Proposed Text for DL subcarrier permutation and UL tile permutation
IEEE 802.16 Presentation Submission Template (Rev. 9)
Document Number:
IEEE C802.16m-09/0582r3
Date Submitted:
2009-03-11
Source:
Taeyoung Kim, Jeongho Park, Kichun Cho, Jaeweon Cho,
Hokyu Choi , Heewon Kang
Voice:
E-mail:
+82-31-279-0202
ty33.kim@samsung.com
E-mail:
jong-kae.fwu@intel.com
E-mail:
savant21@etri.re.kr
Samsung Electronics Co., Ltd.
416 Maetan-3, Suwon, 443-770, Korea
Jong-Kae (JK) Fwu, Minh-Anh Vuong, Huaning Niu, Rongzhen Yang,
Yuval Lomnitz, Wei Guan, Sassan Ahmadi, Hujun Yin
Intel Corporation
Jihyung Kim, Wooram Shin, Dong Seung Kwon
ETRI
Venue:
IEEE 802.16m Session#60, Vancouver, Canada
IEEE 802.16m-09/0012, “Call for Comments on Amendment Working Document”.
Base Contribution:
None
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 a
n 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 IE
EE’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 thi
s 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 >.
Motivation
• In the current IEEE 802.16m Amendment Working
Document (IEEE80216m-09/0010),
– Subcarrier permutation for DL is NOT determined yet
– Tile permutation for UL is NOT determined yet
• This contribution shows the evaluation results to
compare the permutation rules proposed from many
companies (e.g. Intel, LGE, Samsung)
– For Downlink, LLS results under the multicell environment
– For Uplink, Hitting Count Results
3/14
Part 1
Uplink Tile Permutation
4/17
Issue
• Different Permutation Approaches
– Intel/Samsung :
Tile(s,n,t) = LDRU,FPi  n + g(PermSeq(), s, n, t)
PermSeq() is permutation sequence generated with SEED={IDcell*1367} mod 210.
g(PermSeq(),s,n, t) = {PermSeq[(n+s+t) mod LDRU,FPi]+UL_PermBase} mod LDRU,FPi,
UL_PermBase is set to IDcell.
– LG : Tile ( s, n, t )  ( LDRU , FPi  f (n, s)  g ( PermSeq(), s, n, t )  CellID ) mod( 3 * LDRU , FPi )
where
f (n, s )  (5n  7 s ) mod 3
g ( PermSeq(), s, n, t )  PermSeq( f (n, s)  s  OTP  t )
where
 GCD( LDRU , FPi, DTP ) 
PermSeq(i )  {DTP  i  OTP  i 
} mod LDRU , FPi,
LDRU , FPi


where
DTP  (Cell ID mod( LDRU , FPi  1))  1
5/17
and
i  0,1, , LDRU , FPi  1
 Cell ID 
OTP  
 1
 ( LDRU , FPi  1) 
Hitting Count Verification
• Methodology
– Select 2 Sector IDs from total 768
• Total number of cases : combination 2 among 768 = 294528 cases
– Count the number of tiles two sectors share
• Full loading case  obviously 100% hitting
• Not full loading case  worst case is 100% hitting
– Results metric
• Histogram of hitting count
• Average value is not important
• The more high hitting case, the higher IoT. Var.  worse performance
• Assumptions
– Total 48 DRU
– Case 1 : resource loading ratio is 1/6  8 DRU (24 tiles)
– Case 2 : resource loading ratio is 2/6  16 DRU (48 tiles)
Hitting Count Results
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
• Case 2
S/I
52204
23694
26140
41699
28712
27739
27077
19337
15746
11771
7615
5251
3234
1969
1100
652
321
149
67
20
17
10
0
0
4
Intel
61440
36864
30720
49152
12288
24576
12288
12288
0
36864
12288
0
0
0
0
0
0
0
0
0
0
0
0
0
5760
LGE
19133
17202
5250
54952
45987
54400
22186
20500
6387
10667
16814
2099
3980
4019
1072
202
505
518
1676
129
4076
945
212
480
1137
S/I
…
Number of tiles having collision
Number of tiles having collision
• Case 1
Intel
LGE
0
1
2
3
4
5
6
7
8
9
10
4124
2796
2723
5546
5059
5266
7482
7509
8175
10390
10347
0
0
0
12288
0
0
12288
24576
0
0
0
2128
0
0
0
0
512
0
4272
0
949
181
37
38
39
40
41
42
43
44
45
46
47
48
607
453
293
173
108
70
45
14
3
1
3
4
0
0
0
0
0
0
0
0
0
0
0
5760
650
203
0
0
0
8684
291
4397
609
195
418
1209
Part 2
Downlink Subcarrier Permutation
8/17
LLS in Multi-cells environment (1)
• Cell ID Configuration
• Evaluation methodology
– Increasing sector ID
– ISD = 1.5km
31
34
37
35
36
38
40
10
11
49
41
43
13
MS
2
16

22
21
23
55
51
53
24
26
18
52
48
50
5
20
15
49
25
3
19
17
45
47
4
0
– Varying parameter of “radius”
– Select 7 strongest interferers
27
29
8
12
46
28
6
1
14
42
44
7
9
• # of IDcell for desired cell = 0
• Radius is variable, but theta is fixed as
30 degree
30
32
33
– User drop on the desired cell,
which is located in (radius, theta)
• Calculating only path-loss according to
the distance between MS and BSs.
• Not considering shadowing
– Calculate SINR
SINR 
54

P0
7
I
1 k
 N0
56
[ Example]
MS is located in (radius, theta) = (0.75*ISD/2, 30)
 7 strongest interferers(I1~I7) = 8, 19, 5, 4, 12, 22, 29
 SINR=6.36dB
LLS in Multi-cells environment (2)
• Simulation conditions
– Working scenarios
Freq. Partition
# of subbands
# of minibands
# of PRUs in FPi
FP0
6
24
48
FP1 ~ FP3
0
0
0
Scenario #1
– Number of DRUs / LRUs / Miniband allocation
• Half loading in DRUs(Ex. # of DRUs = 8, # of LRUs = 4)
• Half loading in Miniband based CRU
– Assuming random QPSK modulated data bursts are transmitted
– Assuming random sequence for CRU/DRU allocation sequence
– Channel condition: PedB, 3km/h
– MIMO configuration: 2x2 SFBC
– Pilot Structure
• Pilot power = 3 dB
• Interlaced pilot structure
FER vs SINR (1)
• Evaluation Results
– # of DRUs=5
– # of DRUs=4
1.E-01
1.E-01
FER
1.E+00
FER
1.E+00
1.E-02
1.E-02
1.E-03
1.E-03
8.81
8.20
7.59
6.98
6.36
5.68
5.00
4.32
3.64
2.94
2.24
1.52
0.79
0.02
-0.77
-1.58
-2.42
-3.28
-4.16
-5.08
8.81
8.20
7.59
6.98
6.36
5.68
5.00
4.32
3.64
2.94
2.24
1.52
0.79
0.02
-0.77
-1.58
-2.42
-3.28
-4.16
-5.08
SINR [dB]
SINR [dB]
Intel(16QAM )
Intel(QPSK)
Intel(16QAM )
Intel(QPSK)
LGE_M odified(16QAM )
LGE_M odified(QPSK)
LGE_M odified(16QAM )
LGE_M odified(QPSK)
Proposed(16QAM )
Proposed(QPSK)
Proposed(16QAM )
Proposed(QPSK)
FER vs SINR (2)
• Evaluation Results
– # of DRUs=7
– # of DRUs=6
1.E-01
1.E-01
1.E-02
1.E-02
1.E-03
1.E-03
8.81
8.20
7.59
6.98
6.36
5.68
5.00
4.32
3.64
2.94
2.24
1.52
0.79
0.02
-0.77
-1.58
-2.42
-3.28
-4.16
-5.08
8.81
8.20
7.59
6.98
6.36
5.68
5.00
4.32
3.64
2.94
2.24
1.52
0.79
0.02
-0.77
-1.58
-2.42
-3.28
-4.16
-5.08
FER
1.E+00
FER
1.E+00
SINR [dB]
SINR [dB]
Intel(16QAM )
Intel(QPSK)
LGE_M odified(16QAM )
LGE_M odified(QPSK)
Proposed(16QAM )
Proposed(QPSK)
Intel(16QAM )
Intel(QPSK)
LGE_M odified(16QAM )
LGE_M odified(QPSK)
Proposed(16QAM )
Proposed(QPSK)
FER vs SINR (3)
• Evaluation Results
– # of DRUs=9
– # of DRUs=8
1.E-01
1.E-01
FER
1.E+00
FER
1.E+00
1.E-02
1.E-02
1.E-03
1.E-03
8.81
8.20
7.59
6.98
6.36
5.68
5.00
4.32
3.64
2.94
2.24
1.52
0.79
0.02
-0.77
-1.58
-2.42
-3.28
-4.16
-5.08
8.81
8.20
7.59
6.98
6.36
5.68
5.00
4.32
3.64
2.94
2.24
1.52
0.79
0.02
-0.77
-1.58
-2.42
-3.28
-4.16
-5.08
SINR [dB]
SINR [dB]
Intel(16QAM )
Intel(QPSK)
Intel(16QAM )
Intel(QPSK)
LGE_M odified(16QAM )
LGE_M odified(QPSK)
LGE_M odified(16QAM )
LGE_M odified(QPSK)
Proposed(16QAM )
Proposed(QPSK)
Proposed(16QAM )
Proposed(QPSK)
FER vs SINR (4)
• Evaluation Results
– # of DRUs=11
1.E+00
1.E+00
1.E-01
1.E-01
FER
FER
– # of DRUs=10
1.E-02
1.E-02
z
1.E-03
8.81
8.20
7.59
6.98
6.36
5.68
5.00
4.32
3.64
2.94
2.24
1.52
0.79
0.02
-0.77
-1.58
-2.42
-3.28
-4.16
-5.08
8.81
8.20
7.59
6.98
6.36
5.68
5.00
4.32
3.64
2.94
2.24
1.52
0.79
0.02
-0.77
-1.58
-2.42
-3.28
-4.16
-5.08
1.E-03
SINR [dB]
SINR [dB]
Intel(16QAM )
Intel(QPSK)
LGE_M odified(16QAM )
LGE_M odified(QPSK)
Proposed(16QAM )
Proposed(QPSK)
Intel(16QAM )
Intel(QPSK)
LGE_M odified(16QAM )
LGE_M odified(QPSK)
Proposed(16QAM )
Proposed(QPSK)
Conclusion
• Uplink Tile Permutation
– Samsung’s is better in Hitting Count Results
– It will cause better interference averaging
• Downlink Subcarrier Permutation
– Similar performance in both sequential and random case
– Because BS Tx power is constant
• For DL and UL, it is natural to be same formula
and permutation sequence.
Proposed Text for AWD (1)
[Remedy-1: Change the text from line 35 to 38 on the page 31, in 15.3.5.3.3, as follows:]


PermSeq() is the permutation sequence of length LDRU,FPi and is determined by
SEED={IDcell*1367} mod 210. The permutation sequence is generated by the random sequence
generation algorithm specified in Section 15.3.5.3.4. generated by a function or by a lookup
table;
g(PermSeq(),s,m,l,t) is a function (TBD) with value from the set [0, LDRU, FPi -1], which is defined
as follows.;
g(PermSeq(),s,m,l,t) = {PermSeq[{f(m,s)+s+l} mod LDRU,FPi] +DL_PermBase} mod LDRU,FPi,
where DL_PermBase is an integer ranging from 0 to 31(TBD), which is set to preamble IDcell.

16/14
f(m,s) = (m+13·s) mod LSP,l. is a function (TBD) with value from the set [0, LSP,l -1].
Proposed Text for AWD (2)
[Remedy-2: Insert the text in line 48 on the page 31, in 15.3.5.3.3, as follows:]
15.3.5.3.4 Random sequence generation
The permutation sequence generation algorithm with 10-bit SEED (Sn-10, Sn-9,…,Sn-1) shall generate a permutation sequence of size
M by the following process:
1) Initialization
A. Initialize the variables of the first order polynomial equation with the 10-bit seed, SEED.
•
Set d1 = floor(SEED/25) + 1 and d2 = SEED mod 25.
B. Initialize the maximum iteration number, N=4.
C. Initialize an array A with size M with the numbers 0, 1, … , M-1 (i.e. A[0]=0, A[1]=1, … , A[M-1]=M-1).
D. Initialize the counter i to M-1.
E. Initialize x to -1.
2) Repeat the following steps if i > 0
A. Initialize the counter j to 0.
B. Repetition loop as follows,
a.
Increment x and j by 1.
b.
Calculate the output variable of y = {(d1*x + d2) mod 1031} mod M.
c.
Repeat the above step a. and b., if yi and j<N.
C. If y  i, set y = y mod i.
D. Swap the i-th and the y-th elements in the array (i.e. perform the steps Temp= A[i], A[i]= A[y], A[y]=Temp).
E. Decrement i by 1.
3)
PermSeq[i] = A[i], where 0i<M.
17/14
Proposed Text for AWD (3)
[Remedy-3: Change the text in line 1 on the page 42, in 15.3.8.3.3, as follows:]
Tile(s,n,t) = TBD LDRU,FPi  n + g(PermSeq(), s, n, t)
[Remedy-4: Insert the text in line 13 on the page 42, in 15.3.8.3.3, as follows:]


PermSeq() is the permutation sequence of length LDRU,FPi and is determined by
SEED={IDcell*1367} mod 210. The permutation sequence is generated by the random sequence
generation algorithm specified in Section 15.3.5.3.4.
g(PermSeq(),s,n,t) is a function of s, n, t and PermSeq(), which is defined as follows:
g(PermSeq(),s,n, t) = {PermSeq[(n+s+t) mod LDRU,FPi]+UL_PermBase} mod LDRU,FPi,
where UL_PermBase is an integer ranging from 0 to 31(TBD), which is set to preamble IDcell.
18/14
Appendix 1. Parameters for UL SLS
Parameter
Value
Parameter
Value
Carrier frequency (GHz)
2.5 GHz
Site to site distance (m)
1500 m
System bandwidth (MHz)
11.2 MHz
Number of users per sector
10
Reuse factor
1
Channel
Ped B, 3km/h 100%
Frame ( Preamble +DL +UL )
duration
5 ms (TDD, 29:18)
Max power in MS (dBm)
23 dBm
Number of OFDM symbols in
UL Frame
18 symbols
(3 subframes =
6 symbols per subframe)
Antenna type
1x2 SIMO
FFT size (tone)
1024
HARQ
On (Max retrans : 4 / Sync)
Useful tone
864
Target IoT Level
10 dB
Tile structure
6x6 DRU
Link to system mapping
RBIR
Number of LRU
48
Scheduler type
PF
Number of tile per LRU
3 tiles
PF exponent
1.0
Resource assignment block
8 LRU
Penetration loss[dB]
10dB
Number of user per subframe
6 user
Overhead
No control channel, only pilot
Power Control
Open loop power control
UL Target IoT value
10dB
19/#NN
Appendix 2. NI Fluctuation
•
Observed Metric : Histogram of MS’s NI Variance
20/17