SETTING UPLINK ANTENNA TRANSMISSION WEIGHTS IN SOFT HANDOVER
FIELD OFTHE INVENTION
The present invention relates to a method of setting antenna transmission weightings of
user equipment in a wireless telecommunication network, user equipment and a
computer program product.
BACKGROUND
Wireless telecommunications systems are known. In those known systems, radio
coverage is provided to user equipment, for example, mobile telephones, by
geographical area. A base station is located in each geographical area to provide
the required radio coverage. User equipment in the area served by a base station
receives information and data from the base station and transmits information and
data to the base station. In a high-speed packet access (HSPA) telecommunications
network, data and information is sent between user equipment and a base station in
data packets on a radio frequency carrier.
Information and data transmitted by the base station to the user equipment occurs on
radio frequency carriers known as downlink carriers. Information and data transmitted
by user equipment to the base station occurs on radio frequency carriers known as
uplink carriers.
In known wireless HSPAtelecommunication systems, user equipment can move
between geographical base station coverage areas. Service provided to user
equipment is overseen by a radio network controller (RNC). The radio network
controller communicates with user equipment and base stations and determines which
base station, and which cell of the geographical area served by that base station each
user equipment is primarily connected to (known as the "serving cell"). Furthermore, a
radio network controller acts to control and communicate with a base station and user
equipment when user equipment moves from the geographical area served by one
base station to a geographical area served by another base station.
A signal transmitted between user equipment and a base station over a radio channel
typically experiences many propagation paths, for example, due to reflection, before
arriving at a base station receiver. The signal carried on those paths each arrive at a
different time, power and phase at the receiver. The sum of the different signal
propagation paths at the receiver causes the total signal received to attenuate or
amplify depending on the phases of the different received propagation paths.
Changes †o the transmitter position or the transmitter surroundings causes the multiple
propagation path signals to change, leading to fluctuation in the signal at the receiver.
This characteristic is known a s multipath fading or fast fading. Fast fading causes the
signal to fluctuate over even a short time span. A signal may suffer significant
attenuation due to fading or deep fade. A signal that has undergone deep fade may
not be decodable.
Transmit diversity xDiv), is a method according to which a signal can be transmitted
over two or more antennas. If a signal transmitted by a first antenna experiences a
deep fade, the same signal transmitted on another antenna, typically experiences
different radio conditions and propagation paths and may arrive with good quality.
Uplink transmit diversity at user equipment requires more than one antenna to be
provided, and that a signal may be sent to a base station on one or more of those
an†enna(s). The signal arriving at a base station from two antennas may be combined
by the base station and can thus result in a diversity gain at the base station of the
transmitted signal. Further, if the signal transmitted using one antenna is particularly
affected by fast fading, a signal transmitted using another antenna may be less
affected by fast fading.
The requirements of a user having uplink transmission diversity may be contrary to other
requirements in place within a network. Those requirements may conflict, leading to
decreased efficiency of data transfer within a wireless telecommunication network a s a
whole.
Accordingly, it is desired to improve the robustness of a wireless telecommunications
network having uplink transmit diversity functionality.
SUMMARY
According to a first aspect, there is provided a method of setting antenna transmission
weightings of user equipment in a wireless telecommunication network, the user
equipment being operable to transmit on at least two antenna and operable to
communicate simultaneously with at least two base stations from an active set
associated with the user equipment, the method comprising the steps of:
receiving a n indication of preferred antenna transmission weightings from the at
least two base stations in the active set.
calculating a n indication of compromise antenna transmission weightings which
minimise a difference between the indication of compromise antenna transmission
weightings and the indication of preferred antenna transmission weightings; and
setting the antenna transmission weightings in accordance with the indication
of compromise antenna transmission weightings.
The first aspect recognises that transmit diversity (TxDiv), is a method according to
which a signal can be transmitted over two or more antennas, and that method may
allow for improved communication between user equipment and a base station by
offering a chance to overcome or mitigate the likelihood of fast fading degrading a
signal to the extent it is no longer decodable. If a signal transmitted by a first antenna
experiences a deep fade, the same signal transmitted on another antenna, typically
experiences different radio conditions and propagation paths and may arrive with
good quality.
In an implementation of uplink transmit diversity at user equipment, a signal Sis
duplicated and each copy is multiplied by a weight. Signals S i and S are produced by
multiplying weight and V½ respectively by the copy of signal S. Each of those signals
is then transmitted via a different antenna. The signals S i and S.as received at a base
station are combined by the base station and can thus result in a diversity gain at the
base station of signal S.
TxDiv schemes can use various methods to determine the weights W and ½. It is
possible to allow a base station to determine appropriate weights and W directly
rather than allowing user equipment to choose. Since the radio channels that signals
S i and S2 propagate through can be estimated at a base station (BS), the base station
is able to determine the weights V and l ½ that can maximize the received quality of
signal S, after combining signal S i and S2, at that base station. The choice of weights W
and W2 selected by the base station will need to be explicitly signalled to user
equipment.
A HSPAnetwork allows for increased data throughput in the uplink by allowing user
equipment to communicate with more than one base station. That is to say, any
packet transmitted by user equipment may be received at more than one base
station. The network, and specifically RNC, is able to combine the packets received by
different base stations, and thus can help ensure communication between user
equipment and the network asa whole is more robust. That scenario is known as soft
handover (SHO) and comprises a collection of cells, supported by one or more base
stations, communicating with one user equipment.
The cells participating in SHO of a UE belong to a n "Active Set" of the UE. In SP
within the Active Set, there will be one main cell serving the UE, referred to asa "serving
cell". Other cells that are not the serving cells within the active set are called the nonserving
cells. If the cells in the Active Set belong to the same base station asthat which
hosts the serving cell, this is termed "Softer Handover".
In a High Speed Uplink Packet Access (HSUPA) network having uplink transmission
diversity functionality, a n uplink data transmission from user equipment will typically use
SHO, thus allowing for more robust communication with the network asa whole.
Uplink Transmission Diversity fJxDiv) weights are determined based on radio channels
between user equipment and a receiver at a base station. Since the radio channels
between the UE antennas and each of the cells (NB receive antennas) in the Active Set
are different, each cell, or base station, with which user equipment is in communication
will typically request different TxDiv weights to try to optimize communication between
the user equipment and itself.
User equipment is, however, typically operable to apply only one set of weights to its
transmission. If a n overall inappropriate weight is chosen, based on the
recommendation of a single base station in a user equipment active set, the weights
used may degrade the reception of some cells, leading to a n overall loss in network
efficiency.
The first aspect recognises that rather than simply being a slave to a single base station,
it may be possible to balance the requirements of HSPAand a transmit diversity scheme
to optimise the efficiency and operation of the network as a whole. Accordingly, a
method of balancing the transmission weighting requests received from the base
stations in the active set of user equipment may allow some compromise to be
reached, thereby allowing for any detriment to one base station caused by selection of
a weighting which has not been requested by that base station to be compensated for
my moving the selected weighting closer to a weighting selected by another base
station in the active set of the user equipment.
It will be appreciated that various methods may be employed to determine a suitable
compromise, but those methods seek to minimise the difference between the
weightings implemented and the various weightings being requested.
In one embodiment, the step of calculating comprises the step of: minimising the mean
squared distance between the indication of compromise antenna transmission
weightings and the indication of preferred antenna transmission weightings received
from the base stations in the active set.
Accordingly, the difference between all received requested weightings and the
selected weighting is minimised, thereby reaching a numerical compromise between
all base stations in the active set requesting transmission diversity weightings.
In one embodiment, the step of calculating comprises the step of: averaging the
indication of preferred antenna transmission weightings received from the base stations
in the active set. Accordingly, the compromise weightings may simply represent an
average of all requested weightings.
In one embodiment, the step of calculating comprises the step of: calculating the
mode preferred antenna transmission weightings received from the base stations in the
active set. Accordingly, the method selects the most requested weighting from
available possible weightings. If more than one weighting selection is determined to be
the mode, ie if two weightings are selected with an equal frequency, the method may
operate to randomly select one of the modal weighting selections.
In one embodiment, the step of calculating further comprises the steps of: assigning a
weighting to the indication of preferred antenna transmission weightings from each
base station in the active set, the weighting being determined based upon an
indication of the relative importance of each base station in the active set to operation
of the user equipment within the network. Accordingly, those base stations which are
more integral to efficient operation of the user equipment within a wireless
telecommunication network are given priority when determining a compromise
weighting for implementation, thereby mitigating the change that implementation of a
compromise weighting will unnecessarily disrupt the operation of a wireless
communication network.
In one embodiment, the weighting is assigned based upon an indication of received
pilot signal strength from the base stations in the active set at the user equipment.
Accordingly, those base stations which appear to be closest, or appear to have
strongest communication links with the user equipment may be assigned a higher
importance, resulting in the compromise weighting favouring their choice of antenna
weightings. Accordingly, a base station in the active set, but having a weak
communication link with the user equipment has less significance to the calculation of
compromise antenna weightings.
In one embodiment, the weighting takes account of which of the base stations in the
active set is responsible for providing the user equipment serving cell. Accordingly, the
serving cell may be afforded a greater weighting that a non serving cell. Furthermore,
in one embodiment, the serving cell may determine the compromise weightings
implemented by the user equipment.
In one embodiment, the indication of the preferred antenna transmission weightings
comprises a n indication of relative phase shift between transmissions of a signal on the
at least two antenna. It will be appreciated that each base station may be operable
to transmit a desired indication of relative phase and amplitude for each antenna at
said user equipment. Such an arrangement would, however, lead to high signalling
data traffic across the wireless network. The base station may instead transmit
indicators of desired phase and/or amplitude, those indicators comprising, for example,
an index which causes user equipment to look up the relative weightings in a locally
stored code book or similar. Accordingly, the indication may comprise and indication
of relative phase shift between user equipment antenna. In one embodiment, the
indication of the preferred antenna transmission weightings comprises an indication of
an amplitude shift between transmissions of a signal on the at least two antenna.
In one embodiment, the indication of preferred antenna transmission weightings
comprises an indication of antenna selection. Accordingly, a base station may
indicate, and transmit and indication of which antenna is transmitting a signal which is
being received most reliably, and thus experiencing better radio propagation
conditions.
A second aspect provides a computer program product operable, when executed on
a computer, to perform the method of the first aspect.
A third aspect provides user equipment operable to set antenna transmission
weightings in a wireless telecommunication network, the user equipment being
operable to transmit using at least two antenna and operable to communicate
simultaneously with at least two base stations in an active set associated with the user
equipment the user equipment comprising:
reception logic operable to receive an indication of preferred antenna
transmission weightings from the at least two base stations in the active set,
calculation logic operable to calculate an indication of compromise antenna
transmission weightings which minimise a difference between the indication of
compromise antenna transmission weightings and the indication of preferred antenna
transmission weightings; and
implementation logic operable to set the antenna transmission weightings in
accordance with the calculated indication of compromise antenna transmission
weightings.
In one embodiment, the calculation logic is operable to minimise the mean squared
distance between the indication of compromise antenna transmission weightings and
the indication of preferred antenna transmission weightings received from the base
stations in the active set.
In one embodiment, the calculation logic is operable to average the indication of
preferred antenna transmission weightings received from the base stations in the active
set.
In one embodiment, the calculation logic is operable to calculate the mode preferred
antenna transmission weightings received from said base stations in said active set.
In one embodiment, the calculation logic is operable to assign a weighting to the
indication of preferred antenna transmission weightings from each base station in the
active set, the weighting being determined based upon an indication of the relative
importance of each base station in the active set to operation of the user equipment
within the network.
In one embodiment, the weighting is assigned based upon an indication of received
pilot signal strength from the base stations in the active set at the user equipment.
In one embodiment, the weighting takes account of which of the base stations in the
active set is responsible for providing the user equipment serving cell.
In one embodiment the indication of said preferred antenna transmission weightings
comprises a n indication of relative phase shift between transmissions of a signal o n the
at least two antenna.
In one embodiment, the indication of the preferred antenna transmission weightings
comprises a n indication of a n amplitude shift between transmissions of a signal on the
at least two antenna.
In one embodiment, the indication of the preferred antenna transmission weightings
comprises a n indication of antenna selection.
Further particular and preferred aspects are set out in the accompanying independent
and dependent claims. Features of the dependent claims may be combined with
features of the independent claims a s appropriate, and in combinations other than
those explicitly set out in the claims.
BRIEF DESCRIPTION OFTHE DRAWINGS
Embodiments of the present invention will now be described further, with reference to
the accompanying drawings, in which:
Figure 1 illustrates a wireless telecommunications system according to one
embodiment;
Figure 2 illustrates schematically typical propagation paths between a transmitter and
a receiver;
Figure 3 illustrates fast fading of a signal power for user equipment moving at 3 kmph for
a signal operating in 2000 Mhz;
Figure 4 illustrates schematically a n implementation of uplink transmit diversity a t user
equipment according to one embodiment-
Figure 5 illustrates schematically user equipment operating in soft handover;
Figure 6 illustrates schematically a n embodiment of uplink Switched Antenna
Transmission Diversity;
Figure 7 illustrates schematically a set of transmission antenna weights according to one
embodiment
Figure 8 illustrates schematically a set of transmission antenna weights according to one
embodiment
Figure 9 and Figure 10 are plots of average loss .WEIGHT against the index distance DINDEX
according to a first model;
Figure 11 illustrates schematically a difference in weighting vectors;
Figure 12 illustrates the wrap around of codebook phase differences;
Figure 13 illustrates schematically a n active set of base stations for user equipment;
Figure 14 illustrates schematically a n arrangement of five base stations in a n active set
of example user equipment; and
Figure 15 illustrates a compromise weight vector calculated according to one
embodiment.
DETAILLED DESCRIPTION OFTHE DRAWINGS
Figure 1 illustrates a wireless telecommunications system 10 according to one
embodiment. User equipment 50 roam through the wireless telecommunications
system. Base stations 20 are provided which support areas of radio coverage 30. A
number of such base stations 20 are provided and are distributed geographically in
order to provide a wide area of coverage to user equipment 50. When user equipment
is within an area served by a base station 30, communications may be established
between the user equipment and the base station over associated radio links. Each
base station typically supports a number of sectors within the geographical area of
service 30.
Typically a different antenna within a base station supports each associated sector.
Accordingly, each base station 20 has multiple antennas and signals sent through the
different antennas are electronically weighted to provide a sectorised approach. Of
course, it will be appreciated that Figure 1 illustrates a small subset of the total number
of user equipment and base stations that may be present in a typical communications
system.
The radio access network of the wireless communications system is managed by a
radio network controller (RNC) 40. The radio network controller 40 controls operation of
the wireless communications system by communicating with a plurality of base stations
over a backhaul communications link 60. The network controller also communicates
with user equipment 50 via each base station.
A radio network controller 60 maintains a neighbour list which includes information
about geographical relationships between sectors supported by base stations 20. In
addition, the radio network controller 60 maintains location information which provides
information on the location of user equipment 50 within the wireless communication
system 10. The radio network controller is operable to route traffic via circuit switched
and packet switched networks. Hence, a mobile switching centre is provided with
which the radio network controller may communicate. The mobile switching centre
can communicate with a circuit switched network such as a public switched telephone
network (PSTN) 70. Similarly, a network controller can communicate with service
general package radio service support nodes (SGSNs) and a gateway general packet
support node (GGSN). The GGSN can communicate with a packet switched core such
a s for example, the Internet.
User equipment 50 typically transmits information and data to a base station 20 so that
it can be re-routed within a wireless telecommunications network. User equipment
may, for example, need to transmit data to the base station in order to relay text
messages, voice information when a user is using the equipment to make a telephone
call, or other data. The base station 20, in combination with parameters set by the
radio network controller 40, allocates resources to user equipment in a manner that
aims to optimise operation of the wireless telecommunications network 10.
Figure 2 illustrates schematically typical propagation paths between a transmitter, in this
case, user equipment 50 and a receiver on a base station 20. A signal transmitted over
a radio channel typically experiences many propagation paths, for example, due to
reflection, before arriving at a receiver. Those signal paths are represented as S(fi), S ) ,
and S in Figure 2, and each arrive at different time, power and phase at the
receiver. The sum of the different signal propagation paths a t the receiver causes the
total signal received to attenuate o r amplify depending o n the phases of the different
received propagation paths.
Changes to the transmitter position or the transmitter surroundings causes the multiple
propagation path signals to change, leading to fluctuation in the signal at the receiver.
This characteristic is known as multipath fading o r fast fading.
Figure 3 illustrates fast fading of a signal power for user equipment moving a t 3 kmph for
a signal operating in 2000 MHz. Fast fading causes the signal to fluctuate over even a
short time span. A signal may suffer significant attenuation due to fading or deep fade.
A deep fade is shown in Figure 3, and is ringed by a circle. That deep fade attenuates
the signal by 17 dB. A signal that has undergone deep fade may not be decodable.
Transmit diversity (TxDiv), is a method according to which a signal can be transmitted
over two o r more antennas. If a signal transmitted by a first antenna experiences a
deep fade, the same signal transmitted on another antenna, typically experiences
different radio conditions and propagation paths and may arrive with good quality.
Figure 4 illustrates schematically a n example implementation of uplink transmit diversity
at user equipment having two transmit antennas. As shown in Figure 4, a signal S is
duplicated and each copy is multiplied by a weight. Signals S and are produced by
multiplying weight V and ½ respectively by the copy of signal S. Each of those signals
is then transmitted via a different antenna. The signals S and S. as received at a base
station are combined by the base station and can thus result in a diversity gain at the
base station of signal S.
Possible TxDiv schemes use a n open loop method to determine the weights Wi and
and determination of those weights at user equipment is based on received Transmit
Power Control I commands issued by a serving base station CBS). Since TPC is used
for the purpose of power control, the indirect use of TPC to determine the TxDiv weights
does not give a true reflection of the uplink radio condition and the chosen weights
may have impact on non-serving cells, leading to small gains.
It is possible to allow a base station to determine appropriate weights Wi and V
instead of user equipment. The radio channels that signals Si and 2 propagate
through can be estimated at a base station (BS) and once the radio channels are
known, the base station is able to determine weights i and V that can maximize the
received quality of signal S, after combining signal Si and S2. It is expected that such a
scheme will offer higher gain than that offered by open loop TxDiv utilizing TPC
commands. The choice of weights and ½ selected by the base station will need to
be explicitly signalled to user equipment.
A HSPAnetwork allows for increased data throughput in the uplink by allowing user
equipment to communicate with more than one base station. That is to say, any
packet transmitted by user equipment may be received at more than one base
station. The network, and specifically NC, is able to combine the packets received by
different base stations, and thus can help ensure communication between user
equipment and the network as a whole is more robust. That scenario is known a s soft
handover (SHO) and comprises a collection of cells, supported by one or more base
stations, communicating with one user equipment.
The cells participating in SHO of a UE belong to a n "Active Set" of the UE. In HSPA,
within the Active Set, there will be one main cell serving the UE, referred to as a "serving
cell". Other cells that are not the serving cells are called the non-serving cells. If the
cells in the Active Set belong to the same base station a s that which hosts the serving
cell, this is termed "Softer Handover".
In a High Speed Uplink Packet Access (HSUPA) network having uplink transmission
diversity functionality, an uplink data transmission from user equipment will typically with
to use SHO, thus allowing for more robust communication with the network as a whole.
Uplink Transmission Diversity (TxDiv) weights are determined based on radio channels
between user equipment and a receiver at a base station. Since the radio channels
between the UE antennas and each of the cells (NB receive antennas) in the Active Set
are different, each cell will request different TxDiv weights from given user equipment.
User equipment is, however, typically operable to apply only one set of weights to its
transmission. If an overall inappropriate weight is chosen, based on the
recommendation of a single base station in a user equipment active set, the weights
used may degrade the reception of some cells.
Figure 5 illustrates schematically user equipment operating in soft handover. As shown
in Figure 5, NB1 , NB2 and NB3 belong to the Active Set of UE1 . UE1is outside the
coverage of NB4 and hence, NB4 is not in the Active Set of UE1 . In order to implement
uplink transmit diversity at user equipment U 1, each base station in the active set
makes a transmission weighting recommendation, based on optimizing the chances of
receiving a packet successfully at itself. NB1 selects W(l ), NB2 selects W(2) and NB3
selects W(3). The user equipment can only implement one weighting. Aspects address
the issue of how to determine suitable transmission weights for implementation by user
equipment in SHO to use in uplink TxDiv.
UPLINK TRANSMIT DIVERSITY - OVERVIEW
The set of values that the weights and W2. shown in Figure 4, may take is infinite. A
large number of bits would be required to feedback the actual values of the weights
W and W2 to user equipment, causing a huge amount of network traffic. To minimize
unnecessary signalling within the network 10, a finite set of weight values is chosen and
a base station feeds back to user equipment only an index to a set of values (ie, a
reference in a look up table).
Let W be the set of weights, for user equipment having two transmit antennas, as
follows:
Where, n is the index to a pair of weight W and . N is number of possible pairs of
weights that a base station may choose.
A base station will typically evaluate each of the N pair of weights in the set W against
the estimated radio channel to find the best pair of weights to be use in the next uplink
transmission. A base station then signals the "best" index n to user equipment. User
equipment is provided with a look up table comprising the same set of weights, W and
will apply the weights indicated by the index n to the transmit signal for each antenna.
Hence, the number of bits required in the feedback channel is the number of bits
required to represent N indices.
TRANSMIT DIVERSITY SCHEMES
In order to implement transmit diversity at user equipment, both Switched Antenna
TxDiv (SATD) and Beam Forming TxDiv (BFTD) are possible.
Figure 6 illustrates schematically a n embodiment of uplink Switched Antenna
Transmission Diversity. In SATD, user equipment can only transmit via one antenna at a
time and hence the possible values for Vi and ½ are 1 or 0. This avoids having a n
additional costly power amplifier (PA) in the UE. The weights are only logical
representation but from a n implementation point of view, a simple switch a s shown in
Figure 6. Therefore, for SATD, the index size N is 2, which can be indicated with only 1
bit.
In BFTD, a codebook or look up table comprising a set of predetermined relative
weights is used. In close loop TxDiv, a base station operates to select the best
predetermined weights based on radio channel estimations.
For user equipment having a two transmit branch, only the phase difference between
the weight vectors affect the gains. An example implementation is to fix the value of
one weight, i.e. (n) and vary the phase of the other weight, i.e. ½(n) in relation to
Wi(n).
Figure 7 illustrates schematically a set of transmission antenna weights according to one
embodiment. The codebook shown in Figure 7 is a /V=1 6 codebook in which the first
weight Wi(n), is fixed for n= 1to 16 whilst the phase for the second weight varies.
A non-codebook BFTD can be implemented for open loop TxDiv in which the first
weight is fixed and the phase of the second weight, changes by ±. the instruction to
change up or down by a n increment of phase being issued by a base station to user
equipment.
Figure 8 illustrates schematically a set of transmission antenna weights according to one
embodiment, for a 2 branch uplink TxDiv capable user equipment. In the embodiment
illustrated, a first weight is denoted as Vi and is fixed. The phase between Vi and the
second weight ½ is , where can increase or decrease by ^changing ½ to the
dashed vectors shown in Figure 8. In some, open loop, solutions, the UE decides
whether to increase or decrease the phase by based on the TPC. For a close loop
solution, a base signals to user equipment to increase or decrease the phase by 
based on uplink radio channel estimation.
COMPROMISE WEIGHT
In SATD and BFTD, each cell (or NB) in the Active Set of user equipment is expected to
transmit a transmission weight feedback request or indication to the user equipment,
that request indicating the preferred transmit antenna or the preferred weight
according to that base station. Described aspects allow user equipment to derive a
"compromise" weight that benefits a s many cells as possible based on the feedbacks
from all the cells within the user equipment Active Set.
In SATD, a cell or base station needs only indicate the preferred antenna and a simple
"compromise" scheme can be used, whereby the user equipment chooses the
antenna that the majority of cells (or NB) prefers. This will cause the UE to select the
antenna that benefits the most cells.
In BFTD, user equipment using a weight other than a "preferred" base station requested
weight reduces the gain that can be achieved by the network as a whole. In some
cases, the UE using the wrong weight can cause a loss compared to not using an uplink
transmission diversity scheme.
Most generally, let the loss in using the wrong weight LWEIGHT in dB be:
WEIGHT x
Equation 2
Where:
Go is the gain (dB) in using the preferred weight
Gx (dB) is the gain in using a weight other than the preferred weight.
Assuming the indexing correlates to resulting phase shifts, it is also possible to generally
define a n index distance DINDEX as:
INDEX = n x ~ n
Equation 3
Where:
nx is the weight indicated by index n used by the UE
no is the NB preferred weight indicated by index n
Figure 9 and Figure 10 are plots of average loss .WEIGHT VS. the index distance QNDEX for
Pedestrian (A&B) and Vehicular (A&B) radio channels respectively based on
simulations.
In the plots of Figure 8 and Figure 9 the simulations use the codebook shown
schematically in Figure 7 having a codebook size /=1 6 . The plots show that the loss
increases as the index distance increases and the loss is independent of the direction of
change of the index. The index distance is proportional to the difference in the weight
vectors Aw, therefore the loss is dependent upon the difference between the weight
vector used by the UE and the NB preferred weight vector (i.e. Aw).
The .WEIGHT standard deviation (indicated by vertical lines in Figure 9 and Figure 10 ) also
increases as the index distance NDEX, increases (in either direction). This means the
uncertainty of the loss increases with index distance ONDEX.
To determine a suitable compromise weighting, the compromise weights used should
minimise the differences in weight vectors Aw. In one embodiment, minimization of the
differences in weight vectors can b e achieved using a MMSE (Minimum Mean Squared
Error). Other distance minimization techniques for Aw may also b e used.
Figure 1 illustrates schematically a difference in weighting vectors. Figure 11 shows Aw,
for a 2 branch TxDiv system, where the 1 weight is constant. Hence, here Aw
represents the difference in distance between two 2nd weight vectors, W is the weight
vectors selected b y cell and Wc b e the "compromise" weight vectors used by the UE.
The MSE (Mean Squared Error) EMSE is given by:
= ( - w )
Equation 4
The MMSE is found b y minimizing SE with respect to Wc and is given by:
- - W = 0
k=Equation 6
In one embodiment, the compromise weights may b e found b y averaging all the
weights received from cells within the Active Set. Calculation o f Wc may result in a
weight vector that is not within the codebook. If the user equipment is restricted to use
weights defined in the codebook then, Wc needs to b e quantized to the nearest
weight vector in the codebook. In some embodiments, it is possible for user equipment
to use a non-quantised Wc.
In Figure 11, is the phase for vector W t, is the phase for vector Wc and is the
difference in phase between W and Wc. It is noted that is proportional to Aw and
minimizing minimizes Aw. Since is also linearly proportional to the index difference
ONDEX, a n alternate solution is to find the weight o r index n that minimizes DINDEX. An
example o f this is to take the average o f the index indicated b y all the cells in the
user equipment Active Set. However, using this method, the "wrap around" of the
phase needs to be taken account.
Figure 12 illustrates the wrap around of codebook phase differences. The weight
vectors are usually vectors within a circle (or sphere for 3D vectors). The phase , in
radians, ranges from 0 to 2viewed in a clockwise direction from the l-phase (real axis)
and it ranges from 0 to -2viewed in a counter clockwise direction. In Figure 12, in a n
embodiment having a codebook size /=16, the phase for index n=2 is whilst that of
index or
- ) would give the same gain as both points to the same vector (n=l 6). However, the
index number n does not share the same property, for example, in a codebook size
/v=16, the number 6 is not the same as the number -2. If a compromise index nc is
found using the average of , then this wrap around needs to be taken into account.
In a non-codebook based BFTD, each cell (or base station) indicates whether it prefers
to increase or decrease the phase by . Since the loss .WEIGHT is minimized if the
difference between the preferred phase and compromise phase Q.e. UE selected
phase) is minimized, a similar MMSE approach can be used to find a compromise
phase change Sc. which is the average & That is:
= ¾
Equation 7
Where:
is NB I s preferred phase change and it can take the value + or -.
If the user equipment can only increase the phase by £s†ep, then in one
embodiment, the implemented compromise weight calculation method is similar to
that in SATD where the user equipment chooses + -^depending on the majority.
In some embodiments, it is possible for user equipment to change its phase by a
fraction (or non-integer multiple) of . In this case, the user equipment may apply the
value fc calculated in Equation 7 directly to .
The compromise weight methods described above are also applicable for TxDiv with
more than 2 branches. The use of majority vote for SATD can be easily expanded for
higher number of antennas.
For BFTD, the MSE (e.g. average) of the index in Equation 6 is directly applicable for
higher branches of TxDiv. For higher TxDiv branches in non-codebook based BFTD,
each weight corresponding to a n antenna would have its own compromise weight.
That is:
Equation 8
Where:
fc ) the compromise weight for antenna j
&(/) is the preferred increment/decrement of the weight vector for antenna yfrom cell
(or NB) k.
Here it is assumed that each base station will send the preferred phase
increment/decrement for each weight vector corresponding to each antenna.
WEIGHTED AVERAGE
In some embodiments, additional information can be factored when deriving the
compromise weight.
Figure 13 illustrates schematically a n active set of base stations for user equipment. UE1
has 4 cells (base stations) in its Active Set: NB1, NB2, NB3 and NB4. Due to the locations
of the base stations, CPICH (Common Pilot Channel) signal strength received at the UE
from each base station will be different in the downlink.
Furthermore, transmission from UE1 to each base station will experience different path
losses. Since the influence of UE1 on each cell in its Active Set is different, changes to
the Beam Forming Transmission Diversity weightings will have a different impact on
each base station in the active set. Hence, higher priority should be given to the
preferred weight of that base station that will experience the highest impact if user
equipment changes its transmission weightings. A weighted average can therefore be
used, in some embodiments, when deriving the compromise weight. Accordingly, a
weight factor * can be included for each cell (base station) k in Equation 6 and
Equation 7 as follows:
Equation 9
1 K
*=1
Equation 0
The factor a¾ in Equation 9 and Equation 10, is dependent upon how much influence
user equipment has on a NB. An indicator of influence may, in some embodiments, be
the CPICH signal quality received by user equipment (e.g. power or Signal to Noise &
Interference Ratio, SNIR) from each base station. For example in Figure 13, the UE
receives highest CPICH power from NB3, followed by NBl , NB4 and lastly NB2 and thus
an > A .
Factor is also applicable to SATD embodiments. In such embodiments each base
station's "vote" is multiplied by factor a*. Let function F^SATD 0) be:
l if NB k selects antenna = j
lO if NB k selects antenna j
Equation 11
The compromise transmit branch (i.e. antenna) jc is therefore the value / such that the
following equation is met:
Equation 12
where is the total number of transmit branches.
In one embodiment, the weightings may be determined to give a higher value of
weighting factor a to the serving cell. Such an approach is in line with the objective
where a higher weight factor is given to NB that is influenced more by the UE. Since the
serving cell receives essential information from the UE (e.g. HS-DPCCH that is only
transmitted to the serving cell), a degradation in the receive signal due to choosing the
wrong weight (or antenna) would results in higher impact to the system performance
compared to that of a non-serving cell. In the extreme case, =0 if k is not the serving
cell and 1 otherwise, which gives absolute priority to the serving-cell. The level of priority
of the serving cell over the non-serving can be signalled to the UE by the network.
Other factors which influence the reception of the NB can also be used and the
weighting need not be restricted to being calculated based upon received signal
quality or whether a base station is a serving or non-serving cell.
In other words, embodiments allow a compromise transmission weightings to be used
by user equipment utilizing transmission diversity, those compromise transmission
weightings being derived in a manner which minimises differences between the
compromise weightings and the preferred weights indicated by all cells within the
Active Set of the user equipment.
In one embodiment, the compromise weight is found by using a MMSE technique
where the differences in weight vectors is minimized. In one embodiment, the
compromise weight is the average of all the preferred weights of the cells within the
Active Set. In one embodiment the compromise weight is the MMSE or the average of
a representation of the weights of the cells within the Active Set. This representation
can be the phase or the index to the weight. In one embodiment the compromise
weight is the weight that is the majority weight preferred by all cells in the Active Set.
In one embodiment a factor that is proportional to the amount of influence a UE has
over a NB (and vice versa) is applied to the compromise weight calculations.
Example 1
The power of a signal received from a base station at user equipment is dependent
upon the path loss experienced between the base station and user equipment. The
influence of a user equipment of the operation of a base station is dependent upon the
strength of user equipment transmissions reaching the base station. Consequently, if
NB1 receives a signal from user equipment having twice as much power as NB2, then
the UE has twice as much influence over NB1 than it does to NB2. Accordingly, in this
example the weight factor , described above, is proportional to the path loss.
Although path losses in the downlink and uplink can be different due to different carrier
frequencies used in the downlink and uplink, the downlink path loss estimated from the
CPICH power is a good indication of the uplink path loss. The path loss can therefore
b e considered proportional †o the UE received CPICH power from NB k (PJCCPICH (in
nW)).
The weight factor can thus b e calculated according to:
Equation 13
Where: Sk is 2 if is the serving cell (otherwise it is 1).
Figure 1 illustrates schematically a n arrangement o f five base stations namely NB1 ,
NB2, NB3, NB4 and NB5 where NB3 is the serving cell, in a n active set o f example user
equipment UE. In this example, the serving cell is determined to have twice the
influence in comparison with a non-serving cell. The received CPICH powers a t user
equipment UE from each base station and the corresponding weight factors a* are
summarised in Table 1.
Table : Weight factor a* calculations for Example 1
NB P .CPICH Additional Factors
dBm nW s Factored Power (nW)
1 -90 0.001 0 1 0.001 0 0.1 968
2 -96 0.0003 1 0.0003 0.0494
3 -88 0.001 6 2 0.0032 0.6239
4 -93 0.0005 1 0.0005 0.0986
5 -98 0.0002 1 0.0002 0.031 2
Total 0.0051 1
Assuming that UE uses a two transmit branch SATD scheme. The NBs' preferred transmit
branch (i.e. antenna) are a s follows:
Table 2: Calculate compromise transmit branch
3 1 0 0.6239 0.624 0.000
4 0 1 0.0986 0.000 0.099
5 0 1 0.0312 0.000 0.031
Total 0.673 0.327
Thus the compromise transmit branch based on the above calculation is .
Example 2
The same example distribution of base stations a s that shown in Figure 14 is used for a
second example. Again a UE Active Set contains 5 cells: NB1, NB2, NB3, NB4 and NB5
and NB3 is the serving cell. The UE employs a codebook based BFTD with index size
N=8. The weight vectors are summarised in Table 3 .
Table 3 : Codebook weights N=
Index n Weight, W Weight, W
l-Phase Q-Phase l-Phase Q-Phase
1 0.7071 0 0.7071 0.0000
2 0.7071 0 0.5000 -0.5000
3 0.7071 0 0.0000 -0.7071
4 0.7071 0 -0.5000 -0.5000
5 0.7071 0 -0.7071 0.0000
6 0.7071 0 -0.5000 0.5000
7 0.7071 0 0.0000 0.7071
8 0.7071 0 0.5000 0.5000
The weight factor is proportional to the CPICH receive power a s in Equation 3 but in
this example, no additional importance is given to the serving cell, i.e., S = l for all k. The
receive CPICH power and the corresponding weight factorsa* are summarised in Table
Table 4 : Weight factor calculations for Example 2
NB CP CH Additional Factors a
DBm I NW Factored Power (nW)
1 -90 0.0010 1 0.0010 0.1968
2 -96 0.0003 1 0.0003 0.0494
3 -88 0.0016 1 0.0016 0.3119
4 -93 0.0005 1 0.0005 0.0986
5 -98 0.0002 0.0002 0.0312
The NB preferred index its corresponding weight V and compromise weight Wc are
calculated in Table 5 using Equation 9.
Table 5: Calculate compromise weight
NB Peferred Index, Weight (W2) Weighted Vector {akW2)
nk
l-Phase Q-Phase □ l-Phase Q-Phase
1 5 -0.7071 0 0.2861 0.202275645 0
2 4 -0.5 -0.5 0.071 9 0.035927632 -0.03593
3 6 -0.5 0.5 0.4534 0.226688034 0.226688
4 7 0 0.7071 0.1434 0 0.1 0 1378
5 8 0.5 0.5 0.0453 0.022668803 0.022669
Average 0.088444501 0.062961
The weight vectors of each NBW and the corresponding calculated compromise
weight Wc are shown in Figure 15. Here the UE can only select a weight that is within
the codebook and quantising the compromise weight gives Wc = / 6 (6 ) , which is the
same weight used by NB3.
Example 3
The same example distribution of base stations as that shown in Figure 1 is used for a
third example. The UE Active Set contains 5 cells: NBl , NB2, NB3, NB4 and NB5 and NB3
is the serving cell. The UE employs a non-codebook based BFTD where =/4 (or 45°).
The weight factor is proportional to the CPICH receive power as in Equation 13 but here
there is no importance given to the serving cell, i.e., S*=l for all k. The receive CPICH
power and the corresponding weight factor a is same a s that in Table 4.
The NB preferred phase change and the compromise phase change, 5c are
calculated in Table 6 using Equation 10. Here the NB feedback a + 1 or - 1 to indicate a
change of +£and - respectively.
Table 6: Calculate compromise phase change
NB NB Preferred Phase Change Weighted Phase
Indicator, Change
± 1 Phase Change, ±(°)
1 + 1 45 0.2861 12.8727431 9
2 - 1 -45 0.071 9 -3.233486896
3 + 1 45 0.4534 20.40192305
4 - 1 -45 0.1434 -6.451 65455
5 - 1 -45 0.0453 -2.040192305
Average 4.3099
If the UE can apply a weight vector that can be a fraction of , then the phase
changed by +4.3099°. If it cannot apply a fraction of 8. then the rounded phase
change is 0° or no phase change.
A person of skill in the art would readily recognize that steps of various above-described
methods can be performed by programmed computers. Herein, some embodiments
are also intended to cover program storage devices, e.g., digital data storage media,
which are machine or computer readable and encode machine-executable or
computer-executable programs of instructions, wherein said instructions perform some
or all of the steps of said above-described methods. The program storage devices may
be, e.g., digital memories, magnetic storage media such as a magnetic disks and
magnetic tapes, hard drives, or optically readable digital data storage media. The
embodiments are also intended to cover computers programmed to perform said
steps of the above-described methods.
The functions of the various elements shown in the Figures, including any functional
blocks labelled as "processors" or "logic", may be provided through the use of
dedicated hardware as well as hardware capable of executing software in association
with appropriate software. When provided by a processor, the functions may be
provided by a single dedicated processor, by a single shared processor, or by a
plurality of individual processors, some of which may be shared. Moreover, explicit use
of the term "processor" or "controller" or "logic" should not be construed to refer
exclusively to hardware capable of executing software, and may implicitly include,
without limitation, digital signal processor (DSP) hardware, network processor,
application specific integrated circuit (ASIC), field programmable gate array (FPGA),
read only memory (ROM) for storing software, random access memory (RAM), and non
volatile storage. Other hardware, conventional and/or custom, may also be included.
Similarly, any switches shown in the Figures are conceptual only. Their function may be
carried out through the operation of program logic, through dedicated logic, through
the interaction of program control and dedicated logic, or even manually, the
particular technique being selectable by the implementer as more specifically
understood from the context.
It should be appreciated by those skilled in the art that any block diagrams herein
represent conceptual views of illustrative circuitry embodying the principles of the
invention. Similarly, it will be appreciated that any flow charts, flow diagrams, state
transition diagrams, pseudo code, and the like represent various processes which may
be substantially represented in computer readable medium and so executed by a
computer or processor, whether or not such computer or processor is explicitly shown.
The description and drawings merely illustrate the principles of the invention. It will thus
be appreciated that those skilled in the art will be able to devise various arrangements
that, although not explicitly described or shown herein, embody the principles of the
invention and are included within its spirit and scope. Furthermore, all examples recited
herein are principally intended expressly to be only for pedagogical purposes to aid the
reader in understanding the principles of the invention and the concepts contributed
by the inventor(s) to furthering the art, and are to be construed as being without
limitation to such specifically recited examples and conditions. Moreover, all statements
herein reciting principles, aspects, and embodiments of the invention, as well as
specific examples thereof, are intended to encompass equivalents thereof.
CLAIMS
1. A method of setting antenna transmission weightings of user equipment in a
wireless telecommunication network, said user equipment being operable to transmit
on at least two antennas and operable to communicate simultaneously with at least
two base stations from an active set associated with said user equipment, said method
comprising the steps of:
receiving an indication of preferred antenna transmission weightings from said
at least two base stations in said active set,
calculating an indication of compromise antenna transmission weightings which
minimise a difference between said indication of compromise antenna transmission
weightings and said indication of preferred antenna transmission weightings; and
setting said antenna transmission weightings in accordance with said indication
of compromise antenna transmission weightings.
2 . A method according to claim 1, wherein said step of calculating comprises the
step of: minimising the mean squared distance between the indication of compromise
antenna transmission weightings and the indication of preferred antenna transmission
weightings received from the base stations in said active set.
3 . A method according to claim 1, wherein said step of calculating comprises the
step of: averaging the indication of preferred antenna transmission weightings received
from the base stations in said active set.
4 . A method according to claim 1, wherein said step of calculating comprises the
step of: calculating the mode preferred antenna transmission weightings received from
said base stations in said active set.
5. A method according to any preceding claim, wherein said step of calculating
further comprises the steps of: assigning a weighting to said indication of preferred
antenna transmission weightings from each base station in said active set, said
weighting being determined based upon an indication of the relative importance of
each base station in said active set to operation of said user equipment within said
network.
6. A method according to claim 5, wherein said weighting is assigned based upon
an indication of received pilot signal strength from said base stations in said active set
at said user equipment.
7. A method according to claim 5 or claim 6, wherein said weighting takes
account of which of said base stations in said active set is responsible for providing said
user equipment serving cell.
8. A method according to any preceding claim, wherein said indication of said
preferred antenna transmission weightings comprises an indication of relative phase
shift between transmissions of a signal on said at least two antenna.
9. A method according to any preceding claim, wherein said indication of said
preferred antenna transmission weightings comprises an indication of a n amplitude shift
between transmissions of a signal on said at least two antenna.
10. A method according to any preceding claim, wherein said indication of said
preferred antenna transmission weightings comprises an indication of antenna
selection.
11. A computer program product operable, when executed on a computer, to
perform the method of any one of claims 1to 10.
12. User equipment operable to set antenna transmission weightings in a wireless
telecommunication network, said user equipment being operable to transmit using at
least two antenna and operable to communicate simultaneously with at least two base
stations in an active set associated with said user equipment, said user equipment
comprising:
reception logic operable to receive an indication of preferred antenna
transmission weightings from said at least two base stations in said active set,
calculation logic operable to calculate an indication of compromise antenna
transmission weightings which minimise a difference between said indication of
compromise antenna transmission weightings and said indication of preferred antenna
transmission weightings; and
implementation logic operable to set said antenna transmission weightings in
accordance with said calculated indication of compromise antenna transmission
weightings.