Transcript IEEE C802.16m-09/0695

Data Throughput Analysis with A-MAP
Document Number: IEEE C80216m-09/0695
Date Submitted: 2009-3-8
Source:
Yi Hsuan, Hujun Yin Email: {[email protected], [email protected]}
Intel Corporation
Venue:
IEEE Session #60, Vancouver, Canada
Re: Reply comment support for DL control DG comments
Base Contribution:
N/A
Purpose:
To be discussed in TGm in support of two level MCS for assignment A-MAP
Notice:
This document does not represent the agreed views of the IEEE 802.16 Working Group or any of its subgroups. It represents only the views of the participants listed in
the “Source(s)” field above. It is offered as a basis for discussion. It is not binding on the contributor(s), who reserve(s) the right to add, amend or withdraw material
contained herein.
Release:
The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an
IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s
sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this
contribution may be made public by IEEE 802.16.
Patent Policy:
The contributor is familiar with the IEEE-SA Patent Policy and Procedures:
<http://standards.ieee.org/guides/bylaws/sect6-7.html#6> and <http://standards.ieee.org/guides/opman/sect6.html#6.3>.
Further information is located at <http://standards.ieee.org/board/pat/pat-material.html> and <http://standards.ieee.org/board/pat >.
1
Outline
1.
Data throughput and power boosting comparison using fixed MCS AMAP and
variable MCS AMAP.
2.
Data throughput comparison using reuse 1, reuse 3, or both reuse 1 and 3 for AMAP.
2
A-MAP Physical Channels and Allocation
LAMAP distributed LRUs
...
...
Distributed
Localized
Frequency Partition n
A-MAP Region
Non user-specific A-MAP
HARQ Feedback A-MAP
Power control A-MAP
Assignment A-MAP Using MCS1
Assignment A-MAP Using MCS2
Data channels
Nsym symbols
3
Link Simulation Parameters
1.
48 bits are coded and modulated with QPSK ½ TBCC.
2.
ITU-VehA, ITU-PedB in both noise limited and interference limited are considered.
3.
Pilot is 5 dB higher than non-boosted data tones.
4.
Burst data boosts of 0 dB, 2 dB, 4 dB, …, 20 dB are simulated.
5.
2Tx by 2Rx using SFBC
6.
MMSE channel estimation
7.
MMSE MIMO receiver
4
Link Level Simulation (Noise Limited)
5
Link Level Simulation (Interference Limited)
6
Data Power Boosting Gain
• Performance gain from data
tone power boosting is about
the same over different
channels.
• The gains in interference
limited scenarios (no more
than 8 dB) are less than
noise limited scenarios (no
more than 8.5 dB) at higher
boosting power.
• Approximation curve is
5 dB pilot boost over un-boosted data
9
Performance gain in dB at 1% PER
8
7
6
5
4
3
2
1
y  10  (1  e  x /12 )
VehA, noise limited
PedB, noise limited
VehA, interference limited
PedB, interference limited
Approximation used in simulation
2
4
6
8
10
12
Data boost in dB
14
16
18
20
7
DL User SINR Distribution
0
10
CDF
Reuse 3
Reuse 1
-1
10
-2
10
-5
0
5
SINR
10
15
Distribution of 866 m radius cell, EMD baseline path loss, reuse 3 has 3 dB
subcarrier boosting over reuse 1
8
Analysis Methodology
1.
Use the channel capacity equation to approximate the data throughput after resource for AMAP is
taken away. Data throughput is approximated as follows:
S D  BD  log 2 (1  PD  SINR0 )
where BD is the fraction of total bandwidth available for data and PD is the power density change (1
is no change) for data after control channel power boosting or de-boosting. SINR0 is a nominal
number to approximate the averaged data SINR value in the cell.
2.
Given a number of users in the sector, two allocation (DL and UL) AMAPs are associated with
each user. The simulator assigns bandwidth and power required for each allocation AMAP based
on the averaged SINR, MCS of the user and implementation loss.
3.
BD and PD can be calculated after all allocation AMAPs are assigned. The final throughput number
is calculated after 1000 iterations, each with a set of randomly selected user from the cell.
9
Simulation Setup
1.
Link level performance is used to determine the target SINR for a given MCS. To achieve 1%
PER, QPSK ½ with rep 1, 2, 4, 6 needs SINR of 2.9, -0.3, -2.8, and -4 dB respectively.
2.
5% of the weakest users in the cell are excluded in the simulation. 1.5 km ISD cell size is used.
3.
Implementation loss of 3 dB is used to cover performance loss due to implementation and
interference limited scenarios.
4.
10 MHz band and 48 PRUs are used in the simulation.
5.
SINR0=3.
6.
In case of multiple MCSs levels, suppose a user’s SINR (SINR(u)) falls between the target SINRs
for MCS(i) and MCS(i+1), denoted by SINR(i) and SINR(i+1). MCS(i) is chosen if SINR(u)>
(SINR(i)+SINR(i+1))/2 -1. MCS(i+1) is chosen otherwise
10
Data Throughput Comparison for Fixed and Variable MCS
2
1.8
Data throughput (b/s/hz)
1.6
1.4
1.2
1
0.8
0.6
Fixed: QPSK 1/2
Fixed: QPSK 1/2 rep2
Fixed: QPSK 1/2 rep4
Variable: QPSK 1/2 no rep and rep 4
Variable: QPSK 1/2 rep 1, 2, 4, 6
0.4
0.2
0
1
2
3
4
5
6
7
Number of users (each user has two DL control allocations)
8
11
Control Channel Power Boosting Statistics
1
0.9
0.8
0.7
CDF
0.6
0.5
0.4
0.3
Fixed: QPSK 1/2
Fixed: QPSK 1/2 rep2
Fixed: QPSK 1/2 rep4
Variable: QPSK 1/2 no rep and rep 4
Variable: QPSK 1/2 rep 1, 2, 4, 6
0.2
0.1
0
-20
-10
0
10
20
Control channel power boosting
30
40
12
Data Band Power Boosting Statistics
1
0.9
0.8
0.7
Fixed: QPSK 1/2
Fixed: QPSK 1/2 rep2
Fixed: QPSK 1/2 rep4
Variable: QPSK 1/2 no rep and rep 4
Variable: QPSK 1/2 rep 1, 2, 4, 6
CDF
0.6
0.5
0.4
0.3
0.2
0.1
0
-10
-8
-6
-4
-2
Data band power boosting
0
2
13
Conclusion
•
The throughput analysis shows that variable MCS with 2 MCS levels provides
equivalent throughput to 4 MCS levels.
•
Variable MCS has 7% and 26% higher throughput than the best fixed MCS scheme
in 4 users and 8 users cases, respectively.
•
Effect of interference to other cells due to power boosting is not considered in the
analysis.
•
Variable MCS reduces the scale of power boosting or deboosting, reducing potential
issues of power-boosting caused interference.
14
Throughput Analysis with FFR
•
If a cell has both reuse 1 and reuse 3 frequency groups, AMAP, this analysis is to
compare the throughput difference for reuse 1 only AMAP, reuse 3 only AMAP, and
reuse 1 and 3 AMAP.
•
Two nominal SINRs are assumed here: SINR1 for reuse 1 group nominal SINR and
SINR3 for reuse 3 group nominal SINR. SINR1 and SINR3 are assumed to be 3 and
12 (6 dB higher than reuse 1) respectively.
•
There the throughput equation becomes
S D  BD1  log 2 (1  PD  SINR1 )  BD3  log 2 (1  PD  SINR3 )
where BDi is the fraction of total bandwidth available for data in frequency reuse group
i.
•
In the following simulation, 12 reuse 1 PRUs and 12 reuse 3 PRUs are used in a
sector.
15
FFR Assignment
•
In the case that AMAP can be present in both reuse 1 and reuse 3. A user is
assigned to reuse 1 if user SINR is no less than 3 dB below the target SINR.
Otherwise the user is assigned to reuse 3.
16
FFR Sector Throughput
1.4
Sector data throughput (b/s/hz)
1.3
1.2
1.1
1
0.9
Reuse 1 and 3, QPSK 1/2
Reuse 1 only, QPSK 1/2 no rep and rep 4
Reuse 1 only, QPSK 1/2 rep 2
Reuse 3 only, QPSK 1/2
0.8
0.7
1
1.5
2
2.5
3
3.5
Number of users (each user has two DL control allocations)
4
17
Conclusion
•
Using fixed MCS (QPSK ½) and distribute AMAP in both reuse 1 and reuse 3 groups
gives the highest throughput number. For 4 users:
–
–
•
3% higher than reuse 3 only AMAP (QPSK ½)
52% higher than reuse 1 only AMAP (QPSK ½ rep 1 and 4)
Fixed MCS may be applied if each frequency partition has AMAP.
18