mainly reaches the external base station through diffraction at the edge of the body and reflection from the surrounding environment. The multipath component increases, and the amplitude and phase of the received signal exhibit greater fluctuations.

[0047] Multipath characteristics refer to the changes in channel characteristics caused by multipath propagation mechanisms such as reflection, diffraction, and scattering of wireless signals in indoor environments. In this invention, the quantitative characterization of multipath characteristics is achieved using two indicators: signal amplitude stability and phase consistency. The multipath characteristic extraction process is as follows: The peripheral base station collects interactive feature sampling data sequences over multiple consecutive sampling periods. The number of sampling periods is the preset feature extraction window length, which is set to 20 periods in this embodiment, corresponding to a 2-second time window. For the interactive feature sampling data of each sampling period, the peripheral base station calculates the signal amplitude value. Then, the peripheral base station calculates the variance of the changes in all amplitude values ​​within the time window, and uses the reciprocal of the variance as the signal amplitude stability index. The larger the amplitude stability index, the more stable the signal amplitude is within the time window.

[0048] Simultaneously, the peripheral base station calculates the phase value sequence of interactive feature sampling data from continuous sampling periods, statistically analyzes the absolute difference between phase values ​​between adjacent sampling periods, calculates the mean of all absolute difference values, and uses the reciprocal of this mean as the phase consistency index. The larger the phase consistency index, the smoother the change in signal phase within the time window.

[0049] The peripheral base station establishes an orientation determination feature space, with the signal amplitude stability index as the horizontal axis and the phase consistency index as the vertical axis. In this two-dimensional feature space, the peripheral base station pre-sets determination boundaries for facing and departing regions. The facing region is located in the upper right corner of the feature space, corresponding to a situation where both the amplitude stability index and the phase consistency index are high. When the user faces the smart TV, the main propagation path between the user interaction terminal worn on the wrist and the peripheral base station is the direct path (LOS), and the direct component accounts for the majority of the power of the received signal. At this time, the multipath component is relatively weak, and the amplitude and phase fluctuations of the received signal are small. Specifically, the variance of signal amplitude variation within the time window (i.e., the dispersion of amplitude values ​​at different sampling times) is small, and a larger signal amplitude stability index value can be obtained by calculating the reciprocal of the variance. The rate of phase change between adjacent sampling periods is slow, and the mean absolute value of the phase difference between adjacent samples is small, resulting in a larger phase consistency index. The divergence region is located in the lower left of the feature space, corresponding to a lower amplitude stability index or phase consistency index. The judgment boundary can be a linear boundary or a nonlinear boundary. When the user is away from the smart TV, the user's torso is located between the user interaction terminal and the external base station. The direct path is blocked, and the signal mainly reaches the external base station through diffraction on the side of the body and reflection paths from surrounding walls and furniture, resulting in a relatively larger increase in the power of multipath components. At this time, the delays of different propagation paths are different, causing the received signal to exhibit more obvious fading characteristics in time. Specifically, the amplitude of the received signal fluctuates more within the time window, the variance of the change increases, and the signal amplitude stability index decreases. Due to the mutual interference of the phase contributions of multiple paths, the phase change between adjacent sampling periods exhibits more irregular fluctuations, the mean absolute value of the phase difference between adjacent samples increases, and its reciprocal, i.e., the phase consistency index, decreases. In this embodiment, a rectangular area formed by two threshold lines is used as the facing area.

[0050] The peripheral base station maps the signal amplitude stability index and phase consistency index calculated at the current moment to the orientation determination feature space, determines whether the point falls into the facing region or the departing region, and outputs the corresponding viewing orientation status label. When the signal amplitude stability index is higher than the first feature threshold and the phase consistency index is higher than the second feature threshold, it is determined to be in a facing state. When the signal amplitude stability index is lower than the third feature threshold or the phase consistency index is lower than the fourth feature threshold, it is determined to be in a departing state. The first and third feature thresholds correspond to the amplitude stability thresholds for facing and departing determinations, respectively, and the second and fourth feature thresholds correspond to the phase consistency thresholds for facing and departing determinations, respectively. The third feature threshold is less than or equal to the first feature threshold, and the fourth feature threshold is less than or equal to the second feature threshold. The difference between the two sets of thresholds forms a buffer for determination, avoiding frequent state switching near the boundary.

[0051] To further improve the stability of orientation status determination, the external base station performs time-series filtering on viewing orientation status tags across multiple consecutive time points. Specifically, the external base station maintains a sliding window with a length of 10 decision periods. In each decision period, the external base station counts the percentage of facing status tags within the window. When the percentage of facing status tags exceeds a preset threshold, the user's orientation is confirmed as facing; otherwise, it is confirmed as unfacing. In this embodiment, the preset threshold is set to 60%.

[0052] The viewing orientation determination is based on the following physical principle: As a biological tissue with a high water content, the human body has a significant absorption and scattering effect on Bluetooth signals in the 2.4GHz band. When the user interaction terminal is worn on the user's wrist, the wireless channel characteristics between the user interaction terminal and the external base station will change with the orientation of the user's torso relative to the external base station.

[0053] In a typical movie-watching scenario, users usually sit facing the smart TV in a relatively fixed posture, with their hands naturally placed in front of or to the sides of their body. In this situation, the user interface terminal worn on the wrist is located in the area between the user's torso and the smart TV. When the user is facing the smart TV, the direct path between the user interface terminal and the external base station is less obstructed by the user's torso, resulting in a relatively stable signal propagation path. However, when the user is away from the smart TV, the torso is positioned between the user interface terminal and the external base station, obstructing the direct path. The signal then needs to reach the external base station through diffraction or reflection, leading to greater fluctuations in the amplitude and phase of the received signal.

[0054] It should be noted that the orientation state judgment described in this invention is a probabilistic estimation method based on statistical features. By performing time-series analysis and multi-factor fusion of channel features within a preset time window, the accuracy and robustness of the judgment are improved. This method is suitable for movie-watching scenarios where users maintain a relatively stable sitting posture, and by jointly determining the orientation with spatial position parameters, it reduces the impact of misjudgments due to single factors.

[0055] The external base station associates spatial parameters with the viewing orientation status label at the same time to form a complete user status description containing location and orientation information. This user status description serves as input data for subsequent presence status determination.

[0056] Preferably, the specific implementation of the hysteresis mechanism includes: a spatial area centered on the front of the smart TV; setting entry conditions, including a first angle threshold and a first distance threshold, whereby the absolute value of the arrival angle AoA is less than the first angle threshold and the distance interval label does not exceed the first distance threshold, thus indicating that the entry condition is met; setting exit conditions, including a second angle threshold and a second distance threshold, whereby the absolute value of the arrival angle AoA is greater than the second angle threshold or the distance interval label exceeds the second distance threshold, thus indicating that the exit condition is met, wherein the second angle threshold is greater than the first angle threshold and the second distance threshold is less stringent than the first distance threshold; maintaining a current presence status identifier, whereby the current status is an exit status and the entry condition is met continuously for more than a first preset duration, the status is switched to an entry status; and the current status is an entry status and the exit condition is met continuously for more than a second preset duration, thus switching the status to an exit status, avoiding frequent status switching through the threshold difference between the entry and exit conditions and the time constraint of the orientation status; and confirming a valid presence status only when the spatial parameters match the multimedia visual area model and the viewing orientation status is facing.

[0057] Specifically, the peripheral base station matches spatial parameters with the multimedia visual area model and avoids frequent switching of the presence status through a hysteresis mechanism.

[0058] The multimedia viewing area model is defined as the spatial area centered on the front of the smart TV. This area is defined by both an angular range and a distance range. In this embodiment, the angular range is set to an absolute value of the arrival angle AoA not exceeding 45 degrees, and the distance range is set to a distance interval labeled as near or medium distance. The spatial area that meets the above conditions is the viewing area.

[0059] To avoid frequent switching of presence status near the area boundary, the external base station uses a hysteresis mechanism for state management. The core of the hysteresis mechanism is to set different judgment conditions for entering and leaving the presence status.

[0060] The entry conditions include a first angle threshold, a first distance threshold, and an orientation requirement. When the absolute value of the arrival angle AoA is less than the first angle threshold, the distance interval label does not exceed the first distance threshold, and the viewing orientation is "facing," the entry conditions are met. In this embodiment, the first angle threshold is set to 40 degrees, and the first distance threshold is set to mid-distance.

[0061] The departure conditions include a second angle threshold, a second distance threshold, and an orientation change condition. A departure condition is indicated when any of the following conditions are met: the absolute value of the arrival angle AoA is greater than the second angle threshold; the distance interval label exceeds the second distance threshold; or the viewing orientation remains in a divergent state for more than a third preset duration. In this embodiment, the second angle threshold is set to 50 degrees, the second distance threshold is set to a distance, and the third preset duration is set to 5 seconds. The second angle threshold is greater than the first angle threshold, and the second distance threshold is less lenient than the first distance threshold; this threshold difference forms a hysteresis interval.

[0062] The external base station maintains a current presence status identifier, which has two possible values: "entering" and "leaving". When the current status is "leaving" and the entry conditions are met continuously for more than a first preset duration, the external base station switches the status to "entering". When the current status is "entering" and the departure conditions are met continuously for more than a second preset duration, the external base station switches the status to "leaving". In this embodiment, the first preset duration is set to 2 seconds, and the second preset duration is set to 3 seconds.

[0063] The external base station confirms a valid presence status only when the spatial parameters match the multimedia viewing area model and the viewing orientation is in the facing state. This means that the user not only needs to be within the viewing area, but also needs to be facing the smart TV to be considered validly present.

[0064] Preferably, the confidence calculation process includes: collecting spatial parameter samples at multiple consecutive moments within a preset time window, each sample including the angle of arrival (AoA) value, distance interval label, viewing orientation status label, and sampling timestamp; performing consistency analysis on the AoA values ​​within the time window, and counting the number of samples falling within the viewing area angle range as a first proportion of the total number of samples; performing stability analysis on the distance interval labels within the time window, and counting the number of samples remaining within the acceptable distance range as a second proportion of the total number of samples; performing statistical analysis on the viewing orientation status labels within the time window, and counting the number of samples marked as facing the view as a third proportion of the total number of samples; extracting motion state data reported by the user interaction terminal and determining it as stationary. The system considers both static and dynamic states. A first weighting coefficient is assigned to the static state, and a second weighting coefficient is assigned to the dynamic state, with the first weighting coefficient being greater than the second. An orientation weighting coefficient is determined based on the current viewing orientation status label. A third weighting coefficient is assigned when the viewing orientation is facing the viewer, and a fourth weighting coefficient is assigned when the viewing orientation is facing away from the viewer, with the third weighting coefficient being greater than the fourth. The presence confidence value for the current moment is calculated by combining the first ratio, the second ratio, the third ratio, the dynamic state weighting coefficient, and the orientation weighting coefficient. When the confidence value is higher than a preset threshold and the current viewing orientation is facing the viewer, the presence status is confirmed as valid. The output includes the presence status identifier and the confidence value.

[0065] Specifically, the external base station performs time-series fusion of spatial parameters and orientation status within a preset time window to calculate the presence confidence.

[0066] The peripheral base station sets the time window length for time-series fusion, which is set to 5 seconds in this embodiment, corresponding to 50 sampling periods of data. Within this time window, the peripheral base station collects spatial parameter samples at multiple consecutive moments. Each sample includes the angle of arrival (AoA) value, distance interval label, viewing orientation status label, and sampling timestamp.

[0067] The confidence level calculation process includes the following steps: The external base station performs a consistency analysis on the Angle of Arrival (AoA) values ​​within the time window. Specifically, the external base station counts the number of samples falling within the angle range of the viewing area and calculates the proportion of this number to the total number of samples, obtaining the first proportion. The first proportion reflects the stability of the user's location within the viewing area in terms of angle dimension.

[0068] The external base station performs stability analysis on the distance interval labels within the time window. Specifically, the external base station counts the number of samples that remain within the acceptable distance range and calculates the proportion of this number to the total number of samples, obtaining a second proportion. In this embodiment, the acceptable distance range is set to near or medium distance. The second proportion reflects the stability of the user's location within the viewing area in the distance dimension.

[0069] The external base station performs statistical analysis on the viewing orientation status labels within the time window. Specifically, the external base station counts the number of samples marked as facing, calculates the proportion of this number to the total number of samples, and obtains a third proportion. The third proportion reflects the stability of the user's facing orientation in the orientation dimension.

[0070] The peripheral base station extracts motion state data reported by the user interaction terminal. The user interaction terminal detects the user's motion state through a built-in accelerometer and includes this motion state information in the broadcast signal. Motion states are categorized into two types: stationary and moving. The peripheral base station determines motion state weighting coefficients based on the motion state: a first weighting coefficient is assigned to stationary states, and a second weighting coefficient is assigned to moving states. The first weighting coefficient is greater than the second weighting coefficient. In this embodiment, the first weighting coefficient is set to 1.0, and the second weighting coefficient is set to 0.7. This setting reflects the fact that determining the user's presence is more reliable when the user is stationary.

[0071] The peripheral base station determines the orientation weight coefficient based on the viewing orientation status label at the current time. When the viewing orientation status is facing, a third weight coefficient is assigned; when the viewing orientation status is away, a fourth weight coefficient is assigned. The third weight coefficient is greater than the fourth weight coefficient. In this embodiment, the third weight coefficient is set to 1.0 and the fourth weight coefficient is set to 0.3.

[0072] The peripheral base station calculates the presence confidence value at the current moment by combining the first ratio, the second ratio, the third ratio, the motion state weight coefficient, and the orientation weight coefficient. Specifically, the calculation method is as follows: the first ratio, the second ratio, and the third ratio are weighted and averaged with preset weights to obtain a basic confidence value; then, the basic confidence value is multiplied by the motion state weight coefficient and the orientation weight coefficient to obtain the final presence confidence value. In this embodiment, the weights of the first ratio, the second ratio, and the third ratio are set to 0.3, 0.3, and 0.4, respectively, reflecting the important role of orientation state in presence determination.

[0073] When the presence confidence value is higher than a preset threshold and the current viewing orientation is facing, the external base station confirms the presence status is valid. In this embodiment, the preset threshold is set to 0.6. The external base station outputs a determination result including the presence status identifier and the confidence value.

[0074] This invention proposes a peripheral base station presence measurement system and method based on channel measurement. By deploying a peripheral base station with a multi-antenna array on a smart TV and combining it with Bluetooth signals from a wearable user interaction terminal, the system accurately determines the effective presence of a user. The system calculates the angle of arrival (AoA) using multi-antenna interaction feature sampling data and fuses sensing parameters and RSSI to generate distance bucket information, constructing a spatial parameter model including angle and distance. Simultaneously, it extracts multipath features based on signal amplitude stability and phase consistency to determine the user's orientation (facing or away from the TV). Based on this, the peripheral base station matches the spatial parameters with a multimedia visible area model and uses entry / exit condition differences to form a hysteresis mechanism, avoiding frequent switching of presence status. Furthermore, the system performs temporal fusion of angle, distance, orientation, and user movement status within a time window to calculate the presence confidence level, confirming effective presence only when the position is reasonable and the orientation is facing the user.

[0075] Example 2: The system architecture in this embodiment is the same as in Embodiment 1, comprising three parts: an external base station, a user interaction terminal, and a smart TV. The external base station is installed on the top of the smart TV, and the user interaction terminal is worn on the user's wrist. This embodiment focuses on describing the enhanced functions of the system in complex application scenarios, including adaptive calibration capabilities for different deployment environments, a multi-frequency fusion method that leverages Bluetooth frequency hopping characteristics to improve the robustness of feature extraction, and an adaptive window mechanism that dynamically adjusts processing parameters based on user behavior.

[0076] The peripheral base station receives interactive feature sampling data sent by the user interaction terminal through a multi-antenna array, and collects interactive feature sampling data containing phase information and sensing parameters. Based on the interactive feature sampling data, the peripheral base station calculates the angle of arrival (AoA) according to the method described in Embodiment 1, and obtains the angular orientation information of the user interaction terminal relative to the peripheral base station. Based on the sensing parameters, the peripheral base station obtains round-trip time information and received signal strength indication value, and classifies the distance estimation results into three levels—near range, medium range, or long range—by comparing them with a distance bucketing threshold table, and outputs distance interval labels.

[0077] The external base station uses the geometric center of the multi-antenna array as the reference origin and the facing direction as the reference direction. It combines the angle of arrival (AoA) as the azimuth component and the distance interval label as the distance level component to form spatial parameters. These spatial parameters also include sampling timestamps and device identifiers to distinguish measurement data from different times and different devices.

[0078] To address the issue of poor adaptability of fixed thresholds due to differences in multipath characteristics across various indoor environments, this embodiment introduces an environmental adaptive calibration mechanism. This calibration step is performed upon initial system deployment or after a user changes their operating environment.

[0079] It also includes an environment adaptive calibration step, which is performed after the system's initial deployment or after an environmental change. The specific process includes: The peripheral base station enters calibration mode and sends calibration instructions to the user interaction terminal, prompting the user to perform the standard calibration action sequence; During the calibration phase, the user maintains a smart TV orientation, while the peripheral base station continuously collects interactive feature sampling data within a preset calibration time, calculates the statistical distribution of signal amplitude stability index and phase consistency index, extracts the feature mean and feature variance under the orientation state, and forms the orientation feature baseline. During the deviation calibration phase, the user turns around and maintains a posture away from the smart TV. The peripheral base station continuously collects interactive feature sampling data within the preset calibration time, calculates the statistical distribution of signal amplitude stability index and phase consistency index, extracts the feature mean and feature variance under the deviation state, and forms the deviation feature baseline. Based on the feature baseline and the feature baseline, the separation degree of the feature distribution of the two states is calculated. When the separation degree is higher than the preset separation degree threshold, the first feature threshold, the second feature threshold, the third feature threshold and the fourth feature threshold are dynamically determined based on the feature distribution boundary of the two states. The dynamically determined feature thresholds are stored in the external base station, replacing the preset fixed thresholds for subsequent viewing orientation status determination.

[0080] Specifically, when a user initiates the calibration process, the external base station enters calibration mode and sends calibration prompts to the user through the user's interactive terminal, guiding the user to execute the standard calibration action sequence. The calibration process consists of two steps: the orientation calibration phase and the deviation calibration phase.

[0081] During the calibration phase, the user maintains a natural sitting posture facing the smart TV, with hands placed in front of or to the sides of the body, as prompted. The peripheral base station continuously collects interactive feature sampling data within a preset calibration duration; in this embodiment, the calibration duration is set to ten seconds. The peripheral base station calculates the signal amplitude and phase value sequences from the collected IQ data, statistically analyzes the variance of amplitude changes and the mean absolute value of phase difference, and then calculates the signal amplitude stability index and phase consistency index. The peripheral base station performs statistical analysis on multiple sets of feature indices within the calibration duration, extracting the characteristic mean and characteristic variance of the amplitude stability index and the characteristic mean and characteristic variance of the phase consistency index under the facing state, forming the facing feature baseline.

[0082] During the deviation calibration phase, the user turns away from the smart TV as prompted, maintaining a natural standing or sitting posture. The external base station also collects interactive feature sampling data within the ten-second calibration period, calculates the signal amplitude stability index and phase consistency index using the same method, and statistically analyzes the feature mean and feature variance under the deviation state to form a deviation feature baseline.

[0083] After calibration, the peripheral base station calculates the separation degree of the two state feature distributions based on the facing feature baseline and the diverging feature baseline. The separation degree reflects the distinguishability of the two states in the feature space; a higher separation degree occurs when the feature distributions of the facing and diverging states overlap less. The peripheral base station compares the calculated separation degree with a preset separation degree threshold, which is set to 1.5 in this embodiment. When the separation degree is higher than this threshold, it indicates that the two facing states have good distinguishability in the current environment. Based on the boundary positions of the two state feature distributions, the peripheral base station dynamically determines the first, second, third, and fourth feature thresholds for orientation determination.

[0084] Specifically, the peripheral base station uses a certain percentage offset of the average value of the facing state features towards the direction of departure as the facing determination threshold, and a certain percentage offset of the average value of the departure state features towards the facing state as the departure determination threshold. A buffer region is formed between the two sets of thresholds. The dynamically determined feature thresholds are stored in the non-volatile memory of the peripheral base station, replacing the preset fixed thresholds for subsequent viewing orientation state determination. When the separation is lower than the threshold, the peripheral base station prompts the user to adjust the position or re-execute the calibration process.

[0085] To address the issue of frequency-selective fading that Bluetooth signals may encounter at specific frequency points, this embodiment employs a multi-frequency channel feature fusion method to improve the robustness of orientation determination.

[0086] The multipath feature extraction employs a multi-frequency point fusion method, the specific process of which includes: During Bluetooth communication, the peripheral base station records the frequency hopping channel number used for the current data transmission. The frequency hopping channel number corresponds to a specific carrier frequency within the 2.4 GHz band specified by the Bluetooth protocol. The peripheral base station collects interactive feature sampling data on different frequency hopping channels, and independently calculates the signal amplitude stability index and phase consistency index for each frequency hopping channel to form a multi-frequency feature vector; Perform frequency point consistency analysis on the multi-frequency feature vector, and count the proportion of the number of frequency points that exceed the frequency point stability threshold in the amplitude stability index of each frequency point to the first frequency point of the total number of frequency points, and count the proportion of the number of frequency points that exceed the frequency point consistency threshold in the phase consistency index of each frequency point to the second frequency point of the total number of frequency points. Multi-frequency point voting is performed based on the first frequency point ratio and the second frequency point ratio. When both frequency point ratios exceed the preset voting threshold, it is determined to be in a facing state; when any frequency point ratio is lower than the preset voting threshold, it is determined to be in a divergent state. The results of the multi-frequency voting will be output as viewing orientation status labels.

[0087] Specifically, the Bluetooth Low Energy protocol defines forty radio frequency channels in the 2.4 GHz band, thirty-seven of which are data channels, using a frequency hopping mechanism for data transmission. During Bluetooth communication, the peripheral base station records the frequency hopping channel number used for each data transmission; this number corresponds to a specific carrier frequency specified by the Bluetooth protocol.

[0088] The peripheral base station establishes a multi-frequency feature buffer and classifies and stores the collected interactive feature sampling data according to the frequency hopping channel number. When the accumulated sampling data of a certain frequency hopping channel reaches a preset number, the peripheral base station independently calculates the signal amplitude stability index and phase consistency index for that channel. After a period of data accumulation, the peripheral base station obtains a set of feature vectors covering multiple frequency hopping channels, forming a multi-frequency feature vector.

[0089] The peripheral base station performs frequency point consistency analysis on the feature vectors of multiple frequency points. Specifically, the peripheral base station counts the amplitude stability index of each frequency point, determines whether the index value of each frequency point exceeds the frequency point stability threshold, and calculates the proportion of the number of frequency points exceeding the threshold to the total number of frequency points involved in the statistics, obtaining the first frequency point proportion. Similarly, the peripheral base station counts the phase consistency index of each frequency point to see if it exceeds the frequency point consistency threshold, and calculates the proportion of the number of frequency points exceeding the threshold to the total number of frequency points, obtaining the second frequency point proportion.

[0090] The external base station performs multi-frequency point voting based on the proportions of the first and second frequency points. When both frequency point proportions exceed a preset voting threshold, it indicates that the channel characteristics of most frequency points exhibit oriented statistical characteristics, and the external base station determines that the user's orientation is oriented. When any frequency point proportion is lower than the preset voting threshold, it indicates that the channel characteristics of many frequency points exhibit deviating statistical characteristics or are subject to interference, and the external base station determines that the user's orientation is deviating. In this embodiment, the voting threshold is set to 60%, meaning that an oriented status label is only output when more than 60% of the frequency points support oriented determination.

[0091] The multi-frequency voting decision mechanism utilizes the principle of frequency diversity. When some frequencies exhibit abnormal characteristics due to frequency-selective fading or narrowband interference, other normal frequencies can still provide reliable judgment criteria, thereby improving the robustness of the overall judgment.

[0092] To address the issue that a fixed time window cannot simultaneously balance response speed and judgment stability, this embodiment employs a dynamic window adjustment mechanism, adaptively selecting the feature extraction window length based on the channel change rate.

[0093] The length of the time window is dynamically adjusted, and the specific process includes: The peripheral base station calculates the rate of change of channel characteristics within multiple consecutive sampling periods. The rate of change is obtained by the mean of the absolute values ​​of the differences between the statistical signal amplitude stability index and the phase consistency index between adjacent sampling periods. The rate of change is compared with a preset rate classification threshold to determine the current channel change level, which includes fast change level, medium change level and slow change level. The length of the feature extraction time window is dynamically adjusted according to the change level: when the change level is fast, the time window length is set to the first window value; when the change level is medium, the time window length is set to the second window value; when the change level is slow, the time window length is set to the third window value; wherein the first window value is less than the second window value, and the second window value is less than the third window value. Multipath feature extraction and viewing orientation determination are performed within a dynamically adjusted time window.

[0094] Specifically, the peripheral base station continuously monitors the time-varying characteristics of the channel features and calculates the rate of change of the channel features within multiple consecutive sampling periods. Specifically, the peripheral base station statistically analyzes the absolute difference between the signal amplitude stability index and the phase consistency index between adjacent sampling periods, and calculates the average of these two absolute difference values ​​as a comprehensive rate of change index.

[0095] The external base station compares the rate of change indicator with preset rate classification thresholds to determine the current channel change level. The change level is divided into three categories: fast change, medium change, and slow change. When the rate of change is higher than the first rate threshold, it is classified as fast change, indicating that the user may be turning or moving; when the rate of change is between the first and second rate thresholds, it is classified as medium change, indicating that the user is making slight movements; when the rate of change is lower than the second rate threshold, it is classified as slow change, indicating that the user is relatively stationary.

[0096] The external base station dynamically adjusts the length of the feature extraction time window based on the level of change. When the level of change is fast, the external base station sets the time window length to the first window value, which is 1 second in this embodiment, corresponding to 10 sampling periods, to quickly respond to changes in user status. When the level of change is medium, the external base station sets the time window length to the second window value, which is 2 seconds in this embodiment, corresponding to 20 sampling periods. When the level of change is slow, the external base station sets the time window length to the third window value, which is 4 seconds in this embodiment, corresponding to 40 sampling periods, to improve the stability of the judgment by accumulating more samples.

[0097] The peripheral base station performs multipath feature extraction and viewing orientation determination within a dynamically adjusted time window, achieving an adaptive balance between response speed and judgment stability.

[0098] The external base station collects spatial parameter samples within a preset time window and calculates confidence levels based on the viewing orientation. The external base station statistically analyzes the following proportions: the first proportion where the angle of arrival (AoA) falls within the viewing area's angle range; the second proportion where the distance bins remain within an acceptable range; and the third proportion where the user's orientation is facing the viewer. It then extracts motion state weighting coefficients and orientation weighting coefficients to comprehensively calculate the presence confidence value. When the confidence value is higher than a preset threshold and the user's orientation is facing the viewer, a valid presence status is confirmed.

[0099] The peripheral base station manages presence status switching through a hysteresis mechanism, using differentiated threshold settings for entry and exit conditions to avoid frequent status switching. The peripheral base station confirms a valid presence status only when the spatial parameters match the multimedia viewing area model and the viewing orientation is in a facing state.

[0100] When multiple user interaction terminals are detected to be present simultaneously, a presence event is generated. This presence event and the interaction events of the user interaction terminals are then mapped to standard input events and transmitted as keyboard events to the smart TV. Specifically, this includes: When multiple user interaction terminals are detected to be present at the same time, the peripheral base station executes a multi-user arbitration strategy, selects a master control device from multiple present devices according to the arbitration rules, and restricts the access permissions of the interaction events of non-master control devices. The peripheral base station generates presence events based on the presence status and maps the presence events and the interaction events of the user interaction terminal into standard input events; Standard input events are transmitted to the smart TV as keyboard events via the USB HID interface.

[0101] Specifically, the event mapping process includes: monitoring changes in the presence state; when a switch from an "out" state to an "in" state is detected, a user entry event is generated and mapped to a media playback button event; when a switch from an "in" state to an "out" state is detected, a user exit event is generated and mapped to a media pause button event; receiving key interaction data reported by the user interaction terminal, the key interaction data including a key type identifier and a side label; determining whether the user interaction terminal is worn on the left or right hand based on the side label, and performing a lookup operation based on a preset left / right hand function mapping table, the mapping table defining the keys on the left-hand device as corresponding to the back button and volume control keys, and the keys on the right-hand device as corresponding to the directional navigation keys and the confirmation key; generating corresponding standard input event codes based on the lookup results and the key type identifier, the standard input event codes conforming to the USB HID keyboard report format specification.

[0102] Specifically, when the external base station detects that multiple user interaction terminals are simultaneously in a valid presence state, the external base station executes a multi-user arbitration strategy. The purpose of the arbitration strategy is to select a master device from the multiple present devices and restrict the access permissions of interaction events of non-master devices to avoid conflicts caused by simultaneous operation by multiple users.

[0103] The arbitration rules are executed according to the following priority: First, the presence confidence values ​​of each device are compared, and the device with the highest confidence value has priority; if the confidence values ​​are the same, the time when each device entered the presence state is compared, and the device that entered earliest has priority; if the entry times are also the same, the master device is determined according to the preset priority order of the device identifiers. The user interaction terminal selected as the master device has full interaction permissions, and interaction events of non-master devices are ignored or limited to specific functions.

[0104] The peripheral base station generates presence events based on changes in presence status. When it detects a switch from an "out" state to an "in" state, the peripheral base station generates a user "enter" event. When it detects a switch from an "enter" state to an "out" state, the peripheral base station generates a user "leave" event. The peripheral base station maps user "enter" events to media playback button events and user "leave" events to media pause button events.

[0105] The peripheral base station receives button interaction data reported by the user's interactive terminal. This data includes button type identifiers and side identification labels. The peripheral base station determines whether the user's interactive terminal is worn on the user's left or right hand based on the side identification labels and performs a lookup operation based on a preset left / right hand function mapping table. This table defines the button correspondence between different side devices: buttons on the left-hand device correspond to the back button and volume control buttons, while buttons on the right-hand device correspond to the directional navigation buttons and the confirmation button.

[0106] The peripheral base station generates corresponding standard input event codes based on the lookup table results and key type identifiers. These standard input event codes conform to the USB HID keyboard report format specification, including modifier key bytes and key code bytes. The peripheral base station transmits these standard input events as keyboard events to the smart TV via the USB HID interface. Upon receiving the keyboard events, the smart TV executes the corresponding media control operations.

[0107] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A peripheral base station presence measurement method based on channel measurement, characterized in that, include: Collect multimedia interaction signals sent by the user interaction terminal, and collect interaction feature sampling data and perception parameters containing phase information; The angle of arrival (AoA) is calculated based on interactive feature sampling data, distance information is obtained based on sensing parameters, and signal strength indication is fused to generate spatial parameters including angle information and distance partition information. Multipath features are extracted based on interactive feature sampling data to determine the user's viewing orientation. Specifically, when the interactive link quality index is higher than the first feature threshold and the phase consistency index is higher than the second feature threshold, it is determined to be a face-to-face state. When the interactive link quality index is lower than the third feature threshold and / or the phase consistency index is lower than the fourth feature threshold, it is determined to be a divergence state. The spatial parameters are matched with the multimedia visual area model, and the spatial parameters are temporally fused within a preset time window. The confidence level is calculated in combination with the user's viewing orientation. The multimedia visual area model is a spatial area centered on the smart TV display terminal. An entry threshold and an exit threshold are set to form a hysteresis mechanism. When multiple user interaction terminals are detected to be present at the same time, a viewer presence event is generated. The viewer presence event and the interaction events of the user interaction terminals are mapped into multimedia control commands and transmitted to the smart TV.

2. The peripheral base station presence measurement method based on channel measurement according to claim 1, characterized in that: The specific process for calculating the angle of arrival (AoA) includes: receiving signals from the same device through at least two antenna elements in a multi-antenna array, acquiring interactive feature sampling data corresponding to each element, wherein the interactive feature sampling data includes in-phase component I and quadrature component Q; calculating the phase difference between adjacent elements based on the interactive feature sampling data received by different elements, wherein the phase difference is obtained by comparing the phase information of the interactive feature sampling data of different elements; determining the incident angle of the signal according to the phase difference, the physical spacing between antenna elements and the wavelength of the device signal, thereby obtaining the angle of arrival (AoA) value relative to the front direction of the external base station; and repeating the calculation within a preset sampling period to obtain time series data of the angle of arrival (AoA).

3. The peripheral base station presence measurement method based on channel measurement according to claim 1, characterized in that: The specific process for generating distance interval information includes: initiating a sensing parameter request during device connection, receiving sensing parameter response data returned by the user interaction terminal, wherein the sensing parameter response data includes round-trip time information and signal quality parameters; extracting the round-trip time information and the received signal strength indication value at the current moment to form a distance feature vector; comparing the distance feature vector with a preset distance interval threshold table, wherein the distance interval threshold table includes the judgment boundaries of near distance interval, medium distance interval, and far distance interval; classifying the current distance into one of the near distance interval, medium distance interval, and far distance interval levels according to the comparison result, and outputting a distance interval label.

4. The peripheral base station presence measurement method based on channel measurement according to claim 1, characterized in that: The method for obtaining the spatial parameters includes: establishing a multimedia visible area model with the center of the antenna array as the reference origin and the frontal orientation as the reference direction; using the angle of arrival (AoA) as the azimuth component in the multimedia visible area model to characterize the angular orientation of the user interaction terminal relative to the external base station; using the distance interval label obtained based on the sensing parameters as the distance level component in the multimedia visible area model to characterize the relative distance range between the user interaction terminal and the external base station; combining the azimuth component and the distance level component to form spatial parameters, wherein the spatial parameters also include corresponding sampling timestamps and device identifiers; and determining the spatial region where the user interaction terminal is located based on the azimuth component and the distance level component in the spatial parameters, wherein the spatial region is a sector-shaped interval centered on the external base station and jointly defined by the angle range and the distance level.

5. The peripheral base station presence measurement method based on channel measurement according to claim 1, characterized in that: The specific process for determining the viewing orientation status includes: collecting interactive feature sampling data sequences over multiple consecutive sampling periods; calculating the signal amplitude value for the interactive feature sampling data in each sampling period; statistically analyzing the variance of the signal amplitude value within a time window; and using the reciprocal of the variance as a signal amplitude stability index; calculating the phase value sequence for the interactive feature sampling data in consecutive sampling periods; statistically analyzing the mean absolute value difference of the phase values ​​between adjacent sampling periods; and using the reciprocal of the mean absolute value difference as a phase consistency index; establishing an orientation determination feature space, with the signal amplitude stability index as the first dimension and the phase consistency index as the second dimension; setting determination boundaries for facing and departing regions in the feature space; mapping the current signal amplitude stability index and phase consistency index to the feature space; and outputting a viewing orientation status label based on the region it falls into; performing temporal filtering on the viewing orientation status labels over multiple consecutive moments; and confirming the user's orientation as facing when the proportion of facing status labels exceeds a preset threshold, otherwise confirming it as departing.

6. The peripheral base station presence measurement method based on channel measurement according to claim 1, characterized in that: The specific implementation of the hysteresis mechanism includes: a spatial area centered on the front of the smart TV; setting entry conditions, including a first angle threshold and a first distance threshold, where the absolute value of the arrival angle AoA is less than the first angle threshold and the distance interval label does not exceed the first distance threshold, marking it as meeting the entry conditions; setting exit conditions, including a second angle threshold and a second distance threshold, where the absolute value of the arrival angle AoA is greater than the second angle threshold and / or the distance interval label exceeds the second distance threshold, marking it as meeting the exit conditions, wherein the second angle threshold is greater than the first angle threshold and the second distance threshold is less stringent than the first distance threshold; maintaining the current presence status identifier, where the current status is the exit status and the entry conditions are met continuously for more than a first preset time, the status is switched to the entry status; where the current status is the entry status and the exit conditions are met continuously for more than a second preset time, the status is switched to the exit status, avoiding frequent status switching through the threshold difference between the entry and exit conditions and the time constraint of the orientation status; confirming a valid presence status only when the spatial parameters match the multimedia viewing area model and the viewing orientation status is the face-viewing status.

7. The peripheral base station presence measurement method based on channel measurement according to claim 1, characterized in that: The specific process of confidence calculation includes: collecting spatial parameter samples at multiple consecutive moments within a preset time window, each sample including the angle of arrival (AoA) value, distance interval label, viewing orientation status label, and sampling timestamp; performing consistency analysis on the AoA values ​​within the time window, and counting the number of samples falling within the viewing area angle range as a first proportion of the total number of samples; performing stability analysis on the distance interval labels within the time window, and counting the number of samples remaining within the acceptable distance range as a second proportion of the total number of samples; performing statistical analysis on the viewing orientation status labels within the time window, and counting the number of samples marked as facing the view as a third proportion of the total number of samples; extracting motion state data reported by the user interaction terminal, determining it as a stationary state and / or a moving state, assigning a first weight coefficient to the stationary state and a second weight coefficient to the moving state, wherein the first weight coefficient is greater than the second weight coefficient.

8. The peripheral base station presence measurement method based on channel measurement according to claim 7, characterized in that: The specific process of confidence calculation also includes: determining the orientation weight coefficient based on the viewing orientation status label at the current moment; assigning a third weight coefficient when the viewing orientation status is facing the viewer; assigning a fourth weight coefficient when the viewing orientation status is facing away from the viewer; the third weight coefficient being greater than the fourth weight coefficient; calculating the presence confidence value at the current moment by combining the first ratio, the second ratio, the third ratio, the motion state weight coefficient, and the orientation weight coefficient; confirming the presence status is valid when the confidence value is higher than a preset threshold and the current viewing orientation status is facing the viewer; and outputting the determination result including the presence status identifier and the confidence value.

9. A peripheral base station presence metering system based on channel measurement, characterized in that, include: The data analysis module collects multimedia interactive signals sent by the user's interactive terminal, and collects interactive feature sampling data and perception parameters containing phase information; The angle of arrival (AoA) is calculated based on interactive feature sampling data, distance information is obtained based on sensing parameters, and signal strength indication is fused to generate spatial parameters including angle information and distance partition information. The orientation state analysis module extracts multipath features based on interactive feature sampling data to determine the user's viewing orientation state. The hysteresis mechanism analysis module matches spatial parameters with the multimedia visual area model, performs temporal fusion of spatial parameters within a preset time window, and calculates confidence based on the user's viewing orientation. The multimedia visual area model is a spatial area centered on the smart TV display, with entry and exit thresholds set to form a hysteresis mechanism. The data transmission module generates a viewer presence event when it detects that multiple user interaction terminals are present at the same time. It then maps the viewer presence event and the interaction events of the user interaction terminals into multimedia control commands and transmits them to the smart TV.

Images