Performance Evaluation of Pilots for MIMO Mode 5 (CDR)
IEEE 802.16 Presentation Submission Template (Rev. 9)
Document Number:
IEEE S80216m-09_2253
Date Submitted:
2009-11-11
Source:
Wang Zhen, David Mazzarese, Sangheon Kim,
Heewon Kang, Hokyu Choi
Samsung Electronics
[email protected]
Venue:
IEEE 802.16m Session#64, Atlanta, USA
LB30a
Base Contribution:
IEEE C80216m-09_2253
Purpose:
Discussion and approval
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 >.
Introduction
• Problem statement
– 1-stream pilot pattern for CDR (MIMO mode 5) in the
open-loop region is missing
– 1-stream pilot pattern of Fig 489 or 491 is not appropriate
because CDR needs an even number of tones between two
pilot positions in the same OFDM symbol
• Proposed candidates
– CoFIP pilots have been proposed in S80216m09_2238
– We propose to reuse the 2-stream interlaced pilot pattern
by assigning both pilot indices 1 and 2 to a single stream
CoFIP (S80216m09_2238)
P
0
0
P
0
0
0
P
0
0
P
0
0
P
0
P
0
0
P
0
P
0
0
0
P
0
0
P
Pilot pattern Set 0
0
0
0
P
0
0
0
0
0
P
0
P
Pilot pattern Set 1
P denotes pilot tone and 0 denotes null tone
P
0
P
0
0
0
0
0
P
0
P
Pilot pattern Set 2
0
P
0
This Proposal (12 pilots)
Comparison of the 2 proposals
Assumption: 2 receive antennas at AMS
CoFIP (pilots on pilots, pilots
on null tones, data on data)
12 pilots (pilots on pilots, pilots on
data , data on data)
Channel
estimation
6 pilots per stream, interference
from pilots of sector with same
interlace
12 pilots per stream, interference
from all sectors (pilots and data)
Interference
estimation
Direct MMSE Rx: worst-case
estimate on null tones, miss
interference on desired pilot
Indirect MMSE Rx: exact
interference on data after cancellation
of desired pilot
4x4 Interference
covariance
matrix for CDR
Reuse 1 pilot tone and its
conjugate for covariance matrix
of interference: noise correlation
Use 2 adjacent pilot tones for
covariance matrix of interference on
actual CDR pair of data tones
CQI estimation
On null tones, not capturing
interference on same interlace
On data over desired pilots, averaged
over the OL region in time domain
Throughput
108-18 = 90
108-12 = 96
Not straightforward to guess which one is better
Simulation parameters
1. 2Tx antenna – 2Rx antenna, one stream CDR transmission
2. Channel: Modified Ped B, 3 km/h
3. Thermal noise level is 10 dB below the power of the desired signal
4. QPSK ½ CTC , four 18x6 PRUs forms a channel coding block.
5. 2D-MMSE channel estimation. All estimation done within a PRU
6. The CoFIP and a newly proposed pilot pattern are used for CDR
6
Receiver structure
• CoFIP
– Estimate the signal channel in pilot position and the
interference channel in null-tone position
– Generate IC (interference correlation) matrix by stacking the
estimated channel on one pilot tone and its conjugate
– Use Direct MMSE receive filter to combine the received signal
• 12 pilots
– Estimate the signal channel in each pair of pilot tones (totally 6
pairs)
– Generate IC matrix by subtracting the estimated pilot signal
from the received signal for each pair of pilot tones
– Use Indirect MMSE receiver filter to combine the received signal
Full data transmission
• 3cells (1 serving cell & 2 interference cells)
Cell index
Power profile
Pilots trans
Data trans
0
1
on
on
1
1
on
on
2
1
on
on
FER with 1 serving cell & 2 interference
cells
Full data transmit
0
10
CDR CoFIP 3cell full
CDR 12pilot 3cell full
-1
FER
10
-2
10
-3
10
-10
-8
-6
-4
-2
0
2
SINR (dB)
4
6
8
10
CoFIP is good for CDR when there is no interference pilot
on the desired pilot, especially at higher SNR
Goodput with 1 serving cell & 2
interference cells
Goodput
100
90
80
Goodput
70
60
50
40
30
CoFIP 3cell full
12Pilot 3cell full
20
10
-10
-8
-6
-4
-2
0
2
SINR (dB)
4
6
8
10
The channel estimation advantage is not leveraged in the
goodput due to the overhead of null tones
Full data transmission
• 4cells (1 serving cell & 3 interference cells)
Cell index
Power profile
Pilots trans
Data trans
0
1
on
on
1
1
on
on
2
1
on
on
3
0.5
on
on
FER with 1 serving cell & 3 interference
cells
Full data transmit
0
10
-1
FER
10
-2
10
CDR CoFIP 4cell full
CDR 12pilot 4cell full
-3
10
-10
-8
-6
-4
-2
0
2
SINR (dB)
4
6
8
10
CoFIP is not as good for CDR when there is interference pilot on the
desired pilot, but both patterns are similar with 3 interferers
Goodput with 1 serving cell & 3
interference cells
Goodput
100
CoFIP 4cell full
12Pilot 4cell full
90
80
70
Goodput
60
50
40
30
20
10
0
-10
-8
-6
-4
-2
0
2
SINR (dB)
4
6
8
10
With the same channel and interference estimation quality,
goodput is better without null tones
Full data transmission
• 7cells (1 serving cell & 6 interference cells)
Cell index
Power profile
Pilots trans
Data trans
0
1
on
on
1
1
on
on
2
1
on
on
3
0.5
on
on
4
0.5
on
on
5
0.5
on
on
6
0.1
on
on
FER with 1 serving cell & 6 interference
cells (full interference)
Full data transmit
0
10
-1
FER
10
-2
10
CDR CoFIP 7cell full
CDR 12pilot 7cell full
-3
10
-10
-8
-6
-4
-2
0
2
SINR (dB)
4
6
8
10
No more advantage of CoFIP with 6 interferers, which is still underestimating the real-world situation
Goodput with 1 serving cell & 6
interference cells (full interference)
Goodput
100
CoFIP 7cell full
12Pilot 7cell full
90
80
70
Goodput
60
50
40
30
20
10
0
-10
-8
-6
-4
-2
0
2
SINR (dB)
4
6
8
10
Clear goodput advantage of proposed 12 pilots pattern
Empty data transmission
• 7cells (1 serving cell & 6 interference cells)
Cell index
Power profile
Pilots trans
Data trans
0
1
on
on
1
1
on
on
2
1
on
off
3
0.5
on
on
4
0.5
on
off
5
0.5
on
on
6
0.1
on
on
FER with 1 serving cell & 6 interference
cells (partial interference)
Full data VS empty data transmit
0
10
-1
FER
10
-2
10
CDR CoFIP 7cell 2empty
CDR 12pilot 7cell 2empty
-3
10
-10
-8
-6
-4
-2
0
2
SINR (dB)
4
6
8
10
No more advantage of CoFIP with 6 interferers, which is still underestimating the real-world situation
Goodput with 1 serving cell & 6
interference cells (partial interference)
Goodput
100
CoFIP 7cell 2empty
12Pilot 7cell 2empty
90
80
70
Goodput
60
50
40
30
20
10
0
-10
-8
-6
-4
-2
0
2
SINR (dB)
4
6
8
10
Clear goodput advantage of proposed 12 pilots pattern
Conclusion
• 12 pilots pattern is better than CoFIP in terms
of FER in real-world scenarios where
– The number of interferers is greater than 3
– No data is transmitted by some neighbouring ABS
• 12 pilots pattern is better than CoFIP in terms
of goodput
– In all cases
Text Proposal
• Adopt the proposal in the base contribution
C80216m-09_2253