DESCRIPTION
METHOD OF CONFIGURING CROSS-CARRIER CFI
PRIORITY CLAIM
Priority is claimed on Japanese Patent Application No. 2010-87494, filed April 6,
2010, the content of which is incorporated herein by reference.
TECHNICAL FIELD
The present invention relates to a Mobile System.
BACKGROUND ART
A major feature to be introduced for Long Term Evolution (LTE) Rel-10
(LTE-Advanced) is carrier aggregation where two or more component carriers (CCs) are
aggregated in order to support wider transmission bandwidths e.g. up to lOOMHz and for
spectrum aggregation (see Reference 7, below). Once a User Equipment (UE) is
configured with carrier aggregation, the UE is capable of simultaneously receive or
transmit on all the CCs that are aggregated. Thus, the UE may be scheduled over
multiple CCs simultaneously. Further details can be found in Section 5 of Reference 7.
Carrier aggregation has also been recognized as a useful tool for
managing/coordinating intercell interference for heterogeneous network deployments.
For the definition of the heterogeneous deployment as well as the detailed description of
the inter-cell interference issue for the heterogeneous deployment, please see Section 9A
of Reference 7, below.
With reference to FIG 1 which shows one example of carrier aggregation
applies to heterogeneous deployments, a macro UE 11 uses a control signaling on f
and/or f2, and data on f and/or f2. A macro UE 12 uses a control signaling on f and
data on and/or f . A pico UE 13 uses a control signaling on f2, and data on f i and/or f2.
Carrier aggregation, together with cross-carrier scheduling using Carrier Indicator Field
(CIF) (see Section 5.2 of Reference 7) "provides means for coordination of the control
channel interference between cell layers" by "partitioning CCs in each cell layer into two
sets, one set used for data and control and one set used mainly for data and possibly
control signaling with reduced transmission power" (see Section 9A.2.1 of Reference 7
for more details).
An open issue in 3rd Generation Partnership Project (3GPP) for LTE Rel-10 at
the time of writing is the design of Control Format Indicator (CFI) signaling when
cross-carrier scheduling is configured. In Rl- 100835 (see Reference 8, below), it has
been agreed that:
"In case of cross carrier scheduling, a standardized solution will be supported
to provide CFI to the UE for the carriers on which PDSCH is assigned. Details
are FFS."
The reason for the need for a standardized solution to support providing CFI
values to define starting positions of Physical Downlink Shared Channel (PDSCH) on a
cross-CC scheduled CC is as follows:
1. The control region of a subframe for the cross-CC scheduled CC may not be
reliable for Physical Control Format Indicator Channel (PCFICH), Physical
Downlink Control Channel (PDCCH) and Physical Hybrid ARQ Indicator
Channel (PHICH) reception by the UE. Indeed, this is the motivation for
cross-carrier scheduling with C F; and
2. If the span of the control region or the start of data region on the cross-CC
scheduled CC cannot be reliably detected from PCFICH, other means of
providing the CFI information is needed so that the UE can reliably receive its
assigned PDSCH on the cross-CC scheduled CC.
A summary of existing proposals for a standardized solution by other companies
in 3GPP Radio Access Network (RAN)1 is given below:
Semi-static Radio Resource Control (RRC) signaling: Semi-static signaling of
CFI was already proposed but not agreed by a number of companies in Rel-8 for single
carrier operation. This solution would exclude dynamic changes of the PDSCH starting
position on a cross-CC scheduled but has a relatively low signaling cost (see Reference 4
and Reference 5, below) as shown in FIG 2;
Downlink Control Indicator (DCI) Signaling: This solution would allow
dynamic changes of the PDSCH starting position on a Cross-CC scheduled CC. The
cost is additional 2 or 1 or 0 bits in the DCI with CIF depending if and which favor of
CIF + CFI joint coding is selected (see Reference 2 and Reference 3, below) as shown in
FIG 3; and
The UE may assume the same CFI on the PDSCH CC as the one on the PDCCH
CC: This solution does not incur any signaling overhead, but in case the CFIs on PDCCH
and PDSCH CCs are not the same there are some unused or punctured PDSCH Resource
Elements (Res) (see Reference 9, below).
In LTE, there are two main subframe types, namely the normal (or non-
Multicast/Broadcast over Single Frequency Network (MBSFN)) subframe and the
MBSFN subframe. The MBSFN subframe can be used to carry Physical Multicast
Channel (PMCH) or Rel-10 Physical Downlink Shared Channel (PDSCH). Analyses
from companies so far have typically assumed the same subframe type, i.e. the normal
subframe, for all the carriers. However, configuring the same subframe type for all
carriers may not be sensible for the following reasons:
1. For MBSFN subframes carrying PMCH, it may not be sensible to assume all
carriers transmitting PMCH simultaneously;
2. Common MBSFN subframe configuration for all carriers means that no legacy
UEs can be scheduled on any carrier for a particular time instance. This
imposes severe scheduling restriction; and
3. It may be useful to have certain carrier(s) exclusive to Rel- 10 UEs by
disallowing Rel-8/9 UEs from camping on certain carriers. For example, in the
heterogeneous deployment, if a carrier is experiencing from high interference, it
may be beneficial to forbid Rel8/9 UE camping. Since there is no need to serve
Rel-8/9 UEs on the carrier, more subframes can be configured to be MBSFN
subframes for unicast transmission for optimized Rel- 10 performance.
Focusing on heterogeneous network deployment with multiple carriers as shown
in FIG 4, analyses e.g. in Reference 4 and Reference 5, below, show that dynamic
signaling of CFI values is not required. The advantage of dynamic CFI signaling is to
avoid throughput degradation caused by the inflexibility of semi-static or fixed CFI
configuration. However, the effectiveness of dynamic CFI signaling is diminished by
the fact that dynamic coordination of scheduling information on subframe basis among
the enhanced NodeBs (eNBs) is not possible. For example, referring to FIG. 4 and FIG. 5,
the start of data region by the macro eNB in carrier # 1 can vary from subframe to
subframe according to the load of the control channel, but such scheduling information
cannot be conveyed to the pico/femto cells on a subframe basis. Any mismatch of the
starting Orthogonal Frequency Division Multiplexing (OFDM) symbol for data region
between the macro cell and the pico/femto cells causes either waste of bandwidth or
inter-cell interference as shown in FIG 5.
Although sharing of the dynamic scheduling information among the eNBs is not
possible, the neighboring eNBs' MBSFN subframe configuration information, which
only changes in a semi-static manner, can be shared via X2 interface; this is already
possible in Rel-9 (see Reference 6). The implication is as follows:
1. If the subframe type of carrier # 1 of the pico/femto eNB is a normal subframe
but the subframe type of carrier # 1 of the macro eNB is an MBSFN subframe as
shown in Figure 6(a), for carrier # 1 of the pico/femto eNB, the start of the data
region for should adapt to the MBSFN subframe configuration for carrier # 1 of
the macro eNB.
a. The change of subframe type is dynamic in time; therefore, applying a fixed
CFI value for a long period of time as shown in FIG 2 is inefficient.
i. For Frequency Division Duplex (FDD), subframe 0, 4, 5 and 9 must
be normal subframes whereas subframe 1, 2, 3, 6, 7 and 8 can be
MBSFN subframes.
ii. For Time Division Duplex (TDD), DL subframe 0, 1, 2, 5, and 6 are
non-MBSFN subframes whereas DL subframe 3, 4, 7, 8 and 9 can be
MBSFN subframes.
b. Typically, CFI values change between 2 and 3 (all bandwidths other than
1.4MHz), or between 1 and 3 (for 1.4MHz carrier).
i . Let's assume 6 subframes are configured as MBSFN subframes in
macro carrier # 1 for a FDD system. Approximately 10.71%
(12/112) resource is lost on average if CFI=3 is assumed always for
1.4MHz carrier whereas approximately 5.17% (6/116) is lost on
average for all other bandwidths; and
2. If the subframe type of carrier # 1 of the pico/femto eNB is an MBSFN subframe
but the subframe type of carrier # 1 of the macro cell is a normal subframe as shown in
FIG 6B, for carrier # 1 of the pico/femto eNB, the start of the data region should be 4th
OFDM symbol (or 5th OFDM symbol for 1.4MHz) if a conservative approach to
inter-cell interference management is adopted, assuming high load for the macro cell.
a. This implies that the start of the data region for MBSFN subframe in Rel-10
may need to be different from the current Rel-8/9 assumption.
b. The CFI for pico/femto carrier # 1 is UE-specific in general. For example,
in case Rel-8/9 UEs are still supported in the pico/femto carrier #1, or if there are other
Rel-10 UEs not configured for cross-carrier scheduling e.g. if they are close to the
pico/femto eNB, the CFI assumed by different UEs can be different.
The MBSFN subframe configuration of a cell, specified in System Information
Block (SIB)2, can be for one frame (6 bits) or for four consecutive frames (24 bits).
Therefore, the standardized solution of cross-carrier CFI signaling should be able to
specify separate CFI values of each subframe of a frame or of four consecutive frames.
The cross-carrier CFI value should also override any predetermined Rel-8/9 CFI value or
the value signal in PCFICH in the target carrier in case the target carrier is an MBSFN
subframe. This is to address the problem as explained before with reference to FIG 6B.
[List of related art]
[Reference 1] RAN1#60 Chairman's note;
[Reference 2] Rl-101206 "PCIFCH for Cross-carrier Assignment", NTT
DOCOMO;
[Reference 3] Rl-101248 "PCFICH in cross carrier operation", Panasonic;
[Reference 4] Rl -101111 "PCFICH in Carrier Aggregation", Motorola;
[Reference 5] Rl- 100840 "On PCFICH for carrier aggregation", Ericsson,
ST-Ericsson;
[Reference 6] R3-101161 "TS36.423 CR0341R2 Addition of MBSFN
information on X2 interface", CATT,ZTE, CMCC;
[Reference 7] 3GPP TR 36.814 V2.0.0 (2010-3);
[Reference 8] Rl- 100835 "Way forward on PCFICH erroneous detection for
Cross-Carrier Scheduling";
[Reference 9] Rl-10141 1 "PCFICH Issues with Cross-Component Carrier
Scheduling", Nokia Siemens Networks, Nokia.
DISCLOSURE OF INVENTION
The present invention provides a semi-static RRC signaling solution that is able
to signal separate CFI value for each individual subframe over a frame or over multiple
frames, which the UE assumes hold until the next RRC reconfiguration event. This is
in contrast with the existing semi-static signaling proposal whereby only one CFI value,
selected from {1, 2, 3} or a subset thereof, is signaled by RRC (re)configuration.
It is reasonable to assume that the MBSFN subframe configuration can/should
be done differently for each carrier. Furthermore, the MBSFN subframe configuration of
a carrier can also be different for different neighboring eNBs (already possible in LTE
Rel-8/9). The present invention addresses the issues of cross-carrier CFI signaling
design taking into account the possibility of MBSFN subframes (for PMCH or for
unicast transmission) being configured differently for each carrier and for each
neighboring eNB.
Both dynamic DCI signaling approach and semi-static RRC signaling approach
can offer feasible solutions. However, due to the semi-static nature of the MBSFN
subframe configuration, semi-static RRC signaling approach is sufficient. The present
invention provides the (dedicated) RRC signaling to signal separate CFI value to be
assumed by the UE for each individual subframe of one frame, or of four consecutive
frames.
The present invention provides a method of configuring the cross-carrier CFI
such that separate CFI value can be specified for each subframe within an x number of
consecutive frames. The same set of CFI values is applied by the UE for every other x
consecutive frames until it is reconfigured by the eNB.
According to the present invention, the following advantages may be achieved:
1. The invention enables separate CFI value for each subframe of a frame
or multiple frames to be signaled to a UE;
2. The invention can improve the bandwidth utilization of a heterogeneous
network with multiple carriers;
3. The invention can provide improved inter-cell interference coordination
for a heterogeneous network with multiple carriers; and
4. The invention only incurs small cost in terms of signaling to the UE.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG 1 shows a one example of carrier aggregation applies to heterogeneous
deployments.
FIG 2 shows a Semi-static RRC signaling - fixed CFI value between two RRC
(re)configuration.
FIG. 3 shows a DCI Signaling (dynamic signaling, changing CFI value on
subframe basis).
FIG 4 shows a heterogeneous network with two carriers.
FIG. 5 shows a mismatched CFI between the macro carrier and pico/femto
carrier.
FIG 6A shows a MBSFN subframe in macro cell and a normal subframe in
pico/femto cell
FIG 6B shows a normal subframe in macro cell and a MBSFN subframe in
pico/femto cell
FIG 7 shows a CFI value for each subframe which can be different according to
the RRC signaling in an embodiment of the present invention.
EMBODIMENTS FOR CARRYING OUT THE INVENTION
In the event that carrier aggregation and cross-carrier scheduling is configured
by the network for the User Equipment (HE), an embodiment of the invention enables
the information about the start of the data region in a subframe of a carrier (known as the
Control Format Indicator (CFI)) to be delivered by the network to the UE on another
carrier, without requiring the UE to receive and decode the Physical Control Format
Indicator CHannel (PCFICH) on the first carrier, which can be either non-existent or
unreliable for some network deployment scenarios (e.g. heterogeneous deployment).
The invention achieves this by providing the means to signal separate CFI value for each
subframe of a frame or multiple frames to the UE.
The embodiment of the invention can improve the bandwidth utilization of a
heterogeneous deployment with multiple carriers. This can be achieved as follows.
Assume the network has two cell layers with overlapping coverage with the same carrier
frequency (hence interfering with each other) and each cell belongs to different eNB
class as shown in FIG 4. If some subframes of the higher cell layer are configured with
MBSFN subframe, the embodiment of the invention enables the CFI of the lower cell
layer to adapt according to the MBSFN subframe configuration of the higher cell layer,
so that OFDM resources under utilization is minimized.
The embodiment of the invention can provide improved inter-cell interference
coordination capability for a heterogeneous deployment with multiple carriers. This can
be achieved as follows. Assume the network has two cell layers with overlapping
coverage with the same carrier frequency (hence interfering with each other) and each
cell belongs to different eNB class as shown in FIG 4. If some subframes of the lower
cell layer are configured with MBSFN subframe, the embodiment of the invention
enables the CFI value for the MBSFN subframe in the lower cell layer to be overridden,
in order to avoid the interference to the data region of the lower cell layer from the
control region of the higher cell layer and also to avoid the interference to the control
region of the higher cell layer from the data of the lower cell layer.
The embodiment of the invention only incurs a small cost in terms of signaling
to the UE by providing the CFI information through semi-static RRC signaling.
FIG 7 shows that the CFI value for each subframe can be different according to
the RRC signaling. In FIG 7, the CFI pattern is repeated until it is changed by RRC
reconfiguration.
The embodiment provides a (dedicated) RRC signaling that can signal separate
CFI value to be assumed by the UE for each individual subframe of one frame, or of four
consecutive frames. The UE assumes that the same configuration is applied until the
next RRC reconfiguration event as shown in FIG. 7. The aim of that is to achieve
improvement in bandwidth utilization by adapting the CFI of the cross-scheduled carrier
according to the MBSFN subframe configuration of the neighboring cell as shown in FIG
6A. Another aim is to achieve better inter-cell interference coordination by overriding
whenever necessary the CFI for the MBSFN subframe in the cross-scheduled carrier as
shown in FIG. 6B.
There are many ways to design the RRC signaling. Here, we provide several
examples.
Example 1:
As there are 3 possible CFI values, 2 bits are needed to address a CFI value.
There are two subframe groups, called subframe group A and subframe group B. The
RRC signaling to address the CFI values for each subframe group over one frame is 2 + 1
+ 6 = 9 bits, i.e. 2 bits for the CFI value for subframe group A (CFI=1, 2, 3), 1 bit for the
CFI value for subframe group B (CFI=1, 2) and 6 bits to indicate which of the 6
subframes belongs to subframe group B, corresponding to subframes that can potentially
be MBSFN subframes.
6 bits to indicate the subframe that belongs to subframe group B is not enough if
the MBSFN subframe configuration of the interfering cell is done for 4 consecutive
frames. In this case, the number of bits can be 2 + 1 + 24 = 27 bits.
Example 2:
If signaling overhead reduction is desired, the CFI values can be limited as
follows: CFI=3 and CFI=2 for subframe group A and subframe group B, respectively,
since they are considered the typical values (for 1.4MHz carrier, CFI=3 and CFI=1 for
subframe group A and subframe group B, respectively). In this case, only 6 bits (or 24
bits) are needed to indicate which subframes belong to subframe group B.
Note 1: The MBSFN subframe configuration is actually optional. Accordingly,
if there is no MBSFN subframe configured by any eNBs, then semi-static configuration
of CFI to be 1,2 or 3 for the whole time period between two RRC (re)configuration is
sufficient. In this case, all subframes can be set to be subframe A in Example 1. In
Example 2, the design can be modified so that we have 2 + 6 (or 24) = 8 (28) bits, i.e. 2
bits for the CFI value for subframe group A so that all three CFI values can be indicated.
Note that the 6 (or 24) bits can be used to indicate all subframes as subframe group A.
Example 3:
RRC signaling bit width can be flexible according to conditions. If only a
single CFI value is enough (e.g. when there is no MBSFN subframe configured in the
macro cell), then the RRC signaling can be 2 bits (CFI=1, 2, 3). Otherwise, the RRC
signaling can be that given in Example 1 or Example 2.
Note 2: The RRC signaling is not required to be provided to the UE frequently.
This is because the MBSFN subframe configuration or the conditions experienced by the
UE in the cells (which can trigger the need for RRC reconfiguration) do not change
frequently.
We provide a high level description of a possible system operation. Referring
to FIG 4, the following system operation including the following [Step 1] to [Step 7] is
envisioned.
[Step 1]: heterogeneous deployment set up
Macro eNB:
1. The macro eNB has two carriers aggregated (carrier #0 and carrier #1).
Carrier #0 is transmitted with reduced power (small coverage).
Carrier # 1 is transmitted with maximum power (large coverage).
2. The macro eNB sets the MBSFN subframe configuration for carrier #0
and carrier # 1, which can be different.
Pico/femto eNB:
1. The pico/femto eNB has two carriers aggregated (carrier #0 and carrier
#1). Both carriers are transmitted with the same power (same coverage).
2. The pico/femto eNB sets the MBSFN subframe configuration for
carrier #0 and carrier # 1, which can be different.
[Step 2]: Information exchange between the macro eNB and pico/femto eNB
1. The MBSFN subframe configuration information of the eNBs for each
carrier is exchanged via X2 or SI interface.
2. The maximum CFI value to be assumed for the non-MBSFN subframes
for each carrier of each eNB is exchanged via X2 or SI interface.
3. Information exchange is initiated whenever the MBSFN subframe
configuration or the maximum CFI value of a carrier of an eNB changes.
[Step 3]: Pico/femto UE initial access and camping
1. Initial access is performed by the pico/femto UE.
2. The UE camps on the carrier #0 of the pico/femto cell. Carrier #0 is not
inter-cell interference limited, so the UE can reliably receive messages from the eNB in
the control region of carrier #0.
[Step 4]: Carrier aggregation set up and cross-carrier scheduling configuration for the UE
1. Carrier # 1 is configured for the pico/femto UE by the pico/femto eNB
via dedicated RRC signaling on carrier #0. This sets up carrier aggregation for the UE.
2. Cross-carrier scheduling is configured for the pico/femto UE by the
pico/femto eNB via dedicated RRC signaling on carrier #0. This prepares the UE for
detecting the PDCCH on carrier #0 of the pico/femto eNB that assigns PDSCH on carrier
# 1 of the pico/femto eNB.
3. Cross-carrier CFI is configured for the pico/femto UE by the pico/femto
eNB via dedicated RRC signaling on carrier #0. The RRC signaling can be Example 1,
Example 2, Example 3 or others.
[Step 5]: Carrier aggregation activation
1. Carrier # 1 is activated for the pico/femto UE by the pico/femto eNB via
dedicated signaling on carrier #0.
2. The UE starts to attempt to detect PDCCH with CIF on carrier #0 that
schedules PDSCH on carrier #1.
[Step 6]: Cross-carrier PDCCH transmission by the pico/femto eNB
1. The eNB transmits PDCCH with CIF on carrier #0 that assigns PDSCH
on carrier #1.
[Step 7]: Cross-carrier PDCCH reception and PDSCH reception by the pico/femto UE
1. The UE detects the PDCCH with CIF on carrier #0 and decodes the
corresponding PDSCH assignment information.
2. The UE determines the start of the PDSCH OFDM symbol on carrier#l
using the CFI information obtained from Step 4, and attempts to receive and decode
PDSCH accordingly.
As described above, according to the present invention, separate CFI value for
each subframe over one or multiple frames is provided in the manner illustrated in FIG 7.
In the event that carrier aggregation and cross-carrier scheduling is configured
by the network for the User Equipment (UE), the invention enables the information about
the start of the data region in a subframe of a carrier (known as the Control Format
Indicator (CFI)) to be delivered by the network to the UE on another carrier, without
requiring the UE to receive and decode the Physical Control Format Indicator CHannel
(PCFICH) on the first carrier, which can be non-existent or unreliable for some network
deployment scenarios (e.g. heterogeneous deployment).
Thus, the invention provides the following advantages:
1. The invention enables separate CFI value for each subframe of a frame
or multiple frames to be signaled to a UE.
2. The invention can improve the bandwidth utilization of a heterogeneous
network with multiple carriers.
3. The invention can provide improved inter-cell interference coordination
for a heterogeneous network with multiple carriers.
4. The invention only incurs small cost in terms of signaling to the UE.
CLAIMS
1. A method of configuring the cross-carrier Control Format Indicator (CFI) such
that separate CFI value can be specified for each subframe within an x number of
consecutive frames, comprising applying the same set of CFI values by the User
Equipment (UE) for every other x consecutive frames until it is reconfigured by the eNB.
2. The method according to claim 1, wherein x is 1 or 4, corresponding to
Multicast/Broadcast over Single Frequency Network (MBSFN) subframe configuration
of one or four frames.
3. The method according to claim 1, wherein the configuration method is
semi-static RRC signaling.
4. The method according to claim 3, wherein the RRC signaling is dedicated.
5. The method for operating a communication system by using CFI as configured
by the method according to claim 1, comprising:
a step of setting up Macro eNB and pico/femto eNB;
a step of exchanging information between the macro eNB and pico/femto eNB;
a step of initially accessing and camping pico/femto UE;
a step of setting up carrier aggregation and cross-carrier scheduling
configuration for the UE;
a step of activating carrier aggregation;
a step of transmitting cross-carrier Physical Downlink Control Channel
(PDCCH) by the pico/femto eNB; and
a step of receiving cross-carrier PDCCH and Physical Downlink Shared Channel
(PDSCH) by the pico/femto eNB.