Deep convolutional neural network powered terahertz ultra-massive multi-input-multi-output channel estimation method

Patent metadata

Field	Value
Publication number	US-11784685-B2
Application number	US-202217807306-A
Country	US
Kind code	B2
Filing date	Jun 16, 2022
Priority date	Jun 28, 2021
Publication date	Oct 10, 2023
Grant date	Oct 10, 2023

Abstract

A THz UM-MIMO channel estimation method based on the DCNN comprises the steps: the hybrid spherical and planar-wave modeling (HSPM), by taking a sub-array in the antenna array as a unit, employing the PWM within the sub-array, and employing the SWM among the sub-arrays; estimating the channel parameters between the reference sub-arrays at Tx and Rx through a DCNN, including the angles of departure and arrival, the propagation distance and the path gain; deducing the channel parameters between the reference sub-array and other sub-arrays by utilizing the obtained channel parameters and the geometrical relationships among sub-arrays, and recovering the channel matrix; wherein accurate three-dimensional channel modeling is achieved by the HSPM, which possesses high modeling accuracy and low complexity.

First claim

What is claimed is:

1. A Terahertz (THz) ultra-massive multi-input-output (UM-MIMO) channel estimation (CE) method based on a deep convolutional neural network (DCNN), comprising the following steps:
step i), a hybrid spherical and planar-wave modeling (HSPM), which takes a sub-array as a unit, using a planar-wave channel model (PWM) in the sub-array, and models a channel among sub-arrays by a spherical-wave channel model (SWM);
step ii), using a first sub-array at a transceiver end as a reference sub-array, using the DCNN to estimate a departure angle, an angle of arrival, a propagation distance and a path gain between reference sub-arrays according to real values, element imaginary values and element absolute values of channel observation matrix; and
step iii), deriving channel parameters between the reference sub-array and remaining sub-arrays by using the channel parameters and geometric relationship between the sub-arrays, and reconstruct the channel observation matrix.

2. The method according to claim 1 , wherein step i) comprises the following steps:
a) dividing antennas at transmitter (Tx) and receiver (Rx) into K t and K r sub-arrays, respectively, and different sub-arrays have the same multi-path number N p , the amplitude of the channel gain between different sub-arrays is the same, while the phase of the channel gain is changed due to different geometric distances and transceiver angles, to obtain a block structured channel model:

H

H
⁢
S
⁢
P
⁢
M

=

∑

p
=
1

N
p

[

❘
"\[LeftBracketingBar]"

α
p

1
⁢
1

❘
"\[RightBracketingBar]"

⁢

e

-
j

⁢

2
⁢
π

λ

⁢

D
p
11

⁢

a

r
⁢
p

11

(

a
tp
11

)

H

…
⁢

❘
"\[LeftBracketingBar]"

α
p

1
⁢
1

❘
"\[RightBracketingBar]"

⁢

e

-
j

⁢

2
⁢
π

λ

⁢

D
p

1
⁢

K
t

⁢

a
rp

1
⁢

K
t

(

a
tp

1
⁢

K
t

)

H

Assignees

• Univ Shanghai Jiaotong

Inventors

• Han Chong
• Chen Yuhang
• Yan Longfei

Classifications

• H04B7/0452 Primary

  Multi-user MIMO systems · CPC title

• H04B7/043

  using best eigenmode, e.g. beam forming or beam steering · CPC title

• H04L25/0254 Primary

  using neural network algorithms · CPC title

• H04B10/90

  Non-optical transmission systems, e.g. transmission systems employing non-photonic corpuscular radiation · CPC title

• H04B7/0456

  Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting · CPC title