U.S. patent application number 15/707306 was filed with the patent office on 2019-03-21 for system, apparatus and method for reducing audio artifacts in a phase diversity receiver.
The applicant listed for this patent is Silicon Laboratories Inc.. Invention is credited to Alexander August Arthur Hakkola.
Application Number | 20190089391 15/707306 |
Document ID | / |
Family ID | 65721620 |
Filed Date | 2019-03-21 |
![](/patent/app/20190089391/US20190089391A1-20190321-D00000.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00001.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00002.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00003.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00004.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00005.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00006.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00007.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00008.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00009.png)
![](/patent/app/20190089391/US20190089391A1-20190321-D00010.png)
View All Diagrams
United States Patent
Application |
20190089391 |
Kind Code |
A1 |
Hakkola; Alexander August
Arthur |
March 21, 2019 |
System, Apparatus And Method For Reducing Audio Artifacts In A
Phase Diversity Receiver
Abstract
In one embodiment, an apparatus includes: a first radio receiver
to receive and downconvert a first radio frequency (RF) signal to a
first digital signal; a second radio receiver to receive and
downconvert a second RF signal to a second digital signal; a
correlation circuit to receive the first and second digital signals
and determine a correlation between the first and second digital
signals; a weight calculation circuit to determine a first weight
value and a second weight value based at least in part on the
correlation; and a combiner circuit to combine the first and second
digital signals according to the first and second weight
values.
Inventors: |
Hakkola; Alexander August
Arthur; (Austin, TX) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Silicon Laboratories Inc. |
Austin |
TX |
US |
|
|
Family ID: |
65721620 |
Appl. No.: |
15/707306 |
Filed: |
September 18, 2017 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04B 7/0854 20130101;
H04B 7/0845 20130101; H04B 1/126 20130101; H04L 25/0305 20130101;
H04L 27/265 20130101; H04L 25/03866 20130101; H04B 7/0848
20130101 |
International
Class: |
H04B 1/12 20060101
H04B001/12; H04B 7/08 20060101 H04B007/08; H04L 27/26 20060101
H04L027/26; H04L 25/03 20060101 H04L025/03 |
Claims
1. An apparatus comprising: a first radio receiver to receive and
downconvert a first radio frequency (RF) signal to a first digital
signal; a second radio receiver to receive and downconvert a second
RF signal to a second digital signal; a correlation circuit to
receive the first digital signal and the second digital signal and
determine a correlation between the first digital signal and the
second digital signal; a weight calculation circuit to determine a
first weight value and a second weight value based at least in part
on the correlation; and a combiner circuit to combine the first
digital signal and the second digital signal according to the first
weight value and the second weight value.
2. The apparatus of claim 1, wherein the weight calculation circuit
is to determine the first weight value and the second weight value
further based on at least one signal metric associated with the
first digital signal and at least one signal metric associated with
the second digital signal.
3. The apparatus of claim 2, wherein the weight calculation circuit
is to adjust the first weight value and the second weight value
determined further based on the at least one signal metric
associated with the first digital signal and the at least one
signal metric associated with the second digital signal when the
correlation indicates the first digital signal is uncorrelated with
the second digital signal.
4. The apparatus of claim 1, wherein the weight calculation circuit
is to determine the first weight value and the second weight value
for a plurality of samples of the first digital signal and the
second digital signal.
5. The apparatus of claim 4, wherein the weight calculation circuit
is to adjust the first weight value and the second weight value for
a first sample of the plurality of samples of the first digital
signal and a first sample of the plurality of samples of the second
digital signal based at least in part on a phase difference between
the first digital signal and the second digital signal.
6. The apparatus of claim 1, wherein the correlation circuit is to
determine the correlation comprising a cross-correlation between
the first digital signal and the second digital signal.
7. The apparatus of claim 1, wherein, in response to the
correlation indicating that the second digital signal is
uncorrelated with the first digital signal, the weight calculation
circuit is to adjust the first weight value to be substantially
greater than the second weight value, wherein the correlation
results from receipt of first content in the first RF signal and
receipt of second content in the second RF signal, the first
content different than the second content.
8. The apparatus of claim 1, wherein the apparatus comprises a
phase diversity receiver comprising: a first semiconductor die
including the first radio receiver to receive the first RF signal
from a first antenna; and a second semiconductor die including the
second radio receiver to receive the second RF signal from a second
antenna, the second antenna spatially separated from the first
antenna.
9. The apparatus of claim 1, wherein the apparatus comprises a
digital signal processor comprising the correlation circuit and the
weight calculation circuit.
10. At least one non-transitory computer readable medium including
instructions that when executed enable a system to perform a method
comprising: receiving and processing a first radio frequency (RF)
signal from a first antenna into a processed first signal;
receiving and processing a second RF signal from a second antenna
into a processed second signal; determining first signal metric
information based on the processed first signal and determining
second signal metric information based on the processed second
signal; determining a correlation between the processed first
signal and the processed second signal; and combining the processed
first signal and the processed second signal based on the first
signal metric information and the second signal metric information,
and adjusting the combining based on the correlation.
11. The at least one non-transitory computer readable medium of
claim 10, wherein the method further comprises: determining a first
combining ratio based on the first signal metric information and
the second signal metric information; determining a combining ratio
modifier based on the correlation; establishing the first combining
ratio to be a first modified combining ratio in response to the
correlation indicating that the processed first signal is
correlated to the processed second signal to at least a threshold
level; and establishing an adjusted combining ratio to be the first
modified combining ratio in response to the correlation indicating
that the processed first signal is correlated to the second process
signal to less than the threshold level.
12. The at least one non-transitory computer readable medium of
claim 11, wherein the method further comprises: adjusting the first
combining ratio using the combining ratio modifier, the combining
ratio modifier based on a filtered correlation value determined
over a plurality of groups of samples of the processed first signal
and the processed second signal, each of the plurality of groups
including a plurality of samples of the processed first signal and
a plurality of samples of the processed second signal.
13. The at least one non-transitory computer readable medium of
claim 12, wherein the method further comprises determining a first
phase difference value and a second phase difference value based on
at least one of the first signal metric information and the second
signal metric information.
14. The at least one non-transitory computer readable medium of
claim 13, wherein the method further comprises: establishing the
first combining ratio to be an initial combining ratio in response
to a difference between the first phase difference value and the
second phase difference value being less than a first threshold and
the correlation indicating that the processed first signal is
correlated to the processed second signal to at least a second
threshold level; and otherwise, establishing the first modified
combining ratio to be the initial combining ratio.
15. The at least one non-transitory computer readable medium of
claim 14, wherein the method further comprises: generating a first
weight value and a second weight value according to the initial
combining ratio; weighting the processed first signal according to
the first weight value; weighting the processed second signal
according to the second weight value; combining the weighted
processed first signal and the weighted processed second signal;
and outputting the combined signal to a demodulator.
16. The at least one non-transitory computer readable medium of
claim 15, wherein the method further comprises: determining a phase
difference between a first sample of the processed first signal and
a first sample of the processed second signal; and adjusting the
first weight value and the second weight value based on the phase
difference.
17. The at least one non-transitory computer readable medium of
claim 16, wherein the method further comprises combining the
processed first signal and the processed second signal according to
the adjusted first weight value and the adjusted second weight
value.
18. An apparatus comprising: a first radio receiver to receive and
downconvert a first radio frequency (RF) signal from a first
antenna to a first digital signal; a second radio receiver to
receive and downconvert a second RF signal from a second antenna to
a second digital signal; a phase aligner circuit to phase align the
first digital signal and the second digital signal; a correlation
circuit to receive the first digital signal and the second digital
signal and determine a correlation between the first digital signal
and the second digital signal; a weight calculation circuit to
determine a first weight value and a second weight value based at
least in part on the correlation; and a combiner circuit to use the
first weight value and the second weight value to combine the first
phase aligned digital signal and the second phase aligned digital
signal into a combined signal.
19. The apparatus of claim 18, wherein the weight calculation
circuit is to determine a phase difference between a first sample
of the first digital signal and a first sample of the second
digital signal and adjust at least one of the first weight value
and the second weight value based on the phase difference.
20. The apparatus of claim 19, wherein the weight calculation
circuit is to adjust the first weight value and the second weight
value based on at least one signal metric associated with the first
digital signal and at least one signal metric associated with the
second digital signal when the correlation exceeds a threshold.
Description
BACKGROUND
[0001] In certain radio receiver systems, phase diversity is used
to combine signals received from multiple antenna inputs that are
spatially separated, resulting in different channel phase and
condition. Such phase diversity operation is typically used to
decrease signal impairment events. That is, with two channel
phases, the signals can be combined and since the signals are the
same before the channel impairments, audio artifacts can be
minimized. However, in conventional phase diversity combining some
signal information from each antenna source can still cause
impairments. While this combining may be suitable when one antenna
is impaired and the other antenna is not, in the condition of a
highly impaired signal conventional phase diversity operation can
still lead to a resulting signal that includes undesired noise or
other audio artifacts.
SUMMARY OF THE INVENTION
[0002] In one aspect, an apparatus includes: a first radio receiver
to receive and downconvert a first radio frequency (RF) signal to a
first digital signal; a second radio receiver to receive and
downconvert a second RF signal to a second digital signal; a
correlation circuit to receive the first digital signal and the
second digital signal and determine a correlation between the first
digital signal and the second digital signal; a weight calculation
circuit to determine a first weight value and a second weight value
based at least in part on the correlation; and a combiner circuit
to combine the first digital signal and the second digital signal
according to the first weight value and the second weight
value.
[0003] In an embodiment, the weight calculation circuit is to
determine the first weight value and the second weight value
further based on at least one signal metric associated with the
first digital signal and at least one signal metric associated with
the second digital signal. The weight calculation circuit may
adjust the first weight value and the second weight value
determined further based on the at least one signal metric
associated with the first digital signal and the at least one
signal metric associated with the second digital signal when the
correlation indicates the first digital signal is uncorrelated with
the second digital signal. The weight calculation circuit may
determine the first weight value and the second weight value for a
plurality of samples of the first digital signal and the second
digital signal. The weight calculation circuit may adjust the first
weight value and the second weight value for a first sample of the
plurality of samples of the first digital signal and a first sample
of the plurality of samples of the second digital signal based at
least in part on a phase difference between the first digital
signal and the second digital signal.
[0004] In an embodiment, the correlation circuit is to determine
the correlation comprising a cross-correlation between the first
digital signal and the second digital signal. In response to the
correlation indicating that the second digital signal is
uncorrelated with the first digital signal, the weight calculation
circuit may adjust the first weight value to be substantially
greater than the second weight value, where the correlation results
from receipt of first content in the first RF signal and receipt of
second content in the second RF signal, the first content different
than the second content.
[0005] In one embodiment, the apparatus is a phase diversity
receiver that includes: a first semiconductor die including the
first radio receiver to receive the first RF signal from a first
antenna; and a second semiconductor die including the second radio
receiver to receive the second RF signal from a second antenna, the
second antenna spatially separated from the first antenna. The
apparatus may further include a digital signal processor comprising
the correlation circuit and the weight calculation circuit.
[0006] In another aspect, a method includes: receiving and
processing a first RF signal from a first antenna into a processed
first signal; receiving and processing a second RF signal from a
second antenna into a processed second signal; determining first
signal metric information based on the processed first signal and
determining second signal metric information based on the processed
second signal; determining a correlation between the processed
first signal and the processed second signal; and combining the
processed first signal and the processed second signal based on the
first signal metric information and the second signal metric
information, and adjusting the combining based on the
correlation.
[0007] The method may further include: determining a first
combining ratio based on the first signal metric information and
the second signal metric information; determining a combining ratio
modifier based on the correlation; establishing the first combining
ratio to be a first modified combining ratio in response to the
correlation indicating that the processed first signal is
correlated to the processed second signal to at least a threshold
level; and establishing an adjusted combining ratio to be the first
modified combining ratio in response to the correlation indicating
that the processed first signal is correlated to the second process
signal to less than the threshold level. The method also may
include adjusting the first combining ratio using the combining
ratio modifier, the combining ratio modifier based on a filtered
correlation value determined over a plurality of groups of samples
of the processed first signal and the processed second signal, each
of the plurality of groups including a plurality of samples of the
processed first signal and a plurality of samples of the processed
second signal, and determining a first phase difference value and a
second phase difference value based on at least one of the first
signal metric information and the second signal metric
information.
[0008] In an embodiment, the method may further include:
establishing the first combining ratio to be an initial combining
ratio in response to a difference between the first phase
difference value and the second phase difference value being less
than a first threshold and the correlation indicating that the
processed first signal is correlated to the processed second signal
to at least a second threshold level; and otherwise, establishing
the first modified combining ratio to be the initial combining
ratio. The method further may include: generating a first weight
value and a second weight value according to the initial combining
ratio; weighting the processed first signal according to the first
weight value; weighting the processed second signal according to
the second weight value; combining the weighted processed first
signal and the weighted processed second signal; and outputting the
combined signal to a demodulator. The method further may include:
determining a phase difference between a first sample of the
processed first signal and a first sample of the processed second
signal; and adjusting the first weight value and the second weight
value based on the phase difference. The method may further include
combining the processed first signal and the processed second
signal according to the adjusted first weight value and the
adjusted second weight value.
[0009] In another aspect, an apparatus includes: a first radio
receiver to receive and downconvert a first RF signal from a first
antenna to a first digital signal; a second radio receiver to
receive and downconvert a second RF signal from a second antenna to
a second digital signal; a phase aligner circuit to phase align the
first digital signal and the second digital signal; a correlation
circuit to receive the first digital signal and the second digital
signal and determine a correlation between the first digital signal
and the second digital signal; a weight calculation circuit to
determine a first weight value and a second weight value based at
least in part on the correlation; and a combiner circuit to use the
first weight value and the second weight value to combine the first
phase aligned digital signal and the second phase aligned digital
signal into a combined signal.
[0010] In an embodiment, the weight calculation circuit is to
determine a phase difference between a first sample of the first
digital signal and a first sample of the second digital signal and
adjust at least one of the first weight value and the second weight
value based on the phase difference. The weight calculation circuit
may further adjust the first weight value and the second weight
value based on at least one signal metric associated with the first
digital signal and at least one signal metric associated with the
second digital signal when the correlation exceeds a threshold.
BRIEF DESCRIPTION OF THE DRAWINGS
[0011] FIG. 1 is a block diagram of a receiver in accordance with
an embodiment.
[0012] FIG. 2 is a block diagram of a portion of a weight combining
circuit in accordance with an embodiment.
[0013] FIG. 3 is a block diagram of a first combiner weight
modification circuit in accordance with an embodiment.
[0014] FIG. 4 is a block diagram of another weight modification
circuit in accordance with an embodiment.
[0015] FIG. 5 is a block diagram of a mitigation determination
circuit in accordance with an embodiment.
[0016] FIG. 6 is a block diagram of a selection circuit in
accordance with an embodiment.
[0017] FIG. 7 is a block diagram of a first weight generation
circuit in accordance with an embodiment.
[0018] FIG. 8 is a block diagram of a second weight combining
circuit in accordance with an embodiment.
[0019] FIG. 9 is a block diagram of a system in accordance with an
embodiment.
[0020] FIG. 10 is a flow diagram of a method in accordance with an
embodiment.
[0021] FIG. 11 is a flow diagram of a method in accordance with
another embodiment.
[0022] FIG. 12 is a flow diagram of a method in accordance with
another embodiment.
DETAILED DESCRIPTION
[0023] In various embodiments, techniques are provided to improve
operation of a phase diversity receiver when processed signals of
multiple signal paths of the receiver are uncorrelated. Such
uncorrelated signals may occur in a case where directional antennas
of a vehicle each receive a signal from a different transmitter
when tuned to the same frequency. In this case, a phase diversity
receiver without an embodiment herein could continue to combine
these two uncorrelated sources, resulting in bad audio. Embodiments
may be used to reduce audio artifacts when combining two
uncorrelated signals by detecting this situation and dynamically
controlling weighting values of the signals of the multiple paths
to weight towards a processed signal having one or more higher
signal quality metrics. Note that uncorrelated signals may also
exist when at least one of the processed signals has
impairments.
[0024] As will be described herein, a correlation is computed
between demodulated signals of the two paths. When it is determined
that the two signals are not correlated the weighting values may be
adjusted to select signal weighting based on signal quality
metrics. The weighting values also may be determined at least in
part on a phase difference determination between the signals, which
when of a given level, causes weighting values to be adjusted based
on signal metrics. When this phase difference is large enough, a
slower metric may be used to select weightings. In addition,
embodiments may enable the weighting values to be dynamically
adjusted on a sample-by-sample basis, based on a determination of
phase differences between the two processed signals. That is, at a
very fast rate, the weighting values can be modified based on a
current phase difference between the samples of each signal.
[0025] Referring now to FIG. 1, shown is a block diagram of a
receiver in accordance with an embodiment. As illustrated in FIG.
1, receiver 100 is a phase diversity receiver. As one example,
receiver 100 may be a radio receiver incorporated into a vehicle,
such that incoming radio frequency (RF) signals may be received by
way of multiple antennas 110.sub.0, 110.sub.1 and processed in
independent processing paths 105.sub.0, 105.sub.1. After suitable
processing, the resulting processed signals may be combined to
provide improved audio fidelity, particularly in cases of
impairments such as multipath fading and other channel impairments.
Embodiments also may detect when the processed signals of the two
signal paths are uncorrelated and take appropriate control measures
with regard to the diversity combining to ensure that signal
information from a better performing signal path is primarily used.
Stated another way, when an uncorrelation between the two signal
paths is detected, diversity combining may alter the combining
ratio compared with a traditional combiner so that the better
performing signal path may be primarily used for the output. Still
further as described herein, fast and dynamic control of weightings
applied to each of the signal paths can occur on a sample-by-sample
basis.
[0026] For ease of discussion, components within first signal
processing path 105.sub.0 are discussed. Understand that the same
components are present in second signal processing path 105.sub.1.
And thus, reference numerals (without subscript) are intended to
refer to generic components suitable for the various signal paths.
As illustrated, incoming RF signals are received via an antenna
110.sub.0. The received RF signals are provided to an analog front
end circuit 115.sub.0. Various signal processing of these RF
signals may occur in front end circuit 115. As an example, such
processing may include gain control, such as by way of a low noise
amplifier (LNA). In some cases, front end circuit 115 may further
include a filter. The processed signals are provided to a
digitizer, namely an analog-to-digital converter 120.sub.0, which
digitizes the incoming analog signal into digital form. The
resulting digitized signal is provided to a mixer circuit
125.sub.0. Mixer circuit 125.sub.0 may downconvert the digitized
signal from RF to a lower frequency. In one embodiment, mixer
125.sub.0 may downconvert the RF signal to a zero intermediate
frequency (ZIF) level. Of course, downconversion to other
frequencies is possible. Understand that additional processing may
be performed within mixer circuit 125.sub.0. For example,
additional circuitry within this block may include a low pass
filter and a de-rotator, among other circuitry.
[0027] Still with reference to FIG. 1, the downconverted signal
output from mixer circuit 125.sub.0 is provided to a digital
automatic gain control (AGC) circuit 130.sub.0. AGC circuit 130 may
provide a controllable amount of gain to the downconverted signal.
Additional circuitry may be present within this block, including a
channel filter. The output of digital AGC circuit 130.sub.0 is
provided to a phase aligner circuit 135. As illustrated, phase
aligner circuit 135 is configured to receive the resulting signals
from both signal processing paths 105 and perform a phase alignment
to align samples of these two paths in phase.
[0028] After phase alignment, the resulting phase-aligned signals
of the two paths are output to multipliers 140.sub.0, 140.sub.1
which multiply the phase-aligned signals of each of the signal
processing paths with a corresponding weight value (W0 and W1)
received from a weight calculation circuit 170, details of which
are described below. Suffice to say in one example, these weight
values may be fractional values (that collectively sum to one). The
resulting products output by multipliers 140 (namely the
phase-aligned signals each multiplied by a coefficient
corresponding to the weight values) is provided to a summer 145,
which combines the weighted values from the two signal processing
paths to obtain a combined signal. Understand that additional
processing may be performed on the combined signal. For example,
the combined signal may be provided to a demodulator, which may
perform demodulation to output an audio signal to an output device
such as speakers of receiver system 100 (not shown in FIG. 1).
[0029] Note that signal paths 105.sub.0, 105.sub.1 may be
implemented on a single semiconductor die of a single integrated
circuit (IC). In other cases, signal paths 105 may be implemented
on separate die within one or more semiconductor packages. Still
further, note that while signal paths 105.sub.0, 105.sub.1 are
shown generally identical through AGC circuit 130, understand that
the additional components described in receiver 100 may be present
in one or both signals paths, with certain components not used in
one or more the other of signal paths 105 depending upon
implementation.
[0030] Also understand that while some embodiments may implement
the various circuitry shown in FIG. 1 as discrete circuits, in
other cases, the digital circuitry (namely all circuitry after
downconversion in mixer circuit 125) may be implemented within a
programmable execution circuit, such as one or more digital signal
processors (DSPs).
[0031] As described herein, different control techniques for
combining signals of the two signal processing paths may be used,
depending upon various information, including signal metric
information and correlation information. More particularly, each
signal path is coupled to a metric collection circuit 160.sub.0,
160.sub.1 that may determine one or more signal metrics from the
signal information. As seen, metric collection circuits 160 may
receive signal information from corresponding AGC circuits
150.sub.0, 150.sub.1 (which in turn may include channel filters).
In embodiments herein, this signal metric information may include
signal-to-noise (SNR) ratio. Note while described as being a
"signal"-to-noise ratio, in some cases the actual metric may be a
"carrier"-to-noise ratio (CNR) metric. In addition, the signal
metric information may include received signal strength indicator
(RSSI) information.
[0032] Still further as shown in FIG. 1, the signals of the two
paths, prior to phase alignment in phase aligner circuit 135, may
be provided to a cross correlation circuit 155 that performs a
cross correlation of the signals of the two paths. In an
embodiment, the correlation may be calculated as follows:
y=E[(x_1-u_1)*(x_2-u_2)]/(sigma_1*sigma_2): where E is the expected
value; x_1 is signal 1; u_1 is the mean of signal 1; sigma_1 is the
standard deviation of signal 1; x_2 is signal 2; u_2 is the mean of
signal 2; and sigma_2 is the standard deviation of signal 2. Note
that this correlation, which in this embodiment is a cross
correlation, is not a signal metric. That is, a correlation
provides an indication of a level of matching or coherency between
different signals (which in many cases may derive from the same
transmitted content). Nevertheless, understand that as used herein,
the terms "correlation" or "cross correlation" do not refer to
signal metric information, as a correlation value does not provide
any qualitative measure of the involved signals.
[0033] As shown, weight calculation circuit 170 receives these
signal metrics and correlation information, and using this
information and various predetermined values as described herein,
determines appropriate weightings for the two different signal
paths during receiver operation. Understand while shown at this
high level in the embodiment of FIG. 1, many variations and
alternatives are possible. As one example, an embodiment may be
used in an FM-based radio scheme, although other embodiments may be
used for other radio types. Also, understand that for ease of
discussion herein a phase diversity receiver including two
independent signal processing paths is described. It is also
possible for a phase diversity receiver to include additional
independent signal processing paths, where each such independent
signal processing path is coupled to receive an incoming RF signal
from a different antenna input. For example, in a vehicle context,
antennas 110.sub.0, 110.sub.1 may be independently implemented,
e.g., in driver and passenger side view mirrors. In yet other
cases, additional antennas may be spatially separated and adapted
within various portions of a vehicle to provide greater phase
diversity capabilities.
[0034] Referring now to FIG. 2, shown is a block diagram of a
portion of a weight combining circuit in accordance with an
embodiment. More specifically, initial combiner weight circuit 200
may be used to develop a baseline combining ratio for combining
signals of the two paths. Note that in different embodiments,
initial combiner weight circuit 200 may be implemented as a
hardware circuit to perform the various operations described
herein. In other cases, circuit 200 may be implemented using
hardware circuitry such as a general-purpose processor,
microcontroller, digital signal processor or so forth to perform
the operations described herein, e.g., in combination with firmware
and/or software. In yet other implementations, combinations of
hardware, firmware and/or software may be used to implement circuit
200. Similar implementations may be used for the various circuits
described in FIGS. 3 to 8. Initial combiner weight circuit 200 may
be included within weight calculation circuit 170 of FIG. 1.
Understand that the additional circuits described in FIGS. 3 to 8
also may be implemented, in an embodiment, within weight
calculation circuit 170 of FIG. 1.
[0035] With reference to FIG. 2, the baseline combining ratio
(Normal Combining Ratio in FIG. 2) may be a suitable combining
ratio to use when the signals of the two paths are substantially
correlated. In general, initial combiner weight circuit 200
determines the combining ratio based on multiple signal metrics of
the multiple signal paths, namely SNR and RSSI.
[0036] Note that in an embodiment, the signal metrics may be
obtained on a sub-sampling basis. For example, as will be described
herein, some of the processing may be performed on a plurality of
samples of the two signal paths. In such cases, instead of
obtaining metrics for each such sample of a group of samples,
sub-sampled signal quality metrics may be obtained for a given
group of samples undergoing an evaluation. In one example, certain
of the processing described herein may be performed on a group of
16 samples of each of the two signal paths. In this example, it is
possible for a single signal quality metric (namely one SNR value
and one RSSI value) to be used for this group of samples. Thus in
this example, for a given group of samples (e.g., 16 samples),
initial combiner weight circuit 200 may generate a single baseline
combining ratio (Normal Combining Ratio). Of course other examples
are possible.
[0037] As seen, incoming signal metric values for the two signal
paths (namely SNR0 and SNR1) are provided to a summer 210 that
determines a difference and sends this difference to another summer
220, further configured to receive another processed signal metric
value. Specifically, incoming RSSI information (RSSI0, RSSI1) is
provided to corresponding summers 225.sub.0, 225.sub.1. As seen,
summers 225.sub.0, 225.sub.1 further receive a programmable
threshold value (RSSI_THRESH). This threshold may be used to reduce
the influence of this RSSI information when it is of particularly
low value. Thus as illustrated, if the resulting differences from
summers 225 are less than a predetermined level (e.g., 0), the
given RSSI value may be discounted or mitigated in the weight
combining determination. That is, by way of a minimum circuit
230.sub.0, 230.sub.1, a minimum of a predetermined value (e.g., 0)
or the difference output by summer 225 is output, in turn to
another summer 240, in turn coupled to a multiplier 250 that
multiplies the resulting processed RSSI value with a given
coefficient (which in an embodiment may be a predetermined value,
e.g., a first constant (Scalar 1)). Note that this scalar value,
and a plurality of other scalar values described herein may be
constants provided for a particular receiver. For example, these
constants may be set by firmware or another programmable source and
stored in a non-volatile storage. In some cases these scalar values
may be determined based on typical radio constraints and/or
listening preferences. The output of summer 220 is a combining
ratio (Normal Combining Ratio) that may be the maximum ratio of the
two signals, without mitigation, if correlated signals are being
combined.
[0038] Referring now to FIG. 3, shown is a block diagram of a first
combiner weight modification circuit in accordance with an
embodiment. As shown in FIG. 3, circuit 300 may be used to generate
a modifier value for use in modifying the baseline combining ratio
in the context of uncorrelated signals. Note that uncorrelated
signals may occur in different situations. As one example, the
signals in the two signal paths may be uncorrelated where the
signals are obtained from different content. That is, in a phase
diversity receiver the two different antenna may receive RF signals
from two different transmission sources that are outputting
different content, which is thus uncorrelated. In other cases, the
signals in the two paths may be uncorrelated where, although
including the same content, they have substantially different CNR
levels. For example, at least one of the signals may have a CNR
level that is of relatively low level (e.g., within a noise
floor).
[0039] As illustrated, modification circuit 300 is coupled to
receive a correlation value, which in an embodiment is a cross
correlation value. This correlation value is provided to a low pass
filter 310, resulting in a slow correlation signal (Slow
Correlation in FIG. 3). Note that this slow correlation value may
result from low pass filter 310 that has a relatively large time
constant. For example, while the various signal processing
described herein is generally performed on groups of samples (e.g.,
16 individual samples), in embodiments LPF 310 may act to generate
a slow correlation signal over many such blocks of samples. For
example, in some embodiments LPF 310 may generate the slow
correlation value over hundreds of these blocks, thus smoothing
variations in correlation values.
[0040] In one embodiment, the correlation between signals received
in LPF 310 may be generated in a cross correlation circuit (e.g.,
cross correlation circuit 155 of FIG. 1). Note that this cross
correlation may be in units of power (e.g., decibels (dB)). In a
particular embodiment, a highly correlated signal may have a slow
correlation value of approximately zero, whereas a wholly
uncorrelated signal may have a slow correlation value of, e.g.,
approximately 128 (in the instance where this slow correlation
value is an 8-bit value).
[0041] As seen, this slow correlation value is provided to a summer
320, where it is combined with a predetermined value (Scalar 2).
The resulting summed signal is coupled to a maximum circuit 330
that outputs the maximum of the summed signal or a predetermined
value (e.g., 0). The resulting maximum value is multiplied in a
multiplier 340 by a coefficient value (Scalar 3). The resulting
product is transformed into a linear value by way of a
log-to-linear operator 350. This resulting value is multiplied in a
multiplier 360 by a coefficient, namely the output of a multiplier
370, itself generating a product of the slow correlation value and
another predetermined value (Scalar 4). This resulting product of
multiplier 360 is a modifier value (Combining Ratio Modifier) that
provides an indication of how much the baseline combining ratio may
be modified for uncorrelated signals. In general, this modifier
value may result from an approximate curve fitting, where a change
in this slow correlation value is modified in a non-linear
fashion.
[0042] Referring now to FIG. 4, shown is a block diagram of another
weight modification circuit in accordance with an embodiment. As
shown in FIG. 4, circuit 400 may be used to calculate a first
modified combining ratio, which is used to determine a slow
combining ratio, namely a combining ratio based on analysis of a
plurality of samples.
[0043] As illustrated, the baseline combining ratio (Normal
Combining Ratio) is provided as a second input (input B) to a
control circuit 410. A first input to control circuit 410 (namely
input A) is a low pass filtered version of the first modified
combiner ratio output by circuit 400. As such, the output of a low
pass filter 460 provides this slow combining ratio, which in turn
is delayed by a delay circuit 470. A third input to control circuit
410 may be a predetermined value (e.g., 0). In an embodiment,
control circuit 410 may be configured to generate an output based
on the three input values. In a particular embodiment, control
circuit 410 may operate to output a positive or negative
predetermined value or a zero value based on various calculations
performed in control circuit 410. In a particular embodiment,
control circuit 410 is configured to determine: if (A<C) and
(B<C), output--Scalar 5; if (A>C) and (B>C), output Scalar
5; and otherwise output the zero value. Stated another way, control
circuit 410 is configured to determine whether a combining ratio
determined based on signal metrics is pointing to the same signal
path (and the same antenna) as being a better quality as determined
based on a long term average of the combining ratio.
[0044] As further illustrated in FIG. 4, the output of control
circuit 410 is provided to a multiplier 420, where it is combined
with the combining ratio modifier, which is an indication of how
much the normal combining ratio may be modified in response to
identification of uncorrelated signals. The product output by
multiplier 420 is provided to a summer 430, where it is combined
with the baseline combining ratio (Normal Combining Ratio). In
turn, the resulting sum is provided to a selector 440, which in an
embodiment may be implemented as a multiplexer. As illustrated,
multiplexer 440 may be controlled based on an output of another
control circuit 450. As illustrated, control circuit 450 is
configured to receive at a first input (input A) the slow
correlation value, which provides an indication as to a level of
correlation between the two signals. A second input to control
circuit 450 (input B) may be a predetermined value (Scalar 6). In
an embodiment, control circuit 450 may control the selection of
output from multiplexer 440. If the signals are correlated,
multiplexer 440 may be controlled to output the baseline combining
ratio (namely Normal Combining Ratio). Instead, if the signals are
not well correlated, multiplexer 440 may be controlled to output a
modified combining ratio (namely the output of summer 430).
[0045] As such, weight modification circuit 400 may operate based
on determination of correlation level. That is, the slow
correlation value identifies if the signals are correlated or not,
with large values being not correlated and a value of zero being
fully correlated. At multiplier 420, the Combining Ratio Modifier
is multiplied by a positive or negative value to drive the ratio
towards only one antenna, or to not change the Normal Combining
Ratio, based on whether the slow combining ratio and the current
combining ratio (Normal Combining Ratio) both show the same antenna
is the better choice.
[0046] Referring now to FIG. 5, shown is a block diagram of a
mitigation determination circuit in accordance with an embodiment.
More specifically, circuit 500 shown in FIG. 5 may be used to
perform intermediate calculations to determine when to use
mitigations, namely a mitigation of a normal combining ratio when a
minimum signal metric (namely a SNR metric) of the two signal paths
is below a given level. As illustrated, circuit 500 is coupled to
receive signal metrics, namely SNR levels regarding the two signal
paths, which are received in a first minimum circuit 510. The
resulting minimum value of these two SNR values is provided to a
summer 515 where it is combined with a predetermined value (Scalar
7). The resulting sum is provided to another minimum circuit 520,
which outputs the minimum value between this sum and a
predetermined value (e.g., 0). The resulting minimum value is
provided to an absolute value generator 525, and the resulting
value is provided to a multiplier 530, where it is multiplied with
a predetermined value (Scalar 8). The resulting product is provided
to a maximum circuit 535, which outputs the maximum between this
product and another predetermined value (Scalar 9). The resulting
maximum value is then compared to another predetermined value
(Scalar 12) in a minimum circuit 540, with the resulting minimum
value corresponding to a minimum adjusted SNR value
(min_SNR_adj).
[0047] As illustrated, this value is provided as a first input
(input A) to a control circuit 550. Control circuit 550 further
receives as input several predetermined values (Scalar 10 and
Scalar 11, at inputs B and C). In an embodiment, control circuit
550 is configured to perform calculations based on these inputs to
output a first phase difference value (phase diff slow max) and a
second value. In an embodiment, control circuit 550 may generate
the first phase difference value according to: B*2.sup.A. In turn,
control circuit 550 may be configured to generate the second value
according to: C*(2.sup.A-1). As seen, this second output from
control circuit 550 is provided to a maximum circuit 560, which
compares this value to another predetermined value (Scalar 13). The
resulting maximum value is then provided to a summer 570 where it
is summed with another predetermined value (Scalar 14), to generate
a difference corresponding to a second phase difference value,
phase_diff_mult. In an embodiment, this first phase difference
value may move larger as the SNR level of at least one of the
signal paths goes lower. In turn, the second phase difference value
may have a level that proceeds linearly with SNR levels. That is,
as SNR decreases, this phase difference value decreases also.
[0048] Referring now to FIG. 6, shown is a block diagram of a
selection circuit in accordance with an embodiment. More
specifically, selection circuit 600 may be configured to receive
multiple combiner ratios, namely a baseline combiner ratio (Normal
Combining Ratio) and a modified combining ratio (Slow Combining
Ratio) and select one of these values to be used for weighting of
the signals of the different signal paths as described herein. As
illustrated, a first selection circuit 630 may be implemented as a
first multiplexer and a second selection circuit 640 may be
implemented as another multiplexer.
[0049] First selection circuit 630 is controlled by a control
circuit 620. As illustrated, control circuit 620 is configured to
receive a first input (input A) corresponding to a difference
generated in a summer 610 between phase difference values
(phase_diff_slow (generated as described below) and
phase_diff_slow_max). Control circuit 620 is further configured to
receive a second input (a predetermined value, e.g., 0). In an
embodiment, control circuit 620 may control first selection circuit
630 to output the normal combiner ratio where the difference
determined based on the phase difference values is less than a
threshold value (e.g., 0). Otherwise, selection circuit 630 outputs
the slow combining ratio. In turn, selection circuit 640 is
controlled by another control circuit 650. As seen, control circuit
650 is configured to receive the slow correlation value at a first
input (input A) and a predetermined value (Scalar 6) at a second
input (input B). In an embodiment, control circuit 650 is
configured to cause selection circuit 640 to output the slow
combining ratio value if the slow correlation is greater than the
predetermined value. As seen, second selection circuit 640 thus
outputs a second modified combining ratio, which as described
further below is used to generate weight values for weighting the
two signal paths.
[0050] Referring now to FIG. 7, shown is a block diagram of a first
weight generation circuit in accordance with an embodiment. As
illustrated in FIG. 7, circuit 700 may be used to generate a slowly
computed combiner weight. As seen, circuit 700 includes a
conversion circuit 710 configured to receive the second modified
combining ratio. In an embodiment, conversion circuit 710 may
convert this ratio from a log value to a linear value. The
resulting linear value is provided to a multiplier 720 where it is
multiplied by a predetermined value (Scalar 15). The resulting
product is an initial weight value for the first signal path
(Ant0_weight). In turn, an initial weight value for the second
signal path (Ant1_weight) may be generated as a difference between
another predetermined value (Scalar 16) and the first initial
weight value. Note that these initial weight values are slowly
computed weights, namely these weight values are computed for a
group of samples. Although embodiments are not limited, as one
example these weight values may be calculated for a group of 16
samples.
[0051] Referring now to FIG. 8, shown is a block diagram of a
second weight combining circuit in accordance with an embodiment.
More specifically, weight combining circuit 800 is implemented as a
sample-by-sample weight combiner, such that the weight value to be
applied to each of the signal paths may be dynamically controlled
on a sample-by-sample basis based on a phase error between the two
signal paths. In an embodiment, more specifically this phase error
may be a simple phase difference that is determined based on the
digitized signals themselves after Phase alignment (namely
channelized digitized signals). As seen, the initial weight values
generated in first weight generation circuit 700 are provided to a
minimum circuit 810, where the minimum value is output to a
multiplier 815 and a combiner 825.
[0052] Second weight combiner circuit 800 further includes a phase
angle circuit 850 to determine a phase angle for the individual
samples of the two signal paths. As seen, phase angle circuit 850
receives incoming samples and determines a phase angle. In an
embodiment, phase angle circuit 850 may perform a coordinate
rotation digital computer (CORDIC) function on these values and
provide the resulting values to summer 860, which determines a
phase difference between the two samples. This resulting phase
difference is provided to a multiplier 870 where it is multiplied
by the second phase difference value (phase_diff_mult, from FIG.
5).
[0053] As further illustrated in FIG. 8, the phase difference
determined at summer 860 is provided to a multiplier 884, where it
is multiplied with a product of another multiplier 882, which is
configured to multiply a predetermined value (e.g., 2) with a
minimum weight value (weight_min). Note that the product output by
multiplier 884 is provided to a low pass filter 880 that generates
a filtered phase difference value (phase_diff_slow). As described
above, this phase difference value may be compared to another phase
difference value at summer 610 of FIG. 6.
[0054] In response to the product output by multiplier 882
(weight_min*2), filter 880 may operate normally when the combiner
circuit is combining equally. Also the output of filter 880 may be
automatically reduced when the combiner circuit starts to weight
towards either antenna. This is desired because when the combiner
is mainly weighted towards one antenna, there is not as much need
for the filtered phase difference value (phase_diff_slow) to show
that the phases are not aligned, since this is expected.
[0055] Still with reference to FIG. 8, the resulting product output
by multiplier 870 is provided to a minimum circuit 820, along with
the product of multiplier 815. As such this value based on the
sample-based phase difference between the different signal paths
may be used to modify the weighting determined by the second
modified combining ratio (itself based on a slow path
determination). Stated another way, this product output by
multiplier 870, based on sample-by-sample phase differences, can be
used to modify a combining ratio determined based on groups of
blocks of samples, for each sample of the groups of blocks.
[0056] As shown in FIG. 8, minimum circuit 820 provides a minimum
of the product from multiplier 870 and the (scaled) weighting value
output from minimum circuit 810 to summer 825. The sum output by
summer 825 is provided to a maximum circuit 830, which outputs a
maximum value as a first temporary weight value for the first
signal path (WO_tmp). Via a summer 835, which determines a
difference between this weight value and a predetermined value
(Scalar 16), a temporary weight value for the second signal path is
also determined (W1_tmp).
[0057] Note that since it is possible for these weight values to be
related to opposite signal paths, these resulting values are
provided to a swap circuit 890, which in an embodiment may include
logic circuitry and/or multiplexers or so forth. Based upon a swap
signal (generated by minimum circuit 810), resulting final weight
values W0 and W1 are determined. Note that this swap signal is
inactive if minimum circuit 810 identifies the first signal path as
being the minimum value (namely ANT0_weight), no swap occurs.
Instead if minimum circuit 810 identifies the second signal path
(ANT1_weight) as the minimum value, the swap signal is activated.
As such, the temporary weight values are finalized (or swapped
prior to finalization in swap circuit 890).
[0058] The resulting final weight signals (W0 and W1) are provided
as coefficients to combining circuitry of the main signal
processing path to weight the signals of the two signal paths. For
example, with reference back to FIG. 1, these weight values are
provided to multipliers 140.sub.0, 140.sub.1 to act as coefficients
to weight the resulting outputs of phase aligner circuit 135. The
products of these two multipliers are provided to a summer 145 to
combine the two signal paths into a resulting digitized signal that
is then provided, e.g., to a demodulator to demodulate the
resulting signal.
[0059] Referring to FIG. 9, a phase diversity combining technique
can be part of a vehicle infotainment system 900. In other cases,
system 900 may be a multi-function, multi-band radio, cellular
telephone, smartphone, PDA, tablet computer, mobile game device, or
so forth and may play music or book downloads, and may be part of a
wireless link between a multiple antennas and a radio receiver. In
one of the embodiments, the wireless device may be a mobile radio
receiver such as of a car stereo.
[0060] Among its other various functions, system 900 may store
digital content on a storage 930, which may be a flash memory or
hard disk drive, as a few examples. System 900 generally includes
an application subsystem 960 (referred to as a host processor) that
may, for example, receive input from a user interface 962 of the
wireless device 910 (which may be a touchpad, e.g., of a display
970) and display information on display 970. Furthermore,
application subsystem 960 may generally control the retrieval and
storage of content from storage 930. As further seen in FIG. 9,
multiple antennas 980.sub.0-980.sub.1 each may be coupled to a
corresponding tuner 934.sub.0-934.sub.1, which can be coupled
together such that the tuner 934.sub.1 performs the phase diversity
combining described herein, based at least in part on correlation
information of the signals of tuners 934.sub.1, 2. In turn, tuner
934.sub.1 outputs a combined signal to a demodulator/audio decoder
965, which may be directly connected to speakers 940 and 950 for
output of audio data (understand that in some embodiments a
separate audio processor may be integrated between the receiver and
speakers). Note that storage 930 or another non-transitory storage
medium (such as present within tuners 934 themselves) may further
store instructions to perform the phase diversity combining
described herein. Of course, embodiments may be implemented in many
other types of systems.
[0061] Referring now to FIG. 10, shown is a flow diagram of a
method in accordance with an embodiment. More specifically, method
1000 of FIG. 10 is a high level view of the diversity combining
operation described herein. In embodiments, method 1000 may be
performed by hardware circuitry of a receiver, potentially in
combination with firmware and/or software. In some cases, method
1000 may be performed within circuitry as shown in FIG. 1, e.g.,
under control of a microcontroller or other control circuit of a
given entertainment system.
[0062] As illustrated, method 1000 begins by receiving and
processing first and second RF signals from first and second
antennas (block 1010). As discussed above, at least these two RF
signals may be received from at least two spatially separated
antennas to enable diversity combining to be performed. This
processing includes various analog front end processing, down
conversion, digital conversion and additional signal processing.
Next at block 1020 signal metrics may be determined based on these
processed first and second signals. As described herein, these
signal metrics may include SNR and RSSI values for groups of
samples of the two signals.
[0063] Still referring to FIG. 10, at block 1030 a correlation may
be determined between the processed signals. In an embodiment, this
correlation may be implemented as a cross-correlation that is
similarly performed on samples of a block of samples by low pass
filtering and computing over a relatively large number of samples.
Finally, at block 1040 the processed first and second signals may
be combined based on the correlation and the signal metrics. More
specifically as described herein, the signals from these different
signal paths may be combined according to a combining ratio
determined based solely on the signal metric information when the
signals are highly correlated. Instead, in the face of some amount
of uncorrelation, a baseline combining ratio determined based on
the signal metrics may be modified and further adjusted according
to various calculations described herein, and determinations of
phase differences between individual samples of the two signal
paths. Understand while shown at this high level in the embodiment
of FIG. 10, many variations and alternatives are possible.
[0064] Referring now to FIG. 11, shown is a flow diagram of a
method in accordance with another embodiment. More specifically,
method 1100 shown in FIG. 11 is a first part of an overall phase
diversity combining operation. As illustrated, method 1100 begins
by determining a first combining ratio based on signal metrics
(block 1110). For example, for a given plurality of samples SNR and
RSSI information may be leveraged to determine a first or baseline
combining ratio, as described herein. Next at block 1120 a
combining ratio modifier may be determined based at least in part
on correlation information. That is, as described herein for cases
in which signals are not well correlated, some modification to this
first combining ratio may be warranted. Thereafter, the first
combining ratio may be adjusted using this combining ratio modifier
(block 1130). As an example, the combining ratio modifier may be
used to generate an adjustment value that in turn is used to modify
the first combining ratio into this adjusted first combining
ratio.
[0065] Still with reference to FIG. 11, next it is determined
whether the correlation is greater than a threshold level. That is,
this determination at diamond 1140 is whether the signals are
correlated to at least a threshold level. If not, control passes to
block 1160 where the adjusted first combining ratio may be selected
to be a first modified combining ratio. Instead, if the signals are
well correlated (where in a particular embodiment a determined
correlation value may be less than a given threshold), control
passes to block 1150 where the first combining ratio may be
selected to be the first modified combining ratio.
[0066] Referring now to FIG. 12, shown is a flow diagram of a
method in accordance with another embodiment. More specifically,
FIG. 12 shows a further portion of a phase diversity combining
operation in accordance with an embodiment. As illustrated in FIG.
12, method 1200 proceeds after selection of the appropriate source
for the first modified combining ratio (in FIG. 11). More
specifically at block 1210, this first modified combining ratio is
filtered. Note that this filtering operation may be a relatively
long term operation to take into account a large number of groups
of samples (e.g., on the order of hundreds, at least). Next, at
block 1220 first and second phase difference scalars may be
determined based on signal metrics. Next at diamond 1230 a
difference between a first phase difference scalar and a filtered
second phase difference may be determined. Further it is determined
whether this phase difference exceeds a threshold. If this is so,
there is an undesired phase difference between the signals of the
two paths. It is further determined here whether the correlation
exceeds a threshold (meaning that the signals are at least somewhat
uncorrelated). If either of these determinations is made, control
passes to block 1240 where the filtered first modified combining
ratio may be selected. Otherwise, control passes to block 1250
where the first combining ratio may be selected.
[0067] Still with reference to FIG. 12 in any event, control next
passes to block 1260 where initial weight values may be determined
based on the selected combining ratio. Thereafter at block 1270 a
sample-based phase difference may be determined between samples of
the two signals. At block 1280, final weight values may be
generated based on the initial weight values and this sample-based
phase difference. Thereafter at block 1290, the first and second
processed signals may be weighted according to the final weight
values to result in a combined signal, which may then be provided
for further processing (such as demodulation and/or decoding and
output). Thus by using an embodiment, an initial weight value may
be determined for a set of samples, and then sample-by-sample
adjustment may occur based on identification of instantaneous phase
difference between given samples of the first and second signals.
In this way, embodiments provide improved maximal ratio combining
in a phase diversity system.
[0068] While the present invention has been described with respect
to a limited number of embodiments, those skilled in the art will
appreciate numerous modifications and variations therefrom. It is
intended that the appended claims cover all such modifications and
variations as fall within the true spirit and scope of this present
invention.
* * * * *