U.S. patent application number 16/440626 was filed with the patent office on 2020-09-17 for nonlinear self-interference cancellation with sampling rate mismatch.
The applicant listed for this patent is Samsung Electronics Co., Ltd.. Invention is credited to Pranav Dayal, Hamed Maleki, Elina Nayebi, Kee-Bong Song.
Application Number | 20200295793 16/440626 |
Document ID | / |
Family ID | 1000004184630 |
Filed Date | 2020-09-17 |
United States Patent
Application |
20200295793 |
Kind Code |
A1 |
Maleki; Hamed ; et
al. |
September 17, 2020 |
NONLINEAR SELF-INTERFERENCE CANCELLATION WITH SAMPLING RATE
MISMATCH
Abstract
A method for providing nonlinear self-interference cancellation
of a wireless communication device includes: receiving digital
samples of an interfering signal having a first sampling rate and a
corrupted victim signal having a second sampling rate; generating a
kernel vector based on the interfering signal, wherein the kernel
vector has terms of nonlinear self-interference; estimating the
nonlinear self-interference of the corrupted victim signal using
the terms of the nonlinear self-interference; and providing an
estimation of a desired signal by cancelling the nonlinear
self-interference from the corrupted victim signal.
Inventors: |
Maleki; Hamed; (San Diego,
CA) ; Nayebi; Elina; (San Diego, CA) ; Dayal;
Pranav; (San Diego, CA) ; Song; Kee-Bong; (San
Diego, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Samsung Electronics Co., Ltd. |
Suwon-si |
|
KR |
|
|
Family ID: |
1000004184630 |
Appl. No.: |
16/440626 |
Filed: |
June 13, 2019 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62816422 |
Mar 11, 2019 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04L 2025/03375
20130101; H04B 1/1027 20130101; H04L 25/03019 20130101 |
International
Class: |
H04B 1/10 20060101
H04B001/10; H04L 25/03 20060101 H04L025/03 |
Claims
1. A method for providing nonlinear self-interference cancellation
for a wireless communication device, the method comprising:
receiving digital samples of an interfering signal and a corrupted
victim signal; determining that a first sampling rate of the
interfering signal is lower than a second sampling rate of the
corrupted victim signal; generating an interpolated interfering
signal by interpolating the interfering signal to match the first
sampling rate of the interfering signal to the second sampling rate
of the corrupted victim signal; generating a kernel vector based on
the interpolated interfering signal, wherein the kernel vector has
terms of the nonlinear self-interference; estimating nonlinear
self-interference of the corrupted victim signal using the terms of
the nonlinear self-interference; and providing an estimation of a
desired signal by cancelling the nonlinear self-interference from
the corrupted victim signal.
2. The method of claim 1, further comprising: estimating nonlinear
coefficients of the nonlinear self-interference; and generating an
estimated nonlinear self-interference signal based on the kernel
vector and the nonlinear coefficients, wherein the estimation of
the desired signal is provided by canceling the estimated nonlinear
self-interference signal from the corrupted victim signal.
3. The method of claim 1, wherein the nonlinear coefficients are
generated using a least square (LS) scheme or a recursive LS (RLS)
scheme.
4. The method of claim 1, wherein the interpolated interfering
signal is generated by Lagrange interpolation.
5. The method of claim 1, wherein the interfering signal is
received from a transmitter of the wireless communication device,
and the estimation of the desired signal is fed into a receiver of
the wireless communication device.
6. The method of claim 5, wherein the corrupted victim signal is
provided by an analog-to-digital converter (ADC) coupled to the
receiver.
7. A method for providing nonlinear self-interference cancellation
for a wireless communication device, the method comprising:
receiving digital samples of an interfering signal and a corrupted
victim signal; determining that a first sampling rate of the
interfering signal is greater than a second sampling rate of the
corrupted victim signal; generating a kernel vector based on the
digital samples of the interfering signal by matching the first
sampling rate of the interfering signal to the second sampling rate
of the corrupted victim signal; estimating nonlinear coefficients
of the kernel vector using the digital samples of the corrupted
victim signal; generating an estimated nonlinear self-interference
signal based on the kernel vector and the nonlinear coefficients;
and providing an estimation of a desired signal by cancelling the
estimated nonlinear self-interference signal from the corrupted
victim signal.
8. The method of claim 7, further comprising: determining power of
an aliasing portion of the interfering signal is greater than a
threshold; generating frequency-shifted kernels by
frequency-shifting the kernel vector and bringing each aliasing
portion into a desired band associated with the desired signal;
generating a decimated kernel vector by filtering and decimating
the frequency-shifted kernel to match a third sampling rate of the
frequency-shifted kernel to the second sampling rate of the
corrupted victim signal; and generating the nonlinear
self-interference based on the decimated kernel vector.
9. The method of claim 7, further comprising: determining power of
an aliasing portion of the interfering signal is smaller than a
threshold; generating a decimated kernel vector by filtering and
decimating the kernel vector; and generating the nonlinear
self-interference based on the decimated kernel vector.
10. The method of claim 7, wherein the interfering signal is
received from a transmitter of the wireless communication device,
and the estimation of the desired signal is fed into a receiver of
the wireless communication device.
11. The method of claim 10, wherein the corrupted victim signal is
provided by an analog-to-digital converter (ADC) coupled to the
receiver.
12. A wireless communication device comprising: a nonlinear
self-interference cancellation (NSIC) logic and a nonlinear
self-interference coefficients estimator, wherein the NSIC logic
receives digital samples of an interfering signal sampled at a
first sampling rate and a corrupted victim signal sampled at a
second sampling rate, the first sampling rate being greater than
the second sampling rate, generates a kernel vector based on the
digital samples of the interfering signal by matching the first
sampling rate of the interfering signal to the second sampling rate
of the corrupted victim signal, and wherein the nonlinear
self-interference coefficients estimator estimates nonlinear
coefficients of the kernel vector using the digital samples of the
corrupted victim signal, and wherein the NSIC logic generates an
estimated nonlinear self-interference signal based on the kernel
vector and the nonlinear coefficients and generates an estimation
of a desired signal by cancelling the nonlinear self-interference
signal from the corrupted victim signal.
13. The wireless communication device of claim 12 further
comprising a transmitter and a receiver, wherein the interfering
signal is received from the transmitter, and the estimation of the
desired signal is fed into the receiver.
14. The wireless communication device of claim 13, further
comprising an analog-to-digital converter (ADC) coupled to the
receiver, wherein the ADC provides the corrupted victim signal.
15. The wireless communication device of claim 12, wherein the NSIC
logic comprises an interpolator that generates an interpolated
interfering signal by interpolating the interfering signal to match
the first sampling rate of the interfering signal to the second
sampling rate of the corrupted victim signal.
16. (canceled)
17. (canceled)
18. The wireless communication device of claim 12, wherein the NSIC
logic generates the estimation of the desired signal by canceling
the estimated nonlinear self-interference signal from the corrupted
victim signal.
19. The wireless communication device of claim 18, wherein the NSIC
logic further comprises a frequency shifter and a low-pass filter
and decimator.
20. The wireless communication device of claim 19, wherein, if the
NSIC logic determines that power of an aliasing portion of the
interfering signal is greater than a threshold, the frequency
shifter generates frequency-shifted kernels by frequency-shifting
the kernel vector and bringing each aliasing portion into a desired
band associated with the desired signal, and the low-pass filter
and decimator generates a decimated kernel vector by filtering and
decimating the frequency-shifted kernels to match a third sampling
rate of the frequency-shifted kernel to the second sampling rate of
the corrupted victim signal, and wherein the NSIC logic generates
the nonlinear self-interference based on the decimated kernel
vector.
21. The wireless communication device of claim 19, wherein, if the
NSIC logic determines that power of an aliasing portion of the
interfering signal is smaller than a threshold, the low-pass filter
and decimator generates a decimated kernel vector by filtering and
decimating the kernel vector; and wherein the NSIC logic generates
the nonlinear self-interference based on the decimated kernel
vector.
Description
CROSS-REFERENCE TO RELATED APPLICATION(S)
[0001] This application claims the benefits of and priority to U.S.
Provisional Patent Application Ser. No. 62/816,422 filed Mar. 11,
2019, the disclosure of which is incorporated herein by reference
in its entirety.
TECHNICAL FIELD
[0002] The present disclosure is generally related to wireless
communication systems, in particular, to a system and method for
providing nonlinear self-interference cancellation (NSIC) with
sampling rate mismatch.
BACKGROUND
[0003] A wireless communication device such as a mobile device may
include multiple radio access technologies (RATs). Examples of such
RATs include, but are not limited to, long term evolution (LTE),
5th generation new radio (5G NR), Wi-Fi, Bluetooth (BT), and global
navigation satellite system (GNSS).
[0004] An analog circuit including a power amplifier can create
higher order components of an interfering signal, and they may land
into an operating frequency band of the desired signal creating
nonlinear distortion to the desired signal. Nonlinear
self-interference may be caused by simultaneous transmission of the
same carrier, other carriers, or other RATs on the same wireless
communication device where signal reception happens on a given
carrier. There may be different sources of nonlinear
self-interference including harmonic interference from a single
carrier or a particular RAT, inter-modulation distortion (IMD)
between multiple carriers, and interference from reciprocal mixing
of one or more carriers. Typically, nonlinear self-interference
cancellation (NSIC) is applied when both an interfering signal and
a victim (interfered) signal have a same sampling rate.
SUMMARY
[0005] According to one embodiment, a method for providing
nonlinear self-interference cancellation includes: receiving
digital samples of an interfering signal and a corrupted victim
signal; determining that a first sampling rate of the interfering
signal is lower than a second sampling rate of the corrupted victim
signal; generating an interpolated interfering signal by
interpolating the interfering signal to match the first sampling
rate of the interfering signal to the second sampling rate of the
corrupted victim signal; generating a kernel vector based on the
interpolated interfering signal, wherein the kernel vector has
terms of the nonlinear self-interference; estimating nonlinear
self-interference of the corrupted victim signal using the terms of
the nonlinear self-interference; and providing an estimation of a
desired signal by cancelling the nonlinear self-interference from
the corrupted victim signal.
[0006] According to another embodiment, a method for providing
nonlinear self-interference cancellation includes: receiving
digital samples of an interfering signal and a corrupted victim
signal; determining that a first sampling rate of the interfering
signal is greater than a second sampling rate of the corrupted
victim signal; generating a kernel vector based on the interfering
signal, wherein the kernel vector has terms of the nonlinear
self-interference; estimating nonlinear self-interference of the
corrupted victim signal using the terms of the nonlinear
self-interference; and providing an estimation of a desired signal
by cancelling the nonlinear self-interference from the corrupted
victim signal.
[0007] According to another embodiment, a wireless communication
device includes: a nonlinear self-interference cancellation (NSIC)
logic, wherein the NSIC logic is configured to receive an
interfering signal sampled at a first sampling rate and a corrupted
victim signal sampled at a second sampling rate, generate a kernel
vector having terms of nonlinear self-interference, estimate the
nonlinear self-interference of the corrupted victim signal, and
generate an estimation of a desired signal by cancelling the
nonlinear self-interference from the corrupted victim signal.
[0008] The above and other preferred features, including various
novel details of implementation and combination of events, will now
be more particularly described with reference to the accompanying
figures and pointed out in the claims. It will be understood that
the particular systems and methods described herein are shown by
way of illustration only and not as limitations. As will be
understood by those skilled in the art, the principles and features
described herein may be employed in various and numerous
embodiments without departing from the scope of the present
disclosure.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] The accompanying drawings, which are included as part of the
present specification, illustrate the presently preferred
embodiment and together with the general description given above
and the detailed description of the preferred embodiment given
below serve to explain and teach the principles described
herein.
[0010] FIG. 1 is a simplified block diagram of an example system,
according to one embodiment;
[0011] FIG. 2 illustrates a block diagram of an example NSIC
system, according to one embodiment; and
[0012] FIG. 3 illustrates a flow chart of an exemplary nonlinear
self-interference cancellation (NISC) process, according to one
embodiment.
[0013] The figures are not necessarily drawn to scale and elements
of similar structures or functions are generally represented by
like reference numerals for illustrative purposes throughout the
figures. The figures are only intended to facilitate the
description of the various embodiments described herein. The
figures do not describe every aspect of the teachings disclosed
herein and do not limit the scope of the claims.
DETAILED DESCRIPTION
[0014] Each of the features and teachings disclosed herein can be
utilized separately or in conjunction with other features and
teachings to provide nonlinear self-interference cancellation
(NSIC) with sampling rate mismatch. Representative examples
utilizing many of these additional features and teachings, both
separately and in combination, are described in further detail with
reference to the attached figures. This detailed description is
merely intended to teach a person of skill in the art further
details for practicing aspects of the present teachings and is not
intended to limit the scope of the claims. Therefore, combinations
of features disclosed above in the detailed description may not be
necessary to practice the teachings in the broadest sense, and are
instead taught merely to describe particularly representative
examples of the present teachings.
[0015] In the description below, for purposes of explanation only,
specific nomenclature is set forth to provide a thorough
understanding of the present disclosure. However, it will be
apparent to one skilled in the art that these specific details are
not required to practice the teachings of the present
disclosure.
[0016] Some portions of the detailed descriptions herein are
presented in terms of algorithms and symbolic representations of
operations on data bits within a computer memory. These algorithmic
descriptions and representations are used by those skilled in the
data processing arts to effectively convey the substance of their
work to others skilled in the art. An algorithm is here, and
generally, conceived to be a self-consistent sequence of steps
leading to a desired result. The steps are those requiring physical
manipulations of physical quantities. Usually, though not
necessarily, these quantities take the form of electrical or
magnetic signals capable of being stored, transferred, combined,
compared, and otherwise manipulated. It has proven convenient at
times, principally for reasons of common usage, to refer to these
signals as bits, values, elements, symbols, characters, terms,
numbers, or the like.
[0017] It should be borne in mind, however, that all of these and
similar terms are to be associated with the appropriate physical
quantities and are merely convenient labels applied to these
quantities. Unless specifically stated otherwise as apparent from
the below discussion, it is appreciated that throughout the
description, discussions utilizing terms such as "processing,"
"computing," "calculating," "determining," "displaying," or the
like, refer to the action and processes of a computer system, or
similar electronic computing device, that manipulates and
transforms data represented as physical (electronic) quantities
within the computer system's registers and memories into other data
similarly represented as physical quantities within the computer
system memories or registers or other such information storage,
transmission or display devices.
[0018] Moreover, the various features of the representative
examples and the dependent claims may be combined in ways that are
not specifically and explicitly enumerated in order to provide
additional useful embodiments of the present teachings. It is also
expressly noted that all value ranges or indications of groups of
entities disclose every possible intermediate value or intermediate
entity for the purpose of an original disclosure, as well as for
the purpose of restricting the claimed subject matter. It is also
expressly noted that the dimensions and the shapes of the
components shown in the figures are designed to help to understand
how the present teachings are practiced, but not intended to limit
the dimensions and the shapes shown in the examples.
[0019] The present disclosure proposes an equalization technique
when sampling rates of an interfering signal and a desired signal
are different from each other. Herein, a victim signal refers to a
signal that is corrupted or interfered by the interfering signal,
and its nonlinear self-interference is to be equalized or cancelled
by the present equalization technique. In particular, the present
disclosure provides an explicit equalization technique in each of
the cases when the sampling rate of the desired signal is greater
than that of the interfering signal and the sampling rate of the
desired signal is lower than that of the interfering signal.
[0020] FIG. 1 is a simplified block diagram of an example system,
according to one embodiment. The system 100 includes a transmitter
digital processor 151, a receiver digital processor 152, and a
nonlinear self-interference cancellation (NSIC) circuit 155.
[0021] The system 100 transmits digital signals in the following
signal transmission path. The transmitter digital processor 151
sends a digital signal 161 to a digital-to-analog converter (DAC)
115. The DAC 115 converts the digital signal 161 to a baseband
signal 162, and an analog baseband (ABB) filter 114 filters out the
baseband signal 162 to obtain a filtered baseband signal 163. For
example, the ABB filter 114 is a low-pass filter that removes high
frequency components of the baseband signal 162. An up-conversion
mixer 113 mixes the filtered baseband signal 163 with a local
oscillator (LO) signal and generates a radio frequency (RF) signal
164. A power amplifier (PA) 112 amplifies the RF signal 164, a band
pass filter (PBF) 111 filters an amplified RF signal 165, and an
amplified/filtered RF signal 166 is transmitted via an antenna.
[0022] The system 100 receives RF signals in the following signal
reception path. An RF signal 171 is received by an antenna, and a
band pass filter (PBF) 121 filters the RF signal 171 to generate a
band-pass filtered signal 172. A low noise amplifier (LNA) 122
amplifies the band-pass filtered signal 172, and a down-conversion
mixer 123 converts an amplified low-noise signal 173 to a baseband
signal 174 by mixing with a local oscillator (LO) signal. An analog
baseband (ABB) filter 124 filters out the baseband signal 174 to
obtain a filtered baseband signal 175. An analog-to-digital
converter (ADC) 125 converts the filtered baseband signal 175 to a
digital signal 176. The NSIC circuit 155 cancels nonlinear
self-interference component of the digital signal 176, and the
receiver digital processor 152 receives an interference-canceled
digital signal 177.
[0023] An example of the system 100 is a wireless communication
device such as a mobile device. The system 100 may include multiple
radio access technologies (RATs) on the device such as long term
evolution (LTE), 5th generation new radio (5G NR), Wi-Fi, Bluetooth
(BT), and global navigation satellite system (GNSS).
[0024] According to one embodiment, the NSIC circuit 155 performs
nonlinear self-interference cancellation using digital samples of
an interfering signal (e.g., the digital signal 161 input to the
DAC 115) and digital samples of an interfered signal (e.g., the
digital signal 176 output from the ADC 125). However, the sampling
rates of input signals to the NSIC circuit 155 (e.g., the digital
signals 161 and 176) may be different from each other. The system
100 may perform nonlinear self-interference cancellation
differently based on the sampling rates of the interfering signal
and the interfered signal (also referred to as a corrupted victim
signal) that is corrupted by the interfering signal.
[0025] According to one embodiment, the NSIC circuit 155 may
perform nonlinear self-interference cancellation when a sampling
rate of digital samples of the interfering signal (e.g., the
digital signal 161) is lower than a sampling rate of the digital
samples of the corrupted victim signal (e.g., the digital signal
176). In this case, the system 100 may perform interpolation on the
interfering signal to make the sampling rate of the interfering
signal and that of the corrupted victim signal to be the same and
then perform estimation of the nonlinear self-interference
signal.
[0026] According to another embodiment, the NSIC circuit 155 may
perform nonlinear self-interference cancellation when the sampling
rate of digital samples of the interfering signal is greater than
that of the digital samples of the corrupted victim signal. When a
bandwidth of the nonlinear self-interference signal is comparable
to that of the desired signal such that aliasing due to
analog-to-digital conversion is negligible, the NSIC circuit 155
may decimate the digital samples of the interfering signal to make
the sampling rates of the interfering signal and that of the
corrupted victim signal to be the same and then perform estimation
of the nonlinear self-interference signal. When the bandwidth of
the nonlinear self-interference signal is greater than that of the
desired signal such that aliasing of the interfering signal happens
due to analog-to-digital conversion, the NSIC circuit 155 may
frequency shift the interfering signal that occupies a range of
frequencies and decimate each aliasing portion separately to make
the sampling rates of the interfering signal and the corrupted
victim signal the same and then perform estimation of the nonlinear
self-interference signal using in-band and all out-of-band portions
of the interfering signal that fold back into a desired band, i.e.,
a frequency band of the desired signal.
[0027] According to one embodiment, the present system may process
nonlinear self-interference cancellation in real-time. In this
case, the digital samples of the interfering signal and the digital
samples of the corrupted victim signal must be available to the
present system for real-time processing. In general, the digital
samples of the interfering signal and the desired signal are
assumed to be uncorrelated.
[0028] The corrupted victim signal y.sub.r[n] may be represented
as:
y.sub.r[n]=y[n]+d[n]+w[n] (Eq. 1)
where d[n] denotes a desired signal, w[n] denotes an additive
noise, and y[n] denotes a nonlinear self-interference signal that
is to be estimated and canceled by the NSIC circuit 155 of FIG. 1.
The nonlinear self-interference signal y[n] may be generated due to
the transmission of the same carrier, other carriers, or other RATs
on the same device (e.g., the system 100).
[0029] For example, the nonlinear self-interference signal y[n] is
generated after passing an interfering signal x[n] through the
nonlinearity, and it interferes with the desired signal d[n]. The
output of the nonlinearity is referred to as y[n] in (Eq. 1) above
that is the nonlinear self-interference signal of the interfering
signal x[n]. Suppose the digital samples of the interfering signal
x[n] prior to passing through the nonlinearity and the digital
samples y.sub.r[n] are available for real-time processing. However,
a sampling rate F.sub.D of the digital samples of the corrupted
victim signal y.sub.r[n] and a sampling rate F.sub.I of the digital
samples of the interfering signal x[n] may be different from each
other. In other words, the digital samples of the corrupted victim
signal y.sub.r[n] and the digital samples of the interfering signal
x[n] have sampling frequencies of F.sub.D and F.sub.I
respectively.
[0030] According to one embodiment, the NSIC 155 can estimate the
nonlinear self-interference from two or more interfering signals
jointly. The NSIC 155 may apply the nonlinear self-interference
cancelation for the interfering signals together or in a serial
manner.
[0031] It is assumed that the nonlinear self-interference signal
y[n] can be represented (estimated) as follows at the time of
nonlinear self-interference coefficients estimation and
generation:
y.sub.est[n]=z(n).sub.1.times.MC.sub.est.sub.M.times.1' (Eq. 2)
where C.sub.est.sub.M.times.1 is a column vector of nonlinear
coefficients, z(n).sub.1.times.M is a row vector of different
components of the nonlinearity that is referred to as a kernel
vector, M is the number of the nonlinear components. The row vector
z(n) is generated based on the type of nonlinearity.
[0032] FIG. 2 illustrates a block diagram of an example NSIC
system, according to one embodiment. The NSIC system 200 includes
an interpolator 201, a kernel generator 202, a frequency shifter
203, a low-pass filter and decimator 204, a multiplexer 211, a
nonlinear self-interference generator 205, and a nonlinear
self-interference coefficients estimator 206. The NSIC system 200
receives two different digitally sampled signals, for example,
digital samples 251 of the interfering signal x[n] and digital
samples 252 of the corrupted victim signal y.sub.r[n], and provides
an output signal 255 to a demodulator (not shown), for example, the
receiver digital processor 152 of FIG. 1. Referring to (Eq. 1), the
output signal 255 may be treated as an estimation of the desired
signal d[n] when the additive noise w[n] is negligible.
[0033] The NSIC system 200 may correspond to the NSIC circuit 155
of FIG. 1 or at least a portion thereof. The NSIC system 200 may be
implemented in a hardware circuit including a logic component
and/or device, a software running on a processor/controller, a
firmware, or any combination thereof. Although the NSIC circuit 155
is described as a circuit in FIG. 1, the NSIC circuit 155 may be
implemented in a hardware circuit including a logic component
and/or device, a software running on a processor/controller, a
firmware, or any combination thereof similarly to the NSIC system
200. It is understood that the actual implementation of the NSIC
circuit 155 and the NSIC system 200 may be varied without deviating
from the scope of the present disclosure.
[0034] According to one embodiment, the NSIC system 200 may perform
nonlinear self-interference cancellation when a sampling rate
F.sub.I of digital samples 251 of the interfering signal x[n] is
lower than a sampling rate F.sub.D of digital samples 252 of the
corrupted victim signal y.sub.r[n] (when F.sub.D>F.sub.I).
[0035] The interpolator 201 may receive the digital samples 251 of
the interfering signal x[n] and generate interpolated version of
the interfering signal x[n]. The interpolated version of the
interfering signal x[n] is herein referred to as an interpolated
interfering signal x'[n]. The kernel generator 202 may receive the
interpolated interfering signal x'[n] and generate the kernel
vector z(n). The NSIC system 200 later determines the nonlinear
self-interference signal y[n] of (Eq. 1) as a function of the
interpolated interfering signal x'[n]. As an example, the estimated
nonlinear self-interference signal y.sub.est[n] may be represented
as follows in a nonlinear model when there are different harmonics
of signal self-interfering with the corrupted victim signal
y.sub.r[n] in the principal frequency:
y est [ n ] = k = 0 K - 1 l = - L L C k , l x ' [ n - l ] 2 k x ' [
n - l ] ( Eq . 3 ) ##EQU00001##
Then, for this example, the number of the nonlinear components M,
the kernel vector z(n), and the column vector C.sub.est may be
expressed as follows, where 2L+1 is the memory length, and K is the
order of nonlinearity:
M = K ( 2 L + 1 ) z ( n ) = [ x ' [ n + L ] x ' [ n - L ] x ' [ n +
L ] 2 ( K - 1 ) x ' [ n + L ] x ' [ n - L ] 2 ( K - 1 ) x ' [ n - L
] ] 1 .times. K ( 2 L + 1 ) C est = [ c 0 , - L c 0 , L c K - 1 , -
L c K - 1 , L ] K ( 2 L + 1 ) .times. 1 ##EQU00002##
[0036] In the present embodiment, the NSIC system 200 uses the
digital samples 252 of the corrupted victim signal y.sub.r[n] as
well as the kernel vector z(n) for performing nonlinear
self-interference cancellation. By estimating the nonlinear
self-interference signal y[n], the NSIC system 200 may cancel the
nonlinear self-interference y[n] from the corrupted victim signal
y.sub.r[n] to obtain the estimation of the desired signal (e.g.,
the output signal 255).
[0037] In one embodiment, the interpolator 201 may interpolate the
interfering signal x[n] to make the sampling rate of the
interfering signal x[n] to be the same as that of the corrupted
victim signal y.sub.r[n]. As an example of the interpolation
scheme, the interpolator 201 may employ Lagrange interpolation
where its order depends on a size of oversampling. However, it is
understood that the interpolator 201 may employ a different
interpolation scheme without deviating from the scope of the
present disclosure.
[0038] After interpolating the digital samples 251 of the
interfering signal x[n], the kernel generator 202 may generate the
kernel vector z(n) using the interpolated interfering signal x'[n]
that is output from the interpolator 201 to estimate the nonlinear
self-interference signal y[n]. In the present embodiment (when
F.sub.D>F.sub.I), the kernel vector z(n) that is output from the
kernel generator 202 is fed to the nonlinear self-interference
generator 205 and the nonlinear self-interference coefficient
estimator 206 through the multiplexer 211. In one embodiment, the
frequency shifter 203 and the low-pass filter and decimator 204 may
bypass the kernel vector z(n) to the nonlinear self-interference
generator 205 and the nonlinear self-interference coefficients
estimator 206. For the convenience of explanation, the kernel
vector that is fed to the nonlinear self-interference generator 205
and the nonlinear self-interference coefficients estimator 206
through the frequency shifter 203 and the low-pass filter and
decimator 204 is denoted as z''(n).
[0039] The nonlinear self-interference coefficient estimator 206
may generate the estimated nonlinear coefficient column vector
C.sub.est.sub.M.times.1 using the digital samples of the corrupted
victim signal y.sub.r[n] and the kernel vector z(n). Estimation of
the nonlinear coefficient column vector C.sub.est.sub.M.times.1 may
be obtained using various estimation schemes, for example, a least
square (LS) scheme, i.e.,
min C e s t y est - y r 2 , ##EQU00003##
or a recursive LS (RLS) scheme. Regardless of the estimation
scheme, it is desirable to minimize an estimation error of the
nonlinearity output. The interference generator 205 may generate
the estimated nonlinear self-interference signal y.sub.est[n] using
the estimated nonlinear coefficient column vector
C.sub.est.sub.{circumflex over (M)}.times.1 and the kernel vector
z(n).
[0040] According to another embodiment, the NSIC system 200 may
perform nonlinear self-interference cancellation when a sampling
rate of the digital samples 251 of the interfering signal x[n] is
greater than a sampling rate of the digital samples 252 of the
corrupted victim signal y.sub.r[n] (when F.sub.D<F.sub.I). In
this case, the interfering signal x[n] is fed to the kernel
generator 202 bypassing the interpolator 201, and the kernel
generator 202 may generate the kernel vector z'(n) based on the
interfering signal x[n].
[0041] In one embodiment, if an aliasing effect is negligible, for
example, if a bandwidth of the interference signal y[n] is equal to
or smaller than a bandwidth of the desired signal d[n], the NSIC
system 200 may decimate the interfering signal x[n] to make the
sampling rates of the interfering signal x[n] and the corrupted
victim signal y.sub.r[n] to be the same, and then the NISC system
200 may perform estimation of the nonlinear self-interference
signal y[n]. More specifically, the low-pass filter and decimator
204 decimates the kernel vector z'(n) to obtain the kernel vector
z''(n) and use it for estimation using the nonlinear
self-interference generator 205 and the nonlinear self-interference
coefficient estimator 206 as discussed above. In this case, the
frequency shifter 203 is bypassed, and the kernel vector z'(n) is
fed to the low-pass filter and decimator 204, i.e., the kernel
vector z''(n) is a decimated version of the kernel vector
z'(n).
[0042] In another embodiment, if an aliasing effect is negligible,
the NSIC system 200 may interpolate the digital samples 252 of the
corrupted victim signal y.sub.r[n] and generate an interpolated
corrupted victim signal denoted by y'.sub.r[n]. The NSIC system 200
uses the kernel vector z'(n) and the interpolated corrupted victim
signal y'.sub.r[n] to obtain the estimated nonlinear
self-interference signal y.sub.est[n]. For the specific example
where there are different harmonics of signal self-interfering with
the corrupted victim signal y.sub.r[n] in the principal frequency,
the kernel vector z'(n) is obtained using (Eq. 4) below. The NSIC
system 200 may further filter and decimate y.sub.est[n] and obtain
the filtered, decimated signal denoted by y'.sub.est[n] that has
sampling rate equal to F.sub.D. The NSIC system 200 then subtracts
y'.sub.est[n] from y.sub.r[n].
[0043] If the bandwidth of the nonlinear self-interference signal
y[n] is greater than the bandwidth of the desired signal d[n] such
that considerate amount of aliasing of the interfering signal
happens, or in other words the down-conversion in the ADC 125 folds
back out-of-band portions of interference signal into the band of
the desired signal d[n], the NSIC system 200 may generate the
kernel vector z'(n) based on the interfering signal x[n]. For the
specific example where there are different harmonics of signal
self-interfering with the corrupted victim signal y.sub.r[n] in the
principal frequency, the kernel vector z'(n) is constructed as
follows:
z'(n)=[x[n+L] . . . x[n-L] . . . |x[n+L]|.sup.2(K-1)x[n+L] . . .
|x[n-L]|.sup.2(K-1)x[n-L]].sub.1.times.K(2L+1) (Eq. 4)
[0044] When the aliasing effect is significant, the frequency
shifter 203 may frequency shift the kernel vector z'(n) by
.theta..sub.1F.sub.D, .theta..sub.2F.sub.D, . . . ,
.theta..sub.JF.sub.D and generate frequency-shifted kernels denoted
by z'.sub.1(n), . . . , z'.sub.J(n), where F.sub.D is the sampling
frequency of the desired signal d[n], and .theta..sub.j is defined
as:
.theta. j = ( - 1 ) j + 1 j 2 , j = 1 , , J . ( Eq . 5 )
##EQU00004##
[0045] The value of J is determined based on the bandwidth of the
nonlinear self-interference signal y[n], the sampling frequency
F.sub.D, and/or the nonlinearity model for each scenario. It is
noted that j=1 in (Eq. 5) corresponds to the in-band portion of the
kernel vector z'(n), and each of j=2, . . . , J corresponds to a
different out-of-band portion of the kernel vector z'(n) that folds
back into the desired band when down-sampling of the interfering
signal x[n] happens at the ADC 125 of FIG. 1. It is further noted
that frequency shifting of the kernel vector z'(n) by
.theta..sub.jF.sub.D brings the jth aliasing portion of the kernel
vector z'(n) into the desired band.
[0046] The low-pass filter and decimator 204 may further decimate
the frequency-shifted kernel vector z'.sub.j(n), j=1, . . . , J to
make the sampling rates of them to be the same as d[n] or
y.sub.r[n] and obtain a filtered and decimated kernel vector
z''.sub.j(n), j=1, . . . , J. It is noted that in all decimation
stages, the NSIC system 200 may use an anti-aliasing filter before
down-sampling. The stacked version of the decimated kernel vector
may be denoted as z''(n)=[z'.sub.1(n), . . . , z''.sub.J(n)]. Then,
at the time of nonlinearity estimation, the estimated nonlinear
signal used for model fitting may be expressed as follows:
y e s t [ n ] = z '' ( n ) 1 .times. ( M J ) C est ( M J ) .times.
1 ##EQU00005##
where y.sub.est[n] is the estimated version of the nonlinear
self-interference signal y[n], and M is the assumed number of
nonlinear components for each kernel vector z''.sub.j(n) at the
time of estimation. C.sub.est.sub.(MJ).times.1 and consequently the
signal y.sub.est[n] may be estimated via any estimation method such
as LS or RLS. According to one embodiment, the NSIC system 200 may
determine whether to include aliasing terms in nonlinear
self-interference estimation based on various factors such as an
interference power relative to the desired signal d[n], a bandwidth
of the nonlinear self-interference signal y[n] compared to a
sampling rate of the desired signal d[n].
[0047] FIG. 3 illustrates a flow chart of an exemplary nonlinear
self-interference cancellation (NISC) process, according to one
embodiment. First, digital samples of the corrupted victim signal
y.sub.r[n] and digital samples of the interfering signal x[n] are
obtained (301). Then, it is determined whether the sampling rate
F.sub.D of the digital samples of the corrupted victim signal
y.sub.r[n] is greater or lower than the sampling rate F.sub.I of
the digital samples of the interfering signal x[n] (302). If
F.sub.D>F.sub.I, the digital samples of the interfering signal
x[n] is interpolated, and interpolated interfering signal x'[n] is
obtained (303), the kernel vector z(n) is generated (304). If
F.sub.D<F.sub.I, the kernel vector z'(n) is generated using the
interfering signal x[n] (305), and it is determined whether
aliasing is significant or not (306). If aliasing is significant,
frequency-shifted kernel vectors z'.sub.1(n), . . . , z'.sub.J(n)
are generated by frequency shifting the kernel vector z'(n) (308),
and the decimated kernel vector z''(n) is obtained by decimating
z'.sub.j(n), j=1, . . . , J (309). Otherwise, i.e., if aliasing is
insignificant, the output and the input of the frequency shifter
203 is the same, or the frequency shifter 203 may be bypassed, and
the decimated kernel vector z''(n) is obtained by decimating the
kernel vector z'(n) (307).
[0048] The kernel vector z(n) (in the case of F.sub.D>F.sub.I)
or the decimated kernel vector z''(n) (in the case of
F.sub.D<F.sub.I) and the digital samples of the corrupted victim
signal y.sub.r[n] are used to estimate the nonlinear coefficients,
and the estimated nonlinear self-interference signal y.sub.est[n]
is obtained (310). The estimation of the desired signal d[n] is
obtained by canceling the estimated nonlinear self-interference
signal y.sub.est[n] from the digital samples of the corrupted
victim signal y.sub.r[n] (311).
[0049] According to one embodiment, a method for providing
nonlinear self-interference cancellation of a wireless
communication device includes: receiving digital samples of an
interfering signal and a corrupted victim signal; determining that
a first sampling rate of the interfering signal is lower than a
second sampling rate of the corrupted victim signal; generating an
interpolated interfering signal by interpolating the interfering
signal to match the first sampling rate of the interfering signal
to the second sampling rate of the corrupted victim signal;
generating a kernel vector based on the interpolated interfering
signal, wherein the kernel vector has terms of the nonlinear
self-interference; estimating nonlinear self-interference of the
corrupted victim signal using the terms of the nonlinear
self-interference; and providing an estimation of a desired signal
by cancelling the nonlinear self-interference from the corrupted
victim signal.
[0050] The method may further include: estimating nonlinear
coefficients of the nonlinear self-interference; and generating an
estimated nonlinear self-interference signal based on the kernel
vector and the nonlinear coefficients. The estimation of the
desired signal may be provided by canceling the estimated nonlinear
self-interference signal from the corrupted victim signal.
[0051] The nonlinear coefficients may be generated using a least
square (LS) scheme or a recursive LS (RLS) scheme.
[0052] The interpolated interfering signal may be generated by
Lagrange interpolation.
[0053] The interfering signal may be received from a transmitter of
the wireless communication device, and the estimation of the
desired signal may be fed into a receiver of the wireless
communication device.
[0054] The corrupted victim signal may be provided by an
analog-to-digital converter (ADC) coupled to the receiver.
[0055] According to another embodiment, a method for providing
nonlinear self-interference cancellation includes: receiving
digital samples of an interfering signal and a corrupted victim
signal; determining that a first sampling rate of the interfering
signal is greater than a second sampling rate of the corrupted
victim signal; generating a kernel vector based on the interfering
signal, wherein the kernel vector has terms of the nonlinear
self-interference; estimating nonlinear self-interference of the
corrupted victim signal using the terms of the nonlinear
self-interference; and providing an estimation of a desired signal
by cancelling the nonlinear self-interference from the corrupted
victim signal.
[0056] The method may further include: determining power of an
aliasing portion of the interfering signal is greater than a
threshold; generating frequency-shifted kernels by
frequency-shifting the kernel vector and bringing each aliasing
portion into a desired band associated with the desired signal;
generating a decimated kernel vector by filtering and decimating
the frequency-shifted kernels to match a third sampling rate of the
frequency-shifted kernel to the second sampling rate of the
corrupted victim signal; and generating the nonlinear
self-interference based on the decimated kernel vector.
[0057] The method may further include: determining power of an
aliasing portion of the interfering signal is smaller than a
threshold; generating a decimated kernel vector by filtering and
decimating the kernel vector; and generating the nonlinear
self-interference based on the decimated kernel vector.
[0058] The interfering signal may be received from a transmitter of
the wireless communication device, and the estimation of the
desired signal may be fed into a receiver of the wireless
communication device.
[0059] The corrupted victim signal may be provided by an
analog-to-digital converter (ADC) coupled to the receiver.
[0060] According to another embodiment, a wireless communication
device includes: a nonlinear self-interference cancellation (NSIC)
logic, wherein the NSIC logic is configured to receive an
interfering signal sampled at a first sampling rate and a corrupted
victim signal sampled at a second sampling rate, generate a kernel
vector having terms of nonlinear self-interference, estimate
nonlinear self-interference of the corrupted victim signal, and
generate an estimation of a desired signal by cancelling the
nonlinear self-interference from the corrupted victim signal.
[0061] The wireless communication device may further include a
transmitter and a receiver, wherein the interfering signal is
received from the transmitter, and the estimation of the desired
signal is fed into the receiver.
[0062] The wireless communication device may further include an
analog-to-digital converter (ADC) coupled to the receiver, wherein
the ADC provides the corrupted victim signal.
[0063] The NSIC logic may include an interpolator that generates an
interpolated interfering signal by interpolating the interfering
signal to match the first sampling rate of the interfering signal
to the second sampling rate of the corrupted victim signal.
[0064] The NSIC logic may include a kernel generator that generates
the kernel vector having terms of the nonlinear
self-interference.
[0065] The NSIC logic may further include a nonlinear
self-interference generator and a nonlinear self-interference
coefficients estimator, wherein the nonlinear self-interference
coefficients estimator may estimate nonlinear coefficients of the
nonlinear self-interference, and wherein the nonlinear
self-interference generator may generate an estimated nonlinear
self-interference signal based on the kernel vector and the
nonlinear coefficients.
[0066] The NSIC logic may generate the estimation of the desired
signal by canceling the estimated nonlinear self-interference
signal from the corrupted victim signal.
[0067] The NSIC logic may further include a frequency shifter and a
low-pass filter and decimator.
[0068] If the NSIC logic determines that power of an aliasing
portion of the interfering signal is greater than a threshold, the
frequency shifter may generate frequency-shifted kernels by
frequency-shifting the kernel vector and bringing each aliasing
portion into a desired band associated with the desired signal, and
the low-pass filter and decimator may generate a decimated kernel
vector by filtering and decimating the frequency-shifted kernel to
match a third sampling rate of the frequency-shifted kernel to the
second sampling rate of the corrupted victim signal, and the NSIC
logic may generate the nonlinear self-interference based on the
decimated kernel vector.
[0069] If the NSIC logic determines that power of an aliasing
portion of the interfering signal is smaller than a threshold, the
low-pass filter and decimator may generate a decimated kernel
vector by filtering and decimating the kernel vector; and the NSIC
logic may generate the nonlinear self-interference based on the
decimated kernel vector.
[0070] The above example embodiments have been described
hereinabove to illustrate various embodiments of implementing a
system and method for providing a system and method for providing
nonlinear self-interference cancellation (NSIC) with sampling rate
mismatch. Various modifications and departures from the disclosed
example embodiments will occur to those having ordinary skill in
the art. The subject matter that is intended to be within the scope
of the present disclosure is set forth in the following claims.
* * * * *