Systems and methods for optimized beamforming and compression for uplink MIMO cloud radio access networks

Patent metadata

Field	Value
Publication number	US-9537556-B2
Application number	US-201514794684-A
Country	US
Kind code	B2
Filing date	Jul 8, 2015
Priority date	Jul 11, 2014
Publication date	Jan 3, 2017
Grant date	Jan 3, 2017

Abstract

System and method embodiments are provided to optimize uplink multiple-input-multiple-output (MIMO) beamforming for uplink and compression for fronthaul links transmission in cloud radio access network (C-RANs). In an embodiment, cloud-computing based central processor (CP) obtains channel state information for a mobile device (MD) being served by a plurality of access points (APs) in a C-RAN, and generates a channel gain matrix in accordance with the channel state information. A weighted sum-rate maximization model is then established using the channel gain matrix in accordance with power constraints of transmission from the MD to the APs and capacity constraints of fronthaul links connecting the APs to the CP. The CP calculates a transmit beamforming vector for the MD and a quantization noise covariance matrix for the APs jointly by applying a weighted minimum-mean-square-error successive convex approximation algorithm, or separately by applying an approximation algorithm, to solve the weighted sum-rate maximization model.

First claim

What is claimed is:

1. A method for determining a transmit beamformer and quantization noise covariance matrix for uplink multiple-input-multiple-output (MIMO) communications in a cloud radio access network (C-RAN), the method comprising:
obtaining, by a central processor (CP), channel state information associated with a mobile device (MD) being served by a plurality of access points (APs) in the C-RAN;
generating a channel gain matrix in accordance with the channel state information;
establishing a weighted sum-rate maximization model using the channel gain matrix in accordance with power constraints of transmission from the MD to the APs, and capacity constraints of fronthaul links for compressed transmission of received MD signals from the APs to the CP; and
jointly calculating a transmit beamforming vector for the MD and a quantization noise covariance matrix for the APs by applying a weighted minimum-mean-square-error successive convex approximation (WMMSE-SCA) algorithm to solve the weighted sum-rate maximization model.

2. The method of claim 1 , wherein applying the WMMSE-SCA algorithm to solve the weighted sum-rate maximization model includes:
initializing a transmit beamforming vector for the MD, and a quantization noise covariance matrix for the APs;
calculating a receive signal covariance matrix for each of the APs according to the transmit beamforming vector, the quantization noise covariance matrix, the channel gain matrix, and a background noise covariance matrix;
obtaining a minimum-mean-squared-error (MMSE) receive beamforming vector for the CP based on the transmit beamforming vector, the quantization noise covariance matrix, the channel gain matrix, and the background noise covariance matrix;
calculating a weight matrix according to the transmit beamforming vector, the MMSE receive beamforming vector, and the channel gain matrix; and
recalculating the transmit beamforming vector and the quantization noise covariance matrix by solving a convex optimization model based on the receive signal covariance matrix, the MMSE receive beamforming vector, the weight matrix, the power constraints, and the capacity constraints.

3. The method of claim 2 further comprising repeating calculating the weight matrix and recalculating the transmit beamforming vector and the quantization noise covariance matrix until the transmit beamforming vector and the quantization noise covariance matrix converge in value.

4. The method of claim 1 further comprising:
sending the quantization noise covariance matrix from the CP to the APs; and
sending the transmit beamforming vector to the MD.

5. A method for determining a transmit beamformer and a quantization noise covariance matrix for uplink multiple-input-multiple-output (MIMO) communications in a cloud radio access network (C-RAN), the method comprising:
obtaining, by a central processor (CP), channel state information for a mobile device (MD) being served by a plurality of access points (APs) in the C-RAN;
generating a channel gain matrix in accordance with the channel state information;
establishing a weighted sum-rate maximization model using the channel gain matrix in accordance with power constraints of transmission from the MD to the APs, and capacity constraints of fronthaul links for compressed transmission of received MD signals from the APs to the CP; and
separately calculating a transmit beamforming vector for the MD and a quantization noise covariance matrix for the APs by applying an approximation algorithm to solve the weighted sum-rate maximization model.

6. The method of claim 5 , wherein applying the approximation algorithm to solve the weighted sum-rate maximization model includes:
obtaining a transmit beamforming vector for the MD by applying a singular value decomposition of the channel gain matrix;
establishing a relation between quantization noise and the capacity constraints of the fronthaul links, wherein the relation is a function of the transmit beamforming vector and the channel gain matrix; and
determining the quantization noise covariance matrix by applying bisection to the established relation.

7. The method of claim 5 , wherein calculating the transmit beamforming vector includes matching the transmit beamforming vectors to a strongest channel signal vector.

8. The method of claim 5 , wherein calculating the quantization noise covariance matrix includes determining per each antenna at each AP a scalar quantizer with uniform quantization noise levels across all antennas of the AP.

9. The method of claim 5 further comprising applying successive interference cancelation (SIC) at the CP.

10. The method of claim 5 , wherein the transmit beamforming vector is calculated for transmitting signals above a defined signal-to-quantization-noise ratio (SQNR).

11. The method of claim 5 further comprising:
sending the quantization noise covariance matrix from the CP to the APs; and
sending the transmit beamforming vector to the MD.

12. A network component for determining a transmit beamformer and a quantization noise covariance matrix for uplink multiple-input-multiple-output (MIMO) communications in a cloud radio access network (C-RAN), the network component comprising:
a processor;
a non-transitory computer readable storage medium storing programming for execution by the processor, the programming including instructions to:
obtain channel state information for a mobile device (MD) being served by a plurality of access points (APs) in the C-RAN;
generate a channel gain matrix in accordance with the channel state information;
establish a weighted sum-rate maximization model using the channel gain matrix in accordance with power constraints of transmission from the MD to the APs, and capacity constraints of fronthaul links for compressed transmission of received MD signals from the APs to the network component; and
jointly calculate a transmit beamforming vector for the MD and a quantization noise covariance matrix for the APs by applying a weighted minimum-mean-square-error successive convex approximation (WMMSE-SCA) algorithm to solve the weighted sum-rate maximization model.

13. The network component of claim 12 , wherein the instructions to apply the WMMSE-SCA algorithm include instructions to:
initialize a transmit beamforming vector for the MD, and a quantization noise covariance matrix for the APs;
calculate a receive signal covariance matrix for each of the APs according to the transmit beamforming vector, the quantization noise covariance matrix, the channel gain matrix, and a background noise covariance matrix;
obtain a minimum-mean-squared-error (MMSE) receive beamforming vector for the network component based on the transmit beamforming vector, the quantization noise covariance matrix, the channel gain matrix, and the background noise covariance matrix;
calculate a weight matrix according to the transmit beamforming vector, the MMSE receive beamforming vector, and the channel gain matrix; and
recalculate the transmit beamforming vector and the quantization noise covariance matrix by solving a convex optimization model based on the receive signal covariance matrix, the MMSE receive beamforming vector, the weight matrix, the power constraints, and the capacity constraints.

14. The network component of claim 13 , wherein the instructions to apply the WMMSE-SCA algorithm include further instructions to repeat calculating the weight matrix and recalculating the transmit beamforming vector and the quantization noise covariance matrix until the transmit beamforming vector and the quantization noise covariance matrix converge in value.

Assignees

• Huawei Tech Canada Co Ltd
• Governing Council Univ Toronto

Inventors

• Zhou Yuhan
• Yu Wei
• Baligh Mohammadhadi

Classifications

• H04B7/0617 Primary

  for beam forming · CPC title

• H04W84/045

  using private Base Stations, e.g. femto Base Stations, home Node B · CPC title

• H04B7/0404

  the mobile station comprising multiple antennas, e.g. to provide uplink diversity · CPC title

• H04B7/0413

  MIMO systems · CPC title