IP Library › Granted Patent US 11,601,146
Granted Patent B2
US 11,601,146 · App. 17/211,639 · Granted Mar 7, 2023

Wireless devices and systems including examples of compensating power amplifier noise with neural networks or recurrent neural networks

Inventor: Fa-Long Luo (San Jose, CA)
Assignee: MICRON TECHNOLOGY, INC.
H04B1/0475G06N3/0445H03F3/20H03F2200/451H04B2001/045H04B2001/0425H04B2001/0433
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 11,601,146
App. No.
17/211,639
Filed
Mar 24, 2021
Granted
Mar 7, 2023
Kind
B2
Art Unit
2646
USPC
455/114.2
Abstract

Examples described herein include methods, devices, and systems which may compensate input data for nonlinear power amplifier noise to generate compensated input data. In compensating the noise, during an uplink transmission time interval (TTI), a switch path is activated to provide amplified input data to a receiver stage including a recurrent neural network (RNN). The RNN may calculate an error representative of the noise based partly on the input signal to be transmitted and a feedback signal to generate filter coefficient data associated with the power amplifier noise. The feedback signal is provided, after processing through the receiver, to the RNN. During an uplink TTI, the amplified input data may also be transmitted as the RF wireless transmission via an RF antenna. During a downlink TTI, the switch path may be deactivated and the receiver stage may receive an additional RF wireless transmission to be processed in the receiver stage.

Claims (18)

1. An apparatus, comprising:

a plurality of power amplifiers configured to amplify respective signals of a plurality of signals to generate respective amplified signals of a plurality of amplified signals; and

a recurrent neural network (RNN) including a plurality of multiplication or accumulation (MAC) processing units, a plurality of delay units, and a plurality of memory look-up units (MLU units) configured to receive respective ones of the plurality of amplified signals and to generate filter coefficient data for filtering nonlinear power amplifier noise using the respective amplified signals; and

a plurality of switches, each switch configured to activate a respective switch path that provides the respective amplified signal to the RNN.

2. The apparatus of claim 1 , further comprising:

a plurality of processing units configured for bit manipulation and configured to receive the respective amplified signals and to forward the respective amplified signal to the respective MAC processing units of the plurality of MAC processing units, wherein the plurality of processing units implement the RNN in conjunction with the plurality of MAC units, the plurality of delay units, and the plurality of MLU units.

3. The apparatus of claim 2 , wherein the plurality of processing units comprise two or more of a NOT logic unit, an AND logic unit, an OR logic unit, a NOR logic unit, a NAND logic unit, or an XOR logic unit.

4. The apparatus of claim 1 , wherein the apparatus implements an Internet-of-Things (IoT) smart home device, an IoT agricultural device or an IoT industrial device.

5. The apparatus of claim 1 , further comprising:

a plurality of antennas, each antenna configured to transmit a respective amplified signal of the plurality of amplified signals.

6. An apparatus comprising:

a. plurality of power amplifiers configured to amplify respective signals of a plurality of signals to generate respective amplified signals of a plurality of amplified signals; and

a plurality of multiplication or accumulation (MAC) processing units, a plurality of delay units, and a plurality of memory look-up units (MLU units) configured to receive respective ones of the plurality of amplified signals and to generate filter coefficient data for filtering nonlinear power amplifier noise using the respective amplified signals, wherein each MAC processing unit of the plurality of MAC processing units is configured to generate corresponding intermediate processing results based on the respective amplified signals and a delayed version of at least one of the corresponding intermediate processing results being mixed using a plurality of coefficients.

7. The apparatus of claim 6 , wherein at least one MLU of the plurality of MLUs is configured to store and provide the corresponding intermediate processing results.

8. The apparatus of claim 6 , wherein at least one MLU of the plurality of MLUs is configured to provide output data based on the respective amplified signals and the corresponding intermediate processing results being mixed using the plurality of coefficients.

9. An apparatus comprising:

a plurality of power amplifiers configured to amplify respective signals of a plurality of signals to generate respective amplified signals of a plurality of amplified signals; and

a plurality of multiplication or accumulation (MAC) processing units, a plurality of delay units, and a plurality of memory look-up units (MLU units) configured to receive respective ones of the plurality of amplified signals and to generate filter coefficient data for filtering nonlinear power amplifier noise using the respective amplified signals, wherein each delay unit of the plurality of delay units is associated with a respective MAC processing unit and configured to provide the delayed versions of respective outputs of the plurality of MAC processing units to a respective input of the respective MAC processing unit with which the respective delay unit is associated.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 30, 2023
From: MICRON TECHNOLOGY, INC.
To: LODESTAR LICENSING GROUP LLC
Reel/Frame 064751/0750 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 24, 2021
From: LUO, FA-LONG
To: MICRON TECHNOLOGY, INC.
Reel/Frame 055706/0786 →
Continuity (2)
Continuation 16849696 · Apr 15, 2020
Related Publication 20210328608A1 · Oct 21, 2021
Cited By (1)
US 12,199,650