WO2024199534A1 - 信息接收方法、装置、设备及存储介质 - Google Patents
信息接收方法、装置、设备及存储介质 Download PDFInfo
- Publication number
- WO2024199534A1 WO2024199534A1 PCT/CN2024/085391 CN2024085391W WO2024199534A1 WO 2024199534 A1 WO2024199534 A1 WO 2024199534A1 CN 2024085391 W CN2024085391 W CN 2024085391W WO 2024199534 A1 WO2024199534 A1 WO 2024199534A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- node
- information
- perception
- synaesthesia
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/06—Authentication
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
- H04W28/24—Negotiating SLA [Service Level Agreement]; Negotiating QoS [Quality of Service]
Definitions
- the present application relates to the field of wireless communication technology, and in particular to an information receiving method, device, equipment and storage medium.
- the integrated system of interawareness refers to the perception of the direction, distance, speed and other information of the target object while transmitting information in the same system through spectrum sharing, hardware sharing, signal sharing, etc., or the detection, tracking, identification, imaging, etc. of the target object, event or environment.
- the existing perception technology cannot meet the perception performance requirements. Therefore, it is urgent to find a technical solution that can meet the performance indicators of perception.
- embodiments of the present application hope to provide an information receiving method, apparatus, device and storage medium.
- the embodiment of the present application is intended to provide an information receiving method, which is applied to a first node, and the method includes:
- the sensing function includes at least one of the following functions:
- At least one of perceived capability reporting, negotiation, dimension reduction, and configuration wherein the perceived capability configuration is used by the node or provided by the node to other nodes;
- QoS quality of service
- the terminal or application triggers the corresponding modification process
- At least one of collecting, reporting, and processing sensory measurement data At least one of collecting, reporting, and processing sensory measurement data.
- the first message carries at least one of the following:
- the first information is used to indicate at least one of the following:
- the radio measurement parameters of the terminal of the measured object are set;
- the pattern of reflection points of the object under test, the corresponding power density and the receiving antenna gain is the pattern of reflection points of the object under test, the corresponding power density and the receiving antenna gain.
- the perception result information includes at least one of the following:
- the resource information used for sensing includes at least one of the following:
- the activation time range of the resource is the activation time range of the resource
- the send object that the resource can use is
- the training and/or reasoning of the synaesthesia perception model is located in at least one of the following:
- gNB Next-generation base stations
- CU Central Unit
- At least one network element node that controls an application function (AF) using artificial intelligence (AI);
- the method further includes:
- the second information includes at least one of the following:
- the ID of the terminal is the ID of the terminal.
- the second information is used for at least one of the following:
- Authorization information provided to the first node for verification and/or authentication of perception is provided to the first node for verification and/or authentication of perception.
- the type of the perception function includes at least one of the following:
- the perceived accuracy includes at least one of the following:
- performing model training according to the second information to obtain the synaesthesia perception model includes:
- the bandwidth of the current cell is greater than or equal to the first threshold, the channel quality of the current cell is greater than or equal to the second threshold, and the perception accuracy is less than or equal to the third threshold, the reflection points of the target object are collected according to the frequency m, and the training of the synaesthesia perception model of the specific function is performed using the collected reflection points;
- m is a rational number;
- the bandwidth of the current cell is less than or equal to the first threshold, the channel quality of the current cell is less than or equal to the second threshold, and the perception accuracy is greater than or equal to the third threshold, the weight of resource allocation between communication and perception is adjusted to perform model training to obtain the synaesthesia perception model; and/or, the bandwidth size used for perception is adjusted.
- the receiving a first message sent by the second node includes:
- the bandwidth of the current cell is less than or equal to the first threshold, the channel quality of the current cell is less than or equal to the second threshold, and the perceived accuracy is greater than or equal to the third threshold, sending the third message to the second node; the third message is used to request the second node to send the first message to the first node;
- the method further includes:
- the method further includes:
- the third information and the first information are sent to the second node so that the second node can perform model training to obtain the synaesthesia perception model.
- the third information includes at least one of the following:
- the method further includes:
- the synaesthesia perception model is used to perform perception prediction and obtain a prediction result.
- the method further includes:
- Feedback information is sent to a node adjacent to the first node or the second node; the feedback information carries a prediction result; the prediction result includes an updated state of the perceived object and/or an update method of the synaesthesia perception model.
- At least one embodiment of the present application provides an information receiving device, including:
- a receiving unit configured to receive a first message sent by a second node; wherein the first message is used to request the first node to provide first information to the second node; and the first information is used by the second node to perform a perception function;
- the sensing function includes at least one of the following functions:
- At least one of perceived capability reporting, negotiation, dimension reduction, and configuration wherein the perceived capability configuration is used by the node or provided by the node to other nodes;
- QoS quality of service
- the terminal or application triggers the corresponding modification process
- At least one of collecting, reporting, and processing sensory measurement data At least one of collecting, reporting, and processing sensory measurement data.
- At least one embodiment of the present application provides a network device, including a processor and a memory for storing a computer program that can be run on the processor.
- the processor when used to run the computer program, it executes the steps of any one of the methods described above on the network device side.
- At least one embodiment of the present application provides a storage medium having a computer program stored thereon, wherein the computer program implements the steps of any of the above methods when executed by a processor.
- a first node receives a first message sent by a second node; wherein the first message is used to request the first node to provide the second node with first information; the first information is used by the second node to perform a perception function; wherein the perception function includes at least one of the following functions: training a synaesthesia perception model, creating a new synaesthesia perception model, adding a synaesthesia perception model, updating a synaesthesia perception model, and deleting at least one of a synaesthesia perception model; wherein the synaesthesia perception model is used by this node or provided by this node to other nodes; performing Model training, to obtain a synaesthesia perception model, for use by this node or provided by this node to other nodes; to infer the perception result information of the measured object, for use by this node or provided by this node to other nodes; to determine the resource information used for perception, for use by this node or provided by this node to other no
- the first node receives a first message sent by the second node for requesting to provide first information, and subsequently the second node uses the first information to perform a perception function, and the second node's performance of the perception function is related to the synaesthesia perception model.
- the second node uses the content in the first information to optimize the existing perception function and adds the following synaesthesia perception function: determining perceived resource information, performing perceived authorization information verification and/or authentication, performing at least one of perceived capability reporting, negotiation, dimensionality reduction, and configuration, and performing perceived QoS parameter conversion, so as to meet the perception performance requirements.
- FIG1 is a schematic diagram of an implementation flow of an information receiving method according to an embodiment of the present application.
- FIG2 is a schematic diagram of a specific implementation flow of the information receiving method according to an embodiment of the present application.
- FIG3 is a schematic diagram of part of the information included in the first information according to an embodiment of the present application.
- FIG4 is a schematic diagram of a communication perception model obtained by performing model training based on input information in an embodiment of the present application
- FIG5 is a schematic diagram of the structure of an information receiving device according to an embodiment of the present application.
- FIG. 6 is a schematic diagram of the composition structure of a network device according to an embodiment of the present application.
- the integrated synaesthesia system refers to the perception of the target object's direction, distance, speed and other information while transmitting information in the same system through spectrum sharing, hardware sharing, signal sharing and other means, or the detection, tracking, identification and imaging of the target object, event or environment.
- the integration of communication and perception realizes the integration and symbiosis of communication and perception functions in the same spectrum resources through the joint design of air interface and protocol, sharing of software and hardware equipment, so that the wireless network can obtain the perception of target objects or environmental information by analyzing the direct, reflected and scattered wireless communication signals while performing data communication, and realize positioning, ranging, speed measurement, imaging, detection, identification, environmental reconstruction and other functions, bringing a new dimension to improving spectrum utilization and equipment reuse rate.
- the orthogonal frequency division multiplexing (OFDM) signal used by 5G has excellent pulse pressure characteristics and autocorrelation, and is a signal form suitable for the integrated perception and communication system.
- Table 1 is a schematic diagram of the important indicators included in the perception performance.
- the important indicators included in the perception performance include measurement accuracy and resolution; the measurement accuracy may include distance accuracy and angle accuracy; the resolution may include distance resolution and angle resolution.
- the measurement accuracy refers to the accuracy of measuring a single target parameter, which is related to many factors, such as the perfection of the network system itself, radio wave propagation, etc.
- Resolution refers to the ability of the network system to distinguish a specific target from two or more targets, which is related to the characteristics of the network system itself, the characteristics of the target, etc.
- radar perception works in the high frequency band, that is, between 300M and 300GHz, but there is no commercial frequency band for millimeter waves in the current communication system, which is lower than the frequency band of radar. If used in the current low frequency band, for example, 4.9GHz, the accuracy and resolution cannot meet the performance indicators of perception.
- the use of technologies such as oversampling MUSIC algorithm can increase the accuracy level by 2-3 times, but considering the large amount of clutter interference, moving object interference and target scattering ghosting in the low-frequency channel environment, the false alarm probability will increase and the perception performance indicators cannot be met; the false alarm probability refers to the probability that the system judges that the target appears when the target does not exist. Therefore, it is urgent to design a technical solution that uses AI models to improve perception performance indicators.
- formula (1) can be understood as the upper bound of the angle estimation accuracy, that is, the upper limit of the estimation accuracy that can be achieved by using various angle estimation algorithms under the premise of a given antenna array and signal-to-noise ratio.
- a first node receives a first message sent by a second node; wherein the first message is used to request the first node to provide first information to the second node; the first information is used by the second node to perform a perception function; wherein the perception function includes at least one of the following functions: training a synaesthesia perception model, creating a new synaesthesia perception model, adding a synaesthesia perception model, updating a synaesthesia perception model, and deleting at least one of a synaesthesia perception model; wherein the synaesthesia perception model is used by this node or provided by this node to other nodes; model training is performed to obtain a synaesthesia perception model for use by this node or provided by this node to other nodes; perception result information of the object under test is inferred for use by this node or provided by this node to other nodes; resource information for perception is determined for use by this node or provided by this node to other nodes; authorization information verification and/or authentication of perception is performed
- LTE Long Term Evolution
- LTE-A Long Term Evolution-Advanced
- CDMA Code Division Multiple Access
- TDMA Time Division Multiple Access
- FDMA Frequency Division Multiple Access
- OFDMA Orthogonal Frequency Division Multiple Access
- SC-FDMA Single-carrier Frequency-Division Multiple Access
- NR New Radio
- 6G 6th Generation
- the communication system applied in the embodiments of the present application may include network devices, network nodes and terminal devices (also referred to as terminals, communication terminals, etc.); the second node, the third node and the first node in the present application may be network devices, that is, devices that communicate with terminal devices; it may also provide communication coverage within a certain area, and can communicate with terminals located in the area; for example, it may be a base station in each communication system, such as an evolved base station (eNB, Evolutional Node B) in an LTE system, or a base station in a 5G system, NR system or 6G system.
- eNB evolved base station
- LTE Long Term Evolutional Node B
- the first network node may be one or more base stations, transmitting points, receiving points, central units, distribution units, BBUs, RRUs, relays, IABs, smart metasurfaces, communication balloons, flying aircraft base stations, antennas, satellite base stations, terminals using sidelink communication, etc.; it may also be a terminal, or a module with other specific communication functions; it may also be a base station that provides Application servers with perception functions and/or AI functions, etc.; it can also be core network nodes, such as UPF, AF, MME, AMF, devices integrated with the core network, or other network elements used to establish connections.
- core network nodes such as UPF, AF, MME, AMF, devices integrated with the core network, or other network elements used to establish connections.
- service in this document may refer to at least one of the following concepts: service, PDU session, Quality of Service (QoS) flow, stream or service data flow, radio bearer, or logical channel.
- QoS Quality of Service
- the "data” mentioned in this document can be one or more of “data packet”, “Physical Uplink Share Channel transmission (PUSCH)", “Physical Downlink Share Channel transmission (PDSCH)", “data unit”, “PDU set packet”, “sample”, “slice”, “tile”, “stream”, “transmission” and “transmission block”.
- FIG. 1 is a schematic diagram of an implementation flow of an information receiving method according to an embodiment of the present application, which is applied to a first node. As shown in FIG. 1 , the method includes step 101:
- Step 101 receiving a first message sent by a second node; wherein the first message is used to request the first node to provide first information to the second node; and the first information is used by the second node to perform a sensing function;
- the sensing function includes at least one of the following functions:
- At least one of perceived capability reporting, negotiation, dimension reduction, and configuration wherein the perceived capability configuration is used by the node or provided by the node to other nodes;
- the terminal or application triggers the corresponding modification process
- At least one of collecting, reporting, and processing sensory measurement data At least one of collecting, reporting, and processing sensory measurement data.
- information such as the sensing service type or identifier, QoS requirements, and sensing measurement data reporting period are transmitted to other nodes that perform sensing, such as a terminal or a base station.
- the terminal or application needs to trigger the corresponding modification process.
- the measurement node refers to a node used to sense the object under test.
- the second node may be at least one of a sensing function network element SF (sensing Function), an adjacent base station, a CU, other terminals, and a core network element.
- SF sensing Function
- SF is mainly responsible for sensing control and sensing measurement data calculation.
- the first message carries at least one of the following:
- the first information is used to indicate at least one of the following:
- the radio measurement parameters of the terminal of the measured object are set;
- the first message may refer to a request message or the like.
- the resource information determined for perception may refer to the resource information used to clarify the resource information occupied by perception, because the first node needs to coordinate the two functional operations of communication and perception through at least one separation method in the time domain/frequency domain/code domain/spatial domain, so it is necessary to clarify the resource information occupied by the reasonable perception function that is more matched with the user's business needs within a certain resource range.
- the specific resource information occupied by perception includes at least one of the following methods: waveform selection, synaesthesia signal selection, perception signal frame structure design, etc. For example, whether to select a pulse waveform (PW), a linear frequency modulation wave (FMCW) or a communication waveform including an orthogonal frequency division multiplexing (OFDM) waveform.
- PW pulse waveform
- FMCW linear frequency modulation wave
- OFDM orthogonal frequency division multiplexing
- the time domain resource depends on at least one factor such as speed range, speed resolution, detection distance, frame rate, networking interference, etc.
- the frequency domain overhead depends on at least one factor such as distance resolution and networking interference.
- using the first information to determine the perceived resource information may specifically include:
- CP cyclic prefix
- the frame structure with a larger subcarrier interval should be used as much as possible; the lower the Doppler change rate of the terminal, the frame structure with a smaller subcarrier interval should be used as much as possible; this will also affect the length of the CP selection;
- resources may include:
- the activation time range of the resource is the activation time range of the resource
- the send object that the resource can use is
- the authorization information verification and/or authentication for perception can be performed based on the first information.
- the authorization information verification and/or authentication for perception can refer to determining whether the perceived object and/or the object applying for the perception service is legal and/or has reasonable authority; if it is legal, or has the corresponding authority, this node provides the corresponding communication and/or perception service, otherwise, refuses to provide communication and/or perception service. That is, determine whether the applicant has the authority to obtain the functions corresponding to "the power density at the object to be measured, the Doppler change rate of the terminal equipped with the object to be measured, and the distance and angle between the two measurement nodes". Specifically, it can be determined whether the applicant has the authority based on the type of perception function, the accuracy of perception, and the type of object to be perceived.
- the perception capability can be reported, negotiated, reduced in dimension and configured according to the first information.
- the reporting of perception capability can be divided into more reasonable algorithms and/or generate more efficient models in order to make more reasonable and efficient judgments on the sharing of resources and processing capabilities between related nodes in the process of artificial intelligence, resource negotiation and/or computing power sharing.
- the information such as "the power density at the object to be measured, the Doppler change rate of the terminal with the object to be measured, the distance and angle between the two measurement nodes" in the first information, the matching and/or reasonable perception capability is fed back. If the perception capability is insufficient, you can apply for capability downgrade (i.e., dimension reduction) and/or capability negotiation.
- the perceived QoS parameters can be converted according to the first information.
- the perceived QoS parameters come from at least one of the application layer, the requester of the perceived service, the core network, and the neighboring node.
- the perceived QoS parameter conversion can better adapt the node capabilities. For example, the higher the radio measurement accuracy requirement corresponding to the radio measurement parameters of the terminal of the measured object, the higher the Doppler change rate of the terminal of the measured object. The higher the requirement, the higher the perceived QoS is required; anyway, when the first information shows that the radio measurement accuracy requirement corresponding to the radio measurement parameters of the terminal of the measured object becomes lower, the Doppler change rate of the terminal of the measured object becomes lower, etc., the QoS is reduced.
- the first information is used for at least one of the following:
- the perception result information of the object under test is calculated and provided to the first node.
- the synaesthesia perception model may refer to an artificial intelligence (AI, Artificial Intelligence) or machine learning (ML, Machine Learning) model with communication and perception functions.
- AI Artificial Intelligence
- ML Machine Learning
- the first node may provide the first information to the second node, so that the second node may use the first information for training to obtain the synaesthesia perception model.
- the perception result information includes at least one of the following:
- the resource information used for sensing includes at least one of the following:
- the activation time range of the resource is the activation time range of the resource
- the send object that the resource can use is
- the activation time point of the resource is related to the distribution density of the sensing objects and/or the request for the sensing function
- the activation time range of the resource is related to the distribution density of the sensing objects and/or the request for the sensing function
- the geographical scope of resource availability and the distribution density of sensing objects are related to the transmission capacity of the base station
- the altitude range in which resources can be used and the distribution density of the sensing objects are related to the transmission capacity of the base station
- the area in which resources can be used is related to the distribution density of the sensing objects and/or the request for the sensing function
- the activation time point may also be the effective time point.
- the regional scope may refer to at least one combination of geographical scope, national scope, regional scope, base station scope, cell scope, public land mobile network (PLMN, Public Land Mobile Network) scope, and cell type scope.
- PLMN Public Land Mobile Network
- the training and/or reasoning of the synaesthesia perception model is at least one of the following:
- At least one network element node of AI-controlled AF At least one network element node of AI-controlled AF
- the first node itself can also perform model training to obtain the synaesthesia perception model.
- the method further includes:
- the second information includes at least one of the following:
- the ID of the terminal is the ID of the terminal.
- the second information is used for at least one of the following:
- Authorization information provided to the first node for verification and/or authentication of perception is provided to the first node for verification and/or authentication of perception.
- the second information represents perceived demand information.
- the type of the sensing function includes at least one of the following:
- the perceived accuracy includes at least one of the following:
- performing model training according to the second information to obtain the synaesthesia perception model includes:
- the bandwidth of the current cell is greater than or equal to the first threshold, the channel quality of the current cell is greater than or equal to the second threshold, and the perception accuracy is less than or equal to the third threshold, then according to the frequency m
- the reflection points of the target object are collected, and the collected reflection points are used to perform the training of the synaesthesia perception model of specific functions;
- m is a rational number;
- the bandwidth of the current cell is less than or equal to the first threshold, the channel quality of the current cell is less than or equal to the second threshold, and the perception accuracy is greater than or equal to the third threshold, the weight of resource allocation between communication and perception is adjusted to perform model training to obtain the synaesthesia perception model; and/or, the bandwidth size used for perception is adjusted.
- the bandwidth of the current cell is large enough, the environment is simple, the interference is small, and/or the perception accuracy value is low, the perception accuracy requirement can be met in this case, and the reflection points of the target object are collected according to the frequency m, and perception of specific functions is performed.
- the weight of resource allocation between communication and perception is adjusted to perform model training to obtain the synaesthesia perception model; and/or, the bandwidth size used for perception is adjusted.
- an oversampling algorithm is started to increase the sampling density to increase the training samples so as to perform model training to obtain the synaesthesia perception model.
- the first node can judge whether it is necessary to meet higher perception accuracy requirements at present based on the second information sent by the third node.
- the second node performs model training to obtain the synaesthesia perception model to improve the perception accuracy.
- the receiving a first message sent by the second node includes:
- the bandwidth of the current cell is less than or equal to the first threshold, the channel quality of the current cell is less than or equal to the second threshold, and the perceived accuracy is greater than or equal to the third threshold, sending a third message to the second node; the third message is used to request the second node to send the first message to the first node;
- the method further comprises:
- the first information is sent to the second node.
- the terminal may refer to a UE provided with the object to be measured.
- the second node trains a synaesthesia perception model using the first information.
- the method further comprises:
- the third information and the first information are sent to the second node so that the second node can perform model training to obtain the synaesthesia perception model.
- the third information includes at least one of the following:
- the triggering method of the model training includes at least one of the following:
- the type of model training includes at least one of supervised learning, unsupervised learning, semi-supervised learning, federated learning, online training, and offline training.
- the training method includes unilateral training, bilateral training, and multilateral training.
- the specific parameters transmitted by the model include at least one of model generation, model reasoning, model monitoring, and model update, and also include the number of model layers, at least one of the KPIs used for model performance monitoring, and the granularity of model deployment; wherein the granularity of model deployment includes at least one of scenarios, services, configurations, locations, base stations, cells, demands, etc., UEs, UE groups, etc.
- the methods of transmitting model parameters include RRC signaling (for example, creating a new SRBx and defining its corresponding security and integrity protection mechanisms), NAS signaling, and UP transmission (for example, through DRB transmission, that is, a DRB without a corresponding GTP-U tunnel, through the payload of GTP-U data, or the packet header of GTP-data).
- RRC signaling for example, creating a new SRBx and defining its corresponding security and integrity protection mechanisms
- NAS signaling for example, a DRB transmission, that is, a DRB without a corresponding GTP-U tunnel, through the payload of GTP-U data, or the packet header of GTP-data.
- the first node can also collect application data source data; the application data source data includes video, image, location information, etc. of the measured object, and the data can be used as auxiliary input data for model training, so that the second node can combine the first information and the third information to perform model training to obtain the synaesthesia perception model.
- the first node can also use the first information to perform model training to obtain the synaesthesia perception model; or, it can also use the first information and the third information to perform model training to obtain the synaesthesia perception model; or, it can also use the first information, the third information and the application data source data to perform model training to obtain the synaesthesia perception model.
- the method further comprises:
- the synaesthesia perception model is used to perform perception prediction and obtain a prediction result.
- the method further comprises:
- Feedback information is sent to a node adjacent to the first node or the second node; the feedback information carries a prediction result; the prediction result includes an updated state of the perceived object and/or an update method of the synaesthesia perception model.
- the first node receives a first message sent by the second node for requesting to provide first information, and the second node subsequently performs a perception function using the first information.
- the performance of the perception function by the second node is related to a synaesthesia perception model.
- the performance of the synaesthesia perception function by the second node can meet the perception performance indicators.
- FIG. 2 is a schematic diagram of a specific implementation flow of the information receiving method according to an embodiment of the present application. As shown in FIG. 2 , the method includes steps 201 to 206:
- Step 201 The third node (synaesthesia AF) sends a second message, namely, a synaesthesia perception open request message, to the first node (base station node 1 in the RAN set); the second message includes second information; the second message is used to request the first node (base station node 1 in the RAN set) to train a synaesthesia perception model.
- a second message namely, a synaesthesia perception open request message
- the synaesthesia perception model may refer to an AI or ML model with communication and perception functions.
- the second information includes at least one of the following:
- the second information is used for at least one of the following:
- Authorization information provided to the first node for verification and/or authentication of perception is provided to the first node for verification and/or authentication of perception.
- the type of the perception function may include at least one of the following:
- the perceived accuracy may include at least one of the following:
- Range resolution (determined by bandwidth
- the perceived object type may refer to vehicles, animals, birds, etc.
- Step 202 The first node (base station node 1 in the RAN set) performs model training based on the second information to obtain the synaesthesia perception model.
- performing model training according to the second information to obtain the synaesthesia perception model includes at least one of the following:
- the bandwidth of the current cell is greater than or equal to the first threshold, the channel quality of the current cell is greater than or equal to the second threshold, and the perception accuracy is less than or equal to the third threshold, then according to the frequency m Collect the reflection points of the target object, and use the collected reflection points to perform the training of the synaesthesia perception model of specific functions; m is a natural number greater than 0;
- the bandwidth of the current cell is less than or equal to the first threshold, the channel quality of the current cell is less than or equal to the second threshold, and the perception accuracy is greater than or equal to the third threshold, the weight of resource allocation between the communication function and the perception function is adjusted to perform model training to obtain the synaesthesia perception model; and/or, the bandwidth size used for perception is adjusted.
- the bandwidth of the current cell is large enough, the environment is simple, the interference is small, and/or the perception accuracy value is low, the perception accuracy requirement can be met in this case, and the reflection points of the target object are collected according to the frequency m, and perception of specific functions is performed.
- the weight of resource allocation between communication and perception is adjusted to perform model training to obtain the synaesthesia perception model; and/or, the bandwidth size used for perception is adjusted.
- an oversampling algorithm is started to increase the sampling density to increase the training samples so as to perform model training to obtain the synaesthesia perception model.
- step 203 is executed.
- Step 203 The first node (the first base station in the RAN set) sends a third message to the second node (AI controlled AF), i.e., a synaesthesia AI model request message; the third message is used to request the second node (AI controlled AF) to train the synaesthesia model and provide it to the first node.
- AI controlled AF i.e., a synaesthesia AI model request message
- the third message carries at least one of the following:
- Step 204 The second node (AI controlled AF) receives the third message sent by the first node (base station node 1 in the RAN set), and sends a first message to the first node (base station node 1 in the RAN set); the first message is used to request the first node (base station node 1 in the RAN set) to provide first information to the second node (AI controlled AF); the first information represents the open information required for the second node (AI controlled AF) to train the synaesthesia perception model.
- the first message carries at least one of the following:
- the first information includes at least one of the following:
- the Doppler change rate of the UE with the measured object is set;
- the number of reflection points of the object being measured is the number of reflection points of the object being measured.
- the pattern of reflection points of the object under test, the corresponding power density and the receiving antenna gain is the pattern of reflection points of the object under test, the corresponding power density and the receiving antenna gain.
- the incident angle and distance from the target object to the first node refer to the incident angle from the target object to the base station and the distance from the target object to the base station; the distance is derived according to the echo delay from the target object to the base station;
- the radio measurement of the UE provided with the measured object may include RSRP, RSRQ, and SINR; the corresponding sensing function types are positioning and ranging.
- the Doppler change rate of the UE of the object to be measured is set, and the corresponding sensing function type is imaging, detection, and object recognition.
- the power density at the object being measured corresponds to the type of sensing function: imaging, detection, and object recognition.
- the number of reflection points of the measured object corresponds to the type of sensing function, which is detection and object recognition.
- the spacing of the reflection points of the measured object corresponds to the type of sensing function, detection and object recognition.
- the pattern of reflection points of the measured object is detection and object recognition.
- FIG3 is a schematic diagram of part of the information included in the first information of an embodiment of the present application.
- TX1 represents the sending unit of the measurement node 1
- TX2 represents the sending unit of the measurement node 2
- the distance between the two measurement nodes is 4d.
- the second node trains the prediction model using the first information provided by the first node (base station node 1 in the RAN set) to obtain a synaesthesia perception model with communication and perception functions.
- the second node can configure one or more synaesthesia perception models for communication perception, i.e., AI/ML models, for multiple base stations.
- the first node (base station node 1 in the RAN set) can synchronize and interact with the second network device (base station node 2 in the RAN set) at the Xn port through signaling to use an AI/ML model for communication perception.
- the model can provide useful input information to (base station node 1 in the RAN set), such as the predicted state of the target object being measured, as well as the model update method, whether it is online update or offline update.
- the first node (base station node 1 in the RAN set) can also transmit the AI/ML model for model perception over the air interface based on the capabilities of the UE.
- Step 205 The first node (base station node 1 in the RAN set) sends a fourth message to the terminal; the fourth message is used to request to obtain the first information; the first information represents the open information required for training the synaesthesia perception model; the first information returned by the terminal is received, and the first information is sent to the second node (AI controlled AF).
- the fourth message is used to request to obtain the first information
- the first information represents the open information required for training the synaesthesia perception model
- the first information returned by the terminal is received, and the first information is sent to the second node (AI controlled AF).
- the terminal may refer to a UE provided with the object to be measured, and a communication module is provided in the UE.
- the first node (base station node 1 in the RAN set) collects the first information fed back or reflected by the terminal (UE of the object under test).
- the first information includes at least one of the following:
- the Doppler change rate of the UE with the measured object is set;
- the first node (base station node 1 in the RAN set) can also collect application data source data;
- the application data source data includes video, image, location information, etc. of the object being measured, and the data can be used as Auxiliary input data for model training.
- the first node (base station node 1 in the RAN set) sends a fifth message to an adjacent node (base station node 2 in the RAN set); the fifth message is used to request to obtain the third information.
- the third information includes at least one of the following:
- the first node receives the third information required for training the communication perception model returned by the adjacent node (base station node 2 in the RAN set), and sends the third information to the second node (AI controlled AF).
- the second node uses the first information and the third information as input information for model training, and performs model training to obtain the synaesthesia perception model.
- the training of the AI/ML model is located at the second node (AI controls AF).
- the required measurements are used to train the AI/ML model.
- the training of the AI/ML model can also be located at the first node (base station node 1 in the RAN set).
- Step 206 The second node (AI controlled AF) trains a synaesthesia perception model according to the first information, performs perception prediction using the trained synaesthesia perception model, and obtains a prediction result; and returns the prediction result to the first node (base station node 1 in the RAN set).
- AI controlled AF AI controlled AF
- the training of the AI/ML model is located at the first node (base station node 1 in the RAN set), and the first node (base station node 1 in the RAN set) can also perform perception prediction to obtain a prediction result. For example, for the cell of the first node (base station node 1 in the RAN set), perception prediction is performed to obtain a prediction result.
- Figure 4 is a schematic diagram of an embodiment of the present application in which a communication perception model is obtained by model training based on input information.
- the training of the AI/ML model is located at the first node (base station node 1 in the RAN set), and the input information includes the first information, the third information and the application data source data.
- the first node (base station node 1 in the RAN set) can use the first information obtained from the local node as input data based on the AI/ML model:
- the first information includes at least one of the following:
- the incident angle and distance from the target object to the base station (derived from the echo delay from the target to the base station);
- UE radio measurements of the object under test e.g. RSRP, RSRQ, SINR
- the Doppler change rate of the UE of the measured object is the Doppler change rate of the UE of the measured object
- Application data source data from the application layer including video, images, location information, etc. of objects;
- the third information sent by the base station node 2 in the neighboring cell is input information related to the object under test.
- the first node (base station node 1 in the RAN set) can use the following information as output data based on the AI/ML model:
- UE location information of the object being measured e.g. coordinates, serving cell ID, moving speed
- the first node deactivates the current AI/ML model and activates the second type of AI/ML model.
- the second type of AI/ML model activation synchronization information is sent to the adjacent base station.
- the second node (AI controls AF) makes a perception decision based on the prediction result; and sends feedback information to a network device (base station node 1 in the RAN set); the feedback information carries Describe the prediction results.
- the second network device can also send feedback information to the second node (AI controlled AF); the feedback information includes the target state update after the perception, etc.
- the second network device may also send feedback information to the first node (base station node 1 in the RAN set); the feedback information includes the target state update after the perception, etc.
- the first node (base station node 1 in the RAN set) receives a first message sent by the second node (AI controlled AF) for requesting to provide first information.
- the first information can be subsequently sent to the second node (AI controlled AF) so that the second node (AI controlled AF) can use the first information to perform model training to obtain a synaesthesia perception model.
- the synaesthesia perception model can be used to meet the perception performance indicators.
- FIG5 is a schematic diagram of the composition structure of the information receiving device of the embodiment of the present application. As shown in FIG5, the device includes:
- a receiving unit 51 is configured to receive a first message sent by a second node; wherein the first message is used to request the first node to provide first information to the second node; and the first information is used by the second node to perform a sensing function;
- the sensing function includes at least one of the following functions:
- At least one of perceived capability reporting, negotiation, dimension reduction, and configuration wherein the perceived capability configuration is used by the node or provided by the node to other nodes;
- the terminal or application triggers the corresponding modification process
- At least one of collecting, reporting, and processing sensory measurement data At least one of collecting, reporting, and processing sensory measurement data.
- the first message carries at least one of the following:
- the first information is used to indicate at least one of the following:
- the radio measurement parameters of the terminal of the measured object are set;
- the pattern of reflection points of the object under test, the corresponding power density and the receiving antenna gain is the pattern of reflection points of the object under test, the corresponding power density and the receiving antenna gain.
- the first information is used for at least one of the following:
- the perception result information of the object under test is inferred and provided to the first node.
- the perception result information includes at least one of the following:
- the resource information used for sensing includes at least one of the following:
- the activation time range of the resource is the activation time range of the resource
- the training and/or reasoning of the synaesthesia perception model is at least one of the following:
- gNB Next-generation base stations
- CU Central Unit
- At least one network element node that controls an application function (AF) using artificial intelligence (AI);
- the device is further used for:
- the second information includes at least one of the following:
- the ID of the terminal is the ID of the terminal.
- the second information is used for at least one of the following:
- Authorization information provided to the first node for verification and/or authentication of perception is provided to the first node for verification and/or authentication of perception.
- the type of the sensing function includes at least one of the following:
- the perceived accuracy includes at least one of the following:
- the device is further used for:
- the bandwidth of the current cell is greater than or equal to the first threshold, the channel quality of the current cell is greater than or equal to the second threshold, and the perception accuracy is less than or equal to the third threshold, the reflection points of the target object are collected according to the frequency m, and the training of the synaesthesia perception model of the specific function is performed using the collected reflection points;
- m is a rational number;
- the bandwidth of the current cell is less than or equal to the first threshold, the channel quality of the current cell is less than or equal to the second threshold, and the perception accuracy is greater than or equal to the third threshold, the weight of resource allocation between communication and perception is adjusted to perform model training to obtain the synaesthesia perception model; and/or, the bandwidth size used for perception is adjusted.
- the receiving unit 51 is used to:
- the bandwidth of the current cell is less than or equal to the first threshold, the channel quality of the current cell is less than or equal to the second threshold, and the perceived accuracy is greater than or equal to the third threshold, sending the third message to the second node; the third message is used to request the second node to send the first message to the first node;
- the device is further used for:
- the first information is sent to the second node so that the second node can perform model training to obtain the synaesthesia perception model.
- the device is further used for:
- the third information and the first information are sent to the second node so that the second node can perform model training to obtain the synaesthesia perception model.
- the third information includes at least one of the following:
- the device is further used for:
- the synaesthesia perception model is used to perform perception prediction and obtain a prediction result.
- the device is further used for:
- Feedback information is sent to a node adjacent to the first node or the second node; the feedback information carries a prediction result; the prediction result includes an updated state of the perceived object and/or an update method of the synaesthesia perception model.
- the receiving unit 51 can be implemented by a communication interface in an information receiving device.
- the information receiving device provided in the above embodiment only uses the division of the above program modules as an example to illustrate when receiving information.
- the above processing can be assigned to different program modules as needed, that is, the internal structure of the device is divided into different program modules to complete all or part of the processing described above.
- the information receiving device provided in the above embodiment and the information receiving method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.
- the embodiment of the present application further provides a network device, as shown in FIG6 , including:
- Communication interface 61 capable of exchanging information with other devices
- the processor 62 is connected to the communication interface 61 and is used to execute the method provided by one or more technical solutions of the network device side when running the computer program.
- the computer program is stored in the memory 63.
- bus system 64 the various components in the network device 60 are coupled together through the bus system 64. It is understandable that the bus system 64 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 64 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, various buses are marked as the bus system 64 in FIG. 6.
- the memory 63 in the embodiment of the present application is used to store various types of data to support the operation of the network device 60. Examples of such data include: any computer program used to operate on the network device 60.
- the method disclosed in the above embodiment of the present application can be applied to the processor 62, or implemented by the processor 62.
- the processor 62 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 62 or the instruction in the form of software.
- the above-mentioned processor 62 may be a general-purpose processor, a digital data processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.
- the processor 62 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiment of the present application.
- the general-purpose processor may be a microprocessor or any conventional processor, etc.
- the steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to execute, or a combination of hardware and software modules in the decoding processor to execute.
- the software module can be located in a storage medium, which is located in the memory 63.
- the processor 62 reads the information in the memory 63 and completes the steps of the above method in combination with its hardware.
- the network device 60 can be implemented by one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to execute the aforementioned method.
- ASICs application specific integrated circuits
- DSPs digital signal processors
- PLDs programmable logic devices
- CPLDs complex programmable logic devices
- FPGAs field programmable gate arrays
- general-purpose processors controllers, microcontrollers (MCUs), microprocessors, or other electronic components to execute the aforementioned method.
- the memory (memory 63) of the embodiment of the present application can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories.
- the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), Magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disk, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk memory or tape memory.
- Volatile memory can be random access memory (RAM), which is used as an external cache.
- RAM random access memory
- SRAM static random access memory
- SSRAM synchronous static random access memory
- DRAM dynamic random access memory
- SDRAM synchronous dynamic random access memory
- DDRSDRAM double data rate synchronous dynamic random access memory
- ESDRAM enhanced synchronous dynamic random access memory
- SLDRAM synchronous link dynamic random access memory
- DRRAM direct memory bus random access memory
- the embodiment of the present application further provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, for example, a memory storing a computer program, and the computer program can be executed by the processor 62 of the network device 60 to complete the steps described in the aforementioned network device side method.
- the computer-readable storage medium can be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface storage, optical disk, or CD-ROM.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Computer Security & Cryptography (AREA)
- Mobile Radio Communication Systems (AREA)
- Communication Control (AREA)
Abstract
本申请公开了一种信息接收方法、装置、设备及存储介质。其中,所述方法应用于第一节点;所述方法包括:接收第二节点发送的第一消息;第一消息用于请求所述第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功能。
Description
相关申请的交叉引用
本申请基于申请号为2023103425549、申请日为2023年03月31日的中国专利申请提出,并要求该中国专利申请的优先权,该中国专利申请的全部内容在此以全文引入的方式引入本申请。
本申请涉及无线通信技术领域,尤其涉及一种信息接收方法、装置、设备及存储介质。
目前,通信感知一体化是第六代(6G,Sixth Generation)的标志特征之一,通感一体化系统是指在同一个系统中通过频谱共享、硬件共享、信号共享等方式,在进行信息传递的同时感知目标物体的方位、距离、速度等信息,或者,对目标物体、事件或环境等进行检测、跟踪、识别、成像等。目前,考虑到低频信道环境中的大量杂波干扰、移动物体干扰和目标散射重影等网络环境,现有的感知技术无法满足感知性能要求。因此,亟需找到一种能够满足感知的性能指标的技术方案。
发明内容
有鉴于此,本申请实施例期望提供一种信息接收方法、装置、设备及存储介质。
本申请实施例的技术方案是这样实现的:
本申请实施例期望提供一种信息接收方法,应用于第一节点,所述方法包括:
接收第二节点发送的第一消息;其中,所述第一消息用于请求所述第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功
能;
其中,所述感知功能包括以下功能中至少之一:
训练通感感知模型、新建通感感知模型、增加通感感知模型、更新通感感知模型、删除通感感知模型中至少之一;其中,所述通感感知模型由本节点使用或由本节点提供给其他节点;
进行模型训练,得到通感感知模型,以供本节点使用或由本节点提供给其他节点;
推算出被测物体的感知结果信息,以供本节点使用或由本节点提供给其他节点;
确定用于感知的资源信息,以供本节点使用或由本节点提供给其他节点;
进行感知的授权信息验证和/或鉴权;其中,所述授权和/或鉴权信息由本节点使用或由本节点提供给其他节点;
感知的能力上报、协商、降维和配置中的至少一个,其中,感知的能力配置由本节点使用或由本节点提供给其他节点;
选择感知的感知模式,以供本节点使用或由本节点提供给其他节点;
选择感知节点,以供本节点使用或由本节点提供给其他节点;
进行感知服务质量(QoS,Quality of Service)参数转换;其中,所述感知QoS参数由本节点使用或由本节点提供给其他节点;
将感知业务类型或标识、QoS要求和感知测量数据上报周期信息中至少之一传递到执行感知的其他节点;
若感知需求改变,终端或应用触发对应的修改流程;
触发、激活、新建感知业务中的至少一种;
终止、去激活、删除感知业务中的至少一种;
根据感知业务需求控制其他节点执行感知;
收集、上报、处理感知测量数据中的至少一种。
此外,根据本申请的至少一个实施例,所述第一消息,携带有以下至少之一:
所述第一信息的类型;
所述第一信息的发送周期;
所述第一信息的发送数量;
触发所述第一信息的事件信息;
所述第一信息用于指示以下至少之一:
两个测量节点之间的距离和角度;
被测物体到所述第一节点的入射角度、距离;
设置有被测物体的终端的无线电测量参数;
设置有被测物体的终端的多普勒变化率;
被测物体处的功率密度;
被测物体的反射点个数;
被测物体的反射点的间距;
被测物体的反射点的图样、对应的功率密度和接收天线增益。
此外,根据本申请的至少一个实施例,所述感知结果信息,包括以下至少之一:
目标对象的定位;
测距;
测速;
成像;
检测;
物体识别;
虚拟环境重构。
此外,根据本申请的至少一个实施例,所述用于感知的资源信息,包括以下至少之一:
周期、起始位置、持续时间或者每个周期的持续时长中至少一个;
资源的激活时间点;
资源的激活时间范围;
资源可使用的地理范围;
资源可使用的高度范围;
资源可使用的频段范围;
资源可使用的区域范围;
资源可使用的频段范围;
资源可使用的发送对象。
此外,根据本申请的至少一个实施例,所述通感感知模型的训练和/或推理位于以下至少之一:
操作维护管理(OAM,Operation and Maintenance management);
下一代基站(gNB);
无线网络节点;
gNB的中心单元(CU,Central Unit);
人工智能(AI,Artificial Intelligence)控制应用功能(AF,Application Function)的至少一种网元节点;
终端侧的AI服务器;
终端侧。
此外,根据本申请的至少一个实施例,所述方法还包括:
接收第三节点发送的第二消息;其中,所述第二消息包括第二信息;
根据所述第二信息,进行模型训练得到所述通感感知模型;
其中,所述第二信息,包括以下至少之一:
感知功能的类型;
感知的精度;
感知的物体类型;
AF的标识(ID);
终端的ID;
当前小区的带宽;
当前小区的信道质量;
所述第二信息用于以下至少之一:
提供给所述第一节点用于感知的资源信息,
提供给所述第一节点用于感知的授权信息验证和/或鉴权。
此外,根据本申请的至少一个实施例,所述感知功能的类型,包括以下至少之一:
目标对象的定位;
测距;
测速;
成像;
检测;
物体识别;
虚拟环境重构;
所述感知的精度,包括以下至少之一:
距离分辨率;
距离精度;
角度分辨率;
角度精度。
此外,根据本申请的至少一个实施例,所述根据所述第二信息,进行模型训练得到所述通感感知模型,包括:
如果满足当前小区的带宽大于或等于第一门限、当前小区的信道质量大于或等于第二门限和感知的精度小于或等于第三门限中至少之一,则根据频率m进行目标物体的反射点收集,并利用收集的反射点,执行特定功能的通感感知模型的训练;m为有理数;
或者,
如果满足当前小区的带宽小于或等于第一门限、当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则调整通信和感知之间的资源分配的权重,以进行模型训练得到所述通感感知模型;和/或,调整用于感知的带宽大小。
此外,根据本申请的至少一个实施例,所述接收第二节点发送的第一消息,包括:
如果满足当前小区的带宽小于或等于第一门限、当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则向所述第二节点发送所述第三消息;所述第三消息用于请求所述第二节点向所述第一节点发送所述第一消息;
接收所述第二节点发送的所述第一消息。
此外,根据本申请的至少一个实施例,所述方法还包括:
向终端发送第四消息,所述第四消息用于请求获取所述第一信息;
接收所述终端返回的所述第一信息;
向所述第二节点发送所述第一信息,以供所述第二节点进行模型训练得到
所述通感感知模型。
此外,根据本申请的至少一个实施例,所述方法还包括:
向与所述第一节点相邻的节点发送第五消息;所述第五消息用于请求获取第三信息;
接收与所述第一节点相邻的节点返回的所述第三信息;
将所述第三信息和所述第一信息发送给所述第二节点,以供所述第二节点进行模型训练得到所述通感感知模型。
此外,根据本申请的至少一个实施例,所述第三信息,包括以下至少之一:
模型训练的触发方式;
模式训练的类型;
训练的方式;
模型是否需要传递;
模型传递的具体参数;
模型参数传递的方式。
此外,根据本申请的至少一个实施例,所述方法还包括:
利用所述通感感知模型,执行感知预测,得到预测结果。
此外,根据本申请的至少一个实施例,所述方法还包括:
向与所述第一节点相邻的节点或所述第二节点发送反馈信息;所述反馈信息携带有预测结果;所述预测结果包括感知的被测物体更新后的状态和/或所述通感感知模型的更新方式。
本申请的至少一个实施例提供一种信息接收装置,包括:
接收单元,用于用于接收第二节点发送的第一消息;其中,所述第一消息用于请求第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功能;
其中,所述感知功能包括以下功能中至少之一:
训练通感感知模型、新建通感感知模型、增加通感感知模型、更新通感感知模型、删除通感感知模型中至少之一;其中,所述通感感知模型由本节点使用或由本节点提供给其他节点;
进行模型训练,得到通感感知模型,以供本节点使用或由本节点提供给其他节点;
推算出被测物体的感知结果信息,以供本节点使用或由本节点提供给其他节点;
确定用于感知的资源信息,以供本节点使用或由本节点提供给其他节点;
进行感知的授权信息验证和/或鉴权;其中,所述授权和/或鉴权信息由本节点使用或由本节点提供给其他节点;
感知的能力上报、协商、降维和配置中的至少一个,其中,感知的能力配置由本节点使用或由本节点提供给其他节点;
选择感知的感知模式,以供本节点使用或由本节点提供给其他节点;
选择感知节点,以供本节点使用或由本节点提供给其他节点;
进行感知服务质量QoS参数转换;其中,所述感知QoS参数由本节点使用或由本节点提供给其他节点;
将感知业务类型或标识、QoS要求和感知测量数据上报周期信息中至少之一传递到执行感知的其他节点;
若感知需求改变,终端或应用触发对应的修改流程;
触发、激活、新建感知业务中的至少一种;
终止、去激活、删除感知业务中的至少一种;
根据感知业务需求控制其他节点执行感知;
收集、上报、处理感知测量数据中的至少一种。
本申请的至少一个实施例提供一种网络设备,包括处理器和用于存储能够在处理器上运行的计算机程序的存储器,
其中,所述处理器用于运行所述计算机程序时,执行上述网络设备侧任一项所述方法的步骤。
本申请的至少一个实施例提供一种存储介质,其上存储有计算机程序,所述计算机程序被处理器执行时实现上述任一方法的步骤。
本申请实施例提供的信息接收方法、装置、设备及存储介质,第一节点接收第二节点发送的第一消息;其中,所述第一消息用于请求所述第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功能;其中,所述感知功能包括以下功能中至少之一:训练通感感知模型、新建通感感知模型、增加通感感知模型、更新通感感知模型、删除通感感知模型中至少之一;其中,所述通感感知模型由本节点使用或由本节点提供给其他节点;进行
模型训练,得到通感感知模型,以供本节点使用或由本节点提供给其他节点;推算出被测物体的感知结果信息,以供本节点使用或由本节点提供给其他节点;确定用于感知的资源信息,以供本节点使用或由本节点提供给其他节点;进行感知的授权信息验证和/或鉴权;其中,所述授权和/或鉴权信息由本节点使用或由本节点提供给其他节点;感知的能力上报、协商、降维和配置中的至少一个,其中,感知的能力配置由本节点使用或由本节点提供给其他节点;选择感知的感知模式,以供本节点使用或由本节点提供给其他节点;选择感知节点,以供本节点使用或由本节点提供给其他节点;进行感知QoS参数转换;其中,所述感知QoS参数由本节点使用或由本节点提供给其他节点;将感知业务类型或标识、QoS要求和感知测量数据上报周期信息中至少之一传递到执行感知的其他节点;若感知需求改变,终端或应用触发对应的修改流程;触发、激活、新建感知业务中的至少一种;终止、去激活、删除感知业务中的至少一种;根据感知业务需求控制其他节点执行感知;收集、上报、处理感知测量数据中的至少一种。
采用本申请实施例提供的技术方案,所述第一节点接收所述第二节点发送的用于请求提供第一信息的第一消息,后续所述第二节点利用所述第一信息执行感知功能,所述第二节点执行感知功能与通感感知模型有关,如此,在低频信道环境中的大量杂波干扰、移动物体干扰和目标散射重影等网络环境下,所述第二节点利用所述第一信息中的内容,对现有感知功能进了优化,并新增了以下通感感知功能:确定感知的资源信息、进行感知的授权信息验证和/或鉴权、执行感知的能力上报、协商、降维和配置中的至少一个、进行感知QoS参数转换,从而能够满足感知性能要求。
图1是本申请实施例信息接收方法的实现流程示意图;
图2是本申请实施例信息接收方法的具体实现流程示意图;
图3是本申请实施例第一信息包括的部分信息的示意图;
图4是是本申请实施例基于输入的信息进行模型训练得到通信感知模型的示意图;
图5是本申请实施例信息接收装置的组成结构示意图;
图6是本申请实施例网络设备的组成结构示意图。
在对本申请实施例的技术方案进行介绍之前,先对相关技术进行说明。
相关技术中,通信感知一体化是6G的标志特征之一,通感一体化系统是指在同一个系统中通过频谱共享、硬件共享、信号共享等方式,在进行信息传递的同时,感知目标物体的方位、距离、速度等信息,或者,对目标物体、事件或环境等进行检测、跟踪、识别、成像等。
目前,通信感知一体化通过空口及协议联合设计、软硬件设备共享,在相同频谱资源实现通信功能与感知功能的融合共生,使得无线网络在进行数据通信的同时,还能通过分析无线通信信号的直射、反射、散射,获得目标对象或环境信息的感知,实现定位、测距、测速、成像、检测、识别、环境重构等功能,为提升频谱利用率和设备复用率带来一个全新的维度。5G使用的正交频分复用(OFDM,Orthogonal Frequency Division Multiplexing)信号具备优秀的脉压特性与自相关性,是一种适合感知通信一体化系统的信号形式。
表1是感知性能包括的重要指标的示意,如表1所示,感知性能包括的重要指标包括测量精度和分辨率;测量精度可以包括距离精度、角度精度;所述分辨率可以包括距离分辨率、角度分辨率。其中,测量精度是指测量单个目标参数的准确程度,它与许多因素有关,比如网络系统本身的完善性、电波传播等。分辨率是指网络系统能从两个或者两个以上目标中区分特定目标的能力,它与网络系统本身的特性、目标特性等有关。
表1
相关技术中,雷达感知是工作在高频段,即,300M到300GHz之间,而目前通信系统中毫米波还没有商用频段,比雷达的频段低。在目前的低频段使用的话,比如,4.9GHz,则精度和分辨率不能满足感知的性能指标。在带宽不提升的情况下,使用过采样MUSIC算法等技术,虽然能提升2-3倍的精度水平,但是,考虑到低频信道环境中的大量杂波干扰、移动物体干扰和目标散射重影等,会导致虚警概率的提升,无法满足感知的性能指标;所述虚警概率是指目标不存在时系统判断目标出现的概率。因此,亟需设计一种利用AI模型提升感知性能指标的技术方案。
这里,公式(1)可以理解为角度估计精度的上界,即,在给定天线阵列和信噪比的前提下,采用各类角度估计算法所能达到的估计精度的上限。
基于此,本申请实施例中,第一节点接收第二节点发送的第一消息;其中,所述第一消息用于请求所述第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功能;其中,所述感知功能包括以下功能中至少之一:训练通感感知模型、新建通感感知模型、增加通感感知模型、更新通感感知模型、删除通感感知模型中至少之一;其中,所述通感感知模型由本节点使用或由本节点提供给其他节点;进行模型训练,得到通感感知模型,以供本节点使用或由本节点提供给其他节点;推算出被测物体的感知结果信息,以供本节点使用或由本节点提供给其他节点;确定用于感知的资源信息,以供本节点使用或由本节点提供给其他节点;进行感知的授权信息验证和/或鉴权;其
中,所述授权和/或鉴权信息由本节点使用或由本节点提供给其他节点;感知的能力上报、协商、降维和配置中的至少一个,其中,感知的能力配置由本节点使用或由本节点提供给其他节点;选择感知的感知模式,以供本节点使用或由本节点提供给其他节点;选择感知节点,以供本节点使用或由本节点提供给其他节点;进行感知服务质量QoS参数转换;其中,所述感知QoS参数由本节点使用或由本节点提供给其他节点;将感知业务类型或标识、QoS要求和感知测量数据上报周期信息中至少之一传递到执行感知的其他节点;若感知需求改变,终端或应用触发对应的修改流程;触发、激活、新建感知业务中的至少一种;终止、去激活、删除感知业务中的至少一种;根据感知业务需求控制其他节点执行感知;收集、上报、处理感知测量数据中的至少一种。
值得指出的是,本申请实施例所描述的技术不限于长期演进型(Long Term Evolution,LTE)/LTE的演进(LTE-Advanced,LTE-A)系统,还可用于其他无线通信系统,诸如码分多址(Code Division Multiple Access,CDMA)、时分多址(Time Division Multiple Access,TDMA)、频分多址(Frequency Division Multiple Access,FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access,OFDMA)、单载波频分多址(Single-carrier Frequency-Division Multiple Access,SC-FDMA)和其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用,所描述的技术既可用于以上提及的系统和无线电技术,也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio,NR)系统,并且在以下大部分描述中使用NR术语,但是这些技术也可应用于NR系统应用以外的应用,如第6代(6th Generation,6G)通信系统。
示例性的,本申请实施例应用的通信系统可包括网络设备,网络节点和终端设备(也可称为终端、通信终端等等);本申请中的第二节点,第三节点,第一节点可以网络设备,即是与终端设备通信的设备;也可以是为一定区域范围内提供通信覆盖,并且可以与位于该区域内的终端进行通信;比如,可以是各通信系统中的基站,例如LTE系统中的演进型基站(eNB,Evolutional Node B),又例如5G系统,NR系统或6G系统中的基站。第一网络节点可以是一个或多个基站、发射点、接收点、中央单元、分布单元、BBU、RRU、中继、IAB、智能超表面、通信气球、飞行的飞行器基站、天线、卫星基站、利用sidelink通信的终端等;也可以是终端,或者其他具体通信功能的模块;也可以是提供
感知功能和/或AI功能等的应用服务器;也可以是核心网络节点,比如,UPF,AF,MME,AMF,和核心网一体化的设备或者其他用于连接建立的网元。
本文中的“业务”可以表示以下至少一种概念:业务,PDU会话,服务质量(Quality of Service,QoS)流,流(Stream)或业务数据流(service data flow),无线承载,逻辑信道。
本文中所述的“数据”可以是“数据包”,“上行共享物理信道传输(Physical Uplink Share Channel transmission,PUSCH)”、“下行共享物理信道传输(Physical Downlink Share Channel transmission,PDSCH)”、“数据单元”、"PDU集合(PDU set)包”、“采样(sample)”、“切片(slice)”、“瓦片(tile)”、“流(stream)”、“传输(transmission)”和“传输块”等中的一种或多种。
图1是本申请实施例信息接收方法的实现流程示意图,应用于第一节点,如图1所示,所述方法包括步骤101:
步骤101:接收第二节点发送的第一消息;其中,所述第一消息用于请求所述第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功能;
其中,所述感知功能包括以下功能中至少之一:
训练通感感知模型、新建通感感知模型、增加通感感知模型、更新通感感知模型、删除通感感知模型中至少之一;其中,所述通感感知模型由本节点使用或由本节点提供给其他节点;
进行模型训练,得到通感感知模型,以供本节点使用或由本节点提供给其他节点;
推算出被测物体的感知结果信息,以供本节点使用或由本节点提供给其他节点;
确定用于感知的资源信息,以供本节点使用或由本节点提供给其他节点;
进行感知的授权信息验证和/或鉴权;其中,所述授权和/或鉴权信息由本节点使用或由本节点提供给其他节点;
感知的能力上报、协商、降维和配置中的至少一个,其中,感知的能力配置由本节点使用或由本节点提供给其他节点;
选择感知的感知模式,以供本节点使用或由本节点提供给其他节点;
选择感知节点,以供本节点使用或由本节点提供给其他节点;
进行感知QoS参数转换;其中,所述感知QoS参数由本节点使用或由本节点提供给其他节点;
将感知业务类型或标识、QoS要求和感知测量数据上报周期信息中至少之一传递到执行感知的其他节点;
若感知需求改变,终端或应用触发对应的修改流程;
触发、激活、新建感知业务中的至少一种;
终止、去激活、删除感知业务中的至少一种;
根据感知业务需求控制其他节点执行感知;
收集、上报、处理感知测量数据中的至少一种。
这里,为满足感知业务QoS需求,需要进行QoS参数转换,
这里,将感知业务类型或标识、QoS要求和感知测量数据上报周期等信息传递到执行感知的其他节点,如终端或基站。
这里,若感知需求改变,如感知区域、数据上报时间或QoS要求等发生变化,终端或应用需要触发对应的修改流程。
这里,所述测量节点是指用于感知被测物的节点。
这里,所述第二节点可以是感知的新功能网元SF(sensing Function)、相邻的基站、CU、其他的终端、核心网元中的至少一个节点。其中,SF主要负责感知控制和感知测量数据计算。
在一实施例中,所述第一消息,携带有以下至少之一:
所述第一信息的类型;
所述第一信息的发送周期;
所述第一信息的发送数量;
触发所述第一信息的事件信息;
所述第一信息用于指示以下至少之一:
两个测量节点之间的距离和角度;
被测物体到所述第一节点的入射角度、距离;
设置有被测物体的终端的无线电测量参数;
设置有被测物体的终端的多普勒变化率;
被测物体处的功率密度;
被测物体的反射点个数;
被测物体的反射点的间距;
被测物体的反射点的图样、对应的功率密度和接收天线增益;
提供给所述第二节点用于感知的授权信息验证和/或鉴权;
提供给所述第二节点用于感知的能力协商和配置。
可以理解的是,所述第一消息可以是指请求消息等。
这里,所述确定用于感知的资源信息,可以是指用于明确感知所占用的资源信息,因为第一节点需要通过时域/频域/码域/空域中至少一种分离方式来协同通信和感知的两种功能操作,所以,需要在一定的资源范围内明确合理的,和用户的业务需要更匹配的感知功能,所占用的资源信息。具体的感知所占用的资源信息包括以下至少一种方式:波形的选择,通感信号的选择,感知信号帧结构设计等。比如,是选择脉冲波形(PW),线性调频波(FMCW)还是通信波形包括正交频分复用(OFDM)波形。其中,感知信号的帧结构的确定中,比如,是否采用7:3(DDDSUDDSUU)的典型配比,时域资源取决于速度范围、速度分辨率、探测距离、帧率、组网干扰等至少一种因素,而频域开销取决于距离分辨率、组网干扰等至少一种因素。
这里,利用所述第一信息确定感知的资源信息,具体可以包括:
两个测量节点之间的距离越长和/或角度越小,则OFDM的循环前缀(CP)长度越长,如果两个测量节点之间的距离越短和/或角度越大,可以使用短的CP;
被测物体到所述第一节点的入射角度越大和/或距离越长,则OFDM的CP长度越长,如果被测物体到所述第一节点的入射角越小和/或距离越短,可以使用短的CP;
设置有被测物体的终端的无线电测量参数对应的无线电测量精度要求越高,则需要的下行(DL)时隙/时间符号越小;
设置有被测物体的终端的多普勒变化率越高,则尽量用子载波间隔大的帧结构;终端的多普勒变化率越低,则尽量用子载波间隔小的帧结构;也会影响CP的长短选择;
被测物体处的功率密度越高,占用的感知资源越多;
被测物体的反射点个数越多,占用的感知资源越多。
其中,资源可以包括:
资源的激活时间点;
资源的激活时间范围;
资源可使用的地理范围;
资源可使用的高度范围;
资源可使用的频段范围;
资源可使用的区域范围;
资源可使用的频段范围;
资源可使用的发送对象。
这里,在所述第一信息用于提供给所述第二节点用于感知的授权信息验证和/或鉴权的情况下,可以根据所述第一信息进行感知的授权信息验证和/或鉴权。其中,进行感知的授权信息验证和/或鉴权可以是指用于确定感知对象和/或申请感知服务的对象是否合法和/或具有合理权限;如果合法,或具有对应的权限,本节点则提供对应的通信和/或感知服务,否则,拒绝提供通信和/或感知服务。也就是,确定申请方是否有权限获取“被测物体处的功率密度,设置有被测物体的终端的多普勒变化率,两个测量节点之间的距离和角度”对应的功能。具体可以根据感知功能的类型,感知的精度,感知的物体类型,判断是否申请方是否有权限。
这里,在所述第一信息用于提供给所述第二节点用于感知的能力协商和配置的情况下,可以根据所述第一信息进行感知能力的上报、协商、降维和配置。其中,感知能力的上报为了在人工智能,资源协商和/或算力共享的过程中,相关联的节点之间对资源和处理能力的分享有更合理和高效的判断,可以分成更合理的算法和/或生成更高效的模型。比如,根据第一信息中“被测物体处的功率密度,设置有被测物体的终端的多普勒变化率,两个测量节点之间的距离和角度”等信息,反馈匹配的和/或合理的感知能力。如果感知能力不足,可以申请能力的降级(即,降维)和/或能力协商。
这里,可以根据所述第一信息进行感知QoS参数的转换。其中,感知QoS参数来自应用层、感知服务的请求方、核心网、邻节点中至少一种节点,进行感知QoS参数转换可以更好的适配节点的能力。比如,被测物体的终端的无线电测量参数对应的无线电测量精度要求越高,被测物体的终端的多普勒变化率
越高等情况,则要求感知QoS越高;反正,当第一信息显示被测物体的终端的无线电测量参数对应的无线电测量精度要求变低,被测物体的终端的多普勒变化率变低等情况,则QoS降低。
在一实施例中,所述第一信息用于以下至少之一:
推算模型训练得到通感感知模型,并提供给所述第一节点;
推算出被测物体的感知结果信息,并提供给所述第一节点。
提供给所述第二节点用于感知的资源信息;
提供给所述第二节点用于感知的授权信息验证和/或鉴权;
提供给所述第二节点用于感知的能力协商和配置;
提供给所述第二节点用于感知的感知模式选择;
提供给所述第二节点用于感知节点选择。
这里,所述通感感知模型可以是指具有通信和感知功能的人工智能(AI,Artificial Intelligence)或机器学习(ML,Machine Learning)模型。
这里,在一种场景中,所述第一节点可以向所述第二节点提供所述第一信息,以供所述第二节点利用所述第一信息进行训练得到所述通感感知模型。
在一实施例中,所述感知结果信息,包括以下至少之一:
目标对象的定位;
测距;
测速;
成像;
检测;
物体识别;
虚拟环境重构。
在一实施例中,所述用于感知的资源信息,包括以下至少之一:
周期、起始位置、持续时间或者每个周期的持续时长中至少一个;
资源的激活时间点;
资源的激活时间范围;
资源可使用的地理范围;
资源可使用的高度范围;
资源可使用的频段范围;
资源可使用的区域范围;
资源可使用的频段范围;
资源可使用的发送对象。
这里,资源的激活时间点和感知对象的分布密度,和/或申请感知的功能的请求有关;
这里,资源的激活时间范围和感知对象的分布密度,和/或申请感知的功能的请求有关;
这里,资源可使用的地理范围和感知对象的分布密度,和基站的发送能力有关;
这里,资源可使用的高度范围和感知对象的分布密度,和基站的发送能力有关;
这里,资源可使用的区域范围和感知对象的分布密度,和/或申请感知的功能的请求有关;
这里,资源可使用的发送对象和感知对象的类型有关。
这里,所述激活时间点也可以是生效时间点。
这里,所述区域范围可以是指地理范围、国家范围、地区范围、基站范围、小区范围、公共禄蠹移动网络(PLMN,Public Land Mobile Network)范围、小区类型范围中至少一个组合。
在一实施例中,所述通感感知模型的训练和/或推理位于以下至少之一:
OAM;
gNB;
无线网络节点;
gNB的CU;
AI控制AF的至少一种网元节点;
终端侧的AI服务器;
终端侧。
考虑到在实际应用时,在一种场景中,所述第一节点自身也可以进行模型训练得到所述通感感知模型。
基于此,在一实施例中,所述方法还包括:
接收第三节点发送的第二消息;其中,所述第二消息包括第二信息;
根据所述第二信息,进行模型训练得到所述通感感知模型;
其中,所述第二信息,包括以下至少之一:
感知功能的类型;
感知的精度;
感知的物体类型;
AF的ID;
终端的ID;
当前小区的带宽;
当前小区的信道质量;
所述第二信息用于以下至少之一:
提供给所述第一节点用于感知的资源信息,
提供给所述第一节点用于感知的授权信息验证和/或鉴权。
这里,所述第二信息表征感知的需求信息。
在一实施例中,所述感知功能的类型,包括以下至少之一:
目标对象的定位;
测距;
测速;
成像;
检测;
物体识别;
虚拟环境重构;
所述感知的精度,包括以下至少之一:
距离分辨率;
距离精度;
角度分辨率;
角度精度。
在一实施例中,所述根据所述第二信息,进行模型训练得到所述通感感知模型,包括:
如果满足当前小区的带宽大于或等于第一门限、当前小区的信道质量大于或等于第二门限和感知的精度小于或等于第三门限中至少之一,则根据频率m
进行目标物体的反射点收集,并利用收集的反射点,执行特定功能的通感感知模型的训练;m为有理数;
或者,
如果满足当前小区的带宽小于或等于第一门限、当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则调整通信和感知之间的资源分配的权重,以进行模型训练得到所述通感感知模型;和/或,调整用于感知的带宽大小。
也就是说,如果当前小区的带宽足够大,且环境单纯、干扰较小,和/或,感知的精度值较低,在这种情况下可以满足感知的精度需求,则进行根据频率m进行目标物体的反射点收集,并进行特定功能的感知。
如果当前小区的带宽较小,且环境单纯、干扰较小,和/或,感知的精度值较大,在这种情况下为了满足感知的精度需求,则调整通信和感知之间的资源分配的权重,以进行模型训练得到所述通感感知模型;和/或,调整用于感知的带宽大小。
如果无法满足需求,则启动过采样算法,提升采样密度,来增加训练样本,以进行模型训练得到所述通感感知模型。
考虑到在实际应用时,所述第一节点可以根据所述第三节点发送的第二信息,判断当前是否需要满足更高的感知精度需求,在确定需要满足更高的感知精度需求的情况下,由所述第二节点来进行模型训练得到所述通感感知模型,以提高感知的精度。
基于此,在一实施例中,所述接收第二节点发送的第一消息,包括:
如果满足当前小区的带宽小于或等于第一门限、当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则向所述第二节点发送第三消息;所述第三消息用于请求所述第二节点向所述第一节点发送所述第一消息;
接收所述第二节点发送的所述第一消息。
在一实施例中,所述方法还包括:
向终端发送第四消息,所述第四消息用于请求获取所述第一信息:
接收所述终端返回的所述第一信息;
向所述第二节点发送所述第一信息。
这里,所述终端可以是指设置有被测物体的UE。
这里,所述第二节点利用所述第一信息训练通感感知模型。
在一实施例中,所述方法还包括:
向与所述第一节点相邻的节点发送第五消息;所述第五消息用于请求获取第三信息;
接收与所述第一节点相邻的节点返回的所述第三信息;
将所述第三信息和所述第一信息发送给所述第二节点,以供所述第二节点进行模型训练得到所述通感感知模型。
在一实施例中,所述第三信息,包括以下至少之一:
模型训练的触发方式;
模式训练的类型;
训练的方式;
模型是否需要传递;
模型传递的具体参数;
模型参数传递的方式。
可以理解的是,所述模型训练的触发方式,包括以下至少之一:
终端触发;
网络节点触发;
感知服务节点触发。
可以理解的是,所述模式训练的类型,包括监督学习、无监督学习、半监督学习、联邦学习、在线训练、离线训练中至少一种。所述训练的方式,包括单边训练、双边训练、多边训练。
可以理解的是,模型传递的具体参数包括模型生成、模型推理、模型监测、模型更新中至少之一,还包括模型的层数、模型性能监控使用的KPI中至少之一,以及模型部署的粒度;其中,模型部署的粒度包括场景、业务、配置、位置、基站、小区、需求等、UE、UE组等至少之一。
可以理解的是,模型参数传递的方式,包括RRC信令(比如,新建一个SRBx,以及定义其对应的安全,完整性保护机制)、NAS信令、UP传输(比如,通过DRB传输即一种没有GTP-U隧道对应的DRB,通过GTP-U数据的净荷payload,或者GTP-数据的包头)。
可以理解的是,所述第一节点还可以收集应用数据源数据;所述应用数据源数据包括被测物体的视频、图像、位置信息等,该数据可以作为模型训练的辅助输入数据,以供所述第二节点结合所述第一信息、所述第三信息进行模型训练得到所述通感感知模型。
需要说明的是,所述第一节点也可以利用所述第一信息,进行模型训练得到所述通感感知模型;或者,也可以利用所述第一信息和所述第三信息,进行模型训练得到所述通感感知模型;或者,也可以利用所述第一信息、第三信息和所述应用数据源数据,进行模型训练得到所述通感感知模型。
在一实施例中,所述方法还包括:
利用所述通感感知模型,执行感知预测,得到预测结果。
在一实施例中,所述方法还包括:
向与所述第一节点相邻的节点或所述第二节点发送反馈信息;所述反馈信息携带有预测结果;所述预测结果包括感知的被测物体更新后的状态和/或所述通感感知模型的更新方式。
本申请实施例中,具备以下优点:
(1)所述第一节点接收所述第二节点发送的用于请求提供第一信息的第一消息,后续所述第二节点利用所述第一信息执行感知功能,所述第二节点执行感知功能与通感感知模型有关,如此,在低频信道环境中的大量杂波干扰、移动物体干扰和目标散射重影等存在的情况下,所述第二节点执行通感感知功能能够满足感知的性能指标。
(2)针对不同的应用场景、不同的感知功能的类型和感知的精度需求,进行不同算法的选择,以训练得到合适的通感感知模型。
图2是本申请实施例信息接收方法的具体实现流程示意图,如图2所示,所述方法包括步骤201至步骤206:
步骤201:第三节点(通感AF)向第一节点(RAN set中的基站节点1)发送第二消息,即通感感知开放请求消息;所述第二消息包括第二信息;所述第二消息用于请求第一节点(RAN set中的基站节点1)训练通感感知模型。
这里,所述通感感知模型可以是指具有通信和感知功能的AI或ML模型。
这里,所述第二信息包括以下至少之一:
感知功能的类型;
感知的精度;
感知的物体类型;
AF的标识;
终端(UE)的ID;
当前小区的带宽;
当前小区的信道质量;
所述第二信息用于以下至少之一:
提供给所述第一节点用于感知的资源信息,
提供给所述第一节点用于感知的授权信息验证和/或鉴权。
这里,所述感知功能的类型,可以包括以下至少之一:
目标对象的定位;
测距;
测速;
成像;
检测;
物体识别;
虚拟环境重构。
这里,所述感知的精度,可以包括以下至少之一:
距离分辨率(由带宽决定);
距离精度;
角度分辨率;
角度精度。
这里,所述感知的物体类型,可以是指交通工具、动物、鸟类等。
步骤202:第一节点(RAN set中的基站节点1)根据所述第二信息,进行模型训练得到所述通感感知模型。
这里,所述根据所述第二信息,进行模型训练得到所述通感感知模型,包括以下至少之一:
如果满足当前小区的带宽大于或等于第一门限、当前小区的信道质量大于或等于第二门限和感知的精度小于或等于第三门限中至少之一,则根据频率m
进行目标物体的反射点收集,并利用收集的反射点,执行特定功能的通感感知模型的训练;m为大于0的自然数;
或者,
如果满足当前小区的带宽小于或等于第一门限、当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则调整通信功能和感知功能之间的资源分配的权重,以进行模型训练得到所述通感感知模型;和/或,调整用于感知的带宽大小。
也就是说,如果当前小区的带宽足够大,且环境单纯、干扰较小,和/或,感知的精度值较低,在这种情况下可以满足感知的精度需求,则进行根据频率m进行目标物体的反射点收集,并进行特定功能的感知。
如果当前小区的带宽较小,且环境单纯、干扰较小,和/或,感知的精度值较大,在这种情况下为了满足感知的精度需求,则调整通信和感知之间的资源分配的权重,以进行模型训练得到所述通感感知模型;和/或,调整用于感知的带宽大小。
如果无法满足需求,则启动过采样算法,提升采样密度,来增加训练样本,以进行模型训练得到所述通感感知模型。
如果依然还是无法满足需求,则启动AI功能,提升感知精度,即,执行步骤203。
步骤203:第一节点(RAN set中的第一基站)向第二节点(AI控制AF)发送第三消息,即,通感感知AI模型请求消息;所述第三消息用于请求第二节点(AI控制AF)训练通感感知模型,并提供给所述第一节点。
这里,所述第三消息,携带有以下至少之一:
感知功能的类型;
感知的精度;
感知的物体类型。
步骤204:第二节点(AI控制AF)接收第一节点(RAN set中的基站节点1)发送的所述第三消息,并向第一节点(RAN set中的基站节点1)发送第一消息;所述第一消息用于请求第一节点(RAN set中的基站节点1)向第二节点(AI控制AF)提供第一信息;所述第一信息表征所述第二节点(AI控制AF)训练通感感知模型所需的开放信息。
这里,所述第一消息,携带有以下至少之一:
所述第一信息的类型;
所述第一信息的发送周期;
所述第一信息的发送数量;
触发所述第一信息的事件信息;
所述第一信息,包括以下至少之一:
两个测量节点之间的距离和角度;
被测物体到所述第一节点的入射角度、距离;
设置有被测物体的UE的无线电测量参数;
设置有被测物体的UE的多普勒变化率;
被测物体处的功率密度;
被测物体的反射点个数;。
被测物体的反射点的间距;
被测物体的反射点的图样、对应的功率密度和接收天线增益。
这里,以所述第一节点为基站为例,被测物体到所述第一节点的入射角度、距离是指目标被测物体到基站的入射角度,目标被测物体到基站的距离;所述距离根据目标被测物体到基站的回波时延推衍得到;
这里,设置有被测物体的UE的无线电测量,可以包括RSRP、RSRQ、SINR;对应感知功能的类型为定位、测距。
这里,设置有被测物体的UE的多普勒变化率,对应感知功能的类型为成像、检测、物体识别。
这里,被测物体处的功率密度,对应感知功能的类型为成像、检测、物体识别。
这里,被测物体的反射点个数,对应感知功能的类型为检测、物体识别。
这里,被测物体的反射点的间距,对应感知功能的类型为检测、物体识别。
这里,被测物体的反射点的图样,对应的功率密度和接收天线增益,对应感知功能的类型为检测、物体识别。
图3是本申请实施例第一信息包括的部分信息的示意图,如图3所示,TX1表示测量节点1的发送单元,TX2表示测量节点2的发送单元,两个测量节点之间的距离为4d。
这里,第二节点(AI控制AF)利用所述第一节点(RAN set中的基站节点1)提供的所述第一信息对预测模型进行训练,得到具备通信和感知功能的通感感知模型。可选的,第二节点(AI控制AF)可以给多个基站配置一个或多个用于通信感知的通感感知模型,即AI/ML模型。
或者,
第一节点(RAN set中的基站节点1)可以通过信令,在Xn口和第二网络设备(RAN set中的基站节点2)同步和交互用于通信感知的AI/ML模型,该模型可以为(RAN set中的基站节点1)提供有用的输入信息,例如,预测的被测目标物体的状态等,以及模型更新的方式,是线上更新,还是线下更新。
这里,如果需要UE参与,第一节点(RAN set中的基站节点1)也可以根据UE的能力,在空口传递交互用于模型感知的AI/ML模型。
步骤205:第一节点(RAN set中的基站节点1)向终端发送第四消息;所述第四消息用于请求获取所述第一信息;所述第一信息表征训练通感感知模型所需的开放信息;接收所述终端返回的第一信息,并向所述第二节点(AI控制AF)发送所述第一信息。
这里,所述终端可以是指设置有被测物体的UE,该UE中设置有通信模块。
这里,第一节点(RAN set中的基站节点1)收集终端(被测物体的UE)反馈或反射的第一信息。
这里,所述第一信息,包括以下至少之一:
两个测量节点之间的距离和角度;
目标被测物体到基站的入射角度、距离;
设置有被测物体的UE的无线电测量参数;
设置有被测物体的UE的多普勒变化率;
被测物体处的功率密度;
被测物体的反射点个数;
被测物体的反射点的间距;
被测物体的反射点的图样,
对应的功率密度和接收天线增益等。
这里,第一节点(RAN set中的基站节点1)还可以收集应用数据源数据;所述应用数据源数据包括被测物体的视频,图像,位置信息等,该数据可以作
为模型训练的辅助输入数据。
这里,第一节点(RAN set中的基站节点1)向相邻的节点(RAN set中的基站节点2)发送第五消息;所述第五消息用于请求获取第三信息。
这里,所述第三信息,包括以下至少之一:
模型训练的触发方式;
模式训练的类型;
训练的方式;
模型是否需要传递;
模型传递的具体参数;
模型参数传递的方式。
这里,第一节点(RAN set中的基站节点1)接收相邻的节点(RAN set中的基站节点2)返回的训练通信感知模型所需的第三信息,并向所述第二节点(AI控制AF)发送所述第三信息。
这里,所述第二节点(AI控制AF)将所述第一信息和所述第三信息作为模型训练的输入信息,进行模型训练得到所述通感感知模型。
可选的,AI/ML模型的训练位于所述第二节点(AI控制AF)。所需的测量用于训练AI/ML模式。
这里,AI/ML模型的训练也可以位于第一节点(RAN set中的基站节点1)。
步骤206:所述第二节点(AI控制AF)根据所述第一信息,训练通感感知模型,利用训练好的通感感知模型执行感知预测,得到预测结果;并将所述预测结果返回给第一节点(RAN set中的基站节点1)。
这里,AI/ML模型的训练位于第一节点(RAN set中的基站节点1),则第一节点(RAN set中的基站节点1)也可以执行感知预测,得到预测结果。例如,针对第一节点(RAN set中的基站节点1)的小区,执行感知预测,得到预测结果。
图4是本申请实施例基于输入的信息进行模型训练得到通信感知模型的示意图,如图4所示,AI/ML模型的训练位于第一节点(RAN set中的基站节点1),所述输入的信息包括所述第一信息、第三信息和应用数据源数据。
这里,为了预测优化的感知判断的决策,第一节点(RAN set中的基站节点1)可以将从本地节点获取的第一信息作为基于AI/ML模型的输入数据:
其中,所述第一信息,包括以下至少之一:
两个测量节点之间的距离和角度;
目标被测物体到基站的入射角度、距离(根据目标到基站的回波时延推衍);
被测物体的UE无线电测量(例如RSRP、RSRQ、SINR);
被测物体的UE的多普勒变化率;
被测物体处的功率密度;
被测物体的反射点个数;
被测物体的反射点的间距;
被测物体的反射点的图样,对应的功率密度和接收天线增益;
另外,为了优化基于AI/ML的感知的性能,还可以考虑收集以下辅助输入数据:
来自应用层的应用数据源数据,包括物体的视频,图像、位置信息等;
来自邻区的基站节点2发送的第三信息,即与被测对象相关的输入信息。
这里,为了预测优化的感知判断的决策,第一节点(RAN set中的基站节点1)可以将以下信息作为基于AI/ML模型的的输出数据:
被测物体的UE位置信息(例如坐标、服务小区ID、移动速度),
物体长度;
物体宽度;
物体高度;
物体形状;
物体的类型;
物体运动的速度;
物体的运动方向;
物体的运动轨迹。
需要说明的是,如果不能满足感知的精度需求,则第一节点(RAN set中的基站节点1)去激活当前的AI/ML模型,激活第二类型的AI/ML模型,
这里,如果是多侧AI/ML模型训练和推断模式,则发送第二类型AI/ML模型激活同步信息给相邻基站。
这里,所述第二节点(AI控制AF)根据预测结果,做出感知决策;并向一网络设备(RAN set中的基站节点1)发送反馈信息;所述反馈信息携带有所
述预测结果。
这里,如果AI/ML模型的训练位于所述第二节点(AI控制AF),第二网络设备(RAN set中的基站节点2)还可以向所述第二节点(AI控制AF)发送反馈信息;所述反馈信息包括所述感知后的目标状态更新等。
这里,如果AI/ML模型的训练位于第一节点(RAN set中的基站节点1),第二网络设备(RAN set中的基站节点2)还可以向第一节点(RAN set中的基站节点1)发送反馈信息;所述反馈信息包括所述感知后的目标状态更新等。
本示例中,具备以下优点:
(1)第一节点(RAN set中的基站节点1)接收第二节点(AI控制AF)发送的用于请求提供第一信息的第一消息,如此,后续可以将所述第一信息发送给第二节点(AI控制AF),以供第二节点(AI控制AF)利用所述第一信息进行模型训练得到通感感知模型,如此,在低频信道环境中的大量杂波干扰、移动物体干扰和目标散射重影等存在的情况下,可以利用所述通感感知模型满足感知的性能指标。
(2)针对不同的应用场景、不同的感知功能的类型和感知的精度需求,进行不同算法的选择,以训练得到合适的通感感知模型。
(3)与UE和其他网络节点之间进行信令交互,以得到模型训练所需的输入信息。
为实现本申请实施例信息接收方法,本申请实施例还提供一种信息接收装置。图5为本申请实施例信息接收装置的组成结构示意图,如图5所示,所述装置包括:
接收单元51,用于接收第二节点发送的第一消息;其中,所述第一消息用于请求所述第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功能;
其中,所述感知功能包括以下功能中至少之一:
训练通感感知模型、新建通感感知模型、增加通感感知模型、更新通感感知模型、删除通感感知模型中至少之一;其中,所述通感感知模型由本节点使用或由本节点提供给其他节点;
进行模型训练,得到通感感知模型,以供本节点使用或由本节点提供给其
他节点;
推算出被测物体的感知结果信息,以供本节点使用或由本节点提供给其他节点;
确定用于感知的资源信息,以供本节点使用或由本节点提供给其他节点;
进行感知的授权信息验证和/或鉴权;其中,所述授权和/或鉴权信息由本节点使用或由本节点提供给其他节点;
感知的能力上报、协商、降维和配置中的至少一个,其中,感知的能力配置由本节点使用或由本节点提供给其他节点;
选择感知的感知模式,以供本节点使用或由本节点提供给其他节点;
选择感知节点,以供本节点使用或由本节点提供给其他节点;
进行感知QoS参数转换;其中,所述感知QoS参数由本节点使用或由本节点提供给其他节点;
将感知业务类型或标识、QoS要求和感知测量数据上报周期信息中至少之一传递到执行感知的其他节点;
若感知需求改变,终端或应用触发对应的修改流程;
触发、激活、新建感知业务中的至少一种;
终止、去激活、删除感知业务中的至少一种;
根据感知业务需求控制其他节点执行感知;
收集、上报、处理感知测量数据中的至少一种。
在一实施例中,所述第一消息,携带有以下至少之一:
所述第一信息的类型;
所述第一信息的发送周期;
所述第一信息的发送数量;
触发所述第一信息的事件信息;
所述第一信息用于指示以下至少之一:
两个测量节点之间的距离和角度;
被测物体到所述第一节点的入射角度、距离;
设置有被测物体的终端的无线电测量参数;
设置有被测物体的终端的多普勒变化率;
被测物体处的功率密度;
被测物体的反射点个数;
被测物体的反射点的间距;
被测物体的反射点的图样、对应的功率密度和接收天线增益。
在一实施例中,所述第一信息用于以下至少之一:
推算模型训练得到通感感知模型,并提供给所述第一节点;
推算出被测物体的感知结果信息,并提供给所述第一节点。
提供给所述第二节点用于感知的资源信息;
提供给所述第二节点用于感知的授权信息验证和/或鉴权;
提供给所述第二节点用于感知的能力协商和配置;
提供给所述第二节点用于感知的感知模式选择;
提供给所述第二节点用于感知节点选择。
在一实施例中,所述感知结果信息,包括以下至少之一:
目标对象的定位;
测距;
测速;
成像;
检测;
物体识别;
虚拟环境重构。
在一实施例中,所述用于感知的资源信息,包括以下至少之一:
周期、起始位置、持续时间或者每个周期的持续时长中至少一个;
资源的激活时间点;
资源的激活时间范围;
资源可使用的地理范围;
资源可使用的高度范围;
资源可使用的频段范围;
资源可使用的区域范围;
资源可使用的频段范围;
资源可使用的发送对象。
在一实施例中,所述通感感知模型的训练和/或推理位于以下至少之一:
操作维护管理(OAM,Operation and Maintenance management);
下一代基站(gNB);
无线网络节点;
gNB的中心单元(CU,Central Unit);
人工智能(AI,Artificial Intelligence)控制应用功能(AF,Application Function)的至少一种网元节点;
终端侧的AI服务器;
终端侧。
在一实施例中,所述装置还用于:
接收第三节点发送的第二消息;其中,所述第二消息包括第二信息;
根据所述第二信息,进行模型训练得到所述通感感知模型;
其中,所述第二信息,包括以下至少之一:
感知功能的类型;
感知的精度;
感知的物体类型;
AF的ID;
终端的ID;
当前小区的带宽;
当前小区的信道质量;
所述第二信息用于以下至少之一:
提供给所述第一节点用于感知的资源信息,
提供给所述第一节点用于感知的授权信息验证和/或鉴权。
在一实施例中,所述感知功能的类型,包括以下至少之一:
目标对象的定位;
测距;
测速;
成像;
检测;
物体识别;
虚拟环境重构;
所述感知的精度,包括以下至少之一:
距离分辨率;
距离精度;
角度分辨率;
角度精度。
在一实施例中,所述装置还用于:
如果满足当前小区的带宽大于或等于第一门限、当前小区的信道质量大于或等于第二门限和感知的精度小于或等于第三门限中至少之一,则根据频率m进行目标物体的反射点收集,并利用收集的反射点,执行特定功能的通感感知模型的训练;m为有理数;
或者,
如果满足当前小区的带宽小于或等于第一门限、当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则调整通信和感知之间的资源分配的权重,以进行模型训练得到所述通感感知模型;和/或,调整用于感知的带宽大小。
在一实施例中,所述接收单元51,用于:
如果满足当前小区的带宽小于或等于第一门限、当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则向所述第二节点发送所述第三消息;所述第三消息用于请求所述第二节点向所述第一节点发送所述第一消息;
接收所述第二节点发送的所述第一消息。
在一实施例中,所述装置还用于:
向终端发送第四消息,所述第四消息用于请求获取所述第一信息;
接收所述终端返回的所述第一信息;
向所述第二节点发送所述第一信息,以供所述第二节点进行模型训练得到所述通感感知模型。
在一实施例中,所述装置还用于:
向与所述第一节点相邻的节点发送第五消息;所述第五消息用于请求获取第三信息;
接收与所述第一节点相邻的节点返回的所述第三信息;
将所述第三信息和所述第一信息发送给所述第二节点,以供所述第二节点进行模型训练得到所述通感感知模型。
在一实施例中,所述第三信息,包括以下至少之一:
模型训练的触发方式;
模式训练的类型;
训练的方式;
模型是否需要传递;
模型传递的具体参数;
模型参数传递的方式。
在一实施例中,所述装置还用于:
利用所述通感感知模型,执行感知预测,得到预测结果。
在一实施例中,所述装置还用于:
向与所述第一节点相邻的节点或所述第二节点发送反馈信息;所述反馈信息携带有预测结果;所述预测结果包括感知的被测物体更新后的状态和/或所述通感感知模型的更新方式。
实际应用时,所述接收单元51可以由信息接收装置中的通信接口实现。
需要说明的是:上述实施例提供的信息接收装置在进行信息接收时,仅以上述各程序模块的划分进行举例说明,实际应用中,可以根据需要而将上述处理分配由不同的程序模块完成,即将装置的内部结构划分成不同的程序模块,以完成以上描述的全部或者部分处理。另外,上述实施例提供的信息接收装置与信息接收方法实施例属于同一构思,其具体实现过程详见方法实施例,这里不再赘述。
本申请实施例还提供了一种网络设备,如图6所示,包括:
通信接口61,能够与其它设备进行信息交互;
处理器62,与所述通信接口61连接,用于运行计算机程序时,执行上述网络设备侧一个或多个技术方案提供的方法。而所述计算机程序存储在存储器63上。
需要说明的是:所述处理器62和通信接口61的具体处理过程详见方法实施例,这里不再赘述。
当然,实际应用时,网络设备60中的各个组件通过总线系统64耦合在一起。可理解,总线系统64用于实现这些组件之间的连接通信。总线系统64除包括数据总线之外,还包括电源总线、控制总线和状态信号总线。但是为了清楚说明起见,在图6中将各种总线都标为总线系统64。
本申请实施例中的存储器63用于存储各种类型的数据以支持网络设备60的操作。这些数据的示例包括:用于在网络设备60上操作的任何计算机程序。
上述本申请实施例揭示的方法可以应用于所述处理器62中,或者由所述处理器62实现。所述处理器62可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过所述处理器62中的硬件的集成逻辑电路或者软件形式的指令完成。上述的所述处理器62可以是通用处理器、数字数据处理器(DSP,Digital Signal Processor),或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。所述处理器62可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者任何常规的处理器等。结合本申请实施例所公开的方法的步骤,可以直接体现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于存储介质中,该存储介质位于存储器63,所述处理器62读取存储器63中的信息,结合其硬件完成前述方法的步骤。
在示例性实施例中,网络设备60可以被一个或多个应用专用集成电路(ASIC,Application Specific Integrated Circuit)、DSP、可编程逻辑器件(PLD,Programmable Logic Device)、复杂可编程逻辑器件(CPLD,Complex Programmable Logic Device)、现场可编程门阵列(FPGA,Field-Programmable Gate Array)、通用处理器、控制器、微控制器(MCU,Micro Controller Unit)、微处理器(Microprocessor)、或者其他电子元件实现,用于执行前述方法。
可以理解,本申请实施例的存储器(存储器63)可以是易失性存储器或者非易失性存储器,也可包括易失性和非易失性存储器两者。其中,非易失性存储器可以是只读存储器(ROM,Read Only Memory)、可编程只读存储器(PROM,Programmable Read-Only Memory)、可擦除可编程只读存储器(EPROM,Erasable Programmable Read-Only Memory)、电可擦除可编程只读存储器(EEPROM,Electrically Erasable Programmable Read-Only Memory)、
磁性随机存取存储器(FRAM,ferromagnetic random access memory)、快闪存储器(Flash Memory)、磁表面存储器、光盘、或只读光盘(CD-ROM,Compact Disc Read-Only Memory);磁表面存储器可以是磁盘存储器或磁带存储器。易失性存储器可以是随机存取存储器(RAM,Random Access Memory),其用作外部高速缓存。通过示例性但不是限制性说明,许多形式的RAM可用,例如静态随机存取存储器(SRAM,Static Random Access Memory)、同步静态随机存取存储器(SSRAM,Synchronous Static Random Access Memory)、动态随机存取存储器(DRAM,Dynamic Random Access Memory)、同步动态随机存取存储器(SDRAM,Synchronous Dynamic Random Access Memory)、双倍数据速率同步动态随机存取存储器(DDRSDRAM,Double Data Rate Synchronous Dynamic Random Access Memory)、增强型同步动态随机存取存储器(ESDRAM,Enhanced Synchronous Dynamic Random Access Memory)、同步连接动态随机存取存储器(SLDRAM,SyncLink Dynamic Random Access Memory)、直接内存总线随机存取存储器(DRRAM,Direct Rambus Random Access Memory)。本申请实施例描述的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
在示例性实施例中,本申请实施例还提供了一种存储介质,即计算机存储介质,具体为计算机可读存储介质,例如包括存储计算机程序的存储器,上述计算机程序可由网络设备60的处理器62执行,以完成前述网络设备侧方法所述步骤。计算机可读存储介质可以是FRAM、ROM、PROM、EPROM、EEPROM、Flash Memory、磁表面存储器、光盘、或CD-ROM等存储器。
需要说明的是:“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。
另外,本申请实施例所记载的技术方案之间,在不冲突的情况下,可以任意组合。
以上所述,仅为本申请的较佳实施例而已,并非用于限定本申请的保护范围。
Claims (17)
- 一种信息接收方法,应用于第一节点,所述方法包括:接收第二节点发送的第一消息;其中,所述第一消息用于请求所述第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功能;其中,所述感知功能包括以下功能中至少之一:训练通感感知模型、新建通感感知模型、增加通感感知模型、更新通感感知模型、删除通感感知模型中至少之一;其中,所述通感感知模型由本节点使用或由本节点提供给其他节点;进行模型训练,得到通感感知模型,以供本节点使用或由本节点提供给其他节点;推算出被测物体的感知结果信息,以供本节点使用或由本节点提供给其他节点;确定用于感知的资源信息,以供本节点使用或由本节点提供给其他节点;进行感知的授权信息验证和/或鉴权;其中,所述授权和/或鉴权信息由本节点使用或由本节点提供给其他节点;感知的能力上报、协商、降维和配置中的至少一个,其中,感知的能力配置由本节点使用或由本节点提供给其他节点;选择感知的感知模式,以供本节点使用或由本节点提供给其他节点;选择感知节点,以供本节点使用或由本节点提供给其他节点;进行感知服务质量QoS参数转换;其中,所述感知QoS参数由本节点使用或由本节点提供给其他节点;将感知业务类型或标识、QoS要求和感知测量数据上报周期信息中至少之一传递到执行感知的其他节点;若感知需求改变,终端或应用触发对应的修改流程;触发、激活、新建感知业务中的至少一种;终止、去激活、删除感知业务中的至少一种;根据感知业务需求控制其他节点执行感知;收集、上报、处理感知测量数据中的至少一种。
- 根据权利要求1所述的方法,其中,所述第一消息,携带有以下至少之一:所述第一信息的类型;所述第一信息的发送周期;所述第一信息的发送数量;触发所述第一信息的事件信息;所述第一信息用于指示以下至少之一:两个测量节点之间距离和角度;被测物体到所述第一节点的入射角度、距离;设置有被测物体的终端的无线电测量参数;设置有被测物体的终端的多普勒变化率;被测物体处的功率密度;被测物体的反射点个数;被测物体的反射点的间距;被测物体的反射点的图样、对应的功率密度和接收天线增益。
- 根据权利要求1所述的方法,其中,所述感知结果信息,包括以下至少之一:目标对象的定位;测距;测速;成像;检测;物体识别;虚拟环境重构。
- 根据权利要求1所述的方法,其中,所述用于感知的资源信息,包括以下至少之一:周期、起始位置、持续时间或者每个周期的持续时长中至少一个;资源的激活时间点;资源的激活时间范围;资源可使用的地理范围;资源可使用的高度范围;资源可使用的频段范围;资源可使用的区域范围;资源可使用的频段范围;资源可使用的发送对象。
- 根据权利要求1所述的方法,其中,所述通感感知模型的训练和/或推理位于以下至少之一:操作维护管理OAM;下一代基站gNB;无线网络节点;gNB的中心单元CU;人工智能AI控制应用功能AF的至少一种网元节点;终端侧的AI服务器;终端侧。
- 根据权利要求1所述的方法,其中,所述方法还包括:接收第三节点发送的第二消息;其中,所述第二消息包括第二信息;根据所述第二信息,进行模型训练得到所述通感感知模型;其中,所述第二信息,包括以下至少之一:感知功能的类型;感知的精度;感知的物体类型;AF的标识ID;终端的ID;当前小区的带宽;当前小区的信道质量;所述第二信息用于以下至少之一:提供给所述第一节点用于感知的资源信息,提供给所述第一节点用于感知的授权信息验证和/或鉴权。
- 根据权利要求6所述的方法,其中,所述感知功能的类型,包括以下至少之一:目标对象的定位;测距;测速;成像;检测;物体识别;虚拟环境重构;所述感知的精度,包括以下至少之一:距离分辨率;距离精度;角度分辨率;角度精度。
- 根据权利要求6所述的方法,其中,所述根据所述第二信息,进行模型训练得到所述通感感知模型,包括:如果满足当前小区的带宽大于或等于第一门限、当前小区的信道质量大于或等于第二门限和感知的精度小于或等于第三门限中至少之一,则根据频率m进行目标物体的反射点收集,并利用收集的反射点,执行特定功能的通感感知模型的训练;m为有理数;或者,如果满足当前小区的带宽小于或等于第一门限、且当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则调整通信和感知之间的资源分配的权重,以进行模型训练得到所述通感感知模型;和/或,调整用于感知的带宽大小。
- 根据权利要求6所述的方法,其中,所述接收第二节点发送的第一消息,包括:如果满足当前小区的带宽小于或等于第一门限、当前小区的信道质量小于或等于第二门限和感知的精度大于或等于第三门限中至少之一,则向所述第二节点发送第三消息;所述第三消息用于请求所述第二节点向所述第一节点发送所述第一消息;接收所述第二节点发送的所述第一消息。
- 根据权利要求1所述的方法,其中,所述方法还包括:向终端发送第四消息,所述第四消息用于请求获取所述第一信息;接收所述终端返回的所述第一信息;向所述第二节点发送所述第一信息,以供所述第二节点进行模型训练得到所述通感感知模型。
- 根据权利要求10所述的方法,其中,所述方法还包括:向与所述第一节点相邻的节点发送第五消息;所述第五消息用于请求获取第三信息;接收与所述第一节点相邻的节点返回的所述第三信息;将所述第三信息和所述第一信息发送给所述第二节点,以供所述第二节点进行模型训练得到所述通感感知模型。
- 根据权利要求11所述的方法,其中,所述第三信息,包括以下至少之一:模型训练的触发方式;模式训练的类型;训练的方式;模型是否需要传递;模型传递的具体参数;模型参数传递的方式。
- 根据权利要求1所述的方法,其中,所述方法还包括:利用所述通感感知模型,执行感知预测,得到预测结果。
- 根据权利要求13所述的方法,其中,所述方法还包括:向与所述第一节点相邻的节点或所述第二节点发送反馈信息;所述反馈信息携带有预测结果;所述预测结果包括感知的被测物体更新后的状态和/或所述通感感知模型的更新方式。
- 一种信息接收装置,包括:接收单元,用于接收第二节点发送的第一消息;其中,所述第一消息用于请求第一节点向所述第二节点提供第一信息;所述第一信息用于所述第二节点执行感知功能;其中,所述感知功能包括以下功能中至少之一:训练通感感知模型、新建通感感知模型、增加通感感知模型、更新通感感知模型、删除通感感知模型中至少之一;其中,所述通感感知模型由本节点使用或由本节点提供给其他节点;进行模型训练,得到通感感知模型,以供本节点使用或由本节点提供给其他节点;推算出被测物体的感知结果信息,以供本节点使用或由本节点提供给其他节点;确定用于感知的资源信息,以供本节点使用或由本节点提供给其他节点;进行感知的授权信息验证和/或鉴权;其中,所述授权和/或鉴权信息由本节点使用或由本节点提供给其他节点;感知的能力上报、协商、降维和配置中的至少一个,其中,感知的能力配置由本节点使用或由本节点提供给其他节点;选择感知的感知模式,以供本节点使用或由本节点提供给其他节点;选择感知节点,以供本节点使用或由本节点提供给其他节点;进行感知QoS参数转换;其中,所述感知QoS参数由本节点使用或由本节点提供给其他节点;将感知业务类型或标识、QoS要求和感知测量数据上报周期信息中至少之一传递到执行感知的其他节点;若感知需求改变,终端或应用触发对应的修改流程;触发、激活、新建感知业务中的至少一种;终止、去激活、删除感知业务中的至少一种;根据感知业务需求控制其他节点执行感知;收集、上报、处理感知测量数据中的至少一种。
- 一种网络设备,包括处理器和用于存储能够在处理器上运行的计算机程序的存储器,其中,所述处理器用于运行所述计算机程序时,执行权利要求1至14任一项所述方法的步骤。
- 一种计算机可读存储介质,其上存储有计算机程序,所述计算机程序被处理器执行时实现权利要求1至14任一项所述方法的步骤。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310342554.9 | 2023-03-31 | ||
| CN202310342554.9A CN118741508A (zh) | 2023-03-31 | 2023-03-31 | 信息接收方法、装置、设备及存储介质 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2024199534A1 true WO2024199534A1 (zh) | 2024-10-03 |
Family
ID=92862737
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2024/085391 Ceased WO2024199534A1 (zh) | 2023-03-31 | 2024-04-01 | 信息接收方法、装置、设备及存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN118741508A (zh) |
| WO (1) | WO2024199534A1 (zh) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103546888A (zh) * | 2012-07-11 | 2014-01-29 | 华为技术有限公司 | 一种在lte系统中的感知实现方法及装置 |
| CN115278639A (zh) * | 2022-07-22 | 2022-11-01 | 中国联合网络通信集团有限公司 | 感知数据获取方法、装置、设备及存储介质 |
| CN115690558A (zh) * | 2014-09-16 | 2023-02-03 | 华为技术有限公司 | 数据处理的方法和设备 |
| CN115706955A (zh) * | 2021-08-04 | 2023-02-17 | 华为技术有限公司 | 提供通信感知业务的方法、通信装置和系统 |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110831029B (zh) * | 2018-08-13 | 2021-06-22 | 华为技术有限公司 | 一种模型的优化方法和分析网元 |
| CN115776694A (zh) * | 2021-09-06 | 2023-03-10 | 维沃移动通信有限公司 | 服务质量特征参数确定、数据发送方法、装置及设备 |
| CN115802370A (zh) * | 2021-09-10 | 2023-03-14 | 华为技术有限公司 | 一种通信方法及装置 |
| CN115765819A (zh) * | 2022-11-11 | 2023-03-07 | 北京邮电大学 | 面向通信感知一体化的智能超表面的波束赋形方法和装置 |
-
2023
- 2023-03-31 CN CN202310342554.9A patent/CN118741508A/zh active Pending
-
2024
- 2024-04-01 WO PCT/CN2024/085391 patent/WO2024199534A1/zh not_active Ceased
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103546888A (zh) * | 2012-07-11 | 2014-01-29 | 华为技术有限公司 | 一种在lte系统中的感知实现方法及装置 |
| CN115690558A (zh) * | 2014-09-16 | 2023-02-03 | 华为技术有限公司 | 数据处理的方法和设备 |
| CN115706955A (zh) * | 2021-08-04 | 2023-02-17 | 华为技术有限公司 | 提供通信感知业务的方法、通信装置和系统 |
| CN115278639A (zh) * | 2022-07-22 | 2022-11-01 | 中国联合网络通信集团有限公司 | 感知数据获取方法、装置、设备及存储介质 |
Non-Patent Citations (1)
| Title |
|---|
| VIVO CHINA TELECOM CHINA UNICOM: "Introduction of network based Sensing in 5G Advanced ", 3GPP DRAFT; S2-2106008, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), no. 20210730, 10 August 2021 (2021-08-10), FR, XP052053654 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118741508A (zh) | 2024-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP4029305B1 (en) | Apparatus for radio access network data collection | |
| US20230403606A1 (en) | Managing resources in a radio access network | |
| WO2021077365A1 (en) | Methods and apparatuses for interference mitigation and related intelligent network management | |
| CN116709556A (zh) | 联合通信和感测中的资源分配 | |
| US20260023159A1 (en) | Configuration for radio sensing | |
| CN113498071A (zh) | 预测无线通信链路的未来服务质量的方法、装置和程序 | |
| DE102021112407A1 (de) | Verfahren und Vorrichtung zur Positionierung eines Benutzergerätes | |
| CN113905385A (zh) | 无线电资源参数配置 | |
| KR20240007650A (ko) | 사이드링크를 통한 시간 민감 네트워킹 지원 | |
| WO2018054341A1 (zh) | 一种无线网络中的数据传输方法及装置 | |
| CN117632463A (zh) | 一种计算任务的分割方法及相关装置 | |
| US20240397362A1 (en) | Enhanced performance measurements related to one-way uplink packet delay in wireless communications | |
| CN118786357A (zh) | 终端设备的定位中的定时误差校正 | |
| WO2025032229A1 (en) | User equipment and network entity with report configuration message for channel measurement and reporting at the ue | |
| CN121420590A (zh) | 无线网络中的切片编排的网络切片可行性评定 | |
| WO2026012070A1 (zh) | 通信方法和通信装置 | |
| WO2024199534A1 (zh) | 信息接收方法、装置、设备及存储介质 | |
| CN120113268A (zh) | 无线通信中的模型更新技术 | |
| JP7734136B2 (ja) | 端末及び通信方法 | |
| CN120567290B (zh) | 一种多轨道协同的自适应测量参数配置方法及装置 | |
| US20250330831A1 (en) | Perception assisted beam management for wireless communication with vehicles | |
| CN119450516B (zh) | 模型生命周期管理方法、装置、通信设备和存储介质 | |
| US20230275632A1 (en) | Methods for beam coordination in a near-field operation with multiple transmission and reception points (trps) | |
| US20260046647A1 (en) | Data for training of artificial intelligence models for beam prediction | |
| TW202450268A (zh) | 感知無線通訊系統中的階層式分割及感測器資料聚合 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 24778319 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |