Perception result determination method and device, equipment, medium and program product

Abstract

The invention provides a sensing result determination method and device, equipment, a medium and a program product, relates to the technical field of informatization software systems, and is used for guaranteeing the sensing precision and sensing service quality of a sensing network. The method comprises the following steps: in a first statistical period, obtaining a first sensing result of a base station to a sensing target; based on the first perception result and reference data, calculating a first perception calibration value, the reference data comprising at least one of the following items: a real parameter value of a fixed known target, a statistical expected value generated based on a historical perception result, and a historical perception calibration record associated with at least one historical environmental condition; calculating a target correction perception result based on the first perception result and the first perception calibration value; updating the reference data based on the target correction perception result; and in the second statistical period, correcting the sensing result obtained in the second statistical period based on the updated reference data. The method is applied to the self-adaptive modification process of the sensing result in the sensing network.

Description

Technical Field

[0001] This application relates to the field of information software system technology, and in particular to methods, devices, equipment, media and program products for determining perception results. Background Technology

[0002] Sensor integration is an important evolutionary direction for next-generation mobile communication systems, aiming to achieve synergy and resource fusion between communication and sensing functions. In sensor networks, sensing accuracy is a key indicator for ensuring the effectiveness of various sensing applications.

[0003] Currently, solutions to improve the sensing accuracy of sensor networks involve optimizing the hardware design or signal processing algorithms of the base station itself, aiming to achieve better initial sensing performance during system deployment. For example, optimizing the antenna array structure or improving the signal waveform can enhance the initial sensing accuracy.

[0004] However, in actual deployment, sensor networks operate in a complex and dynamic environment. Various external factors, such as environmental changes and weather conditions, continuously affect the sensing process, potentially leading to gradual systematic deviations or accuracy degradation in the sensing results. Consequently, base stations struggle to assess the accuracy of their sensing outputs in real time and autonomously. Therefore, existing solutions cannot guarantee the sensing accuracy and quality of service in dynamic environments. Summary of the Invention

[0005] This application provides a method, apparatus, device, medium, and program product for determining sensing results, used to ensure the sensing accuracy and sensing service quality of a sensor network in a dynamic environment.

[0006] In a first aspect, this application provides a method for determining perception results, comprising: acquiring a first perception result of a base station on a perceived target within a first statistical period, the first perception result including at least one of the following: distance, angle, speed, and position coordinates; calculating a first perception calibration value based on the first perception result and reference data, the reference data including at least one of the following: fixed true parameter values ​​of a known target, statistical expected values ​​generated based on historical perception results, and historical perception calibration records containing at least one historical environmental condition; calculating a target correction perception result based on the first perception result and the first perception calibration value; updating the reference data based on the target correction perception result; and correcting the perception result acquired in the second statistical period based on the updated reference data within a second statistical period.

[0007] The technical solution provided in this application offers at least the following benefits: In each statistical period, raw sensing results are acquired, and then, combined with multi-source reference data including historical experience and objective benchmarks, a dynamic sensing calibration value is calculated. This calibration value is used to correct the raw sensing results in real time, resulting in a more accurate target-corrected sensing result. This correction result is fed back to update the reference data itself, allowing the system experience base or cognitive benchmark to be continuously updated during operation. Thus, in subsequent statistical periods, new sensing results will be corrected based on updated and more accurate reference data. This automatically tracks and compensates for systematic sensing errors caused by environmental changes, hardware drift, and other factors, thereby improving the stability and reliability of the sensing accuracy of the sensor network under long-term complex operating conditions and ensuring the sensing accuracy and quality of sensing services in dynamic environments.

[0008] One possible implementation is that the reference data includes at least one of the actual parameter values ​​and the expected statistical values; the calculation of the first perception calibration value based on the first perception result and the reference data includes: calculating the difference between the first perception result and the reference data to obtain the perception accuracy deviation value within the first statistical period; and determining the first perception calibration value based on the perception accuracy deviation value.

[0009] Another possible implementation is that the reference data includes historical sensing calibration records; the calculation of the first sensing calibration value based on the first sensing result and the reference data includes: determining the current environmental conditions based on sensing network index information within a first statistical period; determining the first sensing calibration value based on the current environmental conditions and historical sensing calibration records; wherein the sensing network index information includes at least one of the following: the signal-to-noise ratio change curve of the sensing signal, the false alarm rate change of the sensing result, and the systematic offset of the angle measurement value.

[0010] Another possible implementation method, which determines the first sensing calibration value based on the current environmental conditions and historical sensing calibration records, includes: when the current environmental conditions are the first environmental conditions, querying the historical sensing calibration records to obtain the matching historical calibration value; and using the matching historical calibration value as the first sensing calibration value.

[0011] Another possible implementation method, which determines the first sensing calibration value based on the current environmental conditions and historical sensing calibration records, includes: when the current environmental conditions are the second environmental conditions, adjusting the weights of different intensity signal components in the first sensing result based on the signal-to-noise ratio change information of the sensing signal within the first statistical period, and calculating the first sensing calibration value based on the weighted sensing result and the historical sensing calibration records.

[0012] Another possible implementation method includes: in the first scenario where the perception network index changes steadily, the target correction perception result of the first statistical period is used as the statistical expected value or perception calibration value for the next statistical period; in the second scenario where the perception network index changes abruptly, the mutation feature information of the second scenario is obtained, and the pre-stored calibration value corresponding to the mutation feature information is called, which is used to correct the perception result of the next statistical period.

[0013] Secondly, this application provides a sensing result determination device, comprising: an acquisition module and a processing module; the acquisition module is configured to acquire a first sensing result of a base station on a sensing target within a first statistical period, the first sensing result including at least one of the following: distance, angle, speed, and position coordinates; the processing module is configured to calculate a first sensing calibration value based on the first sensing result and reference data, the reference data including at least one of the following: fixed true parameter values ​​of a known target, statistical expected values ​​generated based on historical sensing results, and historical sensing calibration records containing at least one historical environmental condition; the processing module is further configured to calculate a target correction sensing result based on the first sensing result and the first sensing calibration value; the processing module is further configured to update the reference data based on the target correction sensing result; and the processing module is further configured to correct the sensing result acquired in the second statistical period based on the updated reference data within a second statistical period.

[0014] One possible implementation is that the reference data includes at least one of the actual parameter values ​​and the expected statistical values; the processing module is specifically used to: calculate the difference between the first perception result and the reference data to obtain the perception accuracy deviation value within the first statistical period; and determine the first perception calibration value based on the perception accuracy deviation value.

[0015] Another possible implementation is that the reference data includes historical sensing calibration records; the processing module is specifically used to: determine the current environmental conditions based on the sensing network index information within the first statistical period; and determine the first sensing calibration value based on the current environmental conditions and the historical sensing calibration records; wherein the sensing network index information includes at least one of the following: the signal-to-noise ratio change curve of the sensing signal, the false alarm rate change of the sensing result, and the systematic offset of the angle measurement value.

[0016] Another possible implementation is that the above processing module is specifically used to: query historical sensing calibration records and obtain matching historical calibration values ​​when the current environmental conditions are the first environmental conditions; and use the matching historical calibration values ​​as the first sensing calibration values.

[0017] Another possible implementation is that the above processing module is specifically used to: when the current environmental condition is the second environmental condition, adjust the weights of different intensity signal components in the first sensing result based on the signal-to-noise ratio change information of the sensing signal within the first statistical period, and calculate the first sensing calibration value based on the weighted sensing result and the historical sensing calibration record.

[0018] In another possible implementation, the above processing module is also used to: in the first scenario where the perception network indicators change smoothly, use the target correction perception result of the first statistical period as the statistical expected value or perception calibration value for the next statistical period; in the second scenario where the perception network indicators change abruptly, obtain the mutation feature information of the second scenario, and call the pre-stored calibration value corresponding to the mutation feature information, which is used to correct the perception result of the next statistical period.

[0019] Thirdly, this application provides an electronic device comprising: a processor and a memory; the memory storing processor-executable instructions; when the processor is configured to execute the instructions, causing the electronic device to implement the method of the first aspect described above.

[0020] Fourthly, this application provides a computer-readable storage medium comprising: computer software instructions; which, when executed in an electronic device, cause the electronic device to implement the method described in the first aspect.

[0021] Fifthly, this application provides a computer program product comprising a computer program; when the computer program is run in an electronic device, the electronic device performs the method described in the first aspect.

[0022] The beneficial effects of the second to fifth aspects mentioned above are described in the corresponding description of the first aspect and will not be repeated here. Attached Figure Description

[0023] Figure 1 This is a schematic diagram illustrating the application environment of a perception result determination method provided in an embodiment of this application; Figure 2 A schematic diagram illustrating a typical principle of sensing error generation in existing solutions; Figure 3 A flowchart illustrating a method for determining a perception result provided in an embodiment of this application; Figure 4 A flowchart illustrating another method for determining the perception result provided in an embodiment of this application; Figure 5 A flowchart illustrating another method for determining the perception result provided in an embodiment of this application; Figure 6A schematic diagram illustrating the overall implementation process of a perception result determination method provided in this application embodiment; Figure 7 This is a schematic diagram of the architecture of a perception result determination system provided in an embodiment of this application; Figure 8 This is a schematic diagram of the composition of a sensing result determination device provided in an embodiment of this application; Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0024] The following is a detailed description of the perception result determination method, apparatus, equipment, medium, and program product provided in this application, with reference to the accompanying drawings.

[0025] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0026] The terms "first" and "second," etc., used in the specification and drawings of this application are used to distinguish different objects or to distinguish different treatments of the same object, rather than to describe a specific order of objects.

[0027] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0028] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0029] To facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with essentially the same function and effect. Those skilled in the art can understand that the terms "first" and "second" are not intended to limit the quantity or execution order.

[0030] In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0031] This application provides a method, apparatus, device, medium, and program product for determining sensing results, which can be applied to integrated communication and sensing networks. An integrated communication and sensing network refers to a new network architecture that deeply integrates wireless communication functions with environmental sensing capabilities such as radar. Utilizing the same set of hardware (such as base station antenna arrays), the same spectrum resources, and signal waveforms, it simultaneously performs data transmission and enables the detection, location, speed measurement, and tracking of targets in the surrounding environment (such as vehicles, drones, and pedestrians). In short, it transforms the base station not only into a communication hub but also into an intelligent sensing node capable of clearly perceiving its surroundings.

[0032] Specific application scenarios include, but are not limited to, the following: Intelligent Transportation Systems: Applied to vehicle-to-everything (V2X) or roadside sensing units, this system continuously and accurately senses and tracks the distance, speed, and location coordinates of traffic participants such as vehicles, pedestrians, and non-motorized vehicles. This solution can adaptively compensate for perception drift caused by weather changes (such as rain, snow, and fog), equipment vibration, or thermal deformation, ensuring the reliability of services such as autonomous driving, traffic flow monitoring, and collision warning.

[0033] Low-altitude security and drone control: Sensor base stations deployed around airports and key areas are used to detect and track the angle, distance, speed, and three-dimensional position of low-altitude aircraft such as drones. This solution can effectively counteract the effects of antenna pointing deviation caused by wind disturbance and signal attenuation caused by complex weather conditions.

[0034] Smart Cities and Environmental Sensing: Sensor network nodes deployed in cities are used to monitor pedestrian density, vehicle speed, and micro-displacements of large facilities (such as bridges) in specific areas. Through adaptive correction, sensing errors caused by urban multipath environments and seasonal weather changes can be suppressed, providing stable and long-term reliable sensing data for urban management, public safety, and infrastructure health monitoring.

[0035] Industrial Internet and Smart Factories: In industrial environments, this solution is used for precise positioning and trajectory tracking of moving targets such as materials and robotic arms. It can compensate for the impact of changes in the electromagnetic environment within the factory and the thermal expansion and contraction of equipment on sensing accuracy, ensuring precise coordination and automated control of production processes.

[0036] Network performance self-optimization: As part of the inherent capabilities of sensor networks, this method can be invoked by the network management system to continuously monitor and evaluate the sensing performance indicators of the entire network or individual stations. By analyzing long-term statistical calibration data, potential problems such as hardware aging and loose installations can be identified, enabling predictive maintenance and reducing operation and maintenance costs.

[0037] Sensing-based communication is an emerging technology whose sensing accuracy is affected by various factors, such as obstacles, weather changes, and environmental changes. These changes primarily degrade channel quality, leading to decreased transmission signal quality or increased interference, thus causing distortion in echo signal detection. However, unlike deterministic communication, the sensed target does not report data, making it impossible to assess the current sensing accuracy during the sensing process. Therefore, sensing accuracy remains an unknown quantity during base station sensing operations.

[0038] The main indicators of sensing accuracy are distance accuracy, angle accuracy, and velocity accuracy. If the entire system is relatively stationary, these parameters will not change. However, over time, some indicators may shift or change abruptly. For example, under the influence of strong winds, the angle of the base station antenna array may deflect slightly, such as rotating horizontally by 0.1 degrees, resulting in an overall shift of 0.1 degrees in the detection angle accuracy. In the actual operation of sensor networks, base station hardware may undergo significant physical changes due to dynamic environmental factors. Figure 2 The diagram illustrates a typical principle of sensing error generation in existing solutions. When the transmitting antenna array deflects—that is, when the base station's (transmitting) antenna array physically deflects (e.g., horizontally deflected by an angle Δθ) due to external influences (such as strong winds)—the direction of its emitted sensing beam changes accordingly. Since the sensing process is unidirectional, the base station cannot detect this change and continues to use its original coordinate system for calculations. This leads to a systematic deviation in its estimation of the position of the same sensing target. There is a significant angular error between the target's actual position and the position sensed by the base station; that is, a deviation occurs between the base station's perceived position and its true position. This deviation phenomenon... Figure 2 The value is marked as target position correction.

[0039] Existing solutions cannot guarantee the sensing accuracy and sensing service quality of sensor networks in dynamic environments.

[0040] To address the aforementioned technical problems, this application provides a method, apparatus, device, medium, and program product for determining sensing results. By constructing a fully adaptive closed loop of sensing-calibration-correction-iteration, the sensing accuracy is assessable and optimizable. By introducing a key variable—the first sensing calibration value—in each statistical period and dynamically calculating this value using multi-source reference data, systematic errors caused by environmental changes such as antenna array deflection due to strong winds and signal-to-noise ratio reduction due to rain, snow, and fog can be quantified and compensated in real time. Furthermore, a feedback mechanism based on the target-corrected sensing results updates the reference data. In subsequent operating cycles, the target-corrected sensing results serve as new statistical indicators for adaptive correction of subsequent results. Continuous iteration ensures optimal accuracy, enabling the network to cope with current environmental interference. The reference data (including statistical expected values ​​and historical sensing calibration records) becomes increasingly rich and accurate over time. Thus, when sensing indicators become abnormal, offset mean correction can be used, enabling the sensor network to monitor its performance in real time and make adaptive adjustments, improving the long-term sensing accuracy stability and service reliability of the sensor network in complex dynamic environments.

[0041] The embodiments provided in this application will now be described in detail with reference to the accompanying drawings.

[0042] The method for determining a sensing result provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown. For example... Figure 1 As shown, the application environment includes a perception result determination device 101 and a front-end device 102. The perception result determination device 101 and the front-end device 102 are interconnected.

[0043] In some embodiments, the perception result determination device 101 may be a server cluster consisting of multiple servers, a single server, a computer, or a processor or processing chip in a server or computer, etc. This application does not limit the specific device form of the perception result determination device 101. Figure 1 The example shown is a single server, with the perception result determination device 101 as an example.

[0044] In some embodiments, the front-end device 102 can be a device with wireless transceiver capabilities, such as a mobile phone, tablet computer, wearable device, in-vehicle device, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), etc. This application embodiment does not limit the specific device form of the front-end device 102. Figure 1 The example shown is a mobile phone, with the front-end device 102 as the illustration.

[0045] In some embodiments, during a statistical period, the perception result determination device 101 receives wireless signals from the front-end device 102 (such as a smartphone, vehicle terminal, drone, etc.) and extracts the original perception results (such as distance, angle, etc.) of the target. Based on the results and internally maintained reference data (including real parameters, statistical expected values, historical calibration records, etc.), the perception result determination device 101 calculates and applies a first perception calibration value to generate a target correction perception result. The perception result determination device 101 updates this correction result to its reference database to optimize the calibration accuracy of subsequent periods. The perception result determination device 101 can send the high-precision correction perception result or control commands generated based on it to the front-end device 102. Communication signals transmitted by the front-end device 102 or reflections of base station signals are used by the perception result determination device 101 for perception calculation. On the other hand, the front-end device 102 receives corrected high-precision perception information services (such as location and speed reminders of itself or surrounding targets) or commands such as resource scheduling and obstacle avoidance control executed based on them from the perception result determination device 101.

[0046] It should be noted that the system architecture described in the embodiments of this application is for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and does not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of system architecture, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0047] See Figure 3 This is a flowchart illustrating a method for determining a perception result provided in an embodiment of this application. Figure 3 As shown, the perception result determination method provided in this application can be implemented by the above-mentioned perception result determination device, specifically including the following steps 201 to 205.

[0048] Step 201: The sensing result determination device acquires the first sensing result of the base station on the sensing target within the first statistical period.

[0049] In some embodiments, the first perception result described above includes at least one of the following: distance, angle, speed, and position coordinates.

[0050] In some embodiments, the above-mentioned sensing result determination device can be a base station, and the sensing computing function is integrated inside the base station.

[0051] In some embodiments, the sensing result determination device may also be an independent network server or cloud platform, which receives raw signal data or preliminarily processed sensing measurement reports reported by the base station and executes subsequent calibration and correction algorithms.

[0052] In some embodiments, the first statistical period refers to a pre-configured or dynamically adjusted time window, such as 1 hour or 10 minutes, used to accumulate data for statistical analysis. During this period, the sensing result determination device controls the base station to transmit sensing signals (typically using communication signals, such as reference signals), and receives echo signals reflected from the sensing target (such as vehicles, drones, pedestrians, etc.) or signals emitted by the target itself. By processing the received signals (e.g., performing Fourier transform, correlation operations, direction-of-arrival estimation, etc.), parameters characterizing the target's state are extracted.

[0053] Specifically, the sensing result determination device can calculate at least one of the following parameters based on the sensing technology used (such as velocity measurement based on the Doppler effect, ranging based on time difference, and angle measurement based on antenna array): Distance: The straight-line distance between the sensing target and the base station, which can be obtained by calculating the round-trip time or phase difference of the signal; Angle: The azimuth or elevation angle of the target relative to the base station, which is usually estimated by the phase difference of the signals received by a multi-antenna array; Velocity: The radial velocity of the target (towards or away from the base station) can be obtained by analyzing the Doppler frequency shift of the received signal; Location coordinates: The two-dimensional or three-dimensional spatial coordinates (such as latitude, longitude, and altitude) of the perceived target can be obtained by combining distance and angle information, and possibly fusing measurements from multiple base stations for triangulation or fingerprint positioning.

[0054] Step 202: The sensing result determination device calculates the first sensing calibration value based on the first sensing result and reference data.

[0055] In some embodiments, the reference data mentioned above includes at least one of the following: the true parameter values ​​of a fixed known target, the statistical expected values ​​generated based on historical sensing results, and historical sensing calibration records containing at least one historical environmental condition.

[0056] In some embodiments, the above-mentioned statistical expected value is based on the statistical average or numerical range of historical perception results for the same type of perception target in the same area.

[0057] In some embodiments, the sensing result determination device may compare or fuse a first sensing result with reference data from one or more sources to calculate a first sensing calibration value for error correction, wherein the reference data represents the understanding of the current state of the sensing system or the impact of the environment. The reference data includes at least one of the following: Fixed true parameter values ​​of known targets: refers to reference objects within the network deployment area whose precise location, height, or motion state (such as being stationary) has been measured and recorded into the system with high precision in advance. For example, the coordinates of a corner reflector installed at a specific location on a rooftop are true parameter values. Statistical expected values ​​generated based on historical perception results: These refer to typical values ​​or reasonable ranges obtained by statistically analyzing (e.g., calculating the mean, median, and distribution interval) a large number of perception results over a period of time, in the same area, and for the same type of target (such as cars). For example, the historical average speed of a road segment on a weekday morning. It contains historical sensing calibration records associated with at least one historical environmental condition: This is an empirical database that stores calibration values ​​that have been effective and verified under different environmental conditions in the past (such as strong winds, moderate rain, and heavy fog). Each record uses the environmental condition as an index tag and is associated with the corresponding historical calibration value (which may be an angle offset, distance compensation value, etc.).

[0058] In some embodiments, the reference data includes at least one of the actual parameter values ​​and the expected statistical values. For example, in combination with... Figure 3 ,like Figure 4 As shown, step 202 above can be specifically implemented as steps 202a and 202b.

[0059] Step 202a: The sensing result determination device calculates the difference between the first sensing result and the reference data to obtain the sensing accuracy deviation value within the first statistical period.

[0060] In some embodiments, the perception result determination device may calculate the difference between the first perception result and the aforementioned reference data (here, the aforementioned true parameter value or statistical expected value). For example, it may calculate the difference between the angle of the currently perceived fixed target and its true angle; or it may calculate the difference between the currently perceived average vehicle speed in a certain area and the historical statistical expected average vehicle speed.

[0061] Step 202b: The sensing result determination device determines the first sensing calibration value based on the sensing accuracy deviation value.

[0062] In some embodiments, the sensing result determination device may further process the above-mentioned difference (i.e., the sensing accuracy deviation value) to determine a first sensing calibration value. This processing method may be to directly take the sensing accuracy deviation value, or to calculate (weighted) average of multiple deviation values ​​within a statistical period. The magnitude and sign of the first sensing calibration value directly reflect the error direction and magnitude of the current sensing system relative to the reference reference.

[0063] Thus, by directly comparing the current sensing results with a highly reliable reference benchmark (such as the actual parameters of a known target or reasonable expectations based on historical data), the sensing accuracy deviation value for the current period can be calculated, and the calibration value can be determined accordingly. This allows for rapid response and correction of systematic measurement offsets. When a fixed known target exists, absolute accuracy calibration can be achieved; when relying on historical statistical expectations, abnormal sensing results that deviate from normal patterns (such as sudden changes in regional average velocity) can be corrected, effectively suppressing sensing drift caused by gradual environmental changes or changes in data distribution.

[0064] In some embodiments, the reference data described above includes the aforementioned historical sensing calibration records. For example, in conjunction with... Figure 3 ,like Figure 5 As shown, step 202 above can be specifically implemented as steps 202c and 202d.

[0065] Step 202c: The sensing result determination device determines the current environmental conditions based on the sensing network index information within the first statistical period.

[0066] In some embodiments, the aforementioned sensing network indicator information includes at least one of the following: signal-to-noise ratio variation curve of the sensing signal, false alarm rate variation of the sensing result, and systematic offset of the angle measurement value.

[0067] In some embodiments, the signal-to-noise ratio (SNR) change curve of the above-mentioned sensing signal: the overall decrease and specific change pattern of the SNR may correspond to weather conditions such as rain, snow, and fog that cause signal attenuation; the false alarm rate change of the above-mentioned sensing results: the abnormal increase in the false alarm rate (the probability of misjudging noise as a target) is a typical feature of rainy and snowy weather; the systematic offset of the above-mentioned angle measurement values: if statistics show that the sensing angles of all targets have an offset that is consistent in direction and similar in magnitude, it strongly suggests that the base station antenna may have been physically deflected due to strong winds.

[0068] Step 202d: The sensing result determination device determines the first sensing calibration value based on the current environmental conditions and historical sensing calibration records.

[0069] In some embodiments, the aforementioned current environmental conditions include at least one of windy weather, rainy or snowy weather, and foggy weather.

[0070] In some embodiments, the sensing result determination device queries historical sensing calibration records based on the determined current environmental conditions to find calibration values ​​that have been used and are valid under the same or similar environmental conditions in the past, and determines a first sensing calibration value based on this.

[0071] In some embodiments, when the current environmental conditions are windy or foggy, the sensing result determination device can directly use the acquired historical calibration value as the first sensing calibration value for semi-static adjustment.

[0072] In some embodiments, when the current environmental conditions are rainy or snowy, the sensing result determination device calculates a first sensing calibration value based on the historical calibration value obtained and combined with the first sensing result after adjusting the weight according to the degree of change in the signal-to-noise ratio of the current sensing signal.

[0073] Thus, by analyzing real-time sensing network metrics (such as signal-to-noise ratio curves, false alarm rates, and angle shifts), the specific environmental conditions (such as strong winds, rain, snow, and heavy fog) can be determined. When the environmental type is identified, matching empirical calibration values ​​can be retrieved from historical calibration records. This allows for full utilization of historical calibration experience accumulated in similar and challenging environments, enabling rapid and accurate compensation for specific environmental interferences (such as signal attenuation due to rain or beam deflection due to strong winds).

[0074] In some embodiments, step 202d can be specifically implemented as steps 202d1 and 202d2.

[0075] Step 202d1: Under the condition that the current environmental conditions are the first environmental conditions, the sensing result determination device queries the historical sensing calibration records and obtains the matching historical calibration values.

[0076] In some embodiments, the first environmental condition described above may include at least one of the following: The aforementioned first environmental condition may include at least one of the following: strong winds or dense fog. Among them, strong winds are characterized by a direct impact on the hardware status of the sensing system. Continuous strong winds may cause physical deformation or attitude deflection of the base station antenna array or supporting structure. This change is deterministic and consistent: once the deflection occurs, its impact on the sensing results of all targets (especially angle measurements) is systematic, directional, and the offset is approximately the same. For example, if the horizontal array as a whole deflects counterclockwise by 0.1 degrees, the sensing azimuth angle of all targets will increase by about 0.1 degrees.

[0077] Foggy weather: Its main characteristic is a significant change in the propagation medium of wireless signals. Dense fog will strongly and uniformly attenuate radio waves, resulting in a significant overall decrease in the power of the received signal (or signal-to-noise ratio SNR). This attenuation is relatively stable in specific areas and time periods, mainly affecting the detection range, ranging accuracy, and weak signal detection capability. It usually does not cause a systematic shift in direction, but it will lead to a general deterioration in sensing performance.

[0078] In some embodiments, the aforementioned matched historical calibration value refers to a calibration parameter in the historical sensing calibration record that is associated with the same or similar characteristic labels as the current environmental conditions and has been verified to be valid in historical use.

[0079] In some embodiments, the perception result determination device compares the current environmental conditions (e.g., identified as "strong wind - northwest wind - level 5") with the historical environmental condition feature tags (such as "strong wind - north wind - level 6", "dense fog - visibility <100 meters") stored in each record in the historical records. The comparison can be based on rules (such as the same weather type and similar intensity level) or classification based on machine learning models. When a historical record is found whose feature tag matches the current condition (is the same or meets a preset similarity threshold), it is considered a successful match, and the perception result determination device can read its stored historical calibration value from that record.

[0080] It is understandable that the historical calibration value mentioned above is a specific numerical value or vector, representing the amount of correction that needs to be applied to compensate for errors in the sensing system under historical environmental conditions. For example, for windy conditions, it might be an angular offset (e.g., +0.12 degrees); for foggy conditions, it might be a distance compensation coefficient (e.g., multiplied by 1.05) or a signal-to-noise ratio compensation threshold adjustment value.

[0081] Step 202d2: The sensing result determination device uses the matched historical calibration value as the first sensing calibration value.

[0082] In some embodiments, when the current environment is determined to be a first environmental condition (such as strong wind or heavy fog), since such environments usually cause deterministic and consistent shifts in perception parameters, the perception result determination device can directly retrieve the matching calibration value from the historical records (e.g., the average angle shift recorded on historical windy days) and use it directly as the first perception calibration value for this time.

[0083] Thus, in specific environments (first environmental conditions, such as strong winds or heavy fog), which can cause deterministic and consistent deviations in sensing parameters, a strategy of directly calling and applying historical calibration values ​​is adopted. This semi-static adjustment method avoids the response delay that may be caused by slow iterations when the environment changes abruptly. It enables the system state to quickly switch to a pre-validated optimal configuration for that environment, thereby immediately and effectively correcting the sensing results and ensuring the continuity and accuracy of sensing services in sudden environments.

[0084] In some embodiments, step 202d above can be specifically implemented as step 202d3 below.

[0085] Step 202d3: Under the condition that the current environmental conditions are the second environmental conditions, the sensing result determination device adjusts the weights of different intensity signal components in the first sensing result based on the signal-to-noise ratio change information of the sensing signal within the first statistical period, and calculates the first sensing calibration value based on the weighted and adjusted sensing result and the historical sensing calibration record.

[0086] In some embodiments, the second environmental condition may include rainy or snowy weather. Rainy or snowy weather, compared to the first environmental condition (such as strong winds or dense fog), has a more complex and dynamic impact mechanism on the sensing system, primarily due to its non-uniform, time-varying, and signal strength-dependent changes in the quality of the sensed signal. Signal attenuation and fluctuations: Raindrops or snowflakes absorb and scatter radio waves, causing signal propagation loss (rain attenuation / snow attenuation). This attenuation is directly related to the precipitation intensity and may change rapidly with time and space, resulting in a dynamic decrease and fluctuation in the received signal power (or signal-to-noise ratio SNR). Introducing clutter and noise: Dense rain and snow themselves become distributed scatterers, generating a large number of clutter echoes that mix with the target echoes, significantly increasing the false alarm rate of the sensing results (misjudging rain and snow clutter as the target). The unevenness of the impact: The impact on signals of different intensities is different. The echo signal of a strong target (such as a large vehicle) may still maintain a high signal-to-noise ratio, while the echo signal of a weak target (such as a pedestrian or a small object) may be submerged in rain, snow noise and clutter.

[0087] In some embodiments, when the current environment is determined to be a second environmental condition (such as rain or snow), since this type of environment mainly causes a deterioration in signal quality (SNR) and affects sensing stability, the sensing result determination device can first assign different weights to signal components of different intensities in the first sensing result based on the current SNR change information (such as the degree of deterioration). For example, it can assign higher weights to echo signals with high SNR and reduce the weights or filter out signals with low SNR. Then, the sensing result determination device uses this weighted and adjusted sensing result (which is more reliable than the original result) as input, and combines it with historical sensing calibration records (such as the baseline compensation value for rainy or snowy days) to obtain the final first sensing calibration value through calculations (such as weighted averaging, filtering, etc.).

[0088] Thus, for certain environmental conditions (secondary environmental conditions, such as rain and snow), which may have a more complex impact on signal quality (such as dynamic degradation of signal-to-noise ratio), the system first intelligently adjusts the weights of signal components of different intensities in the original sensing results based on the current changes in signal-to-noise ratio. For example, it increases the proportion of high signal-to-noise ratio signals and suppresses the influence of low signal-to-noise ratio signals. Then, it combines historical calibration records and calculates the final calibration value based on this purified and enhanced sensing data. In environments where signal quality is generally poor, it prioritizes the more reliable data for calibration decisions, thereby effectively suppressing the influence of noise and interference, reducing the jitter of sensing results, and improving sensing accuracy in environments such as rain and snow that cause signal attenuation and increased noise.

[0089] Step 203: The perception result determination device calculates the target correction perception result based on the first perception result and the first perception calibration value.

[0090] In some embodiments, the target correction perception result is equal to the sum of the first perception result and the first perception calibration value.

[0091] In some embodiments, the base station's sensing result (i.e., the target correction sensing result) can be expressed as: S-correction = S-sensing + S-calibration. Here, S-sensing is the base station's actual calculated value, representing the sensing result for targets within the cell; S-calibration is a dynamic variable given based on changes in the current environment. S-correction is an adaptive result under the current environment, improving sensing accuracy. The sensing result here can be any value among distance, angle, and velocity, or it can be latitude, longitude, and altitude accuracy; both sets are equivalent. S can represent scalars such as distance, angle, and velocity, or it can be a position coordinate vector containing multiple components. S-calibration can be a compensation vector with the same dimension as S-sensing.

[0092] S-sensing is a standard computational quantity, while S-calibration is manually set, and its magnitude can be calculated based on statistical KPIs. For example, in terms of angle accuracy, when it is known that there is an error between the angle statistically obtained by the network and the actual or expected value, the weighted average of this difference is set as the S-calibration. In the comparison process of network statistics, the actual value is the coordinate parameters of a fixed known target. When the base station senses a known target, if the sensed data deviates significantly, it needs to be corrected. The expected value refers to the prediction made based on historical sensed data for the current data. For example, in the same area, the historical average sensed speed is 15m / s, but it has recently increased to 20m / s, so it needs to be corrected. Or, in terms of height statistics, the general height range is usually 100-200m, but it has recently changed to 150-200m, so it also needs to be corrected.

[0093] Depending on the weather conditions, such as strong winds, there may be errors in the angle perception of a known target A. Therefore, it's necessary to calculate Scalibration = Apertured - ATrue. Weather conditions like rain and snow reduce the overall signal-to-noise ratio (SNR) and the power of the received echo signal. Therefore, during the sensing process, it's necessary to increase the proportion of high-power signals and decrease the proportion of low-power signals (the proportion can be adjusted according to the degree of SNR change) to reduce sensing jitter. For example, Scalibration = 0.9 * (Ssensored strong signal calculated value - Ssensored total signal calculated value).

[0094] Alternatively, S-calibration lookup tables can be performed by consulting known historical data. If similar weather conditions are identified in historical statistics and the SNR difference at the same location is not significant, it is considered that the weather conditions are the same. In this case, the correction is made according to the historical data, which is semi-static, except for windy weather.

[0095] Step 204: The perception result determination device corrects the perception result based on the target and updates the reference data.

[0096] In some embodiments, the perception result determination device can update the statistical expectation value by adding the obtained target correction perception result as a new valid data point to the historical perception result dataset used to generate the statistical expectation value. For example, updating the historical data queue used to calculate the regional average vehicle speed.

[0097] In some embodiments, the sensing result determination device can update the historical sensing calibration record: associating the first sensing calibration value calculated and verified to be valid with its corresponding current environmental conditions, and storing it as a new record in the historical sensing calibration record database. For example, recording the effective distance compensation value calculated under a specific rainfall intensity to enrich the experience base.

[0098] Step 205: The sensing result determination device corrects the sensing results obtained in the second statistical period based on the updated reference data within the second statistical period.

[0099] This step demonstrates the continuity and iterative nature of the method. The end of a statistical period and the generation of revised results are intended to better serve the next statistical period.

[0100] In some embodiments, after entering the second statistical cycle, the sensing result determination device can repeatedly execute steps 201 to 204 above. The reference data used to calculate the calibration value in step 202 is parameter data that has been updated by the target correction of the sensing result in step 204. That is, in the new cycle, a new round of sensing calibration is carried out on a more accurate basis of experience and cognition, thereby realizing continuous optimization of sensing accuracy and dynamic adaptation to tracking environmental changes.

[0101] The perception result determination method provided in this application acquires the original perception results in each statistical period, and then calculates a dynamic perception calibration value by combining it with multi-source reference data, including historical experience and objective benchmarks. This calibration value is used to correct the original perception results in real time, resulting in a more accurate target-corrected perception result. This correction result is fed back to update the reference data itself, enabling the system experience base or cognitive benchmark to be continuously updated as the system operates. Thus, in subsequent statistical periods, new perception results will be corrected based on the updated and more accurate reference data. This automatically tracks and compensates for systematic perception errors caused by environmental changes, hardware drift, and other factors, thereby improving the stability and reliability of the sensing accuracy of the sensor network under long-term complex operating conditions and ensuring the sensing accuracy and quality of sensing services in dynamic environments.

[0102] In some embodiments, the method for determining the perception result provided in this application further includes the following steps 301 and 302.

[0103] Step 301: In the first scenario where the perception network indicators change smoothly, the perception result determination device uses the target correction perception result of the first statistical period as the statistical expected value or perception calibration value for the next statistical period.

[0104] In some embodiments, when the perception result determination device detects that the perception network indicators (such as SNR and false alarm rate) are in a stable state, i.e., no drastic changes are detected, it determines the scenario as the first scenario. In this scenario, the perception result determination device can enable a smooth iteration strategy, directly using the target correction perception results generated in the current period (such as the first statistical period) as an important input basis for calculating the statistical expected value in the next period or for calculating the perception calibration value in the next period. This strategy makes the adjustment of calibration parameters continuous and gradual, resulting in a smooth calibration curve, suitable for daily optimization in environments with slow changes, and avoiding result fluctuations.

[0105] Step 302: In the second scenario where the perception network index undergoes a sudden change, the perception result determination device acquires the change feature information of the second scenario and calls the pre-stored calibration value corresponding to the change feature information.

[0106] In some embodiments, the aforementioned pre-stored calibration values ​​are used to correct the sensing results for the next statistical period.

[0107] In some embodiments, when the sensing result determination device detects a drastic change in sensing network indicators (such as a sudden drop in SNR within minutes, a sharp increase in false alarm rate, or a sudden and consistent shift in all target angles), it determines it to be a second scenario (a sudden scenario, such as a sudden fog or strong wind). In this scenario, smooth iteration may be too slow, so the device immediately switches to a semi-static adjustment strategy. The specific process is as follows: Obtaining mutation feature information: Quickly analyze the combination pattern of mutation indicators to identify the corresponding environmental condition type (e.g., identifying it as a fog mode); Retrieving pre-stored calibration values: Directly retrieve pre-stored calibration values ​​that match the environmental condition type from historical sensing calibration records; Applying correction: The pre-stored calibration values ​​are used to correct the sensing results obtained in the next statistical period (or immediately).

[0108] In some embodiments, in subsequent runs, the S-correction will be used as a new statistical metric for adaptive adjustments to the results. Optimal accuracy is ensured through continuous iteration.

[0109] In this process, the statistical period and calibration period can be configured according to requirements. Furthermore, fixed correction values ​​can be saved based on different times and weather conditions, transforming dynamic adjustments into semi-static adjustments.

[0110] For example, in general scenarios, we maintain a 1-hour observation period, calculate the offset value within that hour, and then iterate every hour thereafter, ensuring the entire correction process is linearly correlated with the previous time period, resulting in a smoother calibration curve. However, in some sudden scenarios, gradual iteration is not suitable. For instance, in foggy conditions, optimization is needed for abrupt changes. When the statistically obtained parameters match the characteristics of foggy weather, the correction value is set to the value saved for foggy weather, performing a semi-static adjustment, and then fine-tuning based on real-time statistical values. Specifically, in foggy weather, due to significant signal attenuation and reduced received power, the change curve of SNR and other parameters in the same area can reflect the size of the fog. Similarly, angle calculations will show significant offsets in windy weather. In rainy or snowy weather, the false alarm rate will increase significantly. Various weather conditions can be evaluated using different indicators, and after determining the weather, S-correction can be performed based on the characteristics and historical records of each weather condition.

[0111] Thus, in the first scenario where the sensing network indicators change steadily, a smooth iteration strategy is adopted, using the correction results of the previous cycle as the basis for the calculation of the next cycle. This makes the calibration process continuous and gradual, resulting in a smooth calibration curve, suitable for fine-tuning and optimization of accuracy under normal conditions. In the second scenario where the sensing network indicators change abruptly, a semi-static adjustment strategy is switched to, quickly responding by identifying abrupt change characteristics (such as specific weather patterns) and calling the corresponding pre-stored calibration values. This dual-mode adaptive mechanism can flexibly handle various situations from stable operation to sudden environmental changes, ensuring both optimization efficiency under normal conditions and the ability to quickly restore accuracy in the event of sudden anomalies, thereby guaranteeing the stability and high availability of the sensor network service quality.

[0112] The following describes the perception result determination method of this application embodiment with a specific example, such as... Figure 6 The diagram shown illustrates the overall implementation process of a perception result determination method provided in this application embodiment, with the specific processes as follows: S1 to S7: S1. Initialization and parameter configuration: The sensing result determination device sets the statistical cycle and calibration cycle; initializes reference data, including inputting the real parameter values ​​of fixed known targets, establishing an empty or historical data-based statistical expected value model, and loading the existing historical sensing calibration record experience library; and configures the monitoring thresholds and judgment rules for sensing network indicator information used to identify environmental conditions.

[0113] S2. Enter the statistical period and perform sensing measurement: Within a certain statistical period, the sensing result determination device controls the base station to transmit wireless signals (such as integrated communication and sensing signals) to the area to be sensed, and receives echoes or signals from the sensing target (such as vehicles or drones). The sensing result determination device performs standard signal processing on the received signals, extracts and outputs the first sensing result for the target.

[0114] S3. Determine the environmental scenario and select a calibration strategy.

[0115] Among them, the perception result determines the change pattern of the network indicator information in real time. If the indicator changes smoothly, it is determined to be the first scenario, and then enters the smooth iterative calibration process, that is, the following S4a is executed.

[0116] If an indicator undergoes a sudden change (such as a sharp drop in signal-to-noise ratio, a sharp increase in false alarm rate, or a shift in angle consistency), it is determined to be a second scenario (a sudden scenario, such as a sudden fog or strong wind). Subsequently, it enters a semi-static adjustment and calibration process, that is, it executes the following S4b. This process first quickly identifies the environmental type corresponding to the sudden change characteristics (such as strong wind), and then directly calls the matching pre-stored calibration value from the historical perception calibration record as the current first perception calibration value.

[0117] S4a. In the first scenario, the perception result determination device calculates the first perception calibration value based on the first perception result and reference data.

[0118] The specific calculation methods include: The difference between the first perception result and the actual parameter value or statistical expected value of the fixed known target in the reference data is calculated to obtain the perception accuracy deviation value, and the calibration value is determined based on this.

[0119] Based on the sensing indicators within the current cycle, the current environmental conditions are determined (e.g., light rain). If it is determined to be strong wind or heavy fog (first environmental condition), the historical sensing calibration records are queried to obtain the matching historical calibration values ​​and use them directly as calibration values. If it is determined to be rain or snow (second environmental condition), the weights of signals of different intensities in the first sensing result are first adjusted according to the change in signal-to-noise ratio, and then the calibration values ​​are calculated in combination with historical records.

[0120] S4b In the second scenario, the perception result determination device quickly analyzes the characteristics of the sudden change, identifies the corresponding environmental condition type (such as strong wind or heavy fog), directly queries and calls the pre-stored calibration value that matches the environmental condition type from the historical perception calibration record experience library, and directly uses this value as the first perception calibration value to be used in the current week.

[0121] S5. Generate and apply calibration values ​​to obtain correction results: Perform calculations between the first perception calibration value and the first perception result to obtain the target correction perception result with improved accuracy. This result is the final perception result output to the outside world in this cycle.

[0122] In the first scenario, the first perception calibration value obtained in S4a and the first perception result obtained in S2 are calculated (S correction = S perception + S calibration) to obtain the target correction perception result.

[0123] In the second scenario, the first perception calibration value (i.e. the matched pre-stored calibration value) called and determined by S4b is subjected to the same operation as the first perception result obtained by S2 to obtain the target correction perception result.

[0124] S6. Update internal reference data: The perception result determination device uses the target correction perception result as new valid data and feeds back to update the reference data.

[0125] For example, it can be incorporated into historical datasets to update statistical expectations. Alternatively, the effective first perception calibration value can be correlated with the determined current environmental conditions and stored as a new historical perception calibration record.

[0126] S7. Enter the next statistical cycle and iterate and optimize.

[0127] When the current statistical cycle ends, the sensing result determination device enters the next statistical cycle. In the new cycle, the complete process from S2 to S6 is repeated. The new sensing measurement (S2) is based on the updated, more accurate reference data and richer historical sensing calibration records from S6 to perform a new round of scene judgment (S3) and calibration value determination (S4a or S4b), thereby achieving continuous iteration and adaptive optimization of sensing accuracy.

[0128] It is understandable that the above process forms a complete perception-calibration-correction-learning closed loop from S2 to S7. By continuously running this closed loop, smooth optimization is achieved in the first scenario, and rapid switching and stabilization of the system based on empirical values ​​are initiated in the second scenario. In this way, perception errors introduced by environmental changes and hardware state changes can be automatically tracked and compensated, thereby maintaining and continuously improving the perception accuracy and service quality of the sensor network under long-term, dynamic, and complex operating conditions.

[0129] It should be noted that the descriptions of each step S1 to S7 in this embodiment can be found in the descriptions in the above embodiments, and will not be repeated here.

[0130] Figure 7 This is a schematic diagram of the architecture of a perception result determination system provided in an embodiment of this application. The perception result determination system 800 may include: a perception data acquisition module 801, a calibration value calculation module 802, a correction module 803, a data update module 804, and a periodic iteration module 805.

[0131] The aforementioned sensing data acquisition module 801 is used to acquire the first sensing result of the base station on the sensing target within the first statistical period. The first sensing result includes at least one of the following: distance, angle, speed, and position coordinates. This is applied to step 201 and its related schemes.

[0132] The aforementioned calibration value calculation module 802 is used to calculate a first perception calibration value based on the first perception result and reference data. The reference data includes at least one of the following: the actual parameter values ​​of a fixed known target, statistically expected values ​​generated based on historical perception results, and historical perception calibration records containing at least one historical environmental condition. This is applied to steps 202, 202a and 202b, 202c and 202d, 202d1 and 202d2, and 202d3, and related schemes.

[0133] The aforementioned correction module 803 is used to calculate the target corrected perception result based on the first perception result and the first perception calibration value. It is applied to step 203 and its related schemes.

[0134] The aforementioned data update module 804 is used to update the reference data based on the target correction perception results. It is applied to step 204 and its related schemes.

[0135] The aforementioned periodic iteration module 805 is used to coordinate the system to correct the sensing results obtained in the second statistical period based on the updated reference data. It is applied to step 205 and its related schemes.

[0136] The perception result determination system provided in this application embodiment fully realizes the adaptive closed loop of perception-calibration-correction-update-iteration through the coordinated work of the above modules. The system can automatically cope with the challenges brought about by environmental changes and hardware drift, effectively improve and maintain the perception accuracy and service reliability of the sensor network in the long term.

[0137] It should be noted that for a detailed explanation of the steps performed by each module and their beneficial effects, please refer to the description in the above embodiments, which will not be repeated here.

[0138] As can be seen, the above mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the embodiments of this application provide corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0139] This application embodiment can divide the sensing result determination device into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0140] In some embodiments, this application also provides a perception result determination apparatus. The perception result determination apparatus may include one or more functional modules for implementing the perception result determination method of the above method embodiments.

[0141] For example, Figure 8 This is a schematic diagram illustrating the composition of a sensing result determination device provided in an embodiment of this application. Figure 8 As shown, the sensing result determination device 900 includes an acquisition module 901 and a processing module 902.

[0142] The acquisition module 901 is configured to acquire a first perception result of the base station on the perceived target within a first statistical period. The first perception result includes at least one of the following: distance, angle, speed, and position coordinates. The processing module 902 is configured to calculate a first perception calibration value based on the first perception result and reference data. The reference data includes at least one of the following: the actual parameter value of a fixed known target, a statistical expected value generated based on historical perception results, and a historical perception calibration record containing at least one historical environmental condition. The processing module 902 is further configured to calculate a target correction perception result based on the first perception result and the first perception calibration value. The processing module 902 is further configured to update the reference data based on the target correction perception result. The processing module 902 is further configured to correct the perception result acquired in the second statistical period based on the updated reference data within the second statistical period.

[0143] The sensing result determination device provided in this application acquires the original sensing results in each statistical period, and then calculates a dynamic sensing calibration value by combining it with multi-source reference data including historical experience and objective benchmarks. This calibration value is used to correct the original sensing results in real time, resulting in a more accurate target-corrected sensing result. This corrected result is fed back to update the reference data itself, enabling the system experience base or cognitive benchmark to be continuously updated during operation. Thus, in subsequent statistical periods, new sensing results will be corrected based on the updated and more accurate reference data. This automatically tracks and compensates for systematic sensing errors caused by environmental changes, hardware drift, and other factors, thereby improving the sensing accuracy stability and reliability of the sensor network under long-term complex operating conditions, and ensuring the sensing accuracy and sensing service quality of the sensor network in dynamic environments.

[0144] In some embodiments, the reference data includes at least one of the actual parameter values ​​and the expected statistical values; the processing module 902 is specifically used to: calculate the difference between the first sensing result and the reference data to obtain the sensing accuracy deviation value within the first statistical period; and determine the first sensing calibration value based on the sensing accuracy deviation value.

[0145] In other embodiments, the reference data includes historical sensing calibration records; the processing module 902 is specifically used to: determine the current environmental conditions based on the sensing network index information within the first statistical period; and determine the first sensing calibration value based on the current environmental conditions and the historical sensing calibration records; wherein the sensing network index information includes at least one of the following: the signal-to-noise ratio change curve of the sensing signal, the false alarm rate change of the sensing result, and the systematic offset of the angle measurement value.

[0146] In some other embodiments, the processing module 902 is specifically used to: query historical sensing calibration records and obtain matching historical calibration values ​​when the current environmental conditions are the first environmental conditions; and use the matching historical calibration values ​​as the first sensing calibration values.

[0147] In some other embodiments, the processing module 902 is specifically used to: when the current environmental condition is the second environmental condition, adjust the weights of different intensity signal components in the first sensing result based on the signal-to-noise ratio change information of the sensing signal within the first statistical period, and calculate the first sensing calibration value based on the weighted sensing result and the historical sensing calibration record.

[0148] In some other embodiments, the processing module 902 is further configured to: in a first scenario where the perception network index changes steadily, use the target correction perception result of the first statistical period as the statistical expected value or perception calibration value for the next statistical period; in a second scenario where the perception network index changes abruptly, obtain the mutation feature information of the second scenario and call the pre-stored calibration value corresponding to the mutation feature information, the pre-stored calibration value being used to correct the perception result of the next statistical period.

[0149] It should be noted that the sensing result determination device can implement all the processes implemented in the above method embodiments and achieve the same beneficial effects. To avoid repetition, it will not be described again here.

[0150] In the case where the functions of the integrated modules described above are implemented in hardware, this application provides a possible structural schematic diagram of the electronic device involved in the above embodiments. For example... Figure 9 As shown, the electronic device 90 includes: a processor 92, a communication interface 93, and a bus 94. Optionally, the electronic device 90 may also include a memory 91.

[0151] Processor 92 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0152] Communication interface 93 is used to connect with other devices via a communication network. This communication network can be Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0153] The memory 91 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0154] In one possible implementation, the memory 91 can exist independently of the processor 92. The memory 91 can be connected to the processor 92 via a bus 94 and is used to store instructions or program code. When the processor 92 calls and executes the instructions or program code stored in the memory 91, it can implement the perception result determination method provided in the embodiments of this application.

[0155] In another possible implementation, memory 91 can also be integrated with processor 92.

[0156] Bus 94 can be an Extended Industry Standard Architecture (EISA) bus, etc. Bus 94 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 9 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0157] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the service calling device can be divided into different functional modules to complete all or part of the functions described above.

[0158] This application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be executed by computer instructions instructing related hardware. The program can be stored in the aforementioned computer-readable storage medium, and when executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be any of the foregoing embodiments or memory. The aforementioned computer-readable storage medium can also be an external storage device of the aforementioned service invocation device, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the aforementioned service invocation device. Further, the aforementioned computer-readable storage medium can include both internal storage units of the aforementioned service invocation device and external storage devices. The aforementioned computer-readable storage medium is used to store the aforementioned computer program and other programs and data required by the aforementioned service invocation device. The aforementioned computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0159] This application also provides a computer program product comprising a computer program that, when run on a computer, causes the computer to execute any of the perception result determination methods provided in the above embodiments.

[0160] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A perception result determination method, characterized in that, The method comprises: acquiring a first sensing result of a base station on a sensing target in a first statistical period, the first sensing result comprising at least one of the following: distance, angle, speed, position coordinates; calculating a first sensing calibration value based on the first sensing result and reference data, the reference data comprising at least one of the following: a real parameter value of a fixed known target, a statistical expected value generated based on historical sensing results, a historical sensing calibration record containing historical sensing calibration records associated with at least one historical environmental condition; calculating a target corrected sensing result based on the first sensing result and the first sensing calibration value; updating the reference data based on the target corrected sensing result; in a second statistical period, correcting a sensing result acquired in the second statistical period based on the updated reference data.

2. The perception result determination method of claim 1, wherein, The reference data comprises at least one of the real parameter value and the statistical expected value; The method further comprises: calculating a difference between the first sensing result and the reference data to obtain a sensing accuracy deviation value in the first statistical period; determining the first sensing calibration value based on the sensing accuracy deviation value.

3. The method according to claim 1, wherein, The reference data comprises the historical sensing calibration record; The method further comprises: determining a current environmental condition based on sensing network index information in the first statistical period; determining the first sensing calibration value based on the current environmental condition and the historical sensing calibration record; The sensing network index information comprises at least one of the following: a signal-to-noise ratio change curve of a sensing signal, a false alarm rate change of a sensing result, a systematic deviation of an angle measurement value.

4. The perception result determination method of claim 3, characterized in that, The method further comprises: in a case where the current environmental condition is a first environmental condition, querying the historical sensing calibration record to obtain a matching historical calibration value; using the matching historical calibration value as the first sensing calibration value.

5. The method according to claim 3, wherein The method further comprises: in a case where the current environmental condition is a second environmental condition, adjusting the weight of different intensity signal components in the first sensing result based on signal-to-noise ratio change information of a sensing signal in the first statistical period, and calculating the first sensing calibration value based on the weighted and adjusted sensing result and the historical sensing calibration record.

6. The perception result determination method of claim 1, wherein, The method further comprises: in a first scenario where a sensing network index changes smoothly, using the target corrected sensing result of the first statistical period as a statistical expected value or a sensing calibration value in a next statistical period; in a second scenario where a sensing network index changes abruptly, acquiring mutation feature information of the second scenario, and calling a pre-stored calibration value corresponding to the mutation feature information, the pre-stored calibration value being used to correct a sensing result in a next statistical period.

7. A perception result determination apparatus characterized by comprising: The method comprises: an acquisition module and a processing module; The acquisition module is configured to acquire a first sensing result of a base station on a sensing target in a first statistical period, the first sensing result including at least one of a distance, an angle, a speed, and a position coordinate. The processing module is configured to calculate a first sensing calibration value based on the first sensing result and reference data, the reference data including at least one of a real parameter value of a fixed known target, a statistical expected value generated based on historical sensing results, and a historical sensing calibration record associated with at least one historical environmental condition. The processing module is further configured to calculate a target corrected sensing result based on the first sensing result and the first sensing calibration value. The processing module is further configured to update the reference data based on the target corrected sensing result. The processing module is further configured to correct a sensing result acquired in a second statistical period based on the updated reference data.

8. The perception result determination apparatus according to claim 7, characterized in that, The reference data includes at least one of the real parameter value and the statistical expected value. The processing module is specifically configured to: calculate a difference between the first sensing result and the reference data to obtain a sensing accuracy deviation value in the first statistical period; and determine the first sensing calibration value based on the sensing accuracy deviation value.

9. The perception result determination apparatus according to claim 7, characterized in that, The reference data includes the historical sensing calibration record. The processing module is specifically configured to: determine a current environmental condition based on sensing network index information in the first statistical period; and determine the first sensing calibration value based on the current environmental condition and the historical sensing calibration record. The sensing network index information includes at least one of a signal-to-noise ratio change curve of a sensing signal, a false alarm rate change of a sensing result, and a systematic deviation of an angle measurement value.

10. An electronic device, comprising: The computer device includes a processor and a memory, the processor being coupled to the memory; the memory is configured to store computer instructions, the computer instructions being loaded and executed by the processor to enable the computer device to implement the sensing result determination method in any one of claims 1 to 6.

11. A computer readable storage medium, characterized in that, The computer readable storage medium includes computer execution instructions, when the computer execution instructions run on a computer, enabling the computer to execute the sensing result determination method in any one of claims 1 to 6.

12. A computer program product, characterised in that, The computer program product includes a computer program, when the computer program runs on an electronic device, enabling the electronic device to execute the sensing result determination method in any one of claims 1 to 6.

Images