IP Library Granted Patent US 9,461,724
Granted Patent B1
US 9,461,724 · App. 14/462,505 · Granted Oct 4, 2016

Matrix processing engine for multi-tone MIMO systems

Inventor: Sigurd Schelstraete (Menlo Park, CA)
Assignee: Quantenna Communications, Inc.
H04B7/0456H04B7/0452H04L25/0204H04W84/12
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,461,724
App. No.
14/462,505
Granted
Oct 4, 2016
Kind
B1
Abstract

A wireless transceiver apparatus including: a channel estimator, a discrete matrix transform module, a tone set controller, and a shared matrix transform module. The channel estimator determines a channel matrix for each tone on each communication link. The discrete matrix transform module derives a transform matrix for the communication link(s) on a selected base tone from the associated channel matrix determined by the channel estimator. The tone set controller initializes an estimation of the transform matrix, on a neighboring tone, with the transform matrix derived for the base tone. The shared matrix transform module iteratively estimates the transform matrix on a neighboring tone, from successive perturbations of the initial transform matrix derived for the base tone and the associated channel matrix for the neighboring tone; thereby reducing a number of iterations required to estimate the transform matrix on the neighboring tone.

Claims (36)

1. A wireless access point (WAP) supporting wireless communications with associated stations on a selected communication channel of a wireless local area network (WLAN); and the WAP comprising:

a plurality of antennas;

a plurality of components coupled to one another to form a plurality of the transmit and receive paths each coupled to a corresponding one of the plurality of antennas for multiple-input multiple-output (MIMO) orthogonal frequency division multiplexed (OFDM) communications on the selected communication channel;

a channel estimator coupled to the plurality of components and configured to explicitly sound each of at least two of the associated stations targeted for a multi-user (MU)-MIMO communication link and to determine from feedback received therefrom a channel matrix “H” to each of the targeted stations for corresponding ones of the OFDM tones;

a tone set controller coupled to the channel estimator and configured to identify sets of the selected communication channel's OFDM tones and within each set a base and at least one neighboring tone proximate in frequency with the base tone;

a discrete at transform module configured to derive as precode matrix for each base tone from the corresponding channel matrices for each base tone as determined by the channel estimator, for spatially mapping MU-MIMO transmissions to the at least two stations;

a shared matrix transform module coupled to the discrete matrix transform module and configured to estimate a precode matrix for the at least one a neighboring tone, from the corresponding channel matrices for the at least one neighboring tone and the precede matrix for the base tone of the corresponding set of tones; thereby reducing an amount of time required for the estimation of each neighboring tone's precode matrix by the shared matrix transform module, relative to the amount of time required for the determination of the base tone's precode matrix by the discrete matrix transform module; and

a spatial mapper component of the plurality of components for spatially mapping the MU-MIMO OFDM wireless transmissions utilizing the corresponding base tone precode matrix for the base tone in each set and the corresponding neighboring tone precode matrix for the at least one neighbor tone in each set.

2. The WAP of claim 1 , further comprising:

the shared matrix transform module further configured to iteratively estimate the precede matrix for the at least one neighboring tone in each set, from successive perturbations of the precode matrix determined for the base tone in each set.

3. The WAP of claim 1 , further comprising:

the channel estimator further configured to determine the channel matrices for the base tone in each set and to interpolate the channel matrices the at least one neighboring tone in each set from the channel matrices for the base tone in each set.

4. The WAP of claim 1 , further comprising:

the shared matrix transform module further configured to iteratively estimate the corresponding precode matrix for each of the neighboring tones either serially or in parallel with respect to one another.

5. The WAP of claim 1 , further comprising:

the shared matrix transform module further configured to carry forward an estimated precode matrix from a first neighboring tone to initialize an estimate of the precode matrix for the next neighboring tone.

6. The WAP of claim 1 , further comprising:

the discrete matrix transform module further configured to derive a precode matrix for each base tone from a singular valued decomposition (SVD) of the corresponding channel matrices for each base tone as determined by the channel estimator.

7. A method for operating a wireless access point (WAP) having a plurality of antennas and supporting a multiple-input multiple-output (MIMO) wireless local area network (WLAN) associated stations on a selected one of a plurality of orthogonal frequency-division multiplexed (OFDM) communication channels; the method comprising the acts of:

explicitly sounding each of at least two of the associated stations targeted for a multi-user (MU)-MIMO communication link for feedback of the channel matrix “H” to each of the targeted stations for corresponding ones of the ODFM tones;

identifying sets of the selected communication channel's OFDM tones and within each set a base tone and at least one neighboring tone proximate in frequency with the base tone;

deriving for each base tone, a distinct precode matrix from the corresponding channel matrices for each base tone;

estimating for the at least one neighboring tone a distinct precode matrix from the corresponding channel matrices for the at least one neighboring tone, and the precode matrix the base tone of the corresponding set of tones; and

spatially mapping MU-MIMO OFDM wireless transmissions with the corresponding precode matrix for the base tone in each set derived in the deriving act and the corresponding precode matrix for the at least one neighboring tone in each set estimated in the estimating act.

8. The method for operating the WAP of claim 7 , wherein the estimating act further comprises:

iteratively estimating the precode matrix for the at least one neighboring tone in each set, from successive perturbations of the preemie matrix for the base tone in each set.

9. The method for operating the WAP of claim 7 , wherein the explicit sounding act further comprises:

determining the channel matrices for the selected base tone in each set; and

interpolating the channel matrices for the at least one neighboring tone in each set from the channel matrices for the base tone in each set determined in the determining act.

10. The method for operating the WAP of claim 7 , wherein the at least one neighboring tone comprises a plurality of neighboring tones, and the estimating act further comprises one of the acts of:

estimating the corresponding precode matrix for each of the neighboring tones serially with respect to one another; and

iteratively estimating the corresponding precode matrix for each of the neighboring tones in parallel with respect to one another.

11. The method for operating the WAP of claim 7 , wherein the estimating act further comprises:

carrying forward an estimated precode matrix from a first neighboring tone to initialize an estimate of the precode matrix for the next neighboring tone.

12. The method for operating the WAP of claim 7 , wherein the deriving act further comprises:

deriving for each base tone, a distinct precode matrix from a singular valued decomposition (SVD) of the corresponding channel matrices for each base tone.

Assignments (7)
RELEASE OF SECURITY INTEREST IN PATENTS, RECORDED AT REEL 051426, FRAME 0410 Recorded Jun 22, 2023
From: DEUTSCHE BANK AG NEW YORK BRANCH, AS COLLATERAL AGENT
To: ON SEMICONDUCTOR CONNECTIVITY SOLUTIONS, INC., AS GRANTOR
Reel/Frame 064067/0340 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 8, 2023
From: SEMICONDUCTOR COMPONENTS INDUSTRIES, LLC
To: MAXLINEAR, INC.
Reel/Frame 063572/0701 →
RELEASE OF SECURITY INTEREST Recorded May 2, 2023
From: DEUTSCHE BANK AG NEW YORK BRANCH
To: ON SEMICONDUCTOR CONNECTIVITY SOLUTIONS, INC.
Reel/Frame 063516/0736 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 6, 2023
From: ON SEMICONDUCTOR CONNECTIVITY SOLUTIONS, INC.
To: SEMICONDUCTOR COMPONENTS INDUSTRIES, LLC
Reel/Frame 063280/0591 →
MERGER AND CHANGE OF NAME Recorded Apr 6, 2023
From: RAPTOR OPERATIONS SUB, INC.; QUANTENNA COMMUNICATIONS, INC.
To: ON SEMICONDUCTOR CONNECTIVITY SOLUTIONS, INC.
Reel/Frame 063271/0657 →
PATENT SECURITY AGREEMENT Recorded Dec 26, 2019
From: ON SEMICONDUCTOR CONNECTIVITY SOLUTIONS, INC.
To: DEUTSCHE BANK AG NEW YORK BRANCH, AS COLLATERAL AGENT
Reel/Frame 051426/0410 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 30, 2014
From: SCHELSTRAETE, SIGURD
To: QUANTENNA COMMUNICATIONS, INC.
Reel/Frame 033644/0197 →