Perception method and device, electronic equipment, storage medium and computer program product

Abstract

The invention relates to the technical field of wireless, and particularly provides a sensing method and device, electronic equipment, a storage medium and a computer program product. According to the method and the device, the fusion weight corresponding to each first measurement quantity is determined through the beam information in the symbol-level data, and the first measurement quantity in the symbol-level data is optimized by using the fusion weight to obtain the second measurement quantity, so that multi-target identification is performed based on the second measurement quantity and the sensing result is determined. On one hand, multi-point data optimization is performed in combination with beam information of a more bottom layer, and on the other hand, data optimization and subsequent association identification are directly performed by using symbol-level data after RVA estimation, so that symbol-level data information of a more original layer can be more fully utilized, and the accuracy of data optimization is improved. The problems of information loss and large sensing precision error caused by only data level fusion in the prior art are solved, and the sensing precision is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the wireless technical field, and in particular to a sensing method and device, electronic equipment, storage medium and computer program product. BACKGROUND

[0002] The integrated sensing and communication (ISAC) system is a distributed networking based on communication. Due to the limited distance between stations and the relatively limited transmission power of wireless base stations, the cross-cell / cross-station scenario is frequent between cells. Therefore, compared with the radar sensing system, the integrated sensing and communication system is insufficient in the multi-station cooperative sensing target detection accuracy. The multi-station cooperative sensing target detection and accuracy improvement based on the communication networking structure (i.e., the integrated sensing and communication system) has become an important topic for the current 5G-A / 6G integrated sensing and communication research.

[0003] Please refer to Figure 1 , Figure 1 The figure is a sensing process schematic diagram of the related art integrated sensing and communication system. As shown in Figure 1 , the sensing process includes two parts of sensing signal processing and sensing data processing. The sensing signal processing is to sample the pre-processed (e.g., filtered) sensing echo signal to obtain the masked symbol-level IQ (I represents In-phase and Q represents Quadrature) data, and then to perform the cyclic prefix (CP) processing, Fast Fourier Transform (FFT) processing, and Range, Velocity, and Angle estimation (referred to as RVA estimation). Then, the sensing data processing process is entered. In the sensing data processing process, the point cloud is clustered first to aggregate a group of point clouds into a target position. Then, the multi-cell / cross-station data-level fusion and track association fusion processing are performed, so that the final position and velocity of the target object can be obtained, and the multi-point trajectory at different times is continuously tracked. Finally, the shape and category of the target object are identified, so that the sensing result of the sensing target can be obtained, managed, and tracked. In summary, in the related art, the fusion sensing detection process based on the integrated sensing and communication mainly focuses on the data-level / track-level fusion processing in the multi-station data fusion link. Usually, the position and velocity of the target object are first sensed and output by each single station to form a point track, and then the multi-station data-level / track-level association and fusion are performed.

[0004] However, the perception scheme based on data level / track level association fusion loses a large amount of original information, resulting in a large perception error, affecting the perception accuracy, and especially for small and weak targets such as unmanned aerial vehicles, the "flickering effect" is prone to occur, which cannot meet the perception business requirements. SUMMARY

[0005] The present application is proposed in view of the above problems. The present application provides a perception method and device, electronic equipment, storage medium and computer program product.

[0006] According to one aspect of the present application, a perception method is provided, comprising:

[0007] Obtaining symbol level data corresponding to a perception echo signal to obtain a first measurement quantity and beam information; the first measurement quantity includes: distance, speed, angle;

[0008] Determining a fusion weight corresponding to the first measurement quantity based on the beam information;

[0009] Optimizing the first measurement quantity using the fusion weight to obtain a second measurement quantity;

[0010] Performing multi-target identification based on the second measurement quantity, and determining a perception result of each target object.

[0011] According to another aspect of the present application, a perception device is provided, comprising:

[0012] An obtaining unit, configured to obtain symbol level data corresponding to a perception echo signal to obtain a first measurement quantity and beam information; the first measurement quantity includes: distance, speed, angle;

[0013] A determining unit, configured to determine a fusion weight corresponding to the first measurement quantity based on the beam information;

[0014] An optimizing unit, configured to optimize the first measurement quantity using the fusion weight to obtain a second measurement quantity;

[0015] A processing unit, configured to perform multi-target identification based on the second measurement quantity, and determine a perception result of each target object.

[0016] According to another aspect of the present application, an electronic equipment is provided, comprising a memory, a processor and a computer program stored on the memory, the processor executes the computer program to implement the method of any of the above embodiments.

[0017] According to another aspect of the present application, a computer readable storage medium is provided, which stores a computer program / instruction, the computer program / instruction is executed by a processor to implement the method of any of the above embodiments.

[0018] According to another aspect of the present application, a computer program product is provided, comprising computer programs / instructions which, when executed by a processor, implement the method according to any of the above embodiments.

[0019] As will be described in detail below, according to a sensing method and apparatus, electronic device, storage medium and computer program product according to embodiments of the present application, the present application determines the fusion weight corresponding to each first measurement quantity according to the beam information in the symbol-level data, and optimizes the first measurement quantity in the symbol-level data using the fusion weight to obtain a second measurement quantity, so that multi-target recognition is performed based on the second measurement quantity and a sensing result is determined. On the one hand, the present application performs multi-point data optimization in combination with more bottom-layer beam information, and on the other hand, the present application directly performs data optimization and subsequent correlation recognition using the symbol-level data after RVA estimation, which can more fully utilize the more original-layer symbol-level data information, solves the problem of large sensing precision error due to information loss caused by only data-level fusion in the related art, effectively improves the sensing precision, can avoid the "flickering effect", and meets the sensing business requirements.

[0020] It is to be understood that both the foregoing general description and the following detailed description are exemplary, and are intended to provide further explanation of the subject technology. BRIEF DESCRIPTION OF DRAWINGS

[0021] The foregoing and other objects, features and advantages of the present application will become more apparent from the following detailed description, which proceeds with reference to the accompanying drawings. The drawings are provided to illustrate embodiments of the present application and, together with the detailed description, serve to explain the present application and do not constitute a limitation thereof. In the drawings, like reference numerals refer to like elements or steps throughout.

[0022] Figure 1 A sensing process schematic diagram of a sensing integrated system in the related art.

[0023] Figure 2 A sensing method provided by an embodiment of the present application.

[0024] Figure 3 A symbol-level data acquisition process schematic diagram provided by an embodiment of the present application.

[0025] Figure 4 A normal offset schematic diagram provided by an embodiment of the present application.

[0026] Figure 5 An implementation logic schematic diagram of a sensing method provided by an embodiment of the present application.

[0027] Figure 6This is a flowchart illustrating another sensing method provided in an embodiment of this application.

[0028] Figure 7 This is a structural block diagram of a sensing device provided in an embodiment of this application.

[0029] Figure 8 This is a hardware block diagram of an electronic device provided in an embodiment of this application.

[0030] Figure 9 This is a schematic diagram of a computer-readable storage medium provided in an embodiment of this application. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this application more apparent, exemplary embodiments according to this application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0032] The embodiments of this application are applied to the field of perception, such as perception detection and perception tracking, and there are no particular limitations thereto.

[0033] To address the issue of poor perception accuracy in perception schemes based on data-level / track-level correlation fusion in related technologies, this application provides a novel design concept: advancing the multi-point collaborative perception fusion process by directly weighting and fusing the symbol-level data after RVA estimation of single-point IQ data. Thus, during the fusion process, on the one hand, multi-point weighted fusion is performed by incorporating more fundamental spatial beam feature information; on the other hand, this application directly optimizes data and performs subsequent correlation identification using the symbol-level data after RVA estimation, enabling more full utilization of the original symbol-level data information. Furthermore, this application may further include a beam feedback calibration mechanism, continuously iterating beam calibration during the measurement process, improving perception accuracy while continuously enhancing the original beam measurement effect and target acquisition capability. The following is a detailed description.

[0034] This application provides a sensing method. Please refer to... Figure 2 , Figure 2 This is a flowchart illustrating a sensing method provided in an embodiment of this application. Figure 2 As shown, the method includes:

[0035] S202, acquire the symbol-level data corresponding to the sensed echo signal to obtain the first measurement quantity and beam information; the first measurement quantity includes: distance, velocity, and angle. The first measurement quantity can be denoted as (d, v, θ).

[0036] The symbol level data involved in the present application means the original data obtained by signal processing of the perception echo signal, which carries more comprehensive original information. The acquisition method involved in this step can include directly signal processing of the perception echo signal, or indirectly acquiring the symbol level data obtained by signal processing of the perception echo signal by other devices or equipment. In other words, in the present application, the execution subject of the perception method (marked as a perception device for convenience) can process the perception echo signal by itself, or acquire the processed data of other devices or equipment to obtain the symbol level data corresponding to the perception echo signal.

[0037] Therefore, in a possible embodiment, the step can be implemented by the following manner: signal processing of the perception echo signal to obtain the symbol level data, the symbol level data including: the first measurement and the beam information; and / or receiving the symbol level data reported by the perception device.

[0038] The perception device can sample and signal process the perception echo signal, which can specifically include but is not limited to at least one of the following: RVA estimation, CP removal processing, FFT processing, constant false alarm rate (CFAR) processing, etc.; wherein RVA estimation means distance estimation, velocity estimation and angle estimation; the specific estimation manner is not limited, for example, distance estimation can be realized by R-spectrum estimation, and velocity estimation can be realized by V-spectrum estimation. Details are not described herein. In addition, the present application does not have special limitation on the execution timing of the above-mentioned various signal processing manners, for example Figure 1 as shown, and for example, the perception echo signal can be sequentially subjected to CP removal processing, FFT processing, CFAR processing and RVA estimation processing. Details are not described herein.

[0039] Based on the above processing, the symbol level measurement result (i.e. the first measurement) can be acquired; in addition, the symbol level data that can be acquired by the step can include but is not limited to: beam information, time information (t), cell ID. The beam information (also referred to as beam feature information or perception beam information) can include but is not limited to at least one of the following: beam ID, signal quality. The beam information is used to describe the state of the perception beam, which can be applied to the correction and optimization of the symbol level data in the present application; specifically, the beam information is used to determine the fusion weight, and the first measurement data is optimized accordingly to improve the perception accuracy, which will be described in detail hereinafter.

[0040] In addition, it should be understood that in some possible embodiments, the perception device can also send the symbol level data obtained based on the perception echo signal to other perception devices, so as to facilitate other perception devices to complete the subsequent perception process.

[0041] Furthermore, the embodiments of this application can support sensing processing in single-cell / multi-cell, single-base station / multi-base station, single-spatial beam / multi-spatial beam, and single-channel / multi-channel environments.

[0042] In one exemplary embodiment, the symbol-level data obtained in this step may include, but is not limited to, at least one of the following:

[0043] Symbol-level data from at least one cell;

[0044] Symbol-level data for at least one space beam;

[0045] Symbol-level data in at least one channel environment.

[0046] To facilitate understanding, let's take a sensing scenario using a base station as an example. For illustration, please refer to... Figure 3 , Figure 3 This is a schematic diagram illustrating the symbol-level data acquisition process provided in an embodiment of this application. Figure 3 As shown, a base station can cover multiple cells. During this step, signals can be transmitted and received based on the current beam and channel environment. Then, the sensed echo signals are sampled to obtain I / Q data. Afterward, signal processing is performed on the I / Q data of a single cell to obtain the symbol-level data measured for each cell under the current beam. Repeating this process at different time units (e.g., no limit on time slots, but at least one time unit) yields the symbol-level data for the current base station, current cell, and current spatial beam under the current channel environment at different time units.

[0047] Then, multiple symbol-level estimation data from different cells, different spatial beams, and different channel environments can be processed to achieve symbol-level fusion, and sensing results can be obtained based on the fused data.

[0048] S204, Based on beam information, determine the fusion weight corresponding to the first measurement quantity.

[0049] In this embodiment, the fusion weight can be used to characterize the impact of the sensing beam on symbol-level data; this application applies it to optimize (or correct) the first measurement, so that the optimized second measurement can more accurately reflect the sensing characteristics and improve the sensing accuracy.

[0050] Specifically, the fusion weights involved in the embodiments of this application may include, but are not limited to, at least one of the following: a first weight (denoted as σ) and a second weight (denoted as τ).

[0051] The first weight is positively correlated with the signal quality of the sensing beam. In other words, the better the signal quality of the sensing beam, the more accurate the first measurement quantity corresponding to the sensing beam, and the higher the first weight σ; on the contrary, the worse the signal quality of the sensing beam, the greater the error of the first measurement quantity corresponding to the sensing beam, and the lower the first weight σ.

[0052] The second weight is negatively correlated with the normal offset. The normal offset is the offset of the direction in which the sensing beam captures the target relative to the beam normal. Specifically, the smaller the normal offset (denoted as α), the closer the direction in which the sensing beam captures the target to the beam normal direction, the more concentrated the energy, the higher the gain, the more accurate the first measurement quantity, and the higher the second weight τ; on the contrary, the greater the normal offset α, the farther the direction in which the sensing beam captures the target from the beam normal direction, the more dispersed the energy, the lower the gain, the greater the error of the first measurement quantity, and the lower the second weight τ.

[0053] For ease of understanding, please refer to Figure 4 , Figure 4 The figure of the normal offset provided by the embodiments of the present application is shown. As shown in Figure 4 , the normal of the sensing beam (i.e., the beam normal) is generally the central position of the sensing beam, the energy is concentrated, and the gain is high; thus, the closer the direction in which the sensing beam captures the target to the beam normal, the more accurate the first measurement quantity, and the higher the second weight that can be set; on the contrary, the farther the direction in which the sensing beam captures the target from the beam normal, the greater the error of the first measurement quantity, and the lower the second weight that can be set. As shown in Figure 4 , the present application represents the degree of offset of the direction in which the sensing beam captures the target relative to the beam normal by the normal offset. In specific implementation, if the transmission time of the current beam is t, after the foregoing signal processing process, the RVA symbol-level data can be obtained, and thus the angle θ in the first measurement quantity can also be obtained, and the base station knows the scanning time sequence and the normal direction of each beam, and thus the offset α of the angle θ relative to the beam normal of each beam can be obtained. Thus, the second weight can be determined based on the normal offset α subsequently.

[0054] In addition, in actual scenarios, the sensing object can be in a motion state, for example Figure 4 , as shown, the positions of the same sensing target at t and t+1 are different, the angles of the base station sensing target are different, and the normal offset of the base station sensing target also changes.

[0055] In the present application, when the fusion weight is determined based on the beam information, there can be multiple implementation approaches. Exemplarily, the mapping relationship between the beam information and the fusion weight can be preset in advance, and thus the fusion weight corresponding to the current symbol-level data can be determined based on the beam information determined by the signal processing process and the mapping relationship.

[0056] In an example embodiment, determining the fusion weight corresponding to the first measurement quantity comprises at least one of: determining a first weight corresponding to the first measurement quantity based on a first mapping relationship between the signal quality and the first weight; and determining a second weight corresponding to the first measurement quantity based on a second mapping relationship between the normal offset and the second weight. In this embodiment, the mapping relationship between the different beam information and the respective fusion weight can be preset, so that the fusion weight corresponding to the symbol-level data can be directly determined based on the mapping relationship.

[0057] In addition, the application embodiment does not have special restrictions on the setting mode of the mapping relationship. For example, at least two beam information (i.e., signal quality and / or normal offset) intervals (which can be a closed interval determined by two thresholds, or a single open interval determined by more or less than a threshold, which is not limited) can be preset, and the fusion weight corresponding to each beam information interval can be preset, so as to construct the mapping relationship between them. In this way, when implementing this step, only the beam information interval to which the current symbol-level data belongs needs to be determined, and the fusion weight corresponding to the current symbol-level data can be determined. For another example, a reference value and a unit adjustment step of the beam information (i.e., signal quality and / or normal offset) can also be preset. In this way, when specifically implemented, the difference (for example, the difference of several unit values) between the beam information corresponding to the current symbol-level data and the reference value can be determined, so that the difference is multiplied by the unit adjustment step to determine the corresponding fusion weight. In actual scenarios, the mapping relationship between the two can also have other modes, which are not limited by the application and are not exhaustive.

[0058] In S206, the first measurement quantity is optimized by using the fusion weight to obtain a second measurement quantity.

[0059] As described above, the fusion weight can actually be used to represent the influence of the perception beam on the symbol-level data. In order to improve the perception accuracy, the application uses the fusion weight to optimize the first measurement quantity in the symbol-level data, so that the part of the first measurement quantity with more accurate perception (better signal quality, more concentrated energy, and higher gain) can play a greater component advantage, and the part of the first measurement quantity with a possible large error in perception accuracy can have a lower weight in the perception process, thereby reducing the adverse effect of the part of the first measurement quantity with poor accuracy on the entire perception result. In this way, data optimization is achieved from the level of symbol-level data, which not only retains a large amount of original information, but also assists symbol-level data optimization from the perspective of beam, thereby improving the perception accuracy.

[0060] In an example embodiment, the first measurement quantity is processed by using the fusion weight to obtain the second measurement quantity, including: obtaining the product of the first measurement quantity and the corresponding fusion weight to obtain the second measurement quantity. Specifically, the fusion weight can include: the first weight and / or the second weight, so that when this step is performed, there can be the following possible implementation ways: obtaining the product of the first measurement quantity and the first weight (i.e. (d, v, θ) × σ), obtaining the product of the first measurement quantity and the second weight (i.e. (d, v, θ) × τ), and obtaining the product of the first measurement quantity and the first weight and the second weight (i.e. (d, v, θ) × σ × τ).

[0061] This embodiment is the processing logic for a single first measurement quantity. In actual scenarios, for example, a base station can cover multiple cells, and the symbol-level data of multiple cells can also be optimized and corrected together. Similarly, the symbol-level data of multiple base stations, multiple spatial beams, and multiple single-channel environments can also be processed together.

[0062] For example, in the multi-cell symbol-level data optimization process, when the first measurement quantity is processed by using the fusion weight, the following processing can be performed: the first measurement quantities of each cell are symbol-aligned to obtain at least one first set; then, the product of the first set and the weight vector is obtained to obtain a second set; wherein the weight vector includes the fusion weight corresponding to each first measurement quantity, and the second set includes the second measurement quantity corresponding to each cell.

[0063] Symbol alignment is also called time alignment. For symbol-level data, before subsequent processing, it is necessary to ensure that the symbols are roughly aligned, that is, to ensure that the symbol boundaries are aligned. In specific scenarios, each symbol is about 36us, and the present application can meet the synchronization requirements of 50ns symbol-level fusion. When implemented, time alignment can be achieved based on time interpolation, so that multiple sets of first measurement quantities at the same time ti can be obtained, that is, the first set.

[0064] Specifically, n sets of distance, speed, and angle three-dimensional measurement quantities (i.e. first measurement quantities) of n cells (n is an integer greater than 1) at the same time ti can be aligned, so that an n × 3 matrix can be formed, which is the first set; and the fusion weight of each cell can be represented as a fusion weight vector (or simply a weight vector), so that for n cells, the weight vector can include: the first weight vector and / or the second weight vector. In this way, the second set can be represented as the following cases:

[0065]

[0066] Or,

[0067]

[0068] or,

[0069]

[0070] wherein, is a first vector, is a first weight vector, is a second weight vector.

[0071] By the above processing, the beam information can be used to realize the optimization processing of the symbol-level data, so that the second measurement quantity obtained thereby can more accurately and comprehensively describe the perception feature, and is more conducive to improving the perception accuracy.

[0072] S208, multi-target recognition is performed based on the second measurement quantity, and a perception result of each target object is determined.

[0073] After the optimization correction of the symbol-level data is implemented, a set of n second measurement quantities (which can be denoted as a second set) can be obtained, based on which subsequent perception data processing can be performed to obtain the perception result, that is, multi-target recognition is performed based on the second measurement quantity, and a perception result of each target object is determined.

[0074] For example, clustering analysis can be performed based on the three-dimensional approximation degree of each second measurement quantity to obtain at least one perception target; then, for any one perception target, a perception result is determined based on the second measurement quantity corresponding to the perception target.

[0075] For any two second measurement quantities, the closer the three-dimensional approximation degree between them, the higher the possibility that they are the same perception target, and accordingly, two second measurement quantities with the same or similar (for example, the approximation degree or difference is higher than a preset threshold) three-dimensional estimation value are determined as the same perception target and are assigned a target identifier (i.e., target ID). Through global polling processing of each second measurement quantity, the association and clustering of each perception target can be realized to obtain a set of second measurement quantity data sets (or point cloud clusters, i.e., a set of optimized RVA estimation values) corresponding to each perception target (with a target ID).

[0076] On this basis, each perception target can be processed respectively to determine the perception result of each perception target at the current time.

[0077] The perception result of any one perception target involved in the embodiments of the present application can include but is not limited to at least one of the following: target identifier, time information, position information, and beam control information.

[0078] The target ID is a target ID that can be assigned or generated in a multi-target identification process and is used to uniquely identify a perception target. The time information is used to indicate the current time or each time point in the perception target trajectory process.

[0079] The position information can be used to indicate at least the position of the perception target at the current time. Further, the position information can also be used to indicate each position involved in the trajectory (i.e., the position information can be a set, or can be understood as one or more trajectories). The representation of the position information is not limited in the present application. For example, the position information involved in the embodiments of the present application can include but is not limited to at least one of the following:

[0080] The first coordinate corresponding to the current time, the first coordinate including: distance, speed, angle;

[0081] The second coordinate and speed corresponding to the current time, the second coordinate being a three-dimensional space coordinate.

[0082] The first coordinate corresponding to the current time ti can be represented as: (d ti ,v ti ,θ ti ). The second coordinate corresponding to the current time ti can be represented as: (x ti ,y ti ,z ti ), wherein x, y, and z are perpendicular to each other and form a three-dimensional space coordinate system. It should be understood that when the perception target moves in space, the second coordinate can uniquely identify a position in a three-dimensional space coordinate system. In order to achieve accurate perception or further subsequent application, the perception result can further carry the speed v ti in the perception result when representing the perception result by the second coordinate. In this way, the perception result can clearly and accurately feedback the current position and motion state of the perception target.

[0083] In addition, the beam control information is used to indicate the beam control mode in the subsequent perception process, which is beneficial to optimize the subsequent perception process and improve the perception accuracy. The beam feedback calibration mechanism will be described in detail later.

[0084] In this way, in an exemplary embodiment, the perception result can include but is not limited to one or more of the following: target ID, time information, beam ID, normal offset of the beam, position information.

[0085] In summary, the application determines the fusion weight corresponding to each first measurement quantity through the beam information in the symbol-level data, optimizes the first measurement quantity in the symbol-level data using the fusion weight, obtains the second measurement quantity, and performs multi-target recognition based on the second measurement quantity and determines the perception result. On the one hand, the application optimizes multi-point data in combination with more bottom-layer beam information, and on the other hand, the application directly optimizes data and performs subsequent correlation recognition using the symbol-level data after RVA estimation, which can more fully utilize the more original-level symbol-level data information, solves the problem of large information loss and perception accuracy error in the related art by only data-level fusion, effectively improves the perception accuracy, avoids the "flickering effect", and meets the perception business requirements.

[0086] In addition, as described above, in addition to implementing the above perception method, the application further includes a beam feedback calibration mechanism, which continuously improves the original beam measurement effect and target capture capability while improving the perception accuracy through beam calibration in the continuous iterative measurement process.

[0087] In an exemplary embodiment, the application can further include the following processing:

[0088] For any one perception target, based on the historical perception data and the perception result of the perception target, a predicted value of the perception target at the next time is determined;

[0089] Based on the predicted value, beam control information is determined; wherein the beam control information includes at least one of the following: beam identifier, normal offset of the beam;

[0090] The beam control information is fed back to the beam control module.

[0091] In this embodiment, the application can use any prediction algorithm or prediction model to process the perception result of the perception target, so as to obtain the predicted value of the perception target at the next time. Based on this, the perception device (such as a base station) can determine the beam (or understand as the perception target in the next time may belong to which beam) matched with the perception target (such as the closest or capable of covering the position) in the next time according to this, can determine the beam identifier (i.e. beam ID) of the target beam, and the perception device can also determine the normal offset of the perception target relative to the normal of the target beam based on its position and the normal of the target beam. In this way, the perception device can send these beam control information to the beam control module.

[0092] The beam control module can be used to control the beam direction and angle of the sensing signal, control the sending of the sensing signal and the receiving of the sensing echo signal. The beam control module can be a signal processing part of the sensing device itself, or can be an independent device independent of the sensing device, and there is no limitation. After receiving the beam control information, the beam control module can adjust the beam direction and angle at the next time, so as to realize the accurate alignment of the sensing target, and further improve the sensing accuracy.

[0093] For the convenience of understanding, please refer to Figure 5 , Figure 5 The implementation logic diagram of the sensing method provided by the embodiment of the present application is shown in the figure. In this embodiment, the sensing device is regarded as two parts, the first data processing unit is used to obtain a plurality of single-point symbol-level data corresponding to the sensing echo signal, and the second data processing unit is used to realize symbol-level data optimization and sensing prediction. As shown in Figure 5 , the first data processing unit can realize single-point symbol-level RVA estimation (other processing can also be included, which is not described here), and the obtained first measurement (d, v, θ) and beam information (signal quality, normal offset α) are transmitted to the second data processing unit; the second data processing unit performs the foregoing processing of the present application to obtain the sensing result, i.e. the position information (d ti ,v ti ,θ ti ) corresponding to the current time. In addition, as shown in Figure 5 , the second data processing unit will also feed back the beam prediction direction at the next time to the beam control device through the beam feedback calibration mechanism, to guide the beam scanning / tracking process at the next time. Therefore, the beam control information will act on the single-point symbol-level RVA estimation result at the next time.

[0094] By continuously repeating the process, spatial beam alignment feedback can be provided for the beam at the next time in each sensing process, so as to guide the tracking of the target by the beam at the next time, better align and track the target with the normal direction of the beam, and produce feedback iteration gain. That is, based on the beam feedback calibration mechanism, the present application can continuously iterate the beam calibration in the measurement process, continuously enhance the capture ability of the beam to the target in the early measurement process, optimize the beam scanning efficiency and measurement effect, and thus optimize the detection accuracy.

[0095] In addition, the embodiment of the present application also provides an embodiment, please refer to Figure 6 , Figure 6 The flowchart of another sensing method provided by the embodiment of the present application is shown in the figure. As shown in Figure 6As shown, in the method, first, beam scanning measurement and reception are performed, and then, signal processing (including but not limited to symbol-level RVA estimation) can be performed based on the received sensing echo signals, so that symbol-level data, i.e., a first measurement quantity, can be obtained. In addition, the sensing device can also obtain beam information. Then, fusion weights can be determined based on the beam information, and weighted optimization of the symbol-level data is performed to optimize the first measurement quantity to obtain a second measurement quantity. Then, multi-target recognition is performed based on the second measurement quantity to determine the sensing result. Further, beam control information can be determined based on historical data of the sensing target, and the beam control information is sent to the beam control device, so that the beam control information can be applied to the subsequent beam scanning measurement process to realize beam alignment feedback. The implementation details can be referred to the foregoing, and will not be described here.

[0096] The execution subject of the technical solution provided by the embodiments of the present application is not particularly limited. For example, the sensing method can be applied to a sensing device, a base station (or other network device, network function), a sensing platform, a local processor or a cloud processor, and the like, without being exhaustive.

[0097] In summary, the present application provides a symbol-level data fusion and sensing scheme based on spatial beam feature association. The present application can fully combine the original data information at the symbol level, reduce information loss and estimation error, and make full use of the beam quality and beam offset of multiple cells / multiple points and other related original information to make more accurate measurement quantity estimation. In addition, the present application also incorporates the weighting information of the spatial beam feature, realizes the optimization processing of the fusion factor, selects more effective symbol-level data for fusion, and the weighted optimization process further improves the accuracy of the sensing calculation. Further, the present application also provides a feedback mechanism of beam feature information, by which the beam calibration in the measurement process can be iterated, the beam capture capability for the target in the early measurement process can be continuously enhanced, the beam scanning efficiency and measurement effect can be optimized, and thus the detection precision can be optimized.

[0098] The present application also provides a sensing device. Figure 7 A structural block diagram of a sensing device provided by the embodiments of the present application is shown in Figure 7 As shown, the sensing device 700 includes:

[0099] The acquisition unit 710 is configured to acquire symbol-level data corresponding to the sensing echo signal to obtain a first measurement quantity and beam information. The first measurement quantity includes distance, velocity, and angle.

[0100] The determination unit 720 is configured to determine a fusion weight corresponding to the first measurement quantity based on the beam information.

[0101] An optimization unit 730 is configured to perform optimization processing on the first measurement quantity by using the fusion weight, to obtain a second measurement quantity;

[0102] A processing unit 740 is configured to perform multi-target recognition based on the second measurement quantity, and determine a perception result of each target object.

[0103] In an exemplary embodiment, the fusion weight comprises at least one of a first weight and a second weight.

[0104] The first weight is positively correlated with the signal quality of the perception beam.

[0105] The second weight is negatively correlated with a normal offset, and the normal offset is an offset of a target direction captured by the perception beam relative to a normal line of the beam.

[0106] In an exemplary embodiment, the determination unit 720 is specifically configured to perform at least one of:

[0107] Determine the first weight corresponding to the first measurement quantity based on a first mapping relationship between the signal quality and the first weight.

[0108] Determine the second weight corresponding to the first measurement quantity based on a second mapping relationship between the normal offset and the second weight.

[0109] In an exemplary embodiment, the optimization unit 730 is specifically configured to:

[0110] Obtain a product of the first measurement quantity and the corresponding fusion weight, to obtain the second measurement quantity.

[0111] In an exemplary embodiment, the optimization unit 730 is specifically configured to:

[0112] Perform symbol alignment on the first measurement quantity of each cell, to obtain at least one first set.

[0113] Obtain a product of the first set and a weight vector, to obtain a second set; wherein the weight vector comprises the fusion weight corresponding to each first measurement quantity, and the second set comprises the second measurement quantity corresponding to each cell.

[0114] In an exemplary embodiment, the processing unit 740 is specifically configured to:

[0115] Perform clustering analysis based on a three-dimensional approximation degree of each second measurement quantity, to obtain at least one perception target.

[0116] For any one of the perception targets, determine the perception result based on the second measurement quantity corresponding to the perception target.

[0117] In an example embodiment, the sensing result of any one of the sensing targets comprises at least one of: target identification, time information, position information, beam control information.

[0118] The position information comprises at least one of:

[0119] The first coordinate corresponding to the current time, the first coordinate comprising: distance, speed, angle.

[0120] The second coordinate corresponding to the current time and the speed, the second coordinate being a three-dimensional space coordinate.

[0121] In an example embodiment, the processing unit 740 is further configured to:

[0122] For any one sensing target, based on the historical sensing data and the sensing result of the sensing target, a predicted value of the sensing target at a next time is determined;

[0123] Based on the predicted value, beam control information is determined, wherein the beam control information comprises at least one of: beam identification, normal offset of the beam.

[0124] The beam control information is fed back to a beam control module.

[0125] In an example embodiment, the acquisition unit 710 is specifically configured to:

[0126] The sensing echo signal is processed to obtain the symbol-level data, the symbol-level data comprising: the first measurement quantity and the beam information.

[0127] And / or,

[0128] The symbol-level data reported by the sensing device is received.

[0129] In an example embodiment, the symbol-level data comprises at least one of:

[0130] Symbol-level data of at least one cell;

[0131] Symbol-level data of at least one spatial beam;

[0132] Symbol-level data in at least one channel environment.

[0133] The parts not described in detail are referred to the foregoing, and will not be described again.

[0134] Figure 8A hardware block diagram of an electronic device is provided for embodiments of the present application. The electronic device 800 according to embodiments of the present application comprises at least a memory, a processor and a computer program stored on the memory, the processor executing the computer program to implement the perception method according to any of the embodiments described above.

[0135] Figure 8 The electronic device 800 shown specifically comprises a central processing unit (CPU) 801, a graphics processing unit (GPU) 802 and a memory 803. These units are connected to each other through a bus 804. The central processing unit (CPU) 801 and / or the graphics processing unit (GPU) 802 can be used as the processor described above, and the memory 803 can be used as the memory storing the computer readable instructions described above. In addition, the electronic device 800 can further comprise a communication unit 805, a storage unit 806, an output unit 807, an input unit 808 and an external device 809, and these units are also connected to the bus 804.

[0136] Figure 9 A schematic diagram of a computer readable storage medium is provided for embodiments of the present application. The computer readable storage medium according to embodiments of the present application has computer programs / instructions (including but not limited to computer readable instructions) stored thereon. Specifically, as shown in Figure 9 The computer readable storage medium 900 stores computer readable instructions 901 thereon. The computer programs / instructions are executed by a processor to implement the perception method according to any of the embodiments described above. The computer readable storage medium includes but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory, etc. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, optical disc, magnetic disc, etc.

[0137] The present application further provides a computer program product comprising computer programs / instructions, which are executed by a processor to implement the perception method according to any of the embodiments described above.

[0138] The basic principles of the present application are described above in combination with specific embodiments, but it should be noted that the advantages, advantages, effects, etc. mentioned in the present application are only examples and are not limiting, and these advantages, advantages, effects, etc. cannot be considered as the must-have of each embodiment of the present application. In addition, the above specific details are only for the purpose of example and for the purpose of understanding, and are not limiting, and the above details do not limit the present application to the must-use of the above specific details.

[0139] The block diagrams of the devices, apparatuses, equipment, systems referred to in this application are merely illustrative examples and are not intended to require or imply that the connection, arrangement, configuration must be as shown in the block diagrams. These devices, apparatuses, equipment, systems can be connected, arranged, configured in any manner as will be appreciated by those skilled in the art. Words such as "include," "contain," "have," etc. are open-ended words that are to be interpreted to mean "including but not limited to" and are to be used interchangeably. The words "or" and "and" as used herein are to be interpreted as the word "and / or" and are to be used interchangeably unless the context clearly indicates otherwise. The word "such as" as used herein is to be interpreted as the phrase "such as but not limited to" and is to be used interchangeably.

[0140] Also, as used in this application, the word "or" as used in the context of "at least one of A, B or C" means a disjunctive list of items A, B or C, so that, for example, a list of "at least one of A, B or C" means A or B or C or AB or AC or BC or ABC (i.e. A and B and C). Further, the phrase "example of" does not mean the example described is preferred or better than other examples.

[0141] It is also important to note that the systems and methods of the present application can be embodied in a variety of forms without departing from the spirit or essential characteristics thereof. Thus, the disclosures herein are meant to be taken only by way of example and to not be limiting as to the scope, applicability or configuration of the application. For instance, while the components of the systems and methods of the present application are shown in the drawings and described above as being in a particular arrangement, the various components can be rearranged or otherwise configured without departing from the spirit or essential characteristics of the application.

[0142] Various changes, modifications and alterations in the teachings and techniques described herein can be made without departing from the teachings and techniques defined by the appended claims. Moreover, the scope of the claims of this application is not limited to the specific aspects described above. Processes, machines, manufacture, compositions of matter, means, methods, or steps, presently existing or later to be developed that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Accordingly, the appended claims include within their scope such processes, machines, manufacture, compositions of matter, means, methods, or steps.

[0143] The above description of the disclosed aspects is meant to be illustrative only and not limiting as to the scope, applicability or configuration of the application. Changes can be made in the full intent and scope of equivalents which operate on the same principles as described herein and are encompassed by the following claims. Therefore, to the extent there are variations of the application, which are within the spirit of the application, these are intended to be included within the scope of the exemplary aspects of the application. Accordingly, the application is not to be restricted except in light of the attached claims and their equivalents.

[0144] The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the forms disclosed herein. Although various example aspects and embodiments have been discussed above, those of ordinary skill in the art will appreciate a variety of modifications, alternatives, permutations, additions, and sub-combinations, which fall within the scope of the application.

Claims

1. A sensing method, characterized in that, include: Obtain the symbol-level data corresponding to the sensed echo signal to obtain the first measurement quantity and beam information; The first measured quantities include: distance, speed, and angle; Based on the beam information, determine the fusion weight corresponding to the first measurement quantity; The first measurement is optimized using the fusion weights to obtain the second measurement. Multi-target identification is performed based on the second measurement, and the perception result of each target object is determined.

2. The method according to claim 1, characterized in that, The fusion weights include at least one of the following: a first weight and a second weight; The first weight is positively correlated with the signal quality of the sensing beam; The second weight is negatively correlated with the normal offset; the normal offset is the offset of the target acquisition direction of the sensing beam relative to the beam normal.

3. The method according to claim 2, characterized in that, Determining the fusion weight corresponding to the first measurement based on the beam information includes at least one of the following: Based on the first mapping relationship between signal quality and the first weight, the first weight corresponding to the first measurement quantity is determined; The second weight corresponding to the first measurement is determined based on the second mapping relationship between the normal offset and the second weight.

4. The method according to any one of claims 1-3, characterized in that, The step of optimizing the first measurement using the fusion weights to obtain the second measurement includes: The second measurement is obtained by multiplying the first measurement by the corresponding fusion weight.

5. The method according to claim 1, characterized in that, The step of optimizing the first measurement using the fusion weights to obtain the second measurement includes: The first measurements of each cell are aligned with their symbols to obtain at least one first set; The product of the first set and the weight vector is obtained to obtain the second set; wherein the weight vector includes the fusion weights corresponding to each of the first measurements, and the second set includes the second measurements corresponding to each cell.

6. The method according to claim 1, characterized in that, Multi-target recognition is performed based on the second measurement, and the perception result of each target object is determined, including: Cluster analysis is performed based on the three-dimensional approximation of each second measurement to obtain at least one sensing target; For any one of the sensing targets, the sensing result is determined based on the second measurement quantity corresponding to the sensing target.

7. The method according to claim 1, characterized in that, The perception result of any one of the target objects includes at least one of the following: target identifier, time information, location information, and beam control information; The location information includes at least one of the following: The first coordinate corresponding to the current moment, which includes: distance, velocity, and angle; The second coordinate and velocity at the current moment, where the second coordinate is a three-dimensional spatial coordinate.

8. The method according to claim 1, characterized in that, The method further includes: For any given sensing target, based on the historical sensing data of the sensing target and the sensing results, the predicted value of the sensing target at the next moment is determined; Based on the predicted value, beam control information is determined; wherein, the beam control information includes at least one of the following: beam identifier, beam normal offset; The beam control information is fed back to the beam control module.

9. The method according to claim 1, characterized in that, The acquisition of symbol-level data corresponding to the sensed echo signal includes: The sensed echo signal is processed to obtain the symbol-level data, which includes: the first measurement quantity and the beam information; And / or, Receive the symbol-level data reported by the sensing device.

10. The method according to any one of claims 1, characterized in that, The symbol-level data includes at least one of the following: Symbol-level data from at least one cell; Symbol-level data for at least one space beam; Symbol-level data in at least one channel environment.

11. A sensing device, characterized in that, include: The acquisition unit is used to acquire symbol-level data corresponding to the sensed echo signal to obtain the first measurement quantity and beam information. The first measured quantities include: distance, speed, and angle; The determining unit is used to determine the fusion weight corresponding to the first measurement quantity based on the beam information; An optimization unit is used to optimize the first measurement quantity using the fusion weights to obtain a second measurement quantity; The processing unit is used to perform multi-target recognition based on the second measurement and determine the perception result of each target object.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the method according to any one of claims 1-10.

13. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the method described in any one of claims 1-10.

14. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method described in any one of claims 1-10.

Images