U.S. patent application number 14/692978 was filed with the patent office on 2015-08-13 for method and apparatus for signal quality loss compensation in multiplexing transmission systems.
The applicant listed for this patent is Unwired Planet, LLC. Invention is credited to Jung-fu Thomas Cheng, Stephen J. Grant, Karl J. Molnar, Rajaram Ramesh, Yi-Pin Eric Wang.
Application Number | 20150229436 14/692978 |
Document ID | / |
Family ID | 37452286 |
Filed Date | 2015-08-13 |
United States Patent
Application |
20150229436 |
Kind Code |
A1 |
Grant; Stephen J. ; et
al. |
August 13, 2015 |
METHOD AND APPARATUS FOR SIGNAL QUALITY LOSS COMPENSATION IN
MULTIPLEXING TRANSMISSION SYSTEMS
Abstract
A wireless communication receiver receiving a multiplexed signal
comprising two or more signal streams calculates a received signal
quality for the multiplexed signal as a function of stream-specific
received signal qualities, determines one or more loss parameters
indicative of variations in the stream-specific received signal
qualities, and generates quality feedback based on such
information. In turn, a transmitter controls the selection of one
or more transmission parameters of the multiplexed signal based on
the quality feedback, such that its transmit link adaptations
account for the losses in received signal quality at the receiver
arising from the variations in the stream-specific received signal
qualities. The quality feedback may include calculated loss values,
or parameter/penalties that permit loss calculation, and the method
applies to both code multiplexing and spatial multiplexing.
Inventors: |
Grant; Stephen J.; (Cary,
NC) ; Cheng; Jung-fu Thomas; (Cary, NC) ;
Molnar; Karl J.; (Cary, NC) ; Wang; Yi-Pin Eric;
(Cary, NC) ; Ramesh; Rajaram; (Cary, NC) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Unwired Planet, LLC |
Reno |
NV |
US |
|
|
Family ID: |
37452286 |
Appl. No.: |
14/692978 |
Filed: |
April 22, 2015 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11138724 |
May 26, 2005 |
|
|
|
14692978 |
|
|
|
|
Current U.S.
Class: |
370/252 |
Current CPC
Class: |
H04L 1/0003 20130101;
H04L 5/0057 20130101; H04B 7/0697 20130101; H04L 1/20 20130101;
H04B 7/0691 20130101; H04L 1/0009 20130101; H04L 1/06 20130101;
H04B 7/0678 20130101; H04B 7/061 20130101; H04B 17/309 20150115;
H04W 24/10 20130101; H04L 1/0033 20130101; H04L 1/0026
20130101 |
International
Class: |
H04L 1/00 20060101
H04L001/00; H04W 24/10 20060101 H04W024/10; H04B 7/06 20060101
H04B007/06; H04L 5/00 20060101 H04L005/00; H04B 17/309 20060101
H04B017/309 |
Claims
1. A method of feeding back received signal quality information,
the method comprising: receiving, at a receiver, a multiplexed
signal including two or more signal streams; calculating, at the
receiver, a received signal quality for the multiplexed signal as a
function of two or more different stream-specific received signal
qualities determined for the multiplexed signal; determining, at
the receiver, a loss parameter that represents a loss in the
received signal quality for the multiplexed signal arising from the
two or more different stream-specific received signal qualities,
wherein the determined loss parameter is useable to compensate the
calculated received signal quality for the loss; generating, at the
receiver, received signal quality information including the
calculated received signal quality, or a quantized version thereof,
and the loss parameter; and transmitting, from the receiver, the
received signal quality information as feedback.
2. The method of claim 1, wherein calculating the received signal
quality comprises calculating a mode-specific received signal
quality for each of one or more multiplexing modes defined for the
multiplexed signal.
3. The method of claim 2, wherein determining the loss parameter
comprises, for each mode-specific received signal quality,
determining one or more coding rate-dependent loss parameters.
4. The method of claim 1, wherein calculating the received signal
quality comprises calculating a Channel Quality Indicator (CQI)
value for each of one or more multiplexing modes associated with
the multiplexed signal, and wherein determining the loss parameter
comprises determining one or more coding rate-dependent loss
parameters for each CQI, such that the received signal quality
information comprises mode-specific CQI values and corresponding
coding rate-dependent loss parameters.
5. The method of claim 1, wherein the multiplexed signal comprises
a spatially multiplexed signal transmitted from a different number
of transmit antennas in each of one or more spatial multiplexing
modes, and wherein calculating the received signal quality
comprises calculating a received signal quality for a best set of
transmit antennas in each spatial multiplexing mode.
6. The method of claim 5, wherein determining the loss parameter
comprises, for each spatial multiplexing mode, determining one or
more coding rate-dependent loss parameters.
7. The method of claim 1, wherein calculating the received signal
quality comprises calculating an average signal quality from the
stream-specific received signal qualities.
8. The method of claim 7, wherein determining the loss parameter
comprises determining a maximum spread of the stream-specific
signal qualities, such that the received signal quality information
indicates the average received signal quality and the maximum
spread.
9. The method of claim 7, wherein determining the loss parameter
comprises determining a variation statistic characterizing
variations between the stream-specific signal qualities, such that
the received signal quality information indicates the average
received signal quality and the variation statistic.
10. The method of claim 9, wherein the variation statistic
comprises one or more of a variance of the stream-specific received
signal qualities, a standard deviation of the stream-specific
received signal qualities, and a maximum spread of the
stream-specific received signal qualities.
11. The method of claim 1, wherein determining the loss parameter
comprises determining a loss value representing an effective
reduction in the received signal quality arising from the two or
more different stream-specific received signal qualities.
12. The method of claim 11, wherein determining the loss value
comprises determining the loss value from a loss equation or data
table maintained at the receiver that models signal quality loss as
one or more coding rate-dependent functions of the received signal
quality of the multiplexed signal.
13. The method of claim 1, wherein calculating the received signal
quality comprises determining a set of best transmit antennas for
each of one or more spatial multiplexing modes associated with the
multiplexed signal, and calculating a channel quality indicator
value for each spatial multiplexing mode based on the
stream-specific received signal qualities of the multiplexed signal
corresponding to the best set of transmit antennas.
14. The method of claim 13, wherein determining the loss parameter
comprises determining one or more loss parameters for each spatial
multiplexing mode using the stream-specific received signal
qualities corresponding to the best set of antennas for that
spatial multiplexing mode, and wherein generating the received
signal quality information comprises generating the received signal
quality information as the channel quality indicators and
corresponding loss parameters calculated for the spatial
multiplexing modes.
15. The method of claim 1, wherein the receiver comprises a
Wideband Code Division Multiple Access (W-CDMA) receiver, and
wherein the multiplexed signal comprises a spatially multiplexed
High Speed Downlink Shared Channel (HS-DSCH) signal, and wherein
the received signal quality information comprises a Channel Quality
Indicator (CQI) value and one or more coding rate-dependent loss
parameters, for each of one or more spatial multiplexing modes
associated with the spatially multiplexed HS-DSCH signal.
16. A method of compensating for losses in received signal quality
experienced by a receiver, the method comprising: transmitting,
from a transmitter, a multiplexed signal including two or more
signal streams; receiving, at the transmitter, received signal
quality information as feedback from the receiver, the received
signal quality information including a received signal quality for
the multiplexed signal, or a quantized version thereof, and a loss
parameter that represents a loss in the received signal quality for
the multiplexed signal arising from two or more different
stream-specific received signal qualities of the multiplexed
signal, wherein the loss parameter is useable to compensate the
received signal quality for the loss; and controlling, at the
transmitter, the selection of one or more transmission parameters
of the multiplexed signal based on the received signal quality and
the loss parameter.
17. A wireless communication receiver comprising one or more
processing circuits configured to: receive a multiplexed signal
including two or more signal streams; calculate a received signal
quality for the multiplexed signal as a function of two or more
stream-specific received signal qualities determined for the
multiplexed signal; determine a loss parameter that represents a
loss in the received signal quality for the multiplexed signal
arising from the two or more different stream-specific received
signal qualities, wherein the determined loss parameter is useable
to compensate the calculated received signal quality for the loss;
generate received signal quality information including the
calculated received signal quality, or a quantized version thereof,
and the loss parameter; and transmit the received signal quality
information as feedback.
Description
CROSS-REFERENCE TO RELATED PRIORITY APPLICATIONS
[0001] This is a continuation of application Ser. No. 11/138,724,
filed May 26, 2005, the entire contents of which are incorporated
by reference herein.
BACKGROUND
[0002] The present invention relates to wireless communication
networks, and particularly relates to controlling multiplexed
signal transmission parameters responsive to improved received
signal quality determination.
[0003] It is not uncommon for transmitters in wireless
communication networks to perform transmit link adaptation
responsive to received signal quality feedback from their targeted
receivers. For example, any one or more of the transmit power,
coding rate, modulation format, and the like, may be varied for a
given receiver as function of the received signal quality reported
by that receiver.
[0004] The use of received signal quality as a link adaptation
control input is common for rate-controlled channels, wherein the
network transmitter varies the data rate of a transmitted
communication signal, rather than the transmit power of the signal,
responsive to reported changes in received signal quality at the
targeted receiver(s). The Wideband Code Division Multiple Access
(W-CDMA) standards define a rate-controlled signal known as the
High Speed Downlink Shared Channel (HS-DSCH). The HS-DSCH is a
shared channel that provides high-rate packet data services to a
potentially large number of receivers (users) that share the
channel in time-multiplexed and/or code-multiplexed fashion.
[0005] The HS-DSCH is time-slotted and individual time slots are
dedicated to individual users according to a defined scheduling
algorithm. Users are scheduled according to service needs and other
criteria, but the rate selected for serving a given user on that
user's scheduled time slots generally is selected as a function of
the received signal quality reported by the user. More
particularly, the user reports a channel quality indicator, or some
other representation of received signal quality, and the
transmitter uses the reported value to determine the highest rate
that can be supported by the user at acceptable error rates.
Keeping the transmission error rate relatively low improves the
overall system efficiency, because excessive data re-transmissions
are avoided. Indeed, the effective throughput of the HS-DSCH can
fall off quite rapidly if the transmitter selects data rates that
are higher than are appropriate for the actual reception conditions
at the targeted receivers.
[0006] Thus, if one or more of the targeted receivers "over
reports" its received signal quality, the transmitter likely will
make transmit link adaptations, e.g., the selection of transmit
modulation formats and coding rates, that are inappropriate for the
actual received signal quality at the misreporting receiver. The
adaptation of the HS-DSCH to spatial multiplexing transmitters
represents one circumstance where the targeted receivers are likely
to report higher-than-actual received signal qualities. Similar
over-reporting scenarios arise in code multiplexing transmissions
(e.g., "multi-coding"), and in combinations of code multiplexing
and spatial multiplexing.
[0007] The likelihood of over-reporting signal quality arises as a
consequence of multiplexed transmission, such in spatial
multiplexing transmissions, wherein code symbols belonging to the
same codeword are transmitted as different signal streams and
experience different fading and/or interference between the
transmitter and the targeted receiver. Consequently, the received
signal quality, e.g., the symbol signal-to-noise-plus-interference
ratio (SINR), varies across the codeword at the input to the
receiver's decoder. As an example, with four signal streams
comprising a received multiplexed signal, the receiver experiences
four distinct (stream-specific) received signal qualities. These
variations in stream-specific received signal qualities give rise
to signal quality losses at the receiver. Thus, simply reporting an
average of the stream-specific received signal qualities does not
provide the transmitter with an accurate "picture" of the true
received signal quality at the receiver.
SUMMARY
[0008] At least one method taught herein compensates for losses in
received signal quality experienced by a receiver receiving a
multiplexed signal by controlling the selection of one or more
transmission parameters (e.g., coding rate) for the multiplexed
signal based at least in part on estimating coding rate-dependent
signal quality losses arising at the receiver due to variations in
stream-specific received signal qualities of the multiplexed
signal. By way of non-limiting example, the method permits
modulation and/or coding rate selection for a spatially multiplexed
High-Speed Downlink Shared Channel (HS-DSCH) signal in a Wideband
Code Division Multiple Access (W-CDMA) network to consider such
losses. Of course, the method applies to code multiplexing (e.g.,
"multi-coding") as well as spatial multiplexing, and to
combinations of code and spatial multiplexing.
[0009] In another embodiment, a method of compensating for losses
in received signal quality experienced by a receiver receiving a
multiplexed signal comprises calculating a received signal quality
for the multiplexed signal as a function of stream-specific
received signal qualities determined for the two or more signal
streams of the multiplexed signal, determining a loss in received
signal quality as a function of variations in the stream-specific
received signal qualities, and controlling one or more transmission
parameters of the multiplexed signal as a function of the received
signal quality and the loss in received signal quality. The
transmission parameter(s), for example, may be a modulation format
selection, a coding rate selection, or a combination of modulation
format and coding rate.
[0010] With the above in mind, a wireless communication receiver
may be adapted to receive and process a multiplexed signal
comprising multiple signal streams. In one embodiment, such a
wireless communication receiver comprises one or more processing
circuits configured to calculate a received signal quality for the
multiplexed signal as a function of stream-specific received signal
qualities determined for the multiplexed signal, determine a loss
parameter for the received signal quality as a function of
variations in the stream-specific received signal qualities, and
generate received signal quality information for feedback based on
the received signal quality and the loss parameter.
[0011] The receiver may be configured to calculate the received
signal quality as a mode-specific received signal quality for each
of one or more multiplexing modes defined for the multiplexed
signal. Where the receiver determines mode-specific received signal
qualities, it may be configured to determine one or more coding
rate-dependent loss parameters for each mode-specific received
signal quality, based on the variations in the stream-specific
received signal qualities of the multiplexed signal in the
corresponding multiplexing mode. In such embodiments, the received
signal quality information fed back (directly or indirectly) to the
multiplexing transmitter comprises mode-specific received signal
quality estimates and corresponding loss parameters for each such
mode-specific received signal quality estimate. Note that the loss
parameters may comprise loss values expressing effective losses in
received signal quality, or may comprise values enabling the
transmitter to calculate such losses.
[0012] Regardless of the particular format adopted for loss
parameter feedback, one embodiment of a complementary multiplexing
transmitter comprises radio transmission circuits configured to
support a method of adapting the transmit link for the multiplexed
signal by controlling one or more transmission parameters based on
the feedback described above. Such circuits include one or more
interface circuits configured to receive feedback from the remote
receiver comprising a received signal quality for the multiplexed
signal and a loss parameter indicative of variations in
stream-specific received signal qualities of the multiplexed
signal, and a control circuit configured to control the selection
of one or more transmission parameters of the multiplexed signal as
a function of the received signal quality and the loss
parameter.
[0013] The above control circuit may include, or may be associated
with, a calculation circuit that is configured to determine an
effective received signal quality for the remote receiver based on
estimating a loss in received signal quality arising at the remote
receiver from the variations in the stream-specific received signal
qualities of the multiplexed signal. Where the received signal
quality and the loss parameter comprise a received signal quality
and one or more loss parameters for each of one or more
multiplexing modes associated with the multiplexed signal, the
transmitter may evaluate the effective received signal qualities
for different coding rates in each multiplexing mode, and select
the mode offering the best rate.
[0014] However, those skilled in the art should appreciate that the
present invention is not limited to the above features and
advantages. Indeed, those skilled in the art will recognize
additional features and advantages of the present invention upon
reading the following detailed description of selected embodiments
of the invention, and upon viewing the corresponding drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0015] FIG. 1 is a block diagram of a wireless communication
network and a wireless communication device.
[0016] FIG. 2 is a block diagram of function circuit details for
embodiments of the transceiver node and wireless communication
device illustrated in FIG. 1.
[0017] FIG. 3 is a graph of received signal quality loss curves
that may be embodied in memory as look-up table values or function
values.
[0018] FIG. 4 is a block diagram of a transmitter portion of the
transceiver node illustrated in FIG. 1, for example, for a W-CDMA
embodiment, wherein a HS-DSCH signal is transmitted as a spatially
multiplexed signal.
[0019] FIG. 5 is a graph of a Modulation-and-Coding-Scheme (MCS)
table that may be embodied in memory as look-up table values or
function values.
[0020] FIGS. 6 and 7 are logic flow diagrams for one embodiment of
generating received signal feedback at a remote receiver and
responding to such feedback at a corresponding spatial multiplexing
transmitter, respectively.
[0021] FIG. 8 is a logic flow diagram for an embodiment of transmit
link adaptation at a multiplexing transmitter, responsive to a
particular form of received signal quality feedback.
[0022] FIGS. 9 and 10 are logic flow diagrams for another
embodiment of generating received signal feedback at a remote
receiver and responding to such feedback at a corresponding spatial
multiplexing transmitter, respectively.
[0023] FIGS. 11 and 12 are logic flow diagrams for another
embodiment of generating received signal feedback at a remote
receiver and responding to such feedback at a corresponding spatial
multiplexing transmitter, respectively.
[0024] FIGS. 13 and 14 are logic flow diagrams for another
embodiment of generating received signal feedback at a remote
receiver and responding to such feedback at a corresponding spatial
multiplexing transmitter, respectively.
DETAILED DESCRIPTION
[0025] It should be understood that the methods and corresponding
apparatus taught herein apply to code multiplexed signals and/or
spatially multiplexed signals. Thus, although at least some of the
following discussion is cast within a spatial multiplexing
framework, such details are not limiting.
[0026] With the above point in mind, FIG. 1 partially illustrates
one embodiment of a wireless communication network 10 that
communicatively couples a wireless communication device 12 to one
or more external networks 14. Supporting such communications, the
network 10 comprises a Radio Access Network (RAN) 16 that includes
one or more Radio Network Controllers (RNCs) 18, with each RNC 18
controlling one or more transceiver nodes 20 (sometimes referred to
as "radio base stations" or "base transceiver stations"). The
network 10 further comprises one or more Core Networks (CNs) 22,
which provide communication links with the external network(s)
14.
[0027] By way of non-limiting example, the wireless communication
network 10 may comprise a Wideband Code Division Multiple Access
(W-CDMA) network configured to communicatively couple the wireless
communication device 12 to the Public Switched Telephone Network
(PSTN) and/or the Internet or other Public Data Network. The
wireless communication device 12 thus may comprise a mobile station
or other type of communication device configured for operation
according to the W-CDMA standards. More generally, the wireless
communication device 12, which is also referred to as a "remote
receiver," or, more generally, simply referred to as a "wireless
communication receiver," comprises essentially any type of
communication device configured to receive wireless communication
signals from a supporting communication network.
[0028] In some embodiments, at least one of the transceiver nodes
20 is configured as a code-multiplexing transmitter that is
operative to transmit a code-multiplexed signal to the wireless
communication device 12. However, according to the illustrated
embodiment, at least one of the transceiver nodes 20 is configured
as a spatial multiplexing transmitter that is operative to transmit
a spatially multiplexed signal to the wireless communication device
12.
[0029] In turn, the wireless communication device 12 provides
received signal quality feedback for the multiplexed signal,
thereby enabling the network 10 (e.g., the transmitter of
transceiver node 20) to perform transmit link adaptations based on
the received signal quality feedback. More particularly, the
wireless communication device 12 is configured to provide received
signal quality feedback that reflects variations in the (transmit)
stream-specific received signal qualities corresponding to the
multiplexed signal, as received by the wireless communication
device 12. By providing feedback that explicitly (or implicitly)
indicates stream-specific variations in received signal quality,
the transmit link adaptations performed for the multiplexed signal
may be compensated for the received signal quality losses arising
from such variations.
[0030] By way of non-limiting example, the wireless communication
network 10 may comprise a Wideband Code Division Multiple Access
(W-CDMA) network, and transmitter circuitry included within the
transceiver node 20 may be configured to transmit a High Speed
Downlink Shared Channel (HS-DSCH) signal as the multiplexed signal.
Increasing the effective data rate of the channel stands as one
advantage of spatially multiplexing the HS-DSCH signal.
Complementing this embodiment of the network 10, the wireless
communication device(s) 12 may comprise mobile stations, terminals,
or other type of wireless communication device that is configured
for use in W-CDMA networks.
[0031] Of particular interest herein, the wireless communication
device 12 is configured to provide received signal quality feedback
for a multiplexed signal received by it, wherein that feedback
reflects, directly or indirectly, the variations in stream-specific
received signal qualities for the multiplexed signal. As noted
earlier herein, such variations give rise to a loss in the
effective received signal quality. Thus, reflecting the losses in
the feedback allows the transmit link adaptations being performed
by the network 10 for the multiplexed signal to compensate for the
losses. As is detailed herein, the wireless communication device 12
reflects the variations in stream-specific received signal
qualities either directly, such as by feeding back the
stream-specific received signal qualities or by feeding back
parameterized variation information, or indirectly, such as by
feeding back an effective received signal quality that already is
compensated for the variations.
[0032] FIG. 2 illustrates embodiments of the transceiver node 20
and the wireless communication device 12 that support direct or
indirect feedback of the variations and corresponding transmit link
adaptations. The transceiver node 20 comprises a transmitter 30, a
receiver 32, an RNC interface circuit 34, a plurality of transmit
antennas 36-1 through 36-M, and one or more receiver antennas
38.
[0033] The transmitter 30 comprises a spatial multiplexing
transmitter comprising radio transmission circuits 40 configured to
transmit a multiplexed signal from the multiplexing transmitter to
a remote receiver (e.g., the wireless communication device 12), and
one or more interface circuits 42 configured to receive feedback
from the remote receiver that reflects variations in
stream-specific received signal qualities as calculated by the
remote receiver for the multiplexed signal. These stream-specific
variations relate to the individual ones of the transmit antennas
36 being used to transmit the multiplexed signal to the wireless
communication device 12.
[0034] The transmitter 30 further comprises a calculation circuit
44 configured to determine an effective received signal quality for
the remote receiver based on estimating a loss in received signal
quality arising at the remote receiver because of said variations
in stream-specific received signal qualities, and a transmit link
adaptation control circuit 46 configured to adapt the transmit link
(for the multiplexed signal) by controlling one or more
transmission parameters of the multiplexed signal as a function of
the effective received signal quality.
[0035] In the illustration, the wireless communication device 12
comprises a wireless communication receiver configured to enable a
spatial multiplexing transmitter (e.g., the transmitter 30 of the
transceiver node 20) to compensate transmit link adaptations for
received signal quality losses arising at the receiver because of
variations in stream-specific received signal qualities. To that
end, the wireless communication device 12 comprises one or more
receive/transmit antennas 50, a receiver circuit 52, a transmitter
circuit 54, and one or more (baseband) processing circuits 56 that
include, or are associated with, a calculation circuit 58.
[0036] The calculation circuit 58 is configured to calculate
stream-specific received signal qualities for a multiplexed signal
transmitted from the spatial multiplexing transmitter to the
wireless communication device 12, and to generate received signal
quality feedback that reflects variations in the stream-specific
received signal qualities, for return to the spatial multiplexing
transmitter. Thus, the transmitter 30 of the transceiver node 20
transmits a multiplexed signal to the wireless communication device
12, and the calculation circuit 58 of the wireless communication
device 12 generates received signal quality feedback that reflects
(transmit) stream-specific variations in the received signal
qualities for the multiplexed signal, for return to the transmitter
30.
[0037] By reflecting such variations in the feedback, the transmit
link adaptations performed by the transceiver node 20 for the
multiplexed signal being transmitted to the wireless communication
device 12 can be compensated for the losses arising from the
variations. FIG. 3 illustrates a typical set of loss curves
associated with such variations, wherein each curve in the diagram
plots the loss in effective received signal quality as a function
of the transmit coding rate and the magnitude of the variations in
dBs. For example, one sees from the diagram that for a coding rate
of 0.9 and with a 9 dB variation in the stream-specific received
signal qualities, the wireless communication device 12 suffers a
loss in effective received signal quality of about 5 dB. Of course,
it should be understood that FIG. 3 is merely illustrative, and
that the loss curves applicable to any particular design scenario
will vary.
[0038] FIG. 4 illustrates one scenario, depicting an embodiment of
the transmitter 30 for use in a W-CDMA implementation of the
transceiver 20. The transmitter (radio) resources comprise a
modulator/encoder 60, a 1:N demultiplexer 62, a plurality of
spreading circuits 64, a "best-N" transmit antenna selector 66, and
a plurality of summing circuits 68.
[0039] The (spatially) multiplexed signal, formed as a set of
individually spread substreams is transmitted from the N selected
antennas, as selected by the transmit antenna selector 66 from the
set of M transmit antennas 36, via the combining circuits 68, which
provide composite signals to the transmit antennas 36, representing
the substreams of the multiplexed signal and one or more "other"
signals. Commonly, these other signals comprise pilot signals,
control or broadcast signals, dedicated voice or data signals, etc.
In other words, the other signals represent the collection of
signals in addition to the multiplexed signal that are to be
transmitted from one or more of the transmit antennas 36.
[0040] The particular modulation format and transmit encoding rate
selected by the modulator/encoder 60 for modulating and encoding
the multiplexed input signal (i.e., the HS-DSCH bit stream) is
referred to as the selected "Modulation and Coding Scheme" or
"MCS." The selected combination of modulation format and transmit
encoding rate changes responsive to changing received signal
qualities at the wireless communication devices 12 targeted for
reception of the multiplexed signal. More particularly, for
channels such as the HS-DSCH where a potentially large number of
wireless communication devices 12 (users) are served individually
by the channel according to a time-multiplexed schedule, the MCS
may be changed for each transmission to each user as a function of
the received signal quality feedback from each user.
[0041] Thus, changing the MCS responsive to received signal quality
feedback is a form of transmit link adaptation, and the control
circuit 46 is configured to compensate the MCS selection based on
input from the calculation circuit 44, which, directly or
indirectly, is based on the variations in stream-specific received
signal qualities as measured by the targeted wireless communication
device(s) 12. Therefore, the MCS selections made by the transmitter
30 are compensated for the loss in signal quality arising at the
targeted wireless communication device(s) 12 because of such
variations.
[0042] Mathematically, it may be said that the transmit link
adaptations of the transceiver node 30 for the multiplexed signal
are based on the effective received signal qualities at the
targeted wireless communication device(s) 12, which incorporate the
effects of stream-specific received signal quality variations. That
is, for a given wireless communication device 12, its effective
received signal quality, e.g., its effective
Signal-to-Interference-plus-Noise Ratio (SINR) typically is a
function of two things: (1) the multiple distinct SINRs seen at the
device's decoder input because of the different fading paths from
each of the transmit antennas 36 being used to transmit the
spatially multiplexed signal, and (2) the coding rate and
modulation format used for the multiplexed signal, i.e., the
selected MCS.
[0043] Thus, one definition for the effective received signal
quality for a given wireless communication device 12 simply is the
average of the stream-specific SINRs appearing at the device's
decoder input discounted by a loss factor L.gtoreq.1. The effective
received signal quality thus may be expressed as,
.gamma. eff = .gamma. avg L 1 ##EQU00001##
where the average SINR .gamma..sub.avg is given by,
.gamma. avg = 1 N n = 1 N .gamma. n 2 ##EQU00002##
and where the different values of .gamma..sub.n are the N different
SINRs seen at the targeted receiver's decoder input. As the spread
in SINRs increases, and/or as the transmit coding rate increases,
the loss becomes larger. These effects may be seen in the loss
curves illustrated in FIG. 3.
[0044] The loss effects may be used to control the MCS selections
of the modulator/encoder 60 based on, for example, the MCS
selection table illustrated in FIG. 5. The illustrated axis
corresponds to received signal quality (e.g., SINR in dB), and the
demarcation points on the axis correspond to minimum signal quality
points (switching thresholds) for a plurality of different
MCSs.
[0045] For example, s1 corresponds to the minimum signal quality
threshold for use of the MCS1, s2 corresponds to the minimum signal
quality threshold for the use of MCS2, and so on. In other words,
the signal quality must be above the switching threshold s1 to use
MCS1, and above the switching threshold s2 to use MCS2, and so
on.
[0046] For the illustrated example, point A corresponds to a
hypothetical uncompensated received signal quality, as reported by
the wireless communication device 12. (The signal quality may be
reported as a quantized measurement of SINR, e.g., the device 12
may report a "Channel Quality Indicator," commonly referred to as a
CQI value.) Point B represents the corresponding effective received
signal quality that is obtained by compensating the reported signal
quality for a loss L1, corresponding to the use of MCS1. Similarly,
point C represents the corresponding effective received signal
quality that is obtained by compensating the reported signal
quality for a loss L2, corresponding to the use of MCS2, and point
D represents the corresponding effective received signal quality
that is obtained by compensating the reported signal quality for a
loss L3, corresponding to the use of MCS3.
[0047] The MCS selection process is therefore compensated for the
loss in effective received signal quality experienced at the
wireless communication device 12. More particularly, one sees that,
in the illustrated example, the effective received signal quality
(point B) is above the s1 switching threshold for MCS1. Thus, MCS1
can be selected for use in transmitting the multiplexed signal to
the wireless communication device 12. Similarly, the effective
received signal quality (point C) is above the s2 switching
threshold for MCS2. Thus, MCS2 can be selected for use in
transmitting the multiplexed signal to the wireless communication
device. However, the effective received signal quality (point D) is
below the s3 switching threshold for MCS3. Thus, MCS3 cannot (or
should not) be selected. Therefore, the control circuit 46 will
instruct the modulator/encoder 60 to use either MCS1 or MCS2, and
not MCS3.
[0048] In looking further at the illustrated transmitter
architecture with the above points in mind, a modulated/encoded
version of the HS-DSCH bit stream is output by the
modulator/encoder 60, for input to the 1:N demultiplexer 62, which
demultiplexes that input signal into N substreams, each substream
having a symbol rate of 1/N. The number N is an integer value
representing the number of transmit antennas 36 to be used for
spatially multiplexing the HS-DSCH signal. Each of the N substreams
is spread by a corresponding one of the spreading circuits 64, and
input to the antenna selector 66, which selects N of the available
M antennas 36 for transmitting the individual substreams of the
multiplexed HS-DSCH signal (N.ltoreq.M). In one or more
embodiments, the targeted wireless communication device(s) 12 may
provide pilot strength feedback, or other channel feedback,
indicating the best transmit antennas for each of one or more
spatial multiplexing modes, and the antenna selector 66 may select
the particular combination of N transmit antennas 36 based on such
feedback.
[0049] Of course, the nature of such feedback may be varied
according to different embodiments of the methods described herein.
In one aspect, such variations generally relate to whether the
targeted wireless communication device 12 reflects the variations
in stream-specific received signal qualities by feeding back a
compensated received signal quality, or whether it feeds back the
variations directly, possibly in some parameterized form, such that
the transceiver node generates the compensated received signal
quality.
[0050] For example, FIG. 6 illustrates one embodiment of processing
logic implemented by the wireless communication device 12. In the
illustrated embodiment, the calculation circuit 58 is configured to
calculate a received signal quality for the multiplexed signal as a
function of stream-specific received signal qualities determined
for the multiplexed signal and determine a loss parameter for the
received signal quality as a function of variations in the
stream-specific received signal qualities (Step 100), and generate
received signal quality information for feedback based on the
received signal quality and the loss parameter (Step 102).
[0051] Calculating the received signal quality may comprise
calculating a mode-specific received signal quality for each of one
or more multiplexing modes defined for the multiplexed signal. As
such, determining the loss parameter may comprise determining one
or more coding rate-dependent loss parameters as a function of
variations in the stream-specific received signal qualities for
each mode. In one embodiment, device 12 calculates a Channel
Quality Indicator (CQI) value for each of one or more multiplexing
modes associated with the multiplexed signal, and determines one or
more coding rate-dependent loss parameters for each CQI, such that
the received signal quality information comprises mode-specific CQI
values and corresponding coding rate-dependent loss parameters.
[0052] In cases where the multiplexed signal comprises a spatially
multiplexed signal transmitted from a different number of transmit
antennas in each of one or more spatial multiplexing modes, the
device 12 may calculate a received signal quality for a best set of
transmit antennas in each spatial multiplexing mode. Thus, for each
multiplexing mode, the device 12 may determine one or more coding
rate-dependent loss parameters as a function of variations in the
stream-specific received signal qualities for the best set of
antennas in that multiplexing mode.
[0053] In a specific example, the wireless communication device 12
receives a HS-DSCH signal from the transmitter 30 as a multiplexed
signal transmitted as N substreams on N selected transmit antennas
36. Thus, the wireless communication device 12 determines N
antenna-specific received signal qualities, based on, for example,
using its receiver circuit 52 to measure the SINR of each
substream. The wireless communication device 12 may generate N
antenna-specific quantizations of the measured SINR, e.g., N CQI
values, for feedback to the transmitter 30. Note that the number N
changes for each spatial multiplexing mode, and the wireless
communication device 12 may be configured to return such
information for each available mode. For modes where N=1, i.e., the
signal is transmitted from just one of the antennas 36, the
variation-related loss value is unity, since there are no SINR
variations across multiple antennas.
[0054] FIG. 7 illustrates complementary processing logic
implemented by the transmitter 30 of the transceiver node 20,
wherein transmitter 30 includes one or more processing circuits
implementing a method of compensating for losses in received signal
quality experienced by a receiver receiving a multiplexed signal
comprising two or more signal streams. In at least one embodiment,
the method comprises receiving a received signal quality for the
multiplexed signal and a loss parameter indicative of variations in
stream-specific received signal qualities of the multiplexed signal
(Step 104), and controlling the selection of one or more
transmission parameters of the multiplexed signal as a function of
the received signal quality and the loss parameter (Step 106).
[0055] The transmitter 30 may receive a received signal quality and
a loss parameter for each of one or more multiplexing modes
associated with the multiplexed signal, and thus may calculate
coding rate-dependent signal quality losses for the multiplexed
signal in each multiplexing mode based on the loss parameter
received for that multiplexing mode. This information allows the
transmitter 30 to identify a highest coding rate supportable in
each multiplexing mode based on the coding-rate-dependent signal
quality losses calculated for that multiplexing mode, and select a
multiplexing mode and coding rate to use for the multiplexed signal
based on comparing the highest coding rates between the
multiplexing modes. In at least one embodiment, the transmitter 30
uses the coding rate-dependent losses to determine effective
received signal qualities for the available coding rates in each
multiplexing mode, and compares the effective received signal
qualities to signal quality thresholds defined for the different
coding rates, to identify the coding rate(s) that can be supported
in each mode.
[0056] However, regardless of such details, it should be
appreciated that the received signal information fed back to the
transmitter 30 may comprise mode-specific received signal qualities
and corresponding mode-specific loss parameters, for each of one or
more multiplexing modes (code and/or spatial multiplexing modes)
that are available for transmitting the multiplexed signal. In such
embodiments, the mode-specific received signal qualities are
calculated as functions of the stream-specific received signal
qualities for the multiplexed signal for the corresponding
multiplexing modes, and the mode-specific loss parameters indicate
the variations in the stream-specific received signal qualities for
each multiplexing mode.
[0057] As such, the transmitter 30 may determine effective received
signal qualities in each multiplexing mode based on the
mode-specific received signal quality indicator and a number of
coding rate-dependent signal quality losses calculated from the
mode-specific loss parameter(s) and coding-rate dependent signal
quality loss modeling information. It may then use the effective
received signal qualities to select one of the multiplexing modes
and coding rates based on comparing the effective received signal
qualities to coding-rate dependent signal quality thresholds. For
example, the transmitter 30 may adapt the multiplexed signal's
transmit link based on identifying the combination of transmit
modulation and coding selections having a highest coding rate and
having a corresponding minimum received signal quality requirement
that is below the effective received signal quality calculated for
that combination.
[0058] Regardless of the particular feedback format, the
transmitter 30 may use effective (compensated) received signal
qualities in controlling the one or more transmission parameters of
the multiplexed signal. These effective received signal qualities
generally are reduced from the apparent received signal quality
measured at the remote receiver, as they account for the effective
reduction in received signal quality arising at the remote receiver
because of the variations. See, e.g., the loss curves of FIG. 3 and
the MCS selection line of FIG. 5. Thus, the transmitter 30 may use
loss-compensated effective received signal qualities to make the
appropriate transmit link adaptations for the spatially transmitted
multiplexed signal.
[0059] For example, for a multiplexed HS-DSCH signal, the
transmitter 30 might receive a CQI value for each of one or more
spatial multiplexing modes, along with corresponding variation
information for each such mode, reflecting the spread in
stream-specific SINR or quantized CQI for each such mode. With such
information, the transmitter 30 can be configured to compute an
effective CQI for each mode, and use the effective CQI to determine
the highest-rate MCS that can be supported at a desired maximum
error rate. For example, using a block or bit error rate limit of
10%, the transmitter 30 can determine the highest-rate MCS
selection for each spatial multiplexing mode, compare the
highest-rate supportable in each mode, and select the best one of
them.
[0060] FIG. 8 illustrates one embodiment of the above processing
logic, wherein the transceiver node 20 receives a set of
stream-specific received signal qualities as feedback from a
targeted wireless communication device 12 (Step 110). That is, the
transmitter 30 of the transceiver node 20 receives feedback in the
form of a set of stream-specific received signal quality
measurements made by the wireless communication device 12 for each
of one or more spatial multiplexing modes. For example, for each of
M possible spatial multiplexing modes, for mode "m," there are m
distinct SINRs at the wireless communication device 12, resulting
in a total of
m = 1 M m = 1 2 M ( M + 1 ) ##EQU00003##
SINRs for all M modes.
[0061] As an optional step, the transmitter 30 may be configured to
scale the reported signal qualities to account for differences
between the actual transmit power and code allocations that will be
made by the transmitter 30 for the multiplexed signal, versus the
transmit power and code allocations assumed by the wireless
communication device 12 for its received signal quality
calculations (Step 112).
[0062] That is, the actual received signal quality at the wireless
communication device 12 will depend on actual transmit power and
code allocations made at the transmitter 30 for the multiplexed
signal, which change over time because of changing resource
availabilities. The feedback will be "wrong" to the extent that the
actual allocations are different from those assumed by the wireless
communication device 12, and appropriate scaling up or down by the
transmitter 30 can correct the error. Alternatively, the
transmitter 30 may be configured to transmit information to the
wireless communication device 12 regarding the transmit power and
code allocations, such that the wireless communication device 12
has the correct allocation information, thereby obviating the need
for feedback scaling by the transmitter 30.
[0063] For example, for a spatially multiplexed HS-DSCH signal, the
wireless communication device 12 reports a SINR for the nth antenna
of a given mode m, which is denoted as .gamma..sub.n(m). Each of
these reported SINRs can be scaled up if the assumed allocations
are less than the actual allocations, or scaled down if the assumed
allocations are greater than the actual allocations.
[0064] Continuing with the illustrated processing logic, the
transmitter 30 calculates the received signal quality loss for one
or more MCS selections in each of one or more spatial multiplexing
modes (Step 114). In particular, for the illustrated processing,
the calculation circuit 44 of the transmitter 30 is configured to
calculate an average SINR across the active transmit antennas 36,
for each of the one or more spatial multiplexing modes, based on
the reported stream-specific SINRs. This averaging operation may be
expressed as,
.gamma. avg ( m ) = 1 m n = 1 m .gamma. n ( m ) 3 ##EQU00004##
Then, the transmitter 30 uses an appropriate loss model, e.g., one
based on the curves illustrated in FIG. 3, to determine the
variation-related loss in received signal quality, for each mode.
With the losses thus calculated, the transmitter 30 can determine
the effective SINR for each mode, by compensating the average SINR
determined for that mode by the corresponding loss.
[0065] Thus, one may denote L.sub.i(m) as the SINR loss for the ith
entry in a loss table (stored in memory associated with the
transmitter 30). Accordingly, the calculation circuit 44 and/or
control circuit 46 of the transmitter 30 can be configured to find
the largest-rate MCS such that the effective SINR is greater than
the switch point (quality threshold) for that MCS, i.e,
.gamma..sub.avg(m)/L.sub.i(m).gtoreq.s.sub.i(m), where s.sub.i(m)
is the signal quality switching point for the ith MCS in a given
mode. In general, then, the transmitter 30 is configured to
identify the largest-rate MCS in each mode, such that the effective
received signal quality, as compensated for SINR loss, is greater
than the minimum required signal quality defined for each MCS (Step
116).
[0066] With these operations completed, the transmitter 30 is
configured to compare the best MCS supportable in each spatial
multiplexing mode evaluated, and select the mode having the
largest-rate supportable MCS--i.e., select the "best" mode for
transmitting the multiplexed signal based on comparing the best MCS
supportable in each of the available spatial multiplexing modes
(Step 118). Note that if the largest rate MCS is not needed, the
transmitter 30 may select another mode and/or rate. For example, if
there is insufficient data in a transmit queue at the transmitter
30 for a targeted wireless communication device 12, a non-best mode
might be selected, or a lower-rate MCS within the best mode might
be used.
[0067] While the embodiment immediately above reflected
stream-specific received signal quality variations in the feedback
from the wireless communication device 12 by literally returning
individualized (stream-specific) quality measurements for each of
one or more multiplexing modes, FIG. 9 illustrates an embodiment
wherein the feedback still reflects such variations, but does so
with a smaller amount of feedback. Reducing the amount of feedback
is desirable where reverse link loading is of concern, or simply
where the amount of control and signaling overhead is undesirably
high.
[0068] More particular, FIGS. 9 and 10 illustrate an embodiment of
the wireless communication device 12 generating received signal
quality feedback (FIG. 9) and the transmitter 30 performing
transmit link adaptations responsive to that feedback (FIG. 10).
Broadly, the wireless communication device 12 reflects the
stream-specific variations in the received signal quality it
measures for the multiplexed signal by returning an effective
received signal quality. The effective received signal quality may
be, for example, the average of the stream-specific received signal
qualities scaled by an appropriate loss value.
[0069] Thus, the processing of FIG. 9 begins with the wireless
communication device 12 determining stream-specific received signal
qualities for each of one or more spatial multiplexing modes (Step
120). This may comprise determining distinct SINRs
{.gamma..sub.n(m)}.sub.n=1.sup.m for the best antenna selection for
each possible mode m=1, 2, . . . , M using preconfigured (or agreed
upon) nominal transmit power and code allocations.
[0070] For each mode, the wireless communication device 12 uses the
above stream-specific information to calculate an average received
signal quality (Step 122). For example, for each mode, the wireless
communication device 12 may compute an average SINR across the
particular antennas selected for or corresponding to that mode,
i.e.,
.gamma. avg ( m ) = 1 m n = 1 m .gamma. n ( m ) . ##EQU00005##
Further, for each mode, the wireless communication device 12
calculates a received signal quality loss for each of one or more
MCS selections available for the mode, according to a loss model or
other loss calculation (Step 124). The MCS tables identifying the
available MCS selections for each mode generally should be the same
as those used by the transmitter 30. An appropriate loss model may
comprise one or more look-up tables stored in a memory device
including in the wireless communication device 12, with such table
(or tables) based on the loss curves illustrated in FIG. 3, for
example. Loss calculation also may be based on solving a loss
equation that models loss.
[0071] In any case, for each mode, the wireless communication
device 12 is configured to identify the largest-rate MCS selection
available in the mode, such that an effective received signal
quality is greater than a minimum signal quality associated with
that MCS (Step 126). For example, one may denote L.sub.i(m) as the
SINR loss for the ith entry in a loss table for mode m. (Note that
for the loss for mode m=1, i.e., L.sub.i(1), is always unity, and
thus no loss calculation is required.) Using this loss notation,
the wireless communication device 12 may identify the largest-rate
MCS that can be supported in each mode based on,
.gamma. eff , i ( m ) = .gamma. avg ( m ) L i ( m ) .gtoreq. s i (
m ) 4 ##EQU00006##
where s.sub.i(m) is the switching point--i.e., the signal quality
minimum--for the ith MCS in mode m.
[0072] With the above operations, the wireless communication device
12 then feeds back for each of the M modes, the effective received
signal quality corresponding to the largest-rate MCS identified in
Step 126 (Step 128). For example, the wireless communication device
12 can feed back .gamma..sub.eff,i(m) for each of the M possible
spatial multiplexing modes
[0073] FIG. 10 illustrates complementary processing at the
transmitter 30, wherein it receives feedback reflecting the
stream-specific variations in received signal quality at the
wireless communication device 12 in the form of an effective
received signal quality for each of the M possible spatial
multiplexing modes (Step 130). Optionally, the transmitter 30
scales the reported values up or down, as needed, to account for
any differences between actual transmit power and code allocations
at the transmitter 30, versus the allocations assumed by the
wireless communication device 12 in its received signal quality
computations (Step 132).
[0074] Processing continues with the transmitter 30 simply
"mapping" the reported effective received signal qualities into a
MCS selection table, such as the one used by the wireless
communication device 12, to identify the largest-rate MCS that can
be supported in each mode (Step 134). The transmitter 30 selects
the mode corresponding to the largest-rate MCS that can be
supported, or, if such data rates are not needed, it may select a
lower-rate MCS from that mode, or from another one of the available
modes (Step 136). In any case, transmit link adaptation by the
transmitter 30 for the multiplexed signal is compensated for losses
arising at the wireless communication device 12 from the variations
in stream-specific received signal qualities.
[0075] FIGS. 11 and 12 illustrate another embodiment of the
wireless communication device 12 generating received signal quality
feedback (FIG. 11) and the transmitter 30 performing transmit link
adaptations responsive to that feedback (FIG. 12). Processing in
FIG. 11 begins with the wireless communication device 12
determining stream-specific received signal qualities for each of
one or more spatial multiplexing modes (Step 140). As before, this
step may comprise the wireless communication device 12 estimating
distinct SINRs for each of the particular transmit antennas 36 used
in each of one or more spatial multiplexing modes.
[0076] Processing continues with the wireless communication device
12 calculating an average of the received signal qualities and a
maximum received signal quality "spread" for each mode (Step 142).
The maximum spread for each mode may be calculated as
.DELTA.m=max.sub.n=1.sup.m{.gamma..sub.n}-min.sub.n=1.sup.m{.gamma..sub.n-
} amongst the m individual antenna SINRs for a given mode. The
wireless communication device 12 transmits the received signal
quality information for each mode back to the transceiver node 20,
wherein it is passed along to the transmitter 30 (Step 144).
[0077] The feedback information in this context may comprise an
average SINR and the maximum SINR spread, which may be expressed by
sending the highest and lowest SINRs, or the difference
therebetween. Such information may be sent for each mode and it
should be noted that characterizing the variations in
stream-specific received signal qualities in this manner reduces
the amount of feedback from the wireless communication device 12,
while still allowing the transmitter 30 to compensate its transmit
link adaptations for the losses arising from such variations.
[0078] In understanding one embodiment of such compensation by the
transmitter 30, by way of non-limiting example, one may assume that
M=4 transmit antennas, and the use of spatial multiplexing mode 4
(i.e., the use of all four transmit antennas 36). For such an
example, the wireless communication device 12 may return an average
SINR value, along with the maximum spread in received signal
qualities expressed as the difference between the minimum
stream-specific SINR and the maximum stream-specific SINR measured
by the wireless communication device 12. Since the middle two SINRs
are not fed back to the transmitter 30, i.e., they are "missing"
from the feedback, the transmitter 30 cannot uniquely determine the
variation-related loss in received signal quality for the wireless
communication device 12. However, the transmitter 30 can determine
the loss in a probabilistic sense through the use of confidence
intervals.
[0079] For example, the transmitter 30 may store in its memory a
distribution of SINR losses for given SINR spreads, according to
one or more different modulation formats and different coding
rates, e.g., Quadrature Phase Shift Keying modulation, Quadrature
Amplitude Modulation, etc. In making its loss estimation, the
transmitter 30 may exploit further statistical characteristics that
generally hold for the signal quality loss, such as the fact that
the mean and standard deviation of the SINR losses generally
increase with coding rate, and also increase with the maximum
spread.
[0080] Thus, the transmitter 30 may establish a confidence interval
that is defined such that the probability that the loss falls in
some interval is fixed at a target value, e.g., 90%. Denoting the
mean and standard deviation of the loss as u(r,.DELTA.) and
.sigma.(r,.DELTA.) respectively, where r equals the coding rate and
A equals the maximum SINR spread, the confidence interval is given
by [0,L.sub.P], where
L.sub.P=.mu.(r,.DELTA.)+f.sub.P(r,.DELTA.).sigma.(r,.DELTA.)
In the above expression, f.sub.P(r,.DELTA.) is a multiplier on the
standard deviation that is chosen to ensure that the SINR loss
falls within the confidence interval with probability P, e.g.,
P=90%. The multiplier is a function of the SINR spread .DELTA., the
coding rate r, and the modulation type.
[0081] The values of L.sub.P may be calculated in advance and
stored in a memory table at the transmitter 30. Thus, when the
transmitter 30 receives the SINR spread as feedback, it can use the
spread and the coding rate for a particular MCS to look-up L.sub.P
from the table for a particular mode. That look-up value serves as
a bound on the SINR loss for that mode and MCS. The transmitter 30
thus uses the bound in the MCS selection process, rather than using
an "exact" loss value, knowing that the actual SINR loss at the
wireless communication device 12 rarely will exceed L.
[0082] Alternatively, a single multiplier F.sub.p may be
used--independent of the coding rate and the SINR spread--such that
the probability is P or greater that the loss falls in the interval
[0,L.sub.P+], where,
L.sub.P+=.mu.(r,.DELTA.)+F.sub.P.sigma.(r,.DELTA.) 6
For example, use of a fixed multiplier F.sub.P=1.5 (slightly
greater than an example maximum value of f.sub.P(r,.DELTA.) in the
previously mentioned table of Lp values for given combinations of
coding rate and SINR spread, ensures that the SINR loss falls
within the confidence interval [0,L.sub.P+] with probability 90% or
greater. If a fixed multiplier approach is adopted by the
transmitter 30, then Eq. 6 can be used to generate the look-up
table(s) used by the transmitter 30 for MCS selection.
[0083] Thus, with the above example in mind, FIG. 12 steps through
an embodiment of transmitter link adaptation based on the
transmitter 30 receiving an average signal quality and a
corresponding maximum spread, for each of one or more modes (Step
146). The transmitter 30 optionally may scale the feedback
information, as described elsewhere herein (Step 148). Any such
scaling produces the scale values .gamma..sub.avg(m) and .DELTA.(m)
for each spatial multiplexing mode m.
[0084] Then, for each mode and each entry in the MCS selection
table for that mode, the transmitter 30 looks up the upper bound of
the confidence interval for SINR loss from a pre-calculated table
stored at the transmitter 30. Denote L.sub.P,i(m) (or L.sub.P+,i(m)
if using the fixed multiplier F.sub.P) as the upper bound for the
ith entry in the MCS table for mode-m. Note that for mode-1,
L.sub.P,i(m) or L.sub.P+,i(m) are always unity and no look-up needs
to be performed. The transmitter 30 then finds the largest-rate MCS
such that the lower bound on effective received signal quality
(effective SINR) is greater than the switch point for that MCS,
i.e.,
.gamma. avg ( m ) L P , i ( m ) .gtoreq. s i ( m ) 7
##EQU00007##
Where s.sub.i(m) is the switch point for the ith MCS in the MCS
selection table of mode m (Steps 150 and 152). The transmitter 30
then selects the mode corresponding to the largest-rate MCS
identified in the prior step, or selects a lower-rate MCS in the
same mode, or in a different mode, if the largest-rate MCS is not
needed (Step 154).
[0085] FIGS. 13 and 14 illustrate another embodiment of feedback
generation at the wireless communication device 12 (FIG. 13) and
corresponding transmit link adaptation at the transmitter 30 (FIG.
14). Broadly, the illustrated processing is based on the wireless
communication device 12 feeding back parameterized information that
reflects the variations in stream-specific received signal
qualities at the wireless communication device 12, for the
multiplexed signal in each of one or more spatial multiplexing
modes (Step 160).
[0086] For each such mode, the wireless communication device 12
generates parameterized received signal quality feedback (Step
162). Such parameterization may build on the above-described
maximum spread parameter, or stand as an alternative to that prior
method. For example, rather than sending the maximum spread along
with an average value, the wireless communication device 12 may be
configured to generate and return one or more parameters that
characterize the variations in stream-specific received signal
quality at the wireless communication device 12 (Step 164).
[0087] For example, the wireless communication device 12 may be
configured to compute and return an average value and a standard
deviation value for the stream-specific received signal qualities,
for each of one or more spatial multiplexing modes. Of course,
other statistical values, such as variance, etc., can be sent in
addition to the standard deviation, or as an alternative to it.
[0088] In a similar embodiment, the parameterized feedback may
comprise an average value and a "penalty" value. The penalty value
could be set such that the loss is fixed at some percentile of the
expected losses seen in a typical system. Thus, one or more
pre-configured penalty values could be stored in memory at the
wireless communication device 12, and the wireless communication
device 12 could select an appropriate penalty value for reporting
back to the transmitter 30 based on, for example, the spread in
stream-specific SINRs measured by the wireless communication device
12 for the spatially multiplexed signal.
[0089] Further, the wireless communication device 12 can be
configured to define the loss function for use in calculating the
penalty value, such that it imparts small penalties at low
(average) SINRs. Conversely, the function may produce relatively
larger penalties with increasing average SINRs, to reflect the
larger losses arising from the higher coding rates that generally
are used for the higher signal qualities. In one or more such
embodiments, the switching point step sizes in the MCS selection
tables used by the transmitter 30 can be increased to limit the
impact of loss compensation inaccuracies. That is, the signal
quality spans covered by one or more MCS selections in one or more
MCS tables can be increased, such that it takes a higher
(penalized) signal quality for the next-higher MCS to be selected
by the transmitter 30.
[0090] In a further embodiment of parameterized feedback, the
wireless communication device 12 can be configured to carry out a
parameterization of specific loss curves, given vectors of the
stream-specific received signal qualities calculated by it, i.e., a
vector given as {.gamma..sub.1, .gamma..sub.2, . . . ,
.gamma..sub.M}.
[0091] With such embodiments, the wireless communication device 12
can be configured to fit a linear function to a specific loss curve
according to certain minimum error criteria. The error minimization
criteria could, for example, be defined to conform to the least
square principle. Alternatively, the wireless communication device
12 could be configured to fit a (high-order) polynomial to a
specific loss curve according to minimum error criteria. The
parameterized feedback from the wireless communication device 12 to
the transmitter 30 can be reduced to an average received signal
quality (e.g., average SINR .gamma..sub.avg) and the parameter(s)
to identify the specific loss function. At the transmitter, the
specific loss curve can be reconstructed using the parameter
feedback. In this context, a loss curve expresses the expected SINR
loss (in dB) as a function of modulation and coding rate.
[0092] As an example, the mth parameterized function for four
transmit antennas 36 can be defined as a SINR loss curve with
{ .gamma. 1 , .gamma. 2 , .gamma. 3 , .gamma. 4 } = { .gamma. avg -
3 2 .delta. m , .gamma. avg - 1 2 .delta. m , .gamma. avg + 1 2
.delta. m , .gamma. avg + 3 2 .delta. m } 8 ##EQU00008##
Where the resulting value is expressed in dBs, and where
.delta..sub.m is a spacing parameter for the function set. The
selection of a spacing parameter can be optimized with typical
distribution information regarding the SINRs. For 32 spacing values
corresponding to 32 different specific loss curves, the wireless
communication device 12 can be configured to feedback the average
SINR and a 5-bit spacing value, identifying the particular loss
curve to be used. With the loss curve thus identified, the
transmitter 30 can move along the curve as a function of the coding
rate to determine the SINR loss.
[0093] The best parameterized function (loss curve) from the set of
loss curves defined at the wireless communication device 12 may be
chosen according to a least square principle, or it may be chosen
to be the lowest-indexed one that is higher than the actual SINR
loss curve for all coding rates. Such a criterion guarantees that
the SINR loss is never underestimated, at the expense of
overestimating loss under some circumstances. The best
parameterized function also may be chosen to be the one with the
.delta. value closest to the averaged spacing of the actual SINR
vector. For instance, an actual (sorted) SINR vector for four
antennas, and for a given loss curve in dB may be
{.gamma..sub.1,.gamma..sub.2,.gamma..sub.3,.gamma..sub.4}={.gamma..sub.a-
vg-4.39,.gamma..sub.avg-2.94,.gamma..sub.avg+2.18,.gamma..sub.avg+5.16}
9
[0094] The spacings hence are 1.45, 5.12, and 2.98 dB, with an
average of 3.18 dB. The average may be compared to a table of
.delta. values for the defined loss curves, and the curve
corresponding to the closest .delta. value in the table may be
selected.
[0095] Thus, with the above wide ranging possibilities for
parameterized feedback in mind, FIG. 14 illustrates a general
embodiment of processing operations at the transmitter 30, in which
it compensates its transmit link adaptations for the multiplexed
signal as a function of the parameterized received signal quality
feedback from the wireless communication device 12. Processing thus
begins with the transmitter 30 receiving parameterized feedback
from the wireless communication device 12 (Step 166), and
optionally scaling such information as described elsewhere herein
(Step 168).
[0096] Then, for each mode, the transmitter 30 determines an
effective received signal quality based on the parameterized
information (Step 160). For example, the transmitter 30 scales or
otherwise reduces the average SINR value received for a given mode
for MCS-specific loss values determined for that mode based on the
parameterized feedback (Step 170). The transmitter 30 then
identifies the largest-rate MCS that can be supported by the
corresponding effective received signal quality in each mode (Step
172), and then selects the mode corresponding to the largest-rate
MCS, or makes another selection if the largest-rate MCS is not
needed (Step 174).
[0097] From the number of above-described embodiments, which among
other things illustrates some of the variations in processing at
the wireless communication device 12 and the transmitter 30, those
skilled in the art will appreciate that the present invention
broadly provides methods and apparatus for improving transmit link
adaptations for multiplexed signals, by providing feedback that
reflects the variations in stream-specific received signal quality
for the multiplexed signal at one or more remote receivers. The
feedback may directly convey the variations, or indirectly convey
information regarding the variations, such as by providing
parameterized data, or by providing compensated, effective signal
quality values.
[0098] Further, without regard to the feedback particulars, those
skilled in the art will readily appreciate that the methods taught
herein broadly apply to a range of multiplexing transmissions
systems, including spatial, code, and frequency multiplexing
transmission systems. Broadly, the signal quality loss compensation
methods taught herein may be applied essentially to any
multiplexing transmission system wherein multiple subchannels with
different SINRs are used to transmit information to a receiver.
That is, the methods taught herein directly apply across the range
of these different multiplexing methods, and to combinations of
these methods, such as where code and spatial multiplexing are
combined. As such, the case where multiple transmit antennas are
used can be considered as a special case of multiplexing.
[0099] Thus, the methods of signal quality loss compensation taught
herein apply to Orthogonal Frequency Division Multiplexing (OFDM)
transmissions, wherein multiple subchannels with different SINRs
used to transmit information to a receiver. More particularly, a
common embodiment of an OFDM-based multiplexing transmission system
assigns multiple frequency subchannels to each user. With
frequency-selective fading, the received signal quality on each of
these subchannels can vary widely, and the methods taught herein
thus can be used to determine an effective signal quality, wherein
the effective signal quality reflects a signal quality loss arising
from variations in the substream signal qualities.
[0100] As another example, the methods taught herein may be applied
to transmission systems based on the Global Standard for Mobile
communications (GSM), at least for certain transmission cases in
such systems. More particularly, information may be sent over a GSM
channel based on using different timeslots and hopping over
multiple frequencies. In such cases, the GSM channel is a type of
the multiple-state channels to which the signal quality loss
compensation methods taught herein are applied.
[0101] Thus, those skilled in the art will appreciate that a
multiplexing transmitter, whether based on spatial multiplexing,
code multiplexing (e.g., multi-coded information substreams),
frequency multiplexing, or any combination thereof, can improve its
transmit link adaptations for a multiplexed transmit signal by
compensating those adaptations for losses in received signal
quality arising from received signal quality variations in the
different multiplex streams. With this point in mind, then, it
should be understood that the present invention is not limited by
the foregoing description, nor is it limited by the accompanying
drawings. Instead, the present invention is limited only by the
following claims, and their legal equivalents.
* * * * *