WO2024082195A1 - 一种基于ai模型的终端定位方法及装置 - Google Patents
一种基于ai模型的终端定位方法及装置 Download PDFInfo
- Publication number
- WO2024082195A1 WO2024082195A1 PCT/CN2022/126286 CN2022126286W WO2024082195A1 WO 2024082195 A1 WO2024082195 A1 WO 2024082195A1 CN 2022126286 W CN2022126286 W CN 2022126286W WO 2024082195 A1 WO2024082195 A1 WO 2024082195A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- impulse response
- channel impulse
- model
- module
- bit information
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 117
- 230000004044 response Effects 0.000 claims abstract description 344
- 238000012545 processing Methods 0.000 claims abstract description 68
- 238000004891 communication Methods 0.000 claims abstract description 38
- 238000013139 quantization Methods 0.000 claims description 62
- 238000012549 training Methods 0.000 claims description 44
- 230000006870 function Effects 0.000 claims description 42
- 230000006835 compression Effects 0.000 claims description 28
- 238000007906 compression Methods 0.000 claims description 28
- 230000006837 decompression Effects 0.000 claims description 26
- 238000004590 computer program Methods 0.000 claims description 22
- 230000015654 memory Effects 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 abstract description 12
- 238000013473 artificial intelligence Methods 0.000 description 231
- 238000010586 diagram Methods 0.000 description 16
- 238000012790 confirmation Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 8
- 238000005259 measurement Methods 0.000 description 6
- 239000004065 semiconductor Substances 0.000 description 5
- 229910044991 metal oxide Inorganic materials 0.000 description 4
- 150000004706 metal oxides Chemical class 0.000 description 4
- 239000013589 supplement Substances 0.000 description 4
- 101000827703 Homo sapiens Polyphosphoinositide phosphatase Proteins 0.000 description 3
- 102100023591 Polyphosphoinositide phosphatase Human genes 0.000 description 3
- 101100233916 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) KAR5 gene Proteins 0.000 description 3
- 230000000295 complement effect Effects 0.000 description 3
- 238000010295 mobile communication Methods 0.000 description 3
- 229910000577 Silicon-germanium Inorganic materials 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- JBRZTFJDHDCESZ-UHFFFAOYSA-N AsGa Chemical compound [As]#[Ga] JBRZTFJDHDCESZ-UHFFFAOYSA-N 0.000 description 1
- 101001121408 Homo sapiens L-amino-acid oxidase Proteins 0.000 description 1
- 102100026388 L-amino-acid oxidase Human genes 0.000 description 1
- 101100012902 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) FIG2 gene Proteins 0.000 description 1
- LEVVHYCKPQWKOP-UHFFFAOYSA-N [Si].[Ge] Chemical compound [Si].[Ge] LEVVHYCKPQWKOP-UHFFFAOYSA-N 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
Definitions
- the present disclosure relates to the field of communication technology, and in particular to a terminal positioning method and device based on an AI model.
- the requirements for positioning accuracy are very high.
- AI artificial intelligence
- the AI model After the artificial intelligence (AI) model is trained, it can be deployed on the network side or the terminal side to use the trained AI model for reasoning. During reasoning, the input to the AI model is still the measurement result. Based on the measurement result, the AI model will output the corresponding terminal position.
- the AI model inputs the channel impulse response.
- the terminal needs to obtain the channel impulse response based on the measurement quantity, quantify the channel impulse response, and then feed it back to the network.
- the feedback amount will be relatively large, and therefore more transmission resources will be occupied.
- an embodiment of the present disclosure provides a terminal positioning method based on an AI model, the method being applied to a terminal side, the AI model including a first AI model and a second AI model, the first AI model including a first partial module and a second partial module, the first partial module being deployed on the terminal side, the method including:
- the quantized bit information of the first channel impulse response is sent to the network side.
- the first part of the modules includes a quantization module
- the step of inputting the first channel impulse response into the first part of the modules of the first AI model for processing to obtain quantized bit information of the first channel impulse response includes:
- the input first channel impulse response is quantized based on the quantization module to obtain bit information of the quantized first channel impulse response.
- the first part of the modules includes a compression module and a quantization module
- the step of inputting the first channel impulse response into the first part of the modules of the first AI model for processing to obtain quantized bit information of the first channel impulse response includes:
- the compressed channel impulse response is quantized based on the quantization module to obtain quantized bit information of the first channel impulse response.
- the first part module and the second part module included in the first AI model are obtained by joint training.
- the training objective function of the first AI model is the difference between a first channel impulse response and a second channel impulse response
- the second channel impulse response is a channel impulse response obtained after quantization and dequantization of the first channel impulse response
- the training objective function of the first AI model is the difference between a first channel impulse response and a second channel impulse response
- the second channel impulse response is a channel impulse response obtained after compression, quantization, dequantization, and decompression of the first channel impulse response
- an embodiment of the present disclosure provides a terminal positioning method based on an AI model, the method being applied to a network side, the AI model including a first AI model and a second AI model, the first AI model including a first part module and a second part module, the second part module being deployed on the network side, the method including:
- the second channel impulse response is input into the second AI model for processing to obtain the positioning information of the terminal.
- the second part module includes a dequantization module
- the step of inputting the bit information into the second part module of the first AI model for processing to obtain a second channel impulse response includes:
- the bit information is dequantized based on the dequantization module to obtain the second channel impulse response.
- the second part module includes a dequantization module and a decompression module
- the step of inputting the bit information into the second part module of the first AI model for processing to obtain a second channel impulse response includes:
- the compressed channel impulse response is processed based on the decompression module to obtain the second channel impulse response.
- the training objective function of the first AI model is the difference between a first channel impulse response and a second channel impulse response
- the second channel impulse response is a channel impulse response obtained after quantization and dequantization of the first channel impulse response
- the training objective function of the first AI model is the difference between a first channel impulse response and a second channel impulse response
- the second channel impulse response is a channel impulse response obtained after compression, quantization, dequantization, and decompression of the first channel impulse response
- the first part module and the second part module included in the AI model are obtained through joint training.
- the positioning information is positioning coordinates or parameters required for positioning.
- the parameters required for positioning are any of the following:
- an embodiment of the present disclosure provides a terminal positioning device based on an AI model, the device is applied to a terminal side, the AI model includes a first AI model and a second AI model, the first AI model includes a first part module and a second part module, the first part module is deployed on the terminal side, and the device includes:
- a processing module configured to input the first channel impulse response into the first part module of the first AI model for processing, so as to obtain quantized bit information of the first channel impulse response;
- the sending module is used to send the bit information of the quantized first channel impulse response to the network side.
- the first part of the modules includes a quantization module, and the processing module is further configured to include:
- the input first channel impulse response is quantized based on the quantization module to obtain bit information of the quantized first channel impulse response.
- the first part of the modules includes a compression module and a quantization module, and the processing module is further used to:
- the compressed channel impulse response is quantized based on the quantization module to obtain quantized bit information of the first channel impulse response.
- the first part module and the second part module included in the first AI model are obtained by joint training.
- the training objective function of the first AI model is the difference between a first channel impulse response and a second channel impulse response
- the second channel impulse response is a channel impulse response obtained after quantization and dequantization of the first channel impulse response
- the training objective function of the first AI model is the difference between a first channel impulse response and a second channel impulse response
- the second channel impulse response is a channel impulse response obtained after compression, quantization, dequantization, and decompression of the first channel impulse response
- an embodiment of the present disclosure provides a terminal positioning device based on an AI model, the device is applied to a network side, the AI model includes a first AI model and a second AI model, the first AI model includes a first part module and a second part module, the second part module is deployed on the network side, and the device includes:
- a receiving module used for receiving the quantized bit information of the first channel impulse response sent by the terminal side
- a processing module configured to input the bit information into the second part module of the first AI model for processing to obtain a second channel impulse response
- the processing module is further used to input the second channel impulse response into the second AI model for processing to obtain the positioning information of the terminal.
- the second part of the modules includes a dequantization module
- the processing module is further configured to perform dequantization processing on the bit information based on the dequantization module to obtain the second channel impulse response.
- the second part of modules includes a dequantization module and a decompression module, and the processing module is further used to:
- the compressed channel impulse response is processed based on the decompression module to obtain the second channel impulse response.
- the training objective function of the first AI model is the difference between a first channel impulse response and a second channel impulse response
- the second channel impulse response is a channel impulse response obtained after quantization and dequantization of the first channel impulse response
- the training objective function of the first AI model is the difference between a first channel impulse response and a second channel impulse response
- the second channel impulse response is a channel impulse response obtained after compression, quantization, dequantization, and decompression of the first channel impulse response
- the first part module and the second part module included in the AI model are obtained through joint training.
- the positioning information is positioning coordinates or parameters required for positioning.
- the parameters required for positioning are any of the following:
- an embodiment of the present disclosure provides a computer-readable storage medium for storing instructions used by the above-mentioned AI model-based terminal positioning device.
- the AI model-based terminal positioning device executes the method described in the first or second aspect above.
- an embodiment of the present disclosure further provides a computer program product comprising a computer program, which, when executed on a computer, enables the computer to execute the method described in the first aspect or the second aspect above.
- an embodiment of the present disclosure provides a chip system, which includes at least one processor and an interface, for supporting a communication device to implement the functions involved in the first aspect or the second aspect, for example, determining or processing at least one of the data and information involved in the above method.
- the chip system also includes a memory, which is used to store computer programs and data necessary for the communication device.
- the chip system can be composed of chips, or it can include chips and other discrete devices.
- an embodiment of the present disclosure further provides a computer program, which, when executed on a computer, enables the computer to execute the method described in the first or second aspect above.
- FIG1 is a schematic diagram of the architecture of a communication system provided by an embodiment of the present disclosure.
- FIG2 is a schematic diagram of a process flow of a terminal positioning method based on an AI model provided in an embodiment of the present disclosure
- FIG3 is a schematic diagram of a flow chart of another terminal positioning method based on an AI model provided in an embodiment of the present disclosure
- FIG4 is a schematic diagram of a flow chart of another terminal positioning method based on an AI model provided in an embodiment of the present disclosure
- FIG5 is a schematic diagram of a flow chart of another terminal positioning method based on an AI model provided in an embodiment of the present disclosure
- FIG6 is a schematic diagram of a flow chart of another terminal positioning method based on an AI model provided in an embodiment of the present disclosure
- FIG7 is a schematic diagram of a flow chart of another terminal positioning method based on an AI model provided in an embodiment of the present disclosure
- FIG8 is a schematic diagram of the structure of a terminal positioning device based on an AI model provided in an embodiment of the present disclosure
- FIG9 is a schematic diagram of the structure of another terminal positioning device based on an AI model provided in an embodiment of the present disclosure
- FIG10 is a schematic diagram of the structure of a communication device provided in an embodiment of the present disclosure.
- FIG. 11 is a schematic diagram of the structure of a chip provided in an embodiment of the present disclosure.
- FIG. 1 is a schematic diagram of the architecture of a communication system provided by an embodiment of the present disclosure.
- the communication system may include, but is not limited to, a network device and a terminal device.
- the number and form of devices shown in FIG. 1 are only used as examples and do not constitute a limitation on the embodiment of the present disclosure. In actual applications, two or more network devices and two or more terminal devices may be included.
- the communication system shown in FIG. 1 includes, for example, a network device 11 and a terminal device 12.
- LTE long term evolution
- 5G fifth generation
- NR 5G new radio
- the network device 11 in the embodiment of the present disclosure is an entity on the network side for transmitting or receiving signals.
- the network device 101 may be an evolved NodeB (eNB), a transmission point (TRP), a next generation NodeB (gNB) in an NR system, a base station in other future mobile communication systems, or an access node in a wireless fidelity (WiFi) system.
- eNB evolved NodeB
- TRP transmission point
- gNB next generation NodeB
- WiFi wireless fidelity
- the embodiment of the present disclosure does not limit the specific technology and specific device form adopted by the network device.
- the network device provided in the embodiment of the present disclosure may be composed of a central unit (CU) and a distributed unit (DU), wherein the CU may also be referred to as a control unit.
- CU central unit
- DU distributed unit
- the CU-DU structure may be used to split the protocol layer of the network device, such as a base station, and the functions of some protocol layers are placed in the CU for centralized control, and the functions of the remaining part or all of the protocol layers are distributed in the DU, and the DU is centrally controlled by the CU.
- the terminal device 12 in the disclosed embodiment is an entity on the user side for receiving or transmitting signals, such as a mobile phone.
- the terminal device may also be referred to as a terminal device (terminal), a user equipment (UE), a mobile station (MS), a mobile terminal device (MT), etc.
- the terminal device may be a car with communication function, a smart car, a mobile phone (mobile phone), a wearable device, a tablet computer (Pad), a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control (industrial control), a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in smart grid (smart grid), a wireless terminal device in transportation safety (transportation safety), a wireless terminal device in a smart city (smart city), a wireless terminal device in a smart home (smart home), etc.
- the embodiments of the present disclosure do not limit the specific technology and specific device form adopted by the terminal device.
- the communication system described in the embodiment of the present disclosure is for the purpose of more clearly illustrating the technical solution of the embodiment of the present disclosure, and does not constitute a limitation on the technical solution provided by the embodiment of the present disclosure.
- a person skilled in the art can know that with the evolution of the system architecture and the emergence of new business scenarios, the technical solution provided by the embodiment of the present disclosure is also applicable to similar technical problems.
- the requirements for positioning accuracy are very high.
- the artificial intelligence (AI) model After the artificial intelligence (AI) model is trained, it can be deployed on the network side or the terminal side to use the trained AI model for reasoning. During reasoning, the input to the AI model is still the measurement result. Based on the measurement result, the AI model will output the corresponding terminal position. In the current positioning solution, the AI model inputs the channel impulse response. The terminal needs to obtain the channel impulse response based on the measurement amount, quantify the channel impulse response and feed it back to the network. In this scenario, the feedback amount will be relatively large, so more transmission resources are occupied.
- Figure 2 is a flow chart of a terminal positioning method based on an AI model provided in an embodiment of the present disclosure, and the method is applied to the terminal side. As shown in Figure 2, the method may include but is not limited to the following steps:
- Step S201 input the first channel impulse response into the first partial module of the first AI model for processing to obtain quantized bit information of the first channel impulse response.
- the AI model includes a first AI model and a second AI model, the first AI model includes a first partial module and a second partial module, the first partial module is deployed on the terminal side, and the second partial module is deployed on the network side.
- a first channel impulse response is obtained on the terminal side according to the measured value, and then the first channel impulse response is input into a first part module of a first AI model on the terminal side to obtain quantized bit information of the first channel impulse response. This processing is executed on the terminal side.
- Step S202 sending the quantized bit information of the first channel impulse response to the network side.
- the quantized bit information obtained by the terminal side based on the first part module of the first AI model is sent to the second part module of the first AI model on the network side, so that the second part module on the network side can obtain the positioning information of the terminal based on the bit information quantized by the first channel impulse response.
- the input of the AI model is the first channel impulse response on the terminal side
- the output of the AI model is the positioning information of the terminal executed on the network side.
- the first part module is deployed on the terminal side, and is used to compress the first channel impulse response, and quantize the compressed first channel impulse response to obtain the quantized data of the first channel impulse response.
- the second part module is deployed on the network side, and is used to determine the positioning information of the terminal using the second part module according to the received quantized data of the first channel impulse response.
- bit information refers to one or more bits, or refers to data composed of one or more bits; or in other words, bit information is information in bit form.
- the terminal positioning information includes but is not limited to: positioning coordinates or parameters required for determining the terminal positioning.
- the first part module of the first AI model in the AI model deployed on the terminal side inputs the first channel impulse response into the first part module of the first AI model for processing, obtains quantized bit information of the first channel impulse response, and sends the quantized bit information of the first channel impulse response to the second part module of the first AI model on the network side.
- the first part module of the first AI model on the terminal side assists the second part module of the first AI model on the network side and the second AI model to perform terminal positioning, thereby reducing the input amount of the first channel impulse response on the network side, thereby reducing the transmission resource occupancy on the network side.
- FIG3 is a flow chart of another terminal positioning method based on an AI model provided by the embodiment of the present disclosure. The method is applied to the terminal side. As shown in FIG3, the terminal positioning method based on the AI model may include the following steps:
- Step S301 quantizing the input first channel impulse response based on the quantization module to obtain bit information of the quantized first channel impulse response.
- the AI model includes a first AI model and a second AI model
- the first AI model includes a first partial module and a second partial module.
- the first partial module of the first AI model is deployed on the terminal side, wherein the first partial module includes a quantization module, and the second partial module of the first AI model is deployed on the network side.
- a first channel impulse response is obtained on the terminal side according to the measured value, and then the first channel impulse response is input into a quantization module of a first part module of a first AI model on the terminal side. Based on the quantization module, the input first channel impulse response is quantized to obtain bit information of the quantized first channel impulse response. This processing is executed on the terminal side.
- Step S302 Send the quantized bit information of the first channel impulse response to the network side.
- the quantized bit information obtained by the terminal side based on the first part module of the first AI model is sent to the second part module of the first AI model on the network side, so that the second part module on the network side can obtain the positioning information of the terminal based on the bit information quantized by the first channel impulse response.
- the input of the AI model is the first channel impulse response on the terminal side
- the output of the AI model is the positioning information of the terminal executed on the network side.
- the terminal positioning information includes but is not limited to: positioning coordinates or parameters required for terminal positioning.
- the first part module of the first AI model in the AI model deployed on the terminal side executes the input of the first channel impulse response to the quantization module in the first part module of the first AI model for processing to obtain the quantized bit information of the first channel impulse response, and sends the quantized bit information of the first channel impulse response to the second part module of the first AI model on the network side.
- the quantization module of the first part module of the first AI model on the terminal side assists the second part module of the first AI model on the network side and the second AI model to perform terminal positioning, thereby reducing the input amount of the first channel impulse response on the network side, and thereby reducing the transmission resource occupancy on the network side.
- FIG4 is a flow chart of another terminal positioning method based on an AI model provided by the embodiment of the present disclosure. The method is applied to the terminal side. As shown in FIG4, the terminal positioning method based on the AI model may include the following steps:
- Step S4011 compressing the first channel impulse response based on the compression module to obtain a compressed channel impulse response.
- the AI model comprises a first AI model and a second AI model, wherein the first AI model comprises a first part module and a second part module, and the first part module of the AI model is deployed on the terminal side, wherein the first part module comprises a compression module and a quantization module, and the second part module is deployed on the network side.
- a first channel impulse response is obtained on the terminal side according to the measured amount, and then the first channel impulse response is compressed based on the compression module to obtain a compressed channel impulse response.
- Step S4012 performing quantization processing on the compressed channel impulse response based on the quantization module to obtain quantized bit information of the first channel impulse response.
- the compressed channel impulse response is input into the quantization module of the first part module on the terminal side, and the input compressed channel impulse response is quantized based on the quantization module to obtain the quantized bit information of the first channel impulse response.
- the processing is executed on the terminal side.
- Step S402 sending the quantized bit information of the first channel impulse response to the network side.
- the quantized first channel impulse response bit information obtained by the terminal side based on the quantization module in the first part module is sent to the second part module on the network side, so that the second part module on the network side can obtain the positioning information of the terminal based on the bit information quantized by the first channel impulse response.
- the input of the AI model is the first channel impulse response on the terminal side
- the output of the AI model is the positioning information of the terminal executed by the network side.
- the terminal positioning information includes but is not limited to: positioning coordinates or parameters required for terminal positioning.
- the first part module of the AI model deployed on the terminal side first compresses the first channel impulse response based on the compression module of the first part module, inputs the compressed information into the quantization module of the first part module, and quantizes the input compressed first channel impulse response based on the quantization module to obtain the quantized bit information of the first channel impulse response, and sends the quantized first channel impulse response bit information to the second part module of the AI model on the network side.
- the compression module and quantization module of the first part module of the first AI model on the terminal side assist the second part module and the second AI model of the first AI model on the network side in terminal positioning, thereby further reducing the input amount of the first channel impulse response on the network side, and thereby reducing the transmission resource occupancy on the network side.
- the first AI model deploys the first part module on the terminal side and the second part module on the network side
- the essence is still that the first part module and the second part module constitute a complete first AI model.
- the first part module on the terminal side and the second part module on the network side complement each other, and the combination of the two realizes the positioning of the terminal, because the first part module on the terminal side and the second part module on the network side need to be jointly trained during training, that is, the first AI model is obtained in the same training process.
- the specific training process will not be described one by one in the embodiments of the present disclosure.
- the disclosed embodiment provides another terminal positioning method based on an AI model.
- the training objective function of the first AI model is the difference between the first channel impulse response and the second channel impulse response
- the second channel impulse response is the channel impulse response obtained after quantization and dequantization of the first channel impulse response.
- the training of the first AI model minimizes the difference between the first channel impulse response and the second channel impulse response.
- the minimum difference is that the impulse response of the first channel is completely consistent with the impulse response of the second channel.
- the minimum difference is that the difference between the impulse response of the first channel and the impulse response of the second channel is less than a preset difference.
- the specific embodiment of the present disclosure does not limit the preset difference, and can be flexibly configured according to specific application scenarios.
- the embodiment of the present disclosure provides another terminal positioning method based on an AI model.
- the training objective function of the first AI model is the difference between the first channel impulse response and the second channel impulse response
- the second channel impulse response is the channel impulse response obtained after compression, quantization, dequantization, and decompression of the first channel impulse response.
- the training of the first AI model minimizes the difference between the first channel impulse response and the second channel impulse response.
- Figure 5 is a flow chart of a terminal positioning method based on an AI model provided by an embodiment of the present disclosure, and the method is applied to the network side. As shown in Figure 5, the method may include but is not limited to the following steps:
- Step S501 receiving quantized bit information of a first channel impulse response sent by a terminal side.
- the AI model includes a first AI model and a second AI model.
- the first AI model includes a first partial module and a second partial module.
- the first partial module is deployed on the terminal side, and the second partial module is deployed on the network side.
- the terminal side obtains a first channel impulse response according to the measured amount, and then inputs the first channel impulse response into the first part module of the first AI model on the terminal side to obtain quantized bit information of the first channel impulse response and send it to the network side.
- the processing is performed on the terminal side.
- the network side receives the quantized bit information of the first channel impulse response.
- Step S502 input the bit information into the second part module of the first AI model for processing to obtain a second channel impulse response.
- the received bit information of the quantized first channel impulse response is processed by the second part module of the first AI model to obtain a second channel impulse response.
- the second channel impulse response is an inverse process.
- the bit information sent by the terminal side is obtained by quantization processing, and then the network side needs to perform dequantization to obtain the second channel impulse response.
- Step S503 input the second channel impulse response into the second AI model for processing to obtain the positioning information of the terminal.
- the second channel impulse response obtained after processing the second part module of the first AI model is used as the input of the second AI model, and the second AI model calculates the positioning information of the terminal according to the second channel impulse response.
- the positioning information of the terminal includes but is not limited to: positioning coordinates or parameters required for terminal positioning.
- the parameters required for positioning are any of the following: signal arrival time, signal arrival angle, non-line-of-sight propagation (NLOS) or line-of-sight propagation (LOS).
- the second part model on the network side receives the quantized bit information of the first channel impulse response, and the second part module of the first AI model processes the bit information to obtain the second channel impulse response, and uses the second channel impulse response as the input of the second AI model, which performs the confirmation of the terminal positioning information.
- the first part module of the AI model on the terminal side assists the second part module of the first AI model on the network side and the second AI model to complete the confirmation of the terminal positioning information, which reduces the input amount of the channel impulse response on the network side, thereby reducing the transmission resource occupancy on the network side.
- the present disclosure embodiment provides another terminal positioning method based on an AI model.
- FIG6 is a flow chart of another terminal positioning method based on an AI model provided by the present disclosure embodiment. As shown in FIG6 , the terminal positioning method based on an AI model may include the following steps:
- Step S601 receiving bit information of a first channel impulse response quantized by a terminal side.
- Step S602 Dequantize the bit information based on the dequantization module to obtain the second channel impulse response.
- the received quantized bit information is input into the second part module of the AI model, and the second part module includes a dequantization module.
- the dequantization module dequantizes the quantized bit information to obtain the second channel impulse response.
- the terminal positioning information includes but is not limited to: positioning coordinates or parameters required for terminal positioning.
- the parameters required for positioning are any of the following: signal arrival time, signal arrival angle, NLOS or LOS.
- Step S603 input the second channel impulse response into the second AI model for processing to obtain the positioning information of the terminal.
- the second channel impulse response obtained after processing the second part module of the first AI model is used as the input of the second AI model, and the second AI model calculates the positioning information of the terminal according to the second channel impulse response.
- the second part model on the network side receives the quantized bit information of the first channel impulse response
- the dequantization module of the second part module of the first AI model processes the bit information to obtain the second channel impulse response, and uses the second channel impulse response as the input of the second AI model, and the second AI model performs the confirmation of the terminal positioning information.
- the first part module of the AI model on the terminal side assists the second part module of the first AI model on the network side and the second AI model to complete the confirmation of the terminal positioning information, which reduces the input amount of the channel impulse response on the network side, thereby reducing the transmission resource occupancy on the network side.
- the present disclosure embodiment provides another terminal positioning method based on an AI model.
- FIG. 7 is a flow chart of another terminal positioning method based on an AI model provided by the present disclosure embodiment. As shown in FIG. 7 , the terminal positioning method based on an AI model may include the following steps:
- Step S701 receiving quantized bit information of a first channel impulse response sent by a terminal side.
- Step S702 Dequantize the bit information based on the dequantization module to obtain a compressed channel impulse response.
- a first channel impulse response is obtained on the terminal side according to the measured amount, and then the first channel impulse response is compressed based on the compression module included in the first part module to obtain a compressed channel impulse response.
- the compressed channel impulse response is input into the quantization module of the first part module on the terminal side, and the input compressed channel impulse response is quantized based on the quantization module to obtain the quantized bit information of the first channel impulse response. This processing is performed on the terminal side.
- the quantized bit information of the received first channel impulse response is input into the second module of the AI model.
- the second module includes a dequantization module, which dequantizes the quantized bit information to obtain a compressed channel impulse response.
- Step S703 Processing the compressed channel impulse response based on the decompression module to obtain the second channel impulse response.
- the second part of the modules also includes a decompression module, which decompresses the compressed channel impulse response to obtain a second channel impulse response.
- Step S704 input the second channel impulse response into the second AI model for processing to obtain the positioning information of the terminal.
- the second channel impulse response obtained after processing by the decompression module is input into the second AI model, and the second AI model continues to confirm the positioning information of the terminal.
- the terminal positioning information includes but is not limited to: positioning coordinates or parameters required for terminal positioning.
- the parameters required for positioning are any of the following: signal arrival time, signal arrival angle, NLOS or LOS.
- the second part on the network side receives the bit information, and the dequantization module in the second part performs dequantization to obtain a compressed channel impulse response.
- the decompression module decompresses the compressed channel impulse response, a second channel impulse response is obtained.
- the second channel impulse response is used as the input of the second AI model, and the second AI model performs the confirmation of the terminal positioning information.
- the first part module of the first AI model on the terminal side assists the second part module of the first AI model on the network side and the second AI model to complete the confirmation of the terminal positioning information, which reduces the input amount of the channel impulse response on the network side, thereby reducing the transmission resource occupancy on the network side.
- the first AI model deploys the first part module on the terminal side and the second part module on the network side
- its essence is still a complete first AI model composed of the first part module and the second part module.
- the first part module on the terminal side and the second part module on the network side complement each other and are used in combination with the second AI model to realize the positioning of the terminal, because the first part module on the terminal side and the second part module on the network side need to be jointly trained during training, that is, the first AI model is obtained in the same training process.
- the specific training process will not be described one by one in the embodiments of the present disclosure.
- the embodiment of the present disclosure provides another terminal positioning method based on an AI model.
- the training objective function of the first AI model is the difference between the first channel impulse response and the second channel impulse response
- the second channel impulse response is the channel impulse response obtained after quantization and dequantization of the first channel impulse response.
- the training of the first AI model minimizes the difference between the first channel impulse response and the second channel impulse response.
- the minimum difference is that the impulse response of the first channel is completely consistent with the impulse response of the second channel.
- the minimum difference is that the difference between the impulse response of the first channel and the impulse response of the second channel is less than a preset difference.
- the specific embodiment of the present disclosure does not limit the preset difference, and can be flexibly configured according to specific application scenarios.
- the embodiment of the present disclosure provides another terminal positioning method based on an AI model.
- the training objective function of the first AI model is the difference between the first channel impulse response and the second channel impulse response
- the second channel impulse response is the channel impulse response obtained after compression, quantization, dequantization, and decompression of the first channel impulse response.
- the training of the first AI model minimizes the difference between the first channel impulse response and the second channel impulse response. For instructions on minimizing the difference, please refer to the detailed description of the above embodiment, and the embodiment of the present disclosure will not be repeated here.
- the present disclosure also provides a terminal positioning device based on the AI model. Since the terminal positioning device based on the AI model provided in the embodiments of the present disclosure corresponds to the terminal positioning method based on the AI model provided in the embodiments of Figures 2 to 4 above, the implementation method of the terminal positioning method based on the AI model is also applicable to the terminal positioning device based on the AI model provided in the embodiments of the present disclosure, and will not be described in detail in the embodiments of the present disclosure.
- FIG8 is a schematic diagram of the structure of a terminal positioning device based on an AI model provided in an embodiment of the present disclosure.
- the device is applied to the terminal side, the AI model includes a first AI model and a second AI model, the first AI model includes a first part module and a second part module, the first part module is deployed on the terminal side, as shown in FIG8, the device includes:
- a processing module 81 configured to input the first channel impulse response into the first part module of the first AI model for processing to obtain quantized bit information of the first channel impulse response;
- the sending module 82 is used to send the quantized bit information of the first channel impulse response to the network side.
- the first part module of the first AI model in the AI model deployed on the terminal side inputs the first channel impulse response into the first part module of the first AI model for processing, obtains quantized bit information of the first channel impulse response, and sends the quantized bit information of the first channel impulse response to the second part module of the first AI model on the network side.
- the first part module of the first AI model on the terminal side assists the AI model in terminal positioning, reduces the input amount of the first channel impulse response on the network side, and thereby reduces the transmission resource occupancy on the network side.
- the first part of the modules includes a quantization module
- the processing module 81 is further configured to include:
- the input first channel impulse response is quantized based on the quantization module to obtain bit information of the quantized first channel impulse response.
- the first part of the modules includes a compression module and a quantization module
- the processing module 81 is further used to:
- the compressed channel impulse response is quantized based on the quantization module to obtain quantized bit information of the first channel impulse response.
- the first AI model includes a first part module and a second part module obtained by joint training.
- the training objective function of the first AI model is the difference between the first channel impulse and the second channel impulse response
- the second channel impulse response is the channel impulse response obtained after quantization and dequantization of the first channel impulse response
- the training objective function of the first AI model is the difference between the first channel impulse response and the second channel impulse response
- the second channel impulse response is the channel impulse response obtained after compression, quantization, dequantization, and decompression of the first channel impulse response
- the present disclosure also provides a terminal positioning device based on the AI model. Since the terminal positioning device based on the AI model provided in the embodiments of the present disclosure corresponds to the terminal positioning method based on the AI model provided in the embodiments of Figures 5 to 7 above, the implementation method of the terminal positioning method based on the AI model is also applicable to the terminal positioning device based on the AI model provided in the embodiments of the present disclosure, and will not be described in detail in the embodiments of the present disclosure.
- FIG9 is a schematic diagram of the structure of a terminal positioning device based on an AI model provided in an embodiment of the present disclosure.
- the AI model includes a first AI model and a second AI model, the first AI model includes a first part module and a second part module, the second part module is deployed on the network side, as shown in FIG9, the device includes:
- a receiving module 91 configured to receive quantized bit information of a first channel impulse response sent by a terminal side
- a processing module 92 configured to input the bit information into the second part module of the first AI model for processing to obtain a second channel impulse response
- the processing module 92 is further configured to input the second channel impulse response into the second AI model for processing to obtain the terminal positioning information.
- the second part model on the network side receives the quantized bit information of the first channel impulse response, and the second part module of the first AI model processes the bit information to obtain the second channel impulse response, and uses the second channel impulse response as the input of the second AI model, which performs the confirmation of the terminal positioning information.
- the first part module of the AI model on the terminal side assists the second part module of the AI model on the network side and the second AI model to complete the confirmation of the terminal positioning information, which reduces the input amount of the channel impulse response on the network side, thereby reducing the transmission resource occupancy on the network side.
- the second part module includes a dequantization module
- the processing module 92 is further used to dequantize the bit information based on the dequantization module to obtain the second channel impulse response.
- the second part of modules includes a dequantization module and a decompression module
- the processing module 92 is further used to:
- the compressed channel impulse response is processed based on the decompression module to obtain the second channel impulse response.
- the training objective function of the first AI model is the difference between the first channel impulse response and the second channel impulse response
- the second channel impulse response is the channel impulse response obtained after quantization and dequantization of the first channel impulse response
- the training objective function of the first AI model is the difference between the first channel impulse and the second channel impulse response
- the second channel impulse response is the channel impulse response obtained after compression, quantization, dequantization, and decompression of the first channel impulse response
- the first part module and the second part module included in the AI model are obtained by joint training.
- the positioning information is positioning coordinates or parameters required for positioning.
- the parameters required for positioning are any of the following:
- the communication device 1000 can be a network device, or a terminal device, or a chip, a chip system, or a processor that supports the network device to implement the above method, or a chip, a chip system, or a processor that supports the terminal device to implement the above method.
- the device can be used to implement the method described in the above method embodiment, and the details can be referred to the description in the above method embodiment.
- the communication device 1000 may include one or more processors 1001.
- the processor 1001 may be a general-purpose processor or a dedicated processor, etc. For example, it may be a baseband processor or a central processing unit.
- the baseband processor may be used to process the communication protocol and communication data
- the central processing unit may be used to control the communication device (such as a base station, a baseband chip, a terminal device, a terminal device chip, a DU or a CU, etc.), execute a computer program, and process the data of the computer program.
- the communication device 1000 may further include one or more memories 1002, on which a computer program 1004 may be stored, and the processor 1001 executes the computer program 1004 so that the communication device 1000 performs the method described in the above method embodiment.
- data may also be stored in the memory 1002.
- the communication device 1000 and the memory 1002 may be provided separately or integrated together.
- the communication device 1000 may further include a transceiver 1005 and an antenna 1006.
- the transceiver 1005 may be referred to as a transceiver unit, a transceiver, or a transceiver circuit, etc., for implementing a transceiver function.
- the transceiver 1005 may include a receiver and a transmitter, the receiver may be referred to as a receiver or a receiving circuit, etc., for implementing a receiving function; the transmitter may be referred to as a transmitter or a transmitting circuit, etc., for implementing a transmitting function.
- the communication device 1000 may further include one or more interface circuits 1007.
- the interface circuit 1007 is used to receive code instructions and transmit them to the processor 1001.
- the processor 1001 executes the code instructions to enable the communication device 1000 to execute the method described in the above method embodiment.
- the communication device 1000 is a terminal: the transceiver 1005 is used to execute steps such as step 201 in FIG. 2 .
- the communication device 1000 is a network device: the transceiver 1005 is used to execute steps such as step 402 in FIG. 4 .
- the processor 1001 may include a transceiver for implementing receiving and sending functions.
- the transceiver may be a transceiver circuit, an interface, or an interface circuit.
- the transceiver circuit, interface, or interface circuit for implementing the receiving and sending functions may be separate or integrated.
- the above-mentioned transceiver circuit, interface, or interface circuit may be used for reading and writing code/data, or the above-mentioned transceiver circuit, interface, or interface circuit may be used for transmitting or delivering signals.
- the processor 1001 may store a computer program 1003, and the computer program 1003 runs on the processor 1001, which may enable the communication device 1000 to perform the method described in the above method embodiment.
- the computer program 1003 may be fixed in the processor 1001, in which case the processor 1001 may be implemented by hardware.
- the communication device 1000 may include a circuit that can implement the functions of sending or receiving or communicating in the aforementioned method embodiments.
- the processor and transceiver described in the present disclosure may be implemented in an integrated circuit (IC), an analog IC, a radio frequency integrated circuit RFIC, a mixed signal IC, an application specific integrated circuit (ASIC), a printed circuit board (PCB), an electronic device, etc.
- the processor and transceiver may also be manufactured using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), N-type metal oxide semiconductor (NMOS), P-type metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (BJT), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.
- CMOS complementary metal oxide semiconductor
- NMOS N-type metal oxide semiconductor
- PMOS P-type metal oxide semiconductor
- BJT bipolar junction transistor
- BiCMOS bipolar CMOS
- SiGe silicon germanium
- GaAs gallium arsenide
- the communication device described in the above embodiments may be a network device or a terminal device, but the scope of the communication device described in the present disclosure is not limited thereto, and the structure of the communication device may not be limited by FIG. 10.
- the communication device may be an independent device or may be part of a larger device.
- the communication device may be:
- the IC set may also include a storage component for storing data and computer programs;
- ASIC such as modem
- the communication device can be a chip or a chip system
- the communication device can be a chip or a chip system
- the schematic diagram of the chip structure shown in Figure 11 includes a processor 1101 and an interface 1103.
- the number of processors 1101 can be one or more, and the number of interfaces 1103 can be multiple.
- Interface 1103 is used to execute step 202 in FIG. 2 , step 302 in FIG. 3 , step 402 in FIG. 4 , etc.
- Interface 1103 is used to execute step 501 in FIG. 5 , step 601 in FIG. 6 , etc.
- the chip 1100 further includes a memory 1102, and the memory 1102 is used to store necessary computer programs and data.
- the present disclosure also provides a readable storage medium having instructions stored thereon, which implement the functions of any of the above method embodiments when executed by a computer.
- the present disclosure also provides a computer program product, which implements the functions of any of the above method embodiments when executed by a computer.
- the computer program product includes one or more computer programs.
- the computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
- the computer program can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
- the computer program can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.
- the computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated.
- the available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a high-density digital video disc (DVD)), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.
- a magnetic medium e.g., a floppy disk, a hard disk, a magnetic tape
- an optical medium e.g., a high-density digital video disc (DVD)
- DVD high-density digital video disc
- SSD solid state disk
- At least one in the present disclosure may also be described as one or more, and a plurality may be two, three, four or more, which is not limited in the present disclosure.
- the technical features in the technical feature are distinguished by “first”, “second”, “third”, “A”, “B”, “C” and “D”, etc., and there is no order of precedence or size between the technical features described by the "first”, “second”, “third”, “A”, “B”, “C” and “D”.
- plural refers to two or more than two, and other quantifiers are similar thereto.
- “And/or” describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and/or B may represent: A exists alone, A and B exist at the same time, and B exists alone.
- the character “/” generally indicates that the associated objects before and after are in an “or” relationship.
- the singular forms “a”, “the”, and “the” are also intended to include plural forms, unless the context clearly indicates other meanings.
- the corresponding relationships shown in the tables in the present disclosure can be configured or predefined.
- the values of the information in each table are only examples and can be configured as other values, which are not limited by the present disclosure.
- the corresponding relationships shown in some rows may not be configured.
- appropriate deformation adjustments can be made based on the above table, such as splitting, merging, etc.
- the names of the parameters shown in the titles of the above tables can also use other names that can be understood by the communication device, and the values or representations of the parameters can also be other values or representations that can be understood by the communication device.
- other data structures can also be used, such as arrays, queues, containers, stacks, linear lists, pointers, linked lists, trees, graphs, structures, classes, heaps, hash tables or hash tables.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
本公开提供了一种基于AI模型的终端定位方法及装置,可以应用于通信技术领域,该方法包括:由终端侧部署的AI模型中第一AI模型的第一部分模块在执行将第一信道冲击响应输入所述第一AI模型的第一部分模块进行处理,得到第一信道冲击响应量化后的比特信息,将第一信道冲击响应量化后的比特信息发送给网络侧的第一AI模型的第二部分模块,由终端侧的第一AI模型的第一部分模块辅助网络侧第一AI模型的第二部分模块及第二AI模型进行终端定位,减少了网络侧第一信道冲击响应的输入量,进而减少了网络侧传输资源占用。
Description
本公开涉及通信技术领域,尤其涉及一种基于AI模型的终端定位方法及装置。
相关技术中,例如工业场景中,对定位精度的要求非常高,人工智能(Artificial Intelligence,AI)模型训练完毕,可以在网络侧或者终端侧部署使用训练完毕的AI模型进行推理。推理时对AI模型的输入仍然为测量结果,基于测量结果,AI模型会输出对应的终端位置。
在目前定位方案中,AI模型输入的是信道冲击响应,终端需要根据测量量得到信道冲击响应,将信道冲击响应量化后反馈给网络,该场景下会导致反馈量比较大,因此占用的传输资源比较多。
发明内容
第一方面,本公开实施例提供一种基于AI模型的终端定位方法,所述方法应用于终端侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第一部分模块部署在所述终端侧,所述方法包括:
将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息;
将所述第一信道冲击响应量化后的比特信息发送给网络侧。
在一种实现方式中,所述第一部分模块包括量化模块,所述将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息包括:
基于所述量化模块将输入的所述第一信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
在一种实现方式中,所述第一部分模块包括压缩模块和量化模块,所述将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息包括:
基于所述压缩模块将所述第一信道冲击响应进行压缩处理,得到压缩后的信道冲击响应;
基于所述量化模块对压缩后的信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
在一种实现方式中,所述第一AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
在一种实现方式中,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
在一种实现方式中,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
第二方面,本公开实施例提供一种基于AI模型的终端定位方法,所述方法应用于网络侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第二部分模块部署在所述网络侧,所述方法包括:
接收终端侧发送的第一信道冲击响应量化后的比特信息;
将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应;
将所述第二信道冲击响应输入所述第二AI模型进行处理,得到终端的定位信息。
在一种实现方式中,所述第二部分模块包括解量化模块,所述将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应包括:
基于所述解量化模块将所述比特信息进行解量化处理,得到所述第二信道冲击响应。
在一种实现方式中,所述第二部分模块包括解量化模块和解压缩模块,所述将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应包括:
基于所述解量化模块将所述比特信息进行解量化处理,得到压缩后的信道冲击响应;
基于所述解压缩模块将所述压缩后的信道冲击响应进行处理,得到所述第二信道冲击响应。
在一种实现方式中,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
在一种实现方式中,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
在一种实现方式中,所述AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
在一种实现方式中,所述定位信息为定位坐标或者定位需要使用的参数。
在一种实现方式中,所述定位需要使用的参数为以下任一项:
信号达到时间、信号达到角度、非视距信息NLOS或视距信息LOS。
第三方面,本公开实施例提供一种基于AI模型的终端定位装置,所述装置应用于终端侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第一部分模块部署在所述终端侧,所述装置包括:
处理模块,用于将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息;
发送模块,用于将所述第一信道冲击响应量化后的比特信息发送给网络侧。
在一种实现方式中,所述第一部分模块包括量化模块,所述处理模块,还用于包括:
基于所述量化模块将输入的所述第一信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
在一种实现方式中,所述第一部分模块包括压缩模块和量化模块,所述处理模块,还用于:
基于所述压缩模块将所述第一信道冲击响应进行压缩处理,得到压缩后的信道冲击响应;
基于所述量化模块对压缩后的信道冲击响应进行量化处理,得到所述第一信道冲击响应量 化后的比特信息。
在一种实现方式中,所述第一AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
在一种实现方式中,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
在一种实现方式中,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
第四方面,本公开实施例提供一种基于AI模型的终端定位装置,所述装置应用于网络侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第二部分模块部署在所述网络侧,所述装置包括:
接收模块,用于接收终端侧发送的第一信道冲击响应量化后的比特信息;
处理模块,用于将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应;
所述处理模块,还用于将所述第二信道冲击响应输入所述第二AI模型进行处理,得到终端的定位信息。
在一种实现方式中,所述第二部分模块包括解量化模块,所述处理模块,还用于基于所述解量化模块将所述比特信息进行解量化处理,得到所述第二信道冲击响应。
在一种实现方式中,所述第二部分模块包括解量化模块和解压缩模块,所述处理模块,还用于:
基于所述解量化模块将所述比特信息进行解量化处理,得到压缩后的信道冲击响应;
基于所述解压缩模块将所述压缩后的信道冲击响应进行处理,得到所述第二信道冲击响应。
在一种实现方式中,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
在一种实现方式中,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
在一种实现方式中,所述AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
在一种实现方式中,所述定位信息为定位坐标或者定位需要使用的参数。
在一种实现方式中,所述定位需要使用的参数为以下任一项:
信号达到时间、信号达到角度、非视距信息NLOS或视距信息LOS。
第五方面,本公开实施例提供了一种计算机可读存储介质,用于储存为上述基于AI模型的终端定位装置所用的指令,当所述指令被执行时,使所述基于AI模型的终端定位装置执行上述第一方面或第二方面所述的方法。
第六方面,本公开实施例还提供一种包括计算机程序的计算机程序产品,当其在计算机上运行时,使得计算机执行上述第一方面或第二方面所述的方法。
第七方面,本公开实施例提供了一种芯片系统,该芯片系统包括至少一个处理器和接口,用于支持通信装置实现第一方面或第二方面所涉及的功能,例如,确定或处理上述方法中所涉及的数据和信息中的至少一种。在一种可能的设计中,所述芯片系统还包括存储器,所述存储器,用于保存通信装置必要的计算机程序和数据。该芯片系统,可以由芯片构成,也可以包括芯片和其他分立器件。
第八方面,本公开实施例还提供了一种计算机程序,当其在计算机上运行时,使得计算机执行上述第一方面或第二方面所述的方法。
为了更清楚地说明本公开实施例或背景技术中的技术方案,下面将对本公开实施例或背景技术中所需要使用的附图进行说明。
图1为本公开实施例提供的一种通信系统的架构示意图;
图2为本公开实施例提供的一种基于AI模型的终端定位方法的流程示意图;
图3为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图;
图4为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图;
图5为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图;
图6为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图;
图7为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图;
图8为本公开实施例提供的一种基于AI模型的终端定位装置的结构示意图;
图9为本公开实施例提供的另一种基于AI模型的终端定位装置的结构示意图;
图10为本公开实施例提供的一种通信装置的结构示意图;
图11是本公开实施例提供的芯片的结构示意图。
为了更好的理解本公开实施例公开的一种基于AI模型的终端定位方法及装置,下面首先对本公开实施例适用的通信系统进行描述。
请参见图1,图1为本公开实施例提供的一种通信系统的架构示意图。该通信系统可包括但不限于一个网络设备、和一个终端设备,图1所示的设备数量和形态仅用于举例并不构成对本公开实施例的限定,实际应用中可以包括两个或两个以上的网络设备,两个或两个以上的终端设备。图1所示的通信系统以包括一个网络设备11、和一个终端设备12为例。
需要说明的是,本公开实施例的技术方案可以应用于各种通信系统。例如:长期演进(long term evolution,LTE)系统、第五代(5th generation,5G)移动通信系统、5G新空口(new radio,NR)系统,或者其他未来的新型移动通信系统等。
本公开实施例中的网络设备11是网络侧的一种用于发射或接收信号的实体。例如,网络设备101可以为演进型基站(evolved NodeB,eNB)、传输点(transmission reception point,TRP)、 NR系统中的下一代基站(next generation NodeB,gNB)、其他未来移动通信系统中的基站或无线保真(wireless fidelity,WiFi)系统中的接入节点等。本公开的实施例对网络设备所采用的具体技术和具体设备形态不做限定。本公开实施例提供的网络设备可以是由集中单元(central unit,CU)与分布式单元(distributed unit,DU)组成的,其中,CU也可以称为控制单元(control unit),采用CU-DU的结构可以将网络设备,例如基站的协议层拆分开,部分协议层的功能放在CU集中控制,剩下部分或全部协议层的功能分布在DU中,由CU集中控制DU。
本公开实施例中的终端设备12是用户侧的一种用于接收或发射信号的实体,如手机。终端设备也可以称为终端设备(terminal)、用户设备(user equipment,UE)、移动台(mobile station,MS)、移动终端设备(mobile terminal,MT)等。终端设备可以是具备通信功能的汽车、智能汽车、手机(mobile phone)、穿戴式设备、平板电脑(Pad)、带无线收发功能的电脑、虚拟现实(virtual reality,VR)终端设备、增强现实(augmented reality,AR)终端设备、工业控制(industrial control)中的无线终端设备、无人驾驶(self-driving)中的无线终端设备、远程手术(remote medical surgery)中的无线终端设备、智能电网(smart grid)中的无线终端设备、运输安全(transportation safety)中的无线终端设备、智慧城市(smart city)中的无线终端设备、智慧家庭(smart home)中的无线终端设备等等。本公开的实施例对终端设备所采用的具体技术和具体设备形态不做限定。
可以理解的是,本公开实施例描述的通信系统是为了更加清楚的说明本公开实施例的技术方案,并不构成对于本公开实施例提供的技术方案的限定,本领域普通技术人员可知,随着系统架构的演变和新业务场景的出现,本公开实施例提供的技术方案对于类似的技术问题,同样适用。
相关技术中,例如工业场景中,对定位精度的要求非常高,人工智能(Artificial Intelligence,AI)模型训练完毕,可以在网络侧或者终端侧部署使用训练完毕的AI模型进行推理。推理时对AI模型的输入仍然为测量结果,基于测量结果,AI模型会输出对应的终端位置。在目前定位方案中,AI模型输入的是信道冲击响应,终端需要根据测量量得到信道冲击响应,将信道冲击响应量化后反馈给网络,该场景下会导致反馈量比较大,因此占用的传输资源比较多。
请参见图2,图2为本公开实施例提供的一种基于AI模型的终端定位方法的流程示意图,该方法应用于终端侧。如图2所示,该方法可以包括但不限于如下步骤:
步骤S201,将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息。
所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第一部分模块部署在所述终端侧,第二部分模块部署在网络侧。
在终端侧根据测量量得到第一信道冲击响应,然后将第一信道冲击响应输入至终端侧的第一AI模型的第一部分模块,得到第一信道冲击响应量化后的比特信息,该处理过程在终端侧执行。
步骤S202,将第一信道冲击响应量化后的比特信息发送给网络侧。
将终端侧基于第一AI模型第一部分模块得到的量化后的比特信息发送至网络侧的第一AI模型的第二部分模块,以便由网络侧的第二部分模块根据第一信道冲击响应量化后的比特信息,得到该终端的定位信息,由此可见,本公开实施例提供的方法中,AI模型的输入为终端侧的第一信道冲击响应,AI模型的输出为网络侧执行的终端的定位信息。
在本公开的所有实施例中,该第一部分模块部署在终端侧,用于将第一信道冲击响应进行压缩处理,将压缩处理,并将压缩后的第一信道冲击响应进行量化处理,得到第一信道冲击响应量化后的数据。第二部分模块部署在网络侧,用于根据接收到的第一信道冲击响应量化后的数据,利用第二部分模块确定终端的定位信息。
在本公开的所有实施例中,比特信息是指一个或多个比特,或是指一个或多个比特组成的数据;或是说,比特信息是比特形式的信息。
在一些实施例中,所述终端的定位信息包含但不限于:定位坐标或者用于确定终端定位需要使用的参数。
由终端侧部署的AI模型中第一AI模型的第一部分模块在执行将第一信道冲击响应输入所述第一AI模型的第一部分模块进行处理,得到第一信道冲击响应量化后的比特信息,将第一信道冲击响应量化后的比特信息发送给网络侧的第一AI模型的第二部分模块,由终端侧的第一AI模型的第一部分模块辅助网络侧第一AI模型的第二部分模块及第二AI模型进行终端定位,减少了网络侧第一信道冲击响应的输入量,进而减少了网络侧传输资源占用。
本公开实施例提供了另一种基于AI模型的终端定位方法,图3为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图,该方法应用于终端侧,如图3所示,该基于AI模型的终端定位方法可包括如下步骤:
步骤S301:基于所述量化模块将输入的所述第一信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
本申请实施例中AI模型包括第一AI模型和第二AI模型,而第一AI模型包括第一部分模块和第二部分模块,将第一AI模型的第一部分模块部署在终端侧,其中,第一部分模块包括量化模块,将第一AI模型的第二部分模块部署在网络侧。
在终端侧根据测量量得到第一信道冲击响应,然后将第一信道冲击响应输入至终端侧的第一AI模型的第一部分模块的量化模块中,基于该量化模块将输入的第一信道冲击响应进行量化处理,得到第一信道冲击响应量化后的比特信息,该处理过程在终端侧执行。
步骤S302:将所述第一信道冲击响应量化后的比特信息发送给网络侧。
将终端侧基于第一AI模型第一部分模块得到的量化后的比特信息发送至网络侧的第一AI模型的第二部分模块,以便由网络侧的第二部分模块根据第一信道冲击响应量化后的比特信息,得到该终端的定位信息,由此可见,本公开实施例提供的方法中,AI模型的输入为终端侧的第一信道冲击响应,AI模型的输出为网络侧执行的终端的定位信息。
在一些实施例中,所述终端的定位信息包含但不限于:定位坐标或者终端定位需要使用的参数。
由终端侧部署的AI模型中第一AI模型的第一部分模块在执行将第一信道冲击响应输入 所述第一AI模型的第一部分模块中的量化模块进行处理,得到第一信道冲击响应量化后的比特信息,将第一信道冲击响应量化后的比特信息发送给网络侧的第一AI模型的第二部分模块,由终端侧的第一AI模型的第一部分模块的量化模块辅助网络侧第一AI模型的第二部分模块及第二AI模型进行终端定位,减少了网络侧第一信道冲击响应的输入量,进而减少了网络侧传输资源占用。
本公开实施例提供了另一种基于AI模型的终端定位方法,图4为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图,该方法应用于终端侧,如图4所示,该基于AI模型的终端定位方法可包括如下步骤:
步骤S4011:基于所述压缩模块将所述第一信道冲击响应进行压缩处理,得到压缩后的信道冲击响应。
本申请实施例中AI模型第一AI模型和第二AI模型,其中,第一AI模型包括第一部分模块和第二部分模块,将AI模型的第一部分模块部署在终端侧,其中,第一部分模块包括压缩模块和量化模块,将第二部分模块部署在网络侧。
在终端侧根据测量量得到第一信道冲击响应,然后基于该压缩模块将第一信道冲击响应进行压缩处理,得到压缩后的信道冲击响应。
步骤S4012:基于所述量化模块对压缩后的信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
将压缩后的信道冲击响应输入至终端侧的第一部分模块的量化模块中,基于该量化模块将输入的压缩后的信道冲击响应进行量化处理,得到第一信道冲击响应量化后的比特信息,该处理过程在终端侧执行。
步骤S402:将所述第一信道冲击响应量化后的比特信息发送给网络侧。
将终端侧基于第一部分模块中的量化模块得到的量化后的第一信道冲击响应比特信息发送至网络侧的第二部分模块,以便由网络侧的第二部分模块根据第一信道冲击响应量化后的比特信息,得到该终端的定位信息,由此可见,本公开实施例提供的方法中,AI模型的输入为终端侧的第一信道冲击响应,AI模型的输出为网络侧执行的终端的定位信息。
在一些实施例中,所述终端的定位信息包含但不限于:定位坐标或者终端定位需要使用的参数。
由终端侧部署的AI模型的第一部分模块先基于第一部分模块的压缩模块执行将第一信道冲击响应进行压缩处理,将压缩处理后的输入所述第一部分模块的量化模块中,并基于所述量化模块将输入的压缩后的第一信道冲击响应进行量化处理,得到第一信道冲击响应量化后的比特信息,将量化后的第一信道冲击响应比特信息发送给网络侧的AI模型的第二部分模块,由终端侧的第一AI模型的第一部分模块的压缩模块及量化模块辅助网络侧第一AI模型的第二部分模块及第二AI模型进行终端定位,进一步减少了网络侧第一信道冲击响应的输入量,进而减少了网络侧传输资源占用。
在一些实施例中,虽然第一AI模型分别在终端侧部署了第一部分模块,在网络侧部署了第二部分模块,但是其实质仍然为第一部分模块及第二部分模块构成一个完整的第一AI模型, 终端侧的第一部分模块与网络侧的第二部分模块相辅相成,两者的结合使用实现终端的定位,因为,在训练时需要将终端侧的第一部分模块与网络侧的第二部分模块进行联合训练,即第一AI模型是在同一个训练过程中得到的。具体的训练过程,本公开实施例在此不再进行一一赘述。
本公开实施例提供了另一种基于AI模型的终端定位方法,作为对图2或图3的补充说明,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。第一AI模型的训练使得第一信道冲击响应与第二信道冲击响应之间的差别最小。
作为本公开实施例的一种实现方式,差别最小为第一信道冲击响应与第二信道冲击响应完全一致。作为本公开实施例的另一种实现方式,差别最小为第一信道冲击响应与第二信道冲击响应的差异小于预设差值,具体的本公开实施例对预设差值不做限定,可根据具体应用场景进行灵活配置。
本公开实施例提供了另一种基于AI模型的终端定位方法,作为对图2或图4的补充说明,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。第一AI模型的训练使得第一信道冲击响应与第二信道冲击响应之间的差别最小。有关差别最小的说明可参阅上述实施例的详细描述,本公开实施例在此不再进行赘述。
请参见图5,图5为本公开实施例提供的一种基于AI模型的终端定位方法的流程示意图,该方法应用于网络侧。如图5所示,该方法可以包括但不限于如下步骤:
步骤S501,接收终端侧发送的第一信道冲击响应量化后的比特信息。
所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,第一部分模块部署于终端侧,所述第二部分模块部署在所述网络侧。
在终端侧根据测量量得到第一信道冲击响应,然后将第一信道冲击响应输入至终端侧的第一AI模型的第一部分模块,得到第一信道冲击响应量化后的比特信息,并发送至网络侧,该处理过程在终端侧执行。由网络侧接收第一信道冲击响应量化后的比特信息。
步骤S502:将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应。
将接收到的第一信道冲击响应量化后的比特信息由第一AI模型的第二部分模块进行处理,得到第二信道冲击响应。
本公开实施例中,第二信道冲击响应是一个逆过程,例如,终端侧发送的比特信息是由量化处理得到的,则在网络侧需要执行解量化得到第二信道冲击响应。
步骤S503:将所述第二信道冲击响应输入所述第二AI模型进行处理,得到终端的定位信息。
将第一AI模型的第二部分模块处理后得到的第二信道冲击响应作为第二AI模型的输入,由第二AI模型根据第二信道冲击响应执行终端的定位信息的计算。
在一些实施例中,所述终端的定位信息包含但不限于:定位坐标或者终端定位需要使用的 参数。所述定位需要使用的参数为以下任一项:信号达到时间、信号达到角度、非视距信息(Non-line-of-sight propagation,NLOS)或视距信息(line-of-sight propagation,LOS)。
由终端侧部署的AI模型的第一部分模块将第一信道冲击响应量化后的比特信息发送给网络侧的第一AI模型的第二部分模块后,网络侧的第二部分模型接收第一信道冲击响应量化后的比特信息,并由第一AI模型的第二部分模块对比特信息进行处理得到第二信道冲击响应,将该第二信道冲击响应作为第二AI模型的输入,由第二AI模型执行终端定位信息的确认。该终端定位过程,由终端侧的AI模型的第一部分模块辅助网络侧第一AI模型第二部分模块及第二AI模型完成终端定位信息的确认,减少了网络侧信道冲击响应的输入量,进而减少了网络侧传输资源占用。
本公开实施例提供了另一种基于AI模型的终端定位方法,图6为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图,如图6所示,该基于AI模型的终端定位方法可包括如下步骤:
步骤S601:接收终端侧发送的第一信道冲击响应量化后的比特信息。
步骤S602:基于所述解量化模块将所述比特信息进行解量化处理,得到所述第二信道冲击响应。
将接收到的量化后的比特信息输入到AI模型的第二部分模块中,第二部分模块中包括解量化模块,由该解量化模块对量化后的比特信息进行解量化处理后,得到所述第二信道冲击响应。
在一些实施例中,所述终端的定位信息包含但不限于:定位坐标或者终端定位需要使用的参数。所述定位需要使用的参数为以下任一项:信号达到时间、信号达到角度、NLOS或LOS。
步骤S603:将所述第二信道冲击响应输入所述第二AI模型进行处理,得到终端的定位信息。
将第一AI模型的第二部分模块处理后得到的第二信道冲击响应作为第二AI模型的输入,由第二AI模型根据第二信道冲击响应执行终端的定位信息的计算。
由终端侧部署的AI模型的第一部分模块将第一信道冲击响应量化后的比特信息发送给网络侧的第一AI模型的第二部分模块后,网络侧的第二部分模型接收第一信道冲击响应量化后的比特信息,并由第一AI模型的第二部分模块的解量化模块对比特信息进行处理得到第二信道冲击响应,将该第二信道冲击响应作为第二AI模型的输入,由第二AI模型执行终端定位信息的确认。该终端定位过程,由终端侧的AI模型的第一部分模块辅助网络侧第一AI模型第二部分模块及第二AI模型完成终端定位信息的确认,减少了网络侧信道冲击响应的输入量,进而减少了网络侧传输资源占用。
本公开实施例提供了另一种基于AI模型的终端定位方法,图7为本公开实施例提供的另一种基于AI模型的终端定位方法的流程示意图,如图7所示,该基于AI模型的终端定位方法可包括如下步骤:
步骤S701:接收终端侧发送的第一信道冲击响应量化后的比特信息。
步骤S702:基于所述解量化模块将所述比特信息进行解量化处理,得到压缩后的信道冲 击响应。
在终端侧根据测量量得到第一信道冲击响应,然后基于第一部分模块包括的压缩模块将第一信道冲击响应进行压缩处理,得到压缩后的信道冲击响应。将压缩后的信道冲击响应输入至终端侧的第一部分模块的量化模块中,基于该量化模块将输入的压缩后的信道冲击响应进行量化处理,得到第一信道冲击响应量化后的比特信息,该处理过程在终端侧执行。
在网络侧,将接收到的第一信道冲击响应量化后的比特信息输入到AI模型的第二部分模块中,第二部分模块中包括解量化模块,由该解量化模块对量化后的比特信息进行解量化处理,得到压缩后的信道冲击响应。
步骤S703:基于所述解压缩模块将所述压缩后的信道冲击响应进行处理,得到所述第二信道冲击响应。
第二部分模块中还包括解压缩模块,由该解压缩模块对压缩后的信道冲击响应进行解压缩处理,得到第二信道冲击响应。
步骤S704:将所述第二信道冲击响应输入所述第二AI模型进行处理,得到终端的定位信息。
将解压缩模块处理后得到的第二信道冲击响应,输入到第二AI模型,由第二AI模型继续执行对所述终端的定位信息的确认。
在一些实施例中,所述终端的定位信息包含但不限于:定位坐标或者终端定位需要使用的参数。所述定位需要使用的参数为以下任一项:信号达到时间、信号达到角度、NLOS或LOS。
由终端侧部署的第一AI模型的第一部分模块将压缩、量化后的比特信息发送给网络侧的第一AI模型的第二部分模块后,网络侧的第二部分接收比特信息,并由第二部分中的解量化模块执行解量化,得到压缩后的信道冲击响应,由解压缩模块对压缩后的信道冲击响应解压缩后,得到第二信道冲击响应,该第二信道冲击响应作为第二AI模型的输入,由第二AI模型执行终端定位信息的确认。该终端定位过程,由终端侧的第一AI模型的第一部分模块辅助网络侧第一AI模型第二部分模块及第二AI模型完成终端定位信息的确认,减少了网络侧信道冲击响应的输入量,进而减少了网络侧传输资源占用。
在一些实施例中,虽然第一AI模型分别在终端侧部署了第一部分模块,在网络侧部署了第二部分模块,但是其实质仍然为第一部分模块及第二部分模块共同构成的一个完整的第一AI模型,终端侧的第一部分模块与网络侧的第二部分模块相辅相成,与第二AI模型的结合使用实现对终端的定位,因为,在训练时需要将终端侧的第一部分模块与网络侧的第二部分模块进行联合训练,即第一AI模型是在同一个训练过程中得到的。具体的训练过程,本公开实施例在此不再进行一一赘述。
与终端侧类似的,本公开实施例提供了另一种基于AI模型的终端定位方法,作为对图5或图6的补充说明,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。第一AI模型的训练使得第一信道冲击响应与第二信道冲击响应之间的差别最小。
作为本公开实施例的一种实现方式,差别最小为第一信道冲击响应与第二信道冲击响应完 全一致。作为本公开实施例的另一种实现方式,差别最小为第一信道冲击响应与第二信道冲击响应的差异小于预设差值,具体的本公开实施例对预设差值不做限定,可根据具体应用场景进行灵活配置。
本公开实施例提供了另一种基于AI模型的终端定位方法,作为对图5或图7的补充说明,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。第一AI模型的训练使得第一信道冲击响应与第二信道冲击响应之间的差别最小。有关差别最小的说明可参阅上述实施例的详细描述,本公开实施例在此不再进行赘述。
与上述图2至图4实施例提供的基于AI模型的终端定位方法相对应,本公开还提供一种基于AI模型的终端定位装置,由于本公开实施例提供基于AI模型的终端定位装置与上述图2至图4实施例提供的基于AI模型的终端定位方法相对应,因此在基于AI模型的终端定位方法的实施方式也适用于本公开实施例提供的基于AI模型的终端定位装置,在本公开实施例中不再详细描述。
图8为本公开实施例所提供的一种基于AI模型的终端定位装置的结构示意图。所述装置应用于终端侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第一部分模块部署在所述终端侧,如图8所示,所述装置包括:
处理模块81,用于将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息;
发送模块82,用于将所述第一信道冲击响应量化后的比特信息发送给网络侧。
由终端侧部署的AI模型中第一AI模型的第一部分模块在执行将第一信道冲击响应输入所述第一AI模型的第一部分模块进行处理,得到第一信道冲击响应量化后的比特信息,将第一信道冲击响应量化后的比特信息发送给网络侧的第一AI模型的第二部分模块,由终端侧的第一AI模型的第一部分模块辅助AI模型进行终端定位,减少了网络侧第一信道冲击响应的输入量,进而减少了网络侧传输资源占用。
作为本公开实施例的一种可能实现方式,所述第一部分模块包括量化模块,所述处理模块81,还用于包括:
基于所述量化模块将输入的所述第一信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
作为本公开实施例的一种可能实现方式,所述第一部分模块包括压缩模块和量化模块,所述处理模块81,还用于:
基于所述压缩模块将所述第一信道冲击响应进行压缩处理,得到压缩后的信道冲击响应;
基于所述量化模块对压缩后的信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
作为本公开实施例的一种可能实现方式,所述第一AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
作为本公开实施例的一种可能实现方式,所述第一AI模型的训练目标函数为第一信道冲 击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
作为本公开实施例的一种可能实现方式,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
与上述图5至图7实施例提供的基于AI模型的终端定位方法相对应,本公开还提供一种基于AI模型的终端定位装置,由于本公开实施例提供基于AI模型的终端定位装置与上述图5至图7实施例提供的基于AI模型的终端定位方法相对应,因此在基于AI模型的终端定位方法的实施方式也适用于本公开实施例提供的基于AI模型的终端定位装置,在本公开实施例中不再详细描述。
图9为本公开实施例所提供的一种基于AI模型的终端定位装置的结构示意图。所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第二部分模块部署在所述网络侧,如图9所示,所述装置包括:
接收模块91,用于接收终端侧发送的第一信道冲击响应量化后的比特信息;
处理模块92,用于将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应;
所述处理模块92,还用于将所述第二信道冲击响应输入所述第二AI模型进行处理,得到终端的定位信息。
由终端侧部署的AI模型的第一部分模块将第一信道冲击响应量化后的比特信息发送给网络侧的第一AI模型的第二部分模块后,网络侧的第二部分模型接收第一信道冲击响应量化后的比特信息,并由第一AI模型的第二部分模块对比特信息进行处理得到第二信道冲击响应,将该第二信道冲击响应作为第二AI模型的输入,由第二AI模型执行终端定位信息的确认。该终端定位过程,由终端侧的AI模型的第一部分模块辅助网络侧AI模型第二部分模块及第二AI模型完成终端定位信息的确认,减少了网络侧信道冲击响应的输入量,进而减少了网络侧传输资源占用。
作为本公开实施例的一种可能实现方式,所述第二部分模块包括解量化模块,所述处理模块92,还用于基于所述解量化模块将所述比特信息进行解量化处理,得到所述第二信道冲击响应。
作为本公开实施例的一种可能实现方式,所述第二部分模块包括解量化模块和解压缩模块,所述处理模块92,还用于:
基于所述解量化模块将所述比特信息进行解量化处理,得到压缩后的信道冲击响应;
基于所述解压缩模块将所述压缩后的信道冲击响应进行处理,得到所述第二信道冲击响应。
作为本公开实施例的一种可能实现方式,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
作为本公开实施例的一种可能实现方式,所述第一AI模型的训练目标函数为第一信道冲 击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
作为本公开实施例的一种可能实现方式,所述AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
作为本公开实施例的一种可能实现方式,所述定位信息为定位坐标或者定位需要使用的参数。
作为本公开实施例的一种可能实现方式,所述定位需要使用的参数为以下任一项:
信号达到时间、信号达到角度、非视距信息NLOS或视距信息LOS。
请参见图10,图10为本公开实施例提供的另一种通信装置的结构示意图。图10中,该通信装置1000可以是网络设备,也可以是终端设备,也可以是支持网络设备实现上述方法的芯片、芯片系统、或处理器等,还可以是支持终端设备实现上述方法的芯片、芯片系统、或处理器等。该装置可用于实现上述方法实施例中描述的方法,具体可以参见上述方法实施例中的说明。
通信装置1000可以包括一个或多个处理器1001。处理器1001可以是通用处理器或者专用处理器等。例如可以是基带处理器或中央处理器。基带处理器可以用于对通信协议以及通信数据进行处理,中央处理器可以用于对通信装置(如,基站、基带芯片,终端设备、终端设备芯片,DU或CU等)进行控制,执行计算机程序,处理计算机程序的数据。
可选的,通信装置1000中还可以包括一个或多个存储器1002,其上可以存有计算机程序1004,处理器1001执行所述计算机程序1004,以使得通信装置1000执行上述方法实施例中描述的方法。可选的,所述存储器1002中还可以存储有数据。通信装置1000和存储器1002可以单独设置,也可以集成在一起。
可选的,通信装置1000还可以包括收发器1005、天线1006。收发器1005可以称为收发单元、收发机、或收发电路等,用于实现收发功能。收发器1005可以包括接收器和发送器,接收器可以称为接收机或接收电路等,用于实现接收功能;发送器可以称为发送机或发送电路等,用于实现发送功能。
可选的,通信装置1000中还可以包括一个或多个接口电路1007。接口电路1007用于接收代码指令并传输至处理器1001。处理器1001运行所述代码指令以使通信装置1000执行上述方法实施例中描述的方法。
通信装置1000为终端:收发器1005用于执行图2中的步骤201等步骤。
通信装置1000为网络设备:收发器1005用于执行图4中的步骤402等步骤。
在一种实现方式中,处理器1001中可以包括用于实现接收和发送功能的收发器。例如该收发器可以是收发电路,或者是接口,或者是接口电路。用于实现接收和发送功能的收发电路、接口或接口电路可以是分开的,也可以集成在一起。上述收发电路、接口或接口电路可以用于代码/数据的读写,或者,上述收发电路、接口或接口电路可以用于信号的传输或传递。
在一种实现方式中,处理器1001可以存有计算机程序1003,计算机程序1003在处理器1001上运行,可使得通信装置1000执行上述方法实施例中描述的方法。计算机程序1003可 能固化在处理器1001中,该种情况下,处理器1001可能由硬件实现。
在一种实现方式中,通信装置1000可以包括电路,所述电路可以实现前述方法实施例中发送或接收或者通信的功能。本公开中描述的处理器和收发器可实现在集成电路(integrated circuit,IC)、模拟IC、射频集成电路RFIC、混合信号IC、专用集成电路(application specific integrated circuit,ASIC)、印刷电路板(printed circuit board,PCB)、电子设备等上。该处理器和收发器也可以用各种IC工艺技术来制造,例如互补金属氧化物半导体(complementary metal oxide semiconductor,CMOS)、N型金属氧化物半导体(nMetal-oxide-semiconductor,NMOS)、P型金属氧化物半导体(positive channel metal oxide semiconductor,PMOS)、双极结型晶体管(bipolar junction transistor,BJT)、双极CMOS(BiCMOS)、硅锗(SiGe)、砷化镓(GaAs)等。
以上实施例描述中的通信装置可以是网络设备,或者终端设备,但本公开中描述的通信装置的范围并不限于此,而且通信装置的结构可以不受图10的限制。通信装置可以是独立的设备或者可以是较大设备的一部分。例如所述通信装置可以是:
(1)独立的集成电路IC,或芯片,或,芯片系统或子系统;
(2)具有一个或多个IC的集合,可选的,该IC集合也可以包括用于存储数据,计算机程序的存储部件;
(3)ASIC,例如调制解调器(Modem);
(4)可嵌入在其他设备内的模块;
(5)接收机、终端设备、智能终端设备、蜂窝电话、无线设备、手持机、移动单元、车载设备、网络设备、云设备、人工智能设备等等;
(6)其他等等。
对于通信装置可以是芯片或芯片系统的情况,可参见图11所示的芯片的结构示意图。图11所示的芯片1100包括处理器1101和接口1103。其中,处理器1101的数量可以是一个或多个,接口1103的数量可以是多个。
对于芯片用于实现本公开实施例中终端的功能的情况:
接口1103,用于执行图2中的步骤202;图3中的步骤302;图4中的步骤402等。
对于芯片用于实现本公开实施例中网络的功能的情况:
接口1103,用于执行图5中的步骤501,图6中的步骤601等。
可选的,芯片1100还包括存储器1102,存储器1102用于存储必要的计算机程序和数据。
本领域技术人员还可以了解到本公开实施例列出的各种说明性逻辑块(illustrative logical block)和步骤(step)可以通过电子硬件、电脑软件,或两者的结合进行实现。这样的功能是通过硬件还是软件来实现取决于特定的应用和整个系统的设计要求。本领域技术人员可以对于每种特定的应用,可以使用各种方法实现所述的功能,但这种实现不应被理解为超出本公开实施例保护的范围。
本公开还提供一种可读存储介质,其上存储有指令,该指令被计算机执行时实现上述任一方法实施例的功能。
本公开还提供一种计算机程序产品,该计算机程序产品被计算机执行时实现上述任一方法实施例的功能。
在上述实施例中,可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时,可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机程序。在计算机上加载和执行所述计算机程序时,全部或部分地产生按照本公开实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机程序可以存储在计算机可读存储介质中,或者从一个计算机可读存储介质向另一个计算机可读存储介质传输,例如,所述计算机程序可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line,DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如,软盘、硬盘、磁带)、光介质(例如,高密度数字视频光盘(digital video disc,DVD))、或者半导体介质(例如,固态硬盘(solid state disk,SSD))等。
本领域普通技术人员可以理解:本公开中涉及的第一、第二等各种数字编号仅为描述方便进行的区分,并不用来限制本公开实施例的范围,也表示先后顺序。
本公开中的至少一个还可以描述为一个或多个,多个可以是两个、三个、四个或者更多个,本公开不做限制。在本公开实施例中,对于一种技术特征,通过“第一”、“第二”、“第三”、“A”、“B”、“C”和“D”等区分该种技术特征中的技术特征,该“第一”、“第二”、“第三”、“A”、“B”、“C”和“D”描述的技术特征间无先后顺序或者大小顺序。
进一步可以理解的是,本公开中“多个”是指两个或两个以上,其它量词与之类似。“和/或”,描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。字符“/”一般表示前后关联对象是一种“或”的关系。单数形式的“一种”、“所述”和“该”也旨在包括多数形式,除非上下文清楚地表示其他含义。
进一步可以理解的是,本公开中涉及到的“响应于”、“如果”、“如若”等词语的含义取决于语境以及实际使用的场景,如在此所使用的词语“如若”可以被解释成为“在……时”或“当……时”。
本公开中各表所示的对应关系可以被配置,也可以是预定义的。各表中的信息的取值仅仅是举例,可以配置为其他值,本公开并不限定。在配置信息与各参数的对应关系时,并不一定要求必须配置各表中示意出的所有对应关系。例如,本公开中的表格中,某些行示出的对应关系也可以不配置。又例如,可以基于上述表格做适当的变形调整,例如,拆分,合并等等。上述各表中标题示出参数的名称也可以采用通信装置可理解的其他名称,其参数的取值或表示方式也可以通信装置可理解的其他取值或表示方式。上述各表在实现时,也可以采用其他的数据结构,例如可以采用数组、队列、容器、栈、线性表、指针、链表、树、图、结构体、类、堆、散列表或哈希表等。
Claims (30)
- 一种基于AI模型的终端定位方法,其特征在于,所述方法应用于终端侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第一部分模块部署在所述终端侧,所述方法包括:将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息;将所述第一信道冲击响应量化后的比特信息发送给网络侧。
- 根据权利要求1所述的方法,其特征在于,所述第一部分模块包括量化模块,所述将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息包括:基于所述量化模块将输入的所述第一信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
- 根据权利要求1所述的方法,其特征在于,所述第一部分模块包括压缩模块和量化模块,所述将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息包括:基于所述压缩模块将所述第一信道冲击响应进行压缩处理,得到压缩后的信道冲击响应;基于所述量化模块对压缩后的信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
- 根据权利要求1-3中任一项所述的方法,其特征在于,所述第一AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
- 根据权利要求1或2所述的方法,其特征在于,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
- 根据权利要求1或3所述的方法,其特征在于,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
- 一种基于AI模型的终端定位方法,其特征在于,所述方法应用于网络侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第二部分模块部署在所述网络侧,所述方法包括:接收终端侧发送的第一信道冲击响应量化后的比特信息;将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应;将所述第二信道冲击响应输入所述第二AI模型进行处理,得到终端的定位信息。
- 根据权利要求7所述的方法,其特征在于,所述第二部分模块包括解量化模块,所述将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应包括:基于所述解量化模块将所述比特信息进行解量化处理,得到所述第二信道冲击响应。
- 根据权利要求7所述的方法,其特征在于,所述第二部分模块包括解量化模块和解压缩模块,所述将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应包括:基于所述解量化模块将所述比特信息进行解量化处理,得到压缩后的信道冲击响应;基于所述解压缩模块将所述压缩后的信道冲击响应进行处理,得到所述第二信道冲击响应。
- 根据权利要求7或8所述的方法,其特征在于,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
- 根据权利要求7或9所述的方法,其特征在于,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
- 根据权利要求7-9中任一项所述的方法,其特征在于,所述AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
- 根据权利要求12所述的方法,其特征在于,所述定位信息为定位坐标或者定位需要使用的参数。
- 根据权利要求13所述的方法,其特征在于,所述定位需要使用的参数为以下任一项:信号达到时间、信号达到角度、非视距信息NLOS或视距信息LOS。
- 一种基于AI模型的终端定位装置,其特征在于,所述装置应用于终端侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第一部分模块部署在所述终端侧,所述装置包括:处理模块,用于将第一信道冲击响应输入所述第一AI模型的所述第一部分模块进行处理,得到所述第一信道冲击响应量化后的比特信息;发送模块,用于将所述第一信道冲击响应量化后的比特信息发送给网络侧。
- 根据权利要求15所述的装置,其特征在于,所述第一部分模块包括量化模块,所述处理模块,还用于包括:基于所述量化模块将输入的所述第一信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
- 根据权利要求15所述的装置,其特征在于,所述第一部分模块包括压缩模块和量化模块,所述处理模块,还用于:基于所述压缩模块将所述第一信道冲击响应进行压缩处理,得到压缩后的信道冲击响应;基于所述量化模块对压缩后的信道冲击响应进行量化处理,得到所述第一信道冲击响应量化后的比特信息。
- 根据权利要求15-17中任一项所述的装置,其特征在于,所述第一AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
- 根据权利要求15或16所述的装置,其特征在于,所述第一AI模型的训练目标函数 为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
- 根据权利要求15或16所述的装置,其特征在于,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
- 一种基于AI模型的终端定位装置,其特征在于,所述装置应用于网络侧,所述AI模型包括第一AI模型和第二AI模型,所述第一AI模型包括第一部分模块和第二部分模块,所述第二部分模块部署在所述网络侧,所述装置包括:接收模块,用于接收终端侧发送的第一信道冲击响应量化后的比特信息;处理模块,用于将所述比特信息输入所述第一AI模型的所述第二部分模块进行处理,得到第二信道冲击响应;所述处理模块,还用于将所述第二信道冲击响应输入所述第二AI模型进行处理,得到终端的定位信息。
- 根据权利要求21所述的装置,其特征在于,所述第二部分模块包括解量化模块,所述处理模块,还用于基于所述解量化模块将所述比特信息进行解量化处理,得到所述第二信道冲击响应。
- 根据权利要求21所述的装置,其特征在于,所述第二部分模块包括解量化模块和解压缩模块,所述处理模块,还用于:基于所述解量化模块将所述比特信息进行解量化处理,得到压缩后的信道冲击响应;基于所述解压缩模块将所述压缩后的信道冲击响应进行处理,得到所述第二信道冲击响应。
- 根据权利要求21或22所述的装置,其特征在于,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过量化、解量化处理后得到的信道冲击响应。
- 根据权利要求21或23所述的装置,其特征在于,所述第一AI模型的训练目标函数为第一信道冲击与第二信道冲击响应的差别,所述第二信道冲击响应为所述第一信道冲击响应经过压缩、量化、解量化、解压缩处理后得到的信道冲击响应。
- 根据权利要求21-23中任一项所述的装置,其特征在于,所述AI模型包括的第一部分模块和第二部分模块为联合训练得到的。
- 根据权利要求26所述的装置,其特征在于,所述定位信息为定位坐标或者定位需要使用的参数。
- 根据权利要求27所述的装置,其特征在于,所述定位需要使用的参数为以下任一项:信号达到时间、信号达到角度、非视距信息NLOS或视距信息LOS。
- 一种通信装置,其特征在于,所述装置包括处理器和存储器,所述存储器中存储有计算机程序,所述处理器执行所述存储器中存储的计算机程序,以使所述装置执行如权利要求1至6中任一项所述的方法,或者执行如权利要求7至14中任一项所述的方法。
- 一种计算机可读存储介质,用于存储有指令,当所述指令被执行时,使如权利要求1 至6中任一项所述的方法,或者执行如权利要求7至14中任一项所述的方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202280004296.7A CN118235492A (zh) | 2022-10-19 | 2022-10-19 | 一种基于ai模型的终端定位方法及装置 |
PCT/CN2022/126286 WO2024082195A1 (zh) | 2022-10-19 | 2022-10-19 | 一种基于ai模型的终端定位方法及装置 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2022/126286 WO2024082195A1 (zh) | 2022-10-19 | 2022-10-19 | 一种基于ai模型的终端定位方法及装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2024082195A1 true WO2024082195A1 (zh) | 2024-04-25 |
Family
ID=90736540
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2022/126286 WO2024082195A1 (zh) | 2022-10-19 | 2022-10-19 | 一种基于ai模型的终端定位方法及装置 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN118235492A (zh) |
WO (1) | WO2024082195A1 (zh) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109344177A (zh) * | 2018-09-18 | 2019-02-15 | 图普科技(广州)有限公司 | 一种模型组合方法及装置 |
CN109541548A (zh) * | 2018-11-22 | 2019-03-29 | 西安联丰迅声信息科技有限责任公司 | 一种基于匹配场的空气声呐定位方法 |
CN109685202A (zh) * | 2018-12-17 | 2019-04-26 | 腾讯科技(深圳)有限公司 | 数据处理方法及装置、存储介质和电子装置 |
CN111989948A (zh) * | 2020-07-17 | 2020-11-24 | 北京小米移动软件有限公司 | 定位测量数据上报方法、装置、终端及存储介质 |
CN114189889A (zh) * | 2021-12-03 | 2022-03-15 | 中国信息通信研究院 | 一种无线通信人工智能处理方法和设备 |
-
2022
- 2022-10-19 CN CN202280004296.7A patent/CN118235492A/zh active Pending
- 2022-10-19 WO PCT/CN2022/126286 patent/WO2024082195A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109344177A (zh) * | 2018-09-18 | 2019-02-15 | 图普科技(广州)有限公司 | 一种模型组合方法及装置 |
CN109541548A (zh) * | 2018-11-22 | 2019-03-29 | 西安联丰迅声信息科技有限责任公司 | 一种基于匹配场的空气声呐定位方法 |
CN109685202A (zh) * | 2018-12-17 | 2019-04-26 | 腾讯科技(深圳)有限公司 | 数据处理方法及装置、存储介质和电子装置 |
CN111989948A (zh) * | 2020-07-17 | 2020-11-24 | 北京小米移动软件有限公司 | 定位测量数据上报方法、装置、终端及存储介质 |
CN114189889A (zh) * | 2021-12-03 | 2022-03-15 | 中国信息通信研究院 | 一种无线通信人工智能处理方法和设备 |
Also Published As
Publication number | Publication date |
---|---|
CN118235492A (zh) | 2024-06-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2023216203A1 (zh) | 资源配置方法及装置 | |
CN113841430B (zh) | 一种下行控制信息的对齐方法及其装置 | |
WO2023197187A1 (zh) | 一种信道状态信息的处理方法及装置 | |
WO2024138759A1 (zh) | 全球导航卫星系统gnss定位测量方法及装置 | |
WO2024007171A1 (zh) | 一种算力负载均衡方法及装置 | |
WO2024082195A1 (zh) | 一种基于ai模型的终端定位方法及装置 | |
WO2024082196A1 (zh) | 一种基于ai模型的终端定位方法及装置 | |
US20240244566A1 (en) | Positioning method and apparatus | |
WO2024164123A1 (zh) | 一种反馈方法、装置、设备及存储介质 | |
WO2024020747A1 (zh) | 一种模型的生成方法及装置 | |
WO2024164126A1 (zh) | 一种反馈方法、装置、设备及存储介质 | |
WO2024092828A1 (zh) | 一种连接建立的方法及装置 | |
WO2024168917A1 (zh) | 一种ai模型的注册方法/装置/设备及存储介质 | |
WO2024077427A1 (zh) | 一种psfch传输功率的配置方法及装置 | |
WO2023245498A1 (zh) | 一种al/ml模型的数据采集方法及其装置 | |
WO2024164124A1 (zh) | 一种反馈方法、装置、设备及存储介质 | |
WO2024026639A1 (zh) | 一种波束赋形方法、装置、设备及存储介质 | |
WO2024026889A1 (zh) | 一种数据类型确定方法/装置/设备及存储介质 | |
WO2023216165A1 (zh) | 一种控制智能超表面ris发射参考信号的方法及装置 | |
WO2024065846A1 (zh) | 一种多定时提前组tag和时间对齐定时器tat的配置方法及装置 | |
WO2024016136A1 (zh) | Trp或trp组的相位因子和幅度因子发送/接收方法及其装置 | |
WO2024092818A1 (zh) | 同步源选择方法和装置 | |
WO2024159544A1 (zh) | 一种传输块大小的确定方法及其装置 | |
WO2024016245A1 (zh) | 一种信息指示方法、装置、设备及存储介质 | |
WO2024138460A1 (zh) | 定位方法及装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 202280004296.7 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22962385 Country of ref document: EP Kind code of ref document: A1 |