Reference signals sampling and imputation for enabling parameter estimation via deep learning
AssigneeQUALCOMM Incorporated
InventorsMohamed Fouad Ahmed Marzban, Wooseok Nam, Tao Luo, Taesang Yoo
Life of this patent
Estimating the term…
Abstract
Disclosed is a method for wireless communication. The method comprises receiving, from a network node, a reference signal (RS) of a first type. An input signal of a second type is determined based on the reference signal. Signals of the second type are associated with training a first machine learning (ML) model applied by a UE to estimate a parameter used in wireless communications by the UE. The input signal of the second type is applied to the first ML model to estimate the parameter used in wireless communications by the UE.
Claims22
A method for wireless communication by a user equipment (UE), comprising:
- receiving, from a network node, a reference signal (RS) of a first type;
- determining, based on the reference signal, an input signal of a second type, wherein signals of the second type are associated with training a first machine learning (ML) model applied by the UE to estimate a parameter used in wireless communications by the UE;
- and
- applying the input signal of the second type to the first ML model to estimate the parameter used in wireless communications by the UE,
- wherein the RS of the first type comprises a first quantity of resource elements, REs, and the input signal comprises a second quantity of REs,
- wherein determining the input signal comprises:
- undersampling the reference signal to determine the input signal based on reducing the first quantity of REs to the second quantity of REs, if the first quantity of REs is larger than the second quantity of REs.
A method for wireless communication by a network node, comprising:
- determining a transformation configured to transform a first reference signal (RS) of a first type to an input signal of a second type, wherein signals of the second type are associated with training a first machine learning (ML) model applied by a User Equipment (UE) to estimate a parameter used in wireless communications by the UE;
- transmitting, to the UE, an indication of the transformation;
- and
- transmitting, to the UE, the first reference signal (RS) of the first type,
- wherein the RS of the first type comprises a first quantity of resource elements, REs, and the input signal comprises a second quantity of REs,
- wherein the transformation comprises undersampling the first reference signal to determine the input signal based on reducing the first quantity of REs to the second quantity of REs, if the first quantity of REs is larger than the second quantity of REs.
An apparatus for wireless communication by a user equipment (UE), comprising:
- memory;
- and
- one or more processors coupled with the memory, the memory and the one or more processors configured to:
- receive, from a network node, a reference signal (RS) of a first type;
- determine, based on the reference signal, an input signal of a second type, wherein signals of the second type are associated with training a first machine learning (ML) model applied by the UE to estimate a parameter used in wireless communications by the UE;
- and
- apply the input signal of the second type to the first ML model to estimate the parameter used in wireless communications by the UE,
- wherein the RS of the first type comprises a first quantity of resource elements, REs, and the input signal comprises a second quantity of REs,
- wherein determining the input signal comprises:
- undersampling the reference signal to determine the input signal based on reducing the first quantity of REs to the second quantity of REs, if the first quantity of REs is larger than the second quantity of REs.
An apparatus for wireless communication by a network node, comprising:
- memory;
- and
- one or more processors coupled with the memory, the memory and the one or more processors configured to:
- determine a transformation configured to transform a first reference signal (RS) of a first type to an input signal of a second type, wherein signals of the second type are associated with training a first machine learning (ML) model applied by a User Equipment (UE) to estimate a parameter used in wireless communications by the UE;
- transmit, to the UE, an indication of the transformation;
- and
- transmit, to the UE, the first reference signal (RS) of the first type,
- wherein the RS of the first type comprises a first quantity of resource elements, REs, and the input signal comprises a second quantity of REs,
- wherein the transformation comprises undersampling the first reference signal to determine the input signal based on reducing the first quantity of REs to the second quantity of REs, if the first quantity of REs is larger than the second quantity of REs.
Dependencies are inferred from the claim text.