IEEE C802.16m-08/478r1
Project
Title
IEEE 802.16 Broadband Wireless Access Working Group < http://ieee802.org/16 >
Design on the Synchronization Channel for IEEE802.16m
Date
Submitted
2008-05-05
Source(s) Seunghee Han
Sungho Moon
Jin Sam Kwak
LG Electronics
Re:
Voice: +82-31-450-1935
E-mail: dondai@lge.com
samji@lge.com
msungho@lge.com
LG R&D Complex, 533 Hogye-1dong,
Dongan-gu, Anyang, 431-749, Korea
IEEE 802.16m-08/016r1 "Call for Contributions on Project 802.16m SDD" - Preambles
Abstract
Purpose
Notice
Release
Patent
Policy
This contribution describes many consideration points in the preamble design, and proposes some candidates for IEEE 802.16m preamble for the document C802.16m-08/003r1.
To be discussed and adopted by TGm for use in the IEEE 802.16m SDD
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.
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.
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 >.
Seunghee Han, Sungho Moon, and Jin Sam Kwak
LG Electronics
This document provides various aspects and views in designing a synchronization channel
1
based on discussions and numerical analyses. We briefly describe design criteria and discuss pros and cons in terms of
1
Since there are different views on the location of the IEEE 802.16m synchronization channel, we use the general word
“synchronization channel” throughout this contribution instead of “preamble” .
1

IEEE C802.16m-08/478r1 timing acquisition algorithms, frequency allocation methodologies, and multi-antenna supports. Based on simulations and analysis results, an auto-correlation based CDD-type synchronization channel, which has a sector-common frequency allocation and supports a transparent multi-antenna transmission, is proposed in this contribution.
As a text proposal, we suggest some text and illustrations for the Draft IEEE 802.16m System
Description Document (SDD), C802.16m-08/003r1 [1].
In general, the main objective of synchronization channels (or signals) is to enable MSs to perform the following fundamental functionalities during cell search operations;

Time/frequency synchronization

Detection of cell IDs which can provide best effort services with appropriate link budget
An example of flow chart for cell search procedures is depicted in Figure 1 . In this procedure, we assume that
the synchronization signal has a form of N times (Nx) repeated waveforms in terms of a time-domain amplitude.
The synchronization channel with a signal of N x repeated waveforms in time domain is widely adapted in several OFDM-based systems. The synchronization channel with a signal of N x repeated waveforms in time domain is widely adapted in several OFDM-based systems. For example, the 16e preamble has three times repeated waveforms in time domain [2].
In addition, synchronization channels can be utilized to provide the following supplementary functionalities;

Phase reference for broadcast channel (BCH) decoding to reduce pilot overhead or to provide good channel estimation performance in case that the synchronization channel is time/frequency-coherently located with BCH.

Time/frequency tracking during data communications

Measurement for neighboring cells in scanning modes

Additional system information, if needed
Since the cell search is the earliest step in the cellular communication, its performance would have an effect on other remaining procedures. Therefore, it is natural that the synchronization channel should be optimized even in the worst case.
2

no
Ntrial0
>=
MaxTrial0 ?
yes
Ntrial0=0
SHIFT THE NEXT CARRIER
FREQUENCY (SCAN
FREQUENCY RASTER)
SET INITIAL CARRIER FREQUENCY
(set Ntrial0=0)
COARSE TIMING SYNC
Ntrial0
=
Ntrial0+1 no
>= λ0 ?
yes
FREQUENCY OFFSET SYNC
(set Ntrial1=0)
DETECT CELL ID
DETECT ADDITIONAL INFO
(OPTIONAL)
>=
λ1 ?
yes
FINE TIMING SYNC no Ntrial1
=
Ntrial1+1
IEEE C802.16m-08/478r1
SHIFT THE NEXT CARRIER
FREQUENCY (SCAN
FREQUENCY RASTER) no
Ntrial1
>=
MaxTrial1 ?
yes
Ntrial0=0
DECODE BCH
Figure 1 Example of flow chart for search procedures
Based on above discussions, we summarize the desirable criteria on synchronization channels to provide not only the fundamental functionalities but also the supplementary ones as follows;
1) Good correlation properties
2) Low complexity during cell search, especially in the timing acquisition step
3) Consideration of effects on the frequency offset
4) Low PAPR to extend the cell coverage by power boosting
5) Constant amplitudes for the synchronization sequences in frequency domain in order to provide the optimal capability of channel estimation for BCH decoding
6) Feasibility of channel estimation for each transmit antenna considering the decoding performance on
BCH
7) Feasible synchronization even in the worst case such as a cell-edged MS scenario
3

IEEE C802.16m-08/478r1
The synchronization channel structure may be categorized into two cases according to the timing acquisition algorithms;

Cross-correlation based algorithm in case of

Auto-correlation based algorithm in case of
N

1 rep
N rep

2
In the following, we define N rep
(=1,2,…) as the number of the repetitions corresponding to the signal amplitude periodicity in terms of a time-domain waveform within an OFDM symbol. For example, the periodic insertion in frequency domain with three/two subcarrier spacing result in N
 rep
3 / N
 rep
2 in time domain.
Although the CC (Cross-Correlation) based algorithm can be also applied with a single or a few synchronization signals in case of N
 rep
2 , we will not consider the case due to the ambiguous peaks from the repeated signal waveform. The illustrations of ambiguous peaks using CC based algorithm in case of 2x repeated signal are
shown in Figure 2 .
Aperiodic auto-correlation by CC-based timing detection
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Ambiguous peaks
Desired peak
0
0 800 900 1000 100 200 300 400 500
Time shift
600 700
Figure 2 Ambiguous peaks in 2x rep using CC
The CC based algorithm can be applied when one or a few known synchronization signals are pre-defined. The
CC based timing acquisition algorithm can be implemented by the following equation (1) under the assumptions with a single known signal, a SISO transmission, and no multi-frame averaging. d
 d

R d d N f

1
 with 


M m

1 ( m

1) L

1  

0
*
    
2


/
 
( N

ML ) , (1)
4

IEEE C802.16m-08/478r1 where
  


N n


1

0
P d

 

2
1



,
2  when d

N

0
1
 2
, d

0
, and
( )
/
( )
/
M
/
L
/
N
/ d
ˆ /
N f
represent the known synchronization signal in time domain / the received signal
/ a partition index for the M-partial correlation 2 / the length of the divided sequence due to partial correlation / the length of the total known sequence / a detected timing / the frame length, respectively.
.
The AC (Auto-Correlation) based algorithm using the waveform repetition can be implemented by the following equation (2) assuming a 2x repeated waveform. d
 d

R d d N f

1

with
 

N
 n


0

1

 
R n old
 

*


R r d new

, d
 

0
N
2





/
 
, when d

0
, (2) where R old

 and
 
1
 
*

 
N
2



, R new



 r d

N
2

1
  *





N n


1

0
P d
 

2
1



,
2  when d


N
0
1
 2
, d

0
.

N 1
 
,
Pros and cons for CC and AC based algorithms are summarized as follows;

CC based algorithm

Pros

It can obtain a sharpened peak in timing acquisition under very small frequency offset environment. It means the coarse timing step can be skipped in the synchronization procedure.

Cons

The complexity increases significantly.

In order to achieve the fundamental goal, i.e, cell search, of the synchronization channel, it requires, at least, one additional channel either in time/frequency/code/space domain to carry cell ID information. This is because it is not feasible to adapt multiple correlators for hundreds of hypothesis tests during the timing detection from practical implementation perspective.

Under a large frequency offset environment, the benefit of sharpened peak would disappear due to partial correlation which could result in stubby peak.

AC based algorithm

Pros

The complexity is very small.
 It is possible for the synchronization channel to consist of only a single OFDM symbol. In other
2 It is well known that the partial correlation concept is introduced to reduce the impact from a large initial frequency offset [3]
5

IEEE C802.16m-08/478r1 words, it does not require additional resources or channels.

It can works well regardless of frequency offset effect due to differential operation. (see equation (2))

Cons

Additional fine timing is required after cell ID detection.
Regarding complexity which is directly related to the battery consumption at MSs, a CC based algorithm has much larger complexity than AC based one. For example, assuming 50000 samples during a radio frame with a certain sampling rate, the required number of complex multiplications and additions with N

1024 by
equation (1) and (2) are given by Table 1 .
Table 1 Comparison of the complexity during one radio frame (It doesn ’t count the division by P(d) in the equations (1) and (2))
# of complex multiplications
CC based
51200000
AC based (2x rep.)
100512
Ratio of saving (%)
50939.19 %
# of complex additions 51150000 100511 50889.95 %
Based on the above discussions, we prefer an AC based synchronization channel ( N rep

2 ) .
In this chapter, we discuss two types of synchronization channels, i.e., sector-specific and sector-common frequency allocations. It is noted that the following discussion is based on the auto-correlation based synchronization channel using repeated waveforms as discussed in Chapter 3.
The sector-specific frequency allocation means that each cell or sector segment can be identified according to the frequency allocation of synchronization signals. For example, the current 16e synchronization signal is transmitted in the sector-specific manner in which each part of frequency positions is corresponding to a segment ID to accomplish the frequency reuse of 3. In detail, the preamble sequence of segment ID n is transmitted in the subcarriers at 3 k
 n , where k is a running index 0 to 283 for 1024-FFT. In consequence, each segment can be identified via the frequency position of the synchronization signals.
Meanwhile, the sector-common frequency allocation represents that all cell or sector segments have the same frequency allocations.
The pros and cons for sector-specific and sector-common allocations are summarized as following;

Sector-specific frequency allocation

Pros

The performance of channel estimation by synchronization channel might be improved by the orthogonal positions when synchronization channel is utilized as a phase reference for decoding
BCH.

Cons

It could not work for cell-edged MSs in perspective of time/frequency synchronization due to the destruction of the repeated waveforms. In consequence, this structure could result in worse
6

IEEE C802.16m-08/478r1 synchronization performance than the sector-common frequency allocation.

More hypothesis tests might be required to estimate a frequency offset for all cases of the sector-specific allocation.

With CP-based timing detection,
 the performance will significantly be degraded due to the fact that the CP region is already collapsed by ISI from previous symbol.

It would not be feasible because the current IEEE802.16m system might allow various CP length.

Sector-common frequency allocation

Pros

The above problems from sector-specific frequency allocation can disappear. Moreover, the sector-common frequency allocation can provide a kind of SFN gain in perspective of time/frequency synchronization by construction of repeated waveforms from sectors in a celledged MS scenario in the synchronized network.

Cons

The performance of channel estimation by the synchronization channel might be marginally degraded in comparison to the sector-specific frequency allocation when the synchronization channel is utilized as a phase reference for decoding BCH.
Figure 3
shows an example of a cell-edged MS scenario which is one of the worst cases. The next
Figure 4
and
Figure 5
, respectively, straightforwardly describe the destruction and construction of repeated waveforms
resulting from sector-specific- and sector-common allocations.
Segment 0 Segment 2
Segment 0 Segment 2 se
BS0
Segment 1
MS
Segment 0
BS1
Segment 1
Segment 2
BS2
Segment 1
Figure 3 Example of the worst case for cell-edged MS
7

IEEE C802.16m-08/478r1
Sector 0
Insert only 3m (m is integer)
. . .
3x repetitive waveform
. . .
Time domain waveform
A
A ={a(0) a(1) a(2) … } -(3k+3) -(3k+1)
-(3k+2)
-3 -2 -1
Insert only 3m+1 (m is integer)
Sector 1
. . .
0 1 2 3
. . .
3k 3k+1 3k+2
Frequency index k: integer
Time domain waveform
B
3x repetitive waveform
B
A
·exp(j2 π ·1/3)
A
B
·exp(j2 π ·2/3)
-(3k+3) -(3k+1)
-(3k+2)
-3 -2 -1
Insert only 3m+2 (m is integer)
0 1 2 3
Sector 2
. . .
. . .
3k 3k+1 3k+2
Frequency index k: integer
Time domain waveform
B ={b(0) b(1) b(2) … }
C
3x repetitive waveform
C ={c(0) c(1) c(2)
…
}
C ·exp(j4 π ·1/3) C ·exp(j4 π ·2/3)
-(3k+3) -(3k+1)
-(3k+2)
-3 -2 -1 0 1 2 3 3k 3k+1 3k+2
Frequency index k: integer
At cell-edged MS,
Sector0+Sector1+Sector2
Non-repetitive waveform
. . .
-(3k+3) -(3k+1)
-(3k+2)
-3 -2 -1 0 1 2 3
. . .
3k 3k+1 3k+2
Frequency index k: integer
Time domain waveform
D
D ={d(0) d(1) d(2) … }
Repetitive waveform is destructed!
Figure 4 Destruction of repeated waveforms in the sector-specific allocation (3 x repetitions)
Insert only 2m (m is integer)
2x repetitive waveform
Sector 0
. . .
. . .
Time domain waveform
A ’
A
’
={a
’
(0) a
’
(1) a
’
(2)
…
}
A ’
-(2k+1)
-(2k+2)
-2 -1 0 1 2
2k 2k+1
Frequency index k: integer
Insert only 2m (m is integer)
Sector 2
. . .
. . .
2x repetitive waveform
Sector 1
. . .
. . .
-(2k+1)
-(2k+2)
-2 -1 0 1 2
2k 2k+1
Frequency index k: integer
Insert only 2m (m is integer)
Time domain waveform
B ’
B ’ ={b ’ (0) b ’ (1) b ’ (2) … }
2x repetitive waveform
B ’
Time domain waveform
C ’
C
’
={c
’
(0) c
’
(1) c
’
(2)
…
}
C ’
-(2k+1)
-(2k+2)
-2 -1 0 1 2
2k 2k+1
Frequency index k: integer
At cell-edged MS,
Sector0+Sector1+Sector2
2x repetitive waveform
Time domain waveform
-(2k+1)
-(2k+2)
. . .
. . .
-2 -1 0 1 2
2k 2k+1
Frequency index k: integer
D ’ D ’
D ’ ={d ’ (0) d ’ (1) d ’ (2) … }
Repetitive waveform still remains and link budget may be increased!
Figure 5 Construction of repeated waveforms in the sector-common allocation (2 x repetitions)
8

IEEE C802.16m-08/478r1
While the performance loss of channel estimation from sector-common allocation would be marginal, the problem from sector-specific allocation would be considerably critical. As discussed in Chapter 2, the synchronization at cell-edge, which is the worst case, is a fundamental requirement while allowing a phase reference for decoding BCH via synchronization channel is an optional functionality.
The simulation results for a cell-edged MS scenario as seen in Figure 3
were depicted from Figure 6
to Figure 7 .
The timing detection performances for a sector-specific frequency allocation with 3x repeated waveforms and sector-common one with 2x repeated waveforms for the synchronization channel were shown under the environments of TU6 3km/h and TU6 120km/h. It is noted that the sequence and structure for the sector-specific allocation are the same as that of the IEEE 802.16e synchronization channel. The timing synchronization is regarded as success when the detected timing is ranged within
±
CP_DURATION. The detailed simulation conditions can be found in Annex A. The notation Nsector and I ref
/ I oc
, in these figures, represent the number of sectors which have an effect on the target MS and the ratio of the referenced sector power to other sector powers, in which the signals from other cells were modeled as AWGN, respectively. The referenced sector in this simulation was assumed to be sector 0.
TU6 3km/h, 1Tx/2Rx ant, FO=3ppm
10
0
10
-1
Increase of # of sectors
10
-2
10
-3
10
-4
-12 -10
Increase of # of sectors
-8
Sector-common (2x rep), Nsector=1
Sector-common (2x rep), Nsector=2
Sector-common (2x rep), Nsector=3
Sector-specific (3x rep), Nsector=1
Sector-specific (3x rep), Nsector=2
Sector-specific (3x rep), Nsector=3
-6
Iref/Ioc [dB]
-4 -2 0
Figure 6 TU6 3km/h, 1Tx/2Rx antennas, 3ppm frequency offset
9

IEEE C802.16m-08/478r1
TU6 120km/h, 1Tx/2Rx ant, FO=3ppm
10
0
10
-1
Increase of # of sectors
10
-2
Increase of # of sectors
10
-3
10
-4
-12 -10 -8
Sector-common (2x rep), Nsector=1
Sector-common (2x rep), Nsector=2
Sector-common (2x rep), Nsector=3
Sector-specific (3x rep), Nsector=1
Sector-specific (3x rep), Nsector=2
Sector-specific (3x rep), Nsector=3
-6
Iref/Ioc [dB]
-4 -2 0
Figure 7 TU6 120km/h, 1Tx/2Rx antennas, 3ppm frequency offset
From the simulation results, we observe that

Under a single dominant link scenario with one sector, the performance of timing acquisition for 2x repeated structure is better than 3x repeated one by 1 dB of I ref
/ I oc
.

As increasing the number of sectors at cell edged MS, the acquisition performance is gradually
 degraded in case of the sector-specific frequency allocation due to the destruction of the repeated waveform. Particularly in fully loaded cases which is Nsector=3, the timing acquisition operation does not work at all.
 improved in case of the sector-common frequency allocation due to the construction of the repeated waveform.
Based on above discussion and simulation results, we suggest each synchronization signal is transmitted in a manner of sector-common frequency allocation.
According to desirable design criteria discussed in Chapter 2, we introduce an efficient transmit scheme for the synchronization channel in this chapter in order to provide a channel estimation capability for each antenna. In this chapter, we assume a 2x repeated synchronization channel structure. The introduced scheme can be also applied to the general Nx repeated structures.
1
0

IEEE C802.16m-08/478r1
Figure 8
shows the suggested CDD (Cyclic Delay Diversity) structure in the synchronization channel with the
predefined shifts corresponding to each antenna ports. In order to enable MSs to estimate MIMO channels, it is desirable to have a shifted offset between antennas larger than a CP duration. It can be implemented by a phase rotation in frequency domain as well as a circular shift in time domain. copy
Ant 0
CP A A a_0(0) . . . . . a_0(N-1)
Circular shift by δ copy
Ant 1
CP A A
.
.
.
Ant N_tx-1
Circular shift by (N_tx-1) δ copy
CP A A
(N_tx-1)
δ
Figure 8 Example of Tx operation for the synchronization channel
Figure 9 shows the corresponding example of Rx operation to perform channel estimation for each antenna at an
MS. copy
CP a_0(0)
A
. . . . .
+ copy
A a_0(N-1)
CP A A
δ
+
Circular shift by (N_tx-1) δ copy
+ r(0)
δ
(time domain)
. . . . . . . . . . r(N-1)
R(0)
(frequency domain)
. . . . . . . . . . R(N-1)
DFT(FFT) window
1. received signal at MS
2. perform DFT(FFT)
3. IDFT(IFFT) after multiplying by the sequence for ant0
(R(k)*conj(A_0(k)))
(frequency domain)
CP
(N_tx-1) δ
A A r'(0) . . . . . . . . . . r'(N/2-1) r'(N/2)
(time domain)
. . . . . . . . . . r'(N-1)
. . .
. . .
Eg. window for estimating channel for ant0
Impulse response of the channel for antenna0
Impulse response of the channel for antenna1
Impulse response of the channel for antennaN_tx-1
Figure 9 Example of Rx operation for channel estimation at the synchronization channel
1
1

IEEE C802.16m-08/478r1
The diversity gain from the proposed scheme can be confirmed as seen in Figure 10 and Figure 11 .
CDD, shift offset=CP length, TU6, 3km/h, FO=3ppm
10
0
10
-1
1Tx-2Rx, Nsector=1
2Tx-2Rx, Nsector=1
4Tx-2Rx, Nsector=1
1Tx-2Rx, Nsector=2
2Tx-2Rx, Nsector=2
4Tx-2Rx, Nsector=2
1Tx-2Rx, Nsector=3
2Tx-2Rx, Nsector=3
4Tx-2Rx, Nsector=3
10
-2
10
-3
-4
10
-12 -10 -8 -6
Iref/Ioc [dB]
-4 -2 0
Figure 10 CDD with shift offset of CP length in Sync channel, TU6 3km/h
CDD, shift offset=CP length, TU6, 120km/h, FO=3ppm
10
0
10
-1
1Tx-2Rx, Nsector=1
2Tx-2Rx, Nsector=1
4Tx-2Rx, Nsector=1
1Tx-2Rx, Nsector=2
2Tx-2Rx, Nsector=2
4Tx-2Rx, Nsector=2
1Tx-2Rx, Nsector=3
2Tx-2Rx, Nsector=3
4Tx-2Rx, Nsector=3
10
-2
10
-3
-4
10
-12 -10 -8 -6
Iref/Ioc [dB]
-4 -2 0
Figure 11 CDD with shift offset of CP length in Sync channel, TU6 120km/h
1
2

The proposed CDD scheme on Sync channel has following advantages;
IEEE C802.16m-08/478r1

The synchronization channel can be transmitted to MSs in a transparent manner by adapting CDD.

Multi-antenna diversity gain can be achieved in the synchronization signal transmission.

The channel estimation from each antenna can be available. This can improve the BCH decoding performance when the synchronization channel is time/frequency-coherently located to BCH.
In this contribution, we discussed several aspects on designing the IEEE 802.16m synchronization channel.
Based on the above discussions and analyses, we summarize our preferences on the synchronization channel as follows;
 Auto-correlation based synchronization channel structure
 Sector-common frequency allocation in the synchronization channel
 CDD transmission with predefined antenna-specific shift values in which the shift offsets between arbitrary two antennas have equal to or larger duration than CP length
[1]
IEEE 802.16m-08/003r1, “Draft IEEE 802.16m System Description Document”
[2]
IEEE P802.16Rev2 D2, “DRAFT Standard for Local and metropolitan area networks - Part 16: Air Interface for Broadband Wireless Access Systems,” April, 2008.
[3] Wang, Ottosson, “ Cell Search in W-CDMA ”, IEEE JSAC August 2000, pp. 1470-1482
============= Start of text proposal =================================================
11.x.x Synchronization channel
The synchronization channel is transmitted by one OFDMA symbol in order not only to facilitate time/frequency synchronization and the detection of Cell ID information but also to support the channel estimation for each transmit antenna. The time-domain waveform of each synchronization signal, which is transmitted in a manner of sector-common frequency allocation, should have two times or more repetitions in terms of its amplitude. The CDD scheme in a synchronization channel should be used with antenna-specific shift values. Each shift offset between two antennas should have equal to or larger duration than CP length.
============= End of text proposal =================================================
1
3

IEEE C802.16m-08/478r1
1
4

The simulation parameters used in this paper are summarized in Table 2 .
IEEE C802.16m-08/478r1
Parameter
Carrier frequency
System bandwidth
Sampling factor
Sampling frequency
Subcarrier spacing
FFT size
CP config.
Number of used subcarriers
Number of guard subcarriers
Frame duration & Sync channel periodicity
Table 2 Simulation parameters
Explanations
2.5GHz
10MHz
28/25
11.2MHz
10.9375kHz
1024
1/16 CP (= 5.71us)
852 for Sync channel, 840 for data channel
172 for Sync channel, 184 for data channel
5ms
Number of antennas
Tx
Rx
1, 2, 4
2
Multi-antenna transmission scheme
CDD with 5.71us shift offsets in case of 2Tx and 4Tx
Repetition factor
Sync channel sequence
Power boosting on Sync channel compared to other data channels
Modeling of other data channels
Timing acquisition algorithm
Channel
Mobility
Number of sectors
Modeling of other sectors (cells)
Referenced sector
Averaging duration
Number of sequences (cell IDs) in
Sync channel
Frequency offset
Sector-specific
Sector-common
Sector-specific
Sector-common
Sector-specific
Sector-common
3x repetition
2x repetition
Binary sequence defined in [2] for preamble
Binary random generated sequence
8.0 [2]
8.0*284/426
Randomly generated QPSK signals
Auto-correlation based
TU6
3km/h, 120km/h
1, 2, 3
AWGN
Sector 0
5ms
114
3ppm
1
5

IEEE C802.16m-08/478r1
The criterion on successful timing detection in this simulation is illustrated in
Figure 12
.
1 OFDM symbol duration
CP
Range for successful timing acquisition ( ±CP)
Figure 12 Criterion of successful timing acquisition for simulation
1
6