Date post: | 19-Jan-2016 |
Category: |
Documents |
Upload: | gyles-hampton |
View: | 220 times |
Download: | 0 times |
Performance Evaluation of SU OL MIMO Schemes Proposed for IEEE 802.16m
IEEE 802.16 Presentation Submission Template (Rev. 9) Document Number:
IEEE C80216m-09_0137r1Date Submitted:
2009-01-05Source:
David Mazzarese, Sangheon Kim, Sangwoo Lee, Bruno Clerckx [email protected] Choi, Heewon KangSamsung Electronics
Kaushik Josiam, Zhouyue Pi, Farooq Khan [email protected] Telecommunications America
Venue:IEEE 802.16m Session#59, San Diego, US Call for Comments (tgmsdd) on IEEE 802.16m-08/003r6 System Description Document (SDD) Topic: request for details
Base Contribution:IEEE C80216m-09_0137r1
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 >.
Overview
• This contribution provides an analysis of the two classes of OL SU MIMO schemes proposed for more than 2 Tx– SFBC, SM with codebook-based precoder and
precoded pilots.– SFBC, SM, DSFBC, hybrid SFBC+SM with antenna
hopping precoder and non-precoded pilots.
• Based on the performance and complexity analysis, we recommend to adopt SFBC and SM with codebook-based precoder and precoded pilots.
Candidate OL SU MIMO Schemes
Current baseline in the SDD, using codebook-based precoding (RBF, PC) and precoded pilots.
Schemes that are listed as FSS in the SDD, based on antenna hopping and using non-precoded pilots.
Rate 1 Rank 1 effective SIMO
SFBC
SFBC
Rate 2 Rank 2 SM Rank 2 SM
DSFBC
Rate 3 Rank 3 SM Rank 3 hybrid SFBC+SM
RBF: random beamformingPC: precoder cycling (as simulated in this contribution using the DFT-based codebook of C80216m-08/1187)
Simulation EnvironmentParameters Values
Carrier Frequency 2.35GHz
System BW 10 MHz
Number of Antennas 4 Tx and 2 Rx Antennas
Data Burst Size 4 RUs
Permutation
Paired tone based LDRU - 4RUs over 24 PRUs PRU based LDRU - 4RUs over 48 PRUs
Modulation and Coding QPSK and 16QAM ½ with LTE Turbo Coding
Channel Model Uncorrelated channel Modified Veh A 30 and 120 km/h Modified Ped B 3 km/h
Channel Estimation Single and double PRUs level MMSE
MIMO Detector MMSE and MLD
4/6
Complexity Discussion
• Receiver decoding complexity– Decoder complexity– Number of decoding algorithms
• Channel estimation– Size of MMSE interpolation in frequency domain– Number of channel estimation algorithms
• Pilot overhead– CQI measurement pilots– Demodulation pilots
• System operation– Multiplexing of MIMO schemes in the same subframe
1. Receiver decoding complexity
• Several proponents are convinced that a maximum-likelihood decoder (MLD) has affordable complexity and therefore should be used for up to 2 streams.
• As a result, vertical encoding (i.e. SCW) has been adopted as the baseline in the SDD (horizontal encoding or MCW is FFS).
• Thus schemes aiming at introducing some structure in the STC to approach the MLD performance with an MMSE receiver are not necessary.
• The computational complexity of rate 2 MLD compares favorably with a 4x4 matrix inversion (MMSE receiver for DSFBC)
• Codebook-based schemes for rate > 2 all require the same type of receiver (linear MMSE), whereas DSFBC and hybrid SFBC+SM each introduce an additional receiver type.
Receiver complexity is in favor of codebook-based schemes
2. Channel Estimation
• The pilot patterns have been evaluated on the assumption that 2D-MMSE interpolation is performed within 1 RU.
• This is because one cannot guarantee that an MS would be able to find 2 or more physically adjacent RUs with the same type of pilots.
• Whenever an MS has that possibility it could exploit it, but this scenario cannot be assumed as the baseline performance, and it requires that an MS be equipped with multiple algorithms for channel estimation, which is complex.
• Both antenna hopping and codebook-based precoding with precoder cycling can benefit from multiple-RU channel estimation, as long as the cycling period is larger than 1 RU (e.g. 4 RUs).
• Codebook-based precoding schemes with precoded pilots allow to benefit from the increased pilot density per antenna of the 2-stream pilot pattern even when the BS is equipped with 4 or 8 Tx antennas.
Channel estimation performance is in favor of codebook-based schemes
3. Pilot Overhead
• Codebook-based precoding schemes with precoded pilots allow to adapt the number of pilot streams to the number of streams of the MIMO transmission.
• The advantage of precoded pilots will be seen in the goodput vs. SNR, even though antenna hopping schemes may show a better BLER vs. SNR in certain propagation environments and certain type of permutation.
• The non-precoded (common) pilots used with antenna hopping cannot be reused for CQI measurements since they will only be present on subcarriers allocated to OL MIMO transmissions, whereas all MS should be able to measure the CQI on any subband.
• Therefore a MIMO midamble with as many pilots as the number of Tx antennas at the BS is always required for all types of proposed schemes.
Pilot overhead is in favor of codebook-based schemes
4. System Operation• Antenna hopping schemes allow to use any MIMO scheme on different data
subcarriers within the same PRU, based on the same non-precoded pilots (using 4 pilots streams with 4Tx).
• Codebook-based schemes with precoded pilots constraint the use of MIMO schemes using the same precoder on different data subcarriers within the same PRU.
• Given that the USCCH will be transmitted in the DRU permutation in every subframe (most likely using SFBC), we can derive the following specific constraints– For AH schemes: USCCH must be decoded with the 4 streams pilot pattern that
has lower density per antenna than the 2 streams pilot pattern, even at the cell-edge.
– For codebook-based schemes: there is an upper limit of 2 for the MIMO rank and rate in tone-based DRU
• Limiting MIMO to SFBC and SM rank 2 in DRU only incurs not to optimize tone-based allocation usage by high SNR rank 3 and 4 users with somewhat high velocity. These users can still benefit from frequency diversity in PRU-based distributed allocations in CRU.
System operation constraint in paired-tone based DRU seems more acceptable for codebook-based
schemes than for antenna hopping schemes
Link-Level Simulation Results
Rate 2
DSFBC+AH, SM+AH, and SM+PC
PRU based-LDRU, QPSK, mVehA 30km/h
12/6
Goodput BLER
PRU based-LDRU, 16QAM, mVehA 30km/h
13/6
Goodput BLER
Tone based-LDRU, QPSK, mVehA 30km/h
14/6
Goodput BLER
Tone based-LDRU, 16QAM, mVehA 30km/h
15/6
Goodput BLER
PRU based-LDRU, QPSK, mPedB 3km/h
16/6
Goodput BLER
PRU based-LDRU, 16QAM, mPedB 3km/h
17/6
Goodput BLER
Tone based-LDRU, QPSK, mPedB 3km/h
18/6
Goodput BLER
Tone based-LDRU, 16QAM, mPedB 3km/h
19/6
Goodput BLER
PRU based-LDRU, QPSK, mVehA 120km/h
20/6
Goodput BLER
PRU based-LDRU, 16QAM, mVehA 120km/h
21/6
Goodput BLER
Tone based-LDRU, QPSK, mVehA 120km/h
22/6
Goodput BLER
Tone based-LDRU, 16QAM, mVehA 120km/h
23/6
Goodput BLER
Rate 1
SFBC+SM+AH, SFBC+SM+PC
Results of rate-1 schemesQPSK, SFBC+PC, SFBC+AH (1/6)
rate 1 SFBC schemes (mPedB 3km/h, QPSK 1/2, tDRU and pDRU, uncorrelatedchannel)
1.0E-03
1.0E-02
1.0E-01
1.0E+00
-1 0 1 2 3 4 5SNR[dB]
BLE
R
SFBC+PC,tDRU
SFBC+PC,pDRU
SFBC+AH, pDRU
SFBC+AH, tDRU
tDRU = tone-based LDRUpDRU = PRU-based DRU (e.g. based on CRU permutation)
Results of rate-1 schemesQPSK, SFBC+PC, SFBC+AH (2/6)
rate 2 SFBC schemes (mVehA 30km/h, QPSK 1/2, tDRU and pDRU, uncorrelated channel)
1.0E-03
1.0E-02
1.0E-01
1.0E+00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00
SNR[dB]
BLE
R
SFBC AH tDRU
SFBC AH pDRU
SFBC PC tDRU
SFBC PC pDRU
Rate 1rate 2 SFBC schemes (mVehA 30km/h, QPSK 1/2, tDRU and pDRU, uncorrelated channel)
1.0E-03
1.0E-02
1.0E-01
1.0E+00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00
SNR[dB]
BLE
R
SFBC AH tDRU
SFBC AH pDRU
SFBC PC tDRU
SFBC PC pDRU
Rate 1
Results of rate-1 schemesQPSK, SFBC+PC, SFBC+AH (3/6)
rate 2 SFBC schemes (mVehA 120km/h, QPSK 1/2, tDRU and pDRU, uncorrelated channel)
1.0E-03
1.0E-02
1.0E-01
1.0E+00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00
SNR[dB]
BLE
R
SFBC AH tDRU
SFBC AH pDRU
SFBC PC tDRU
SFBC PC pDRU
Rate 1rate 2 SFBC schemes (mVehA 120km/h, QPSK 1/2, tDRU and pDRU, uncorrelated channel)
1.0E-03
1.0E-02
1.0E-01
1.0E+00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00
SNR[dB]
BLE
R
SFBC AH tDRU
SFBC AH pDRU
SFBC PC tDRU
SFBC PC pDRU
Rate 1
Results of rate-1 schemesQPSK, SFBC+PC, SFBC+AH (4/6)
rate 1 SFBC schemes (mPedB 3km/h, QPSK 1/2, tDRU and pDRU, uncorrelatedchannel)
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
2.00
- 1.00 - 0.50 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00SNR[ dB]
Go
od
ut
SFBC+AH,tDRU
SFBC+AH,pDRU
SFBC+PC, tDRU
SFBC+PC, pDRU
Results of rate-1 schemesQPSK, SFBC+PC, SFBC+AH (5/6)
rate 1 SFBC schemes (mPedB 3km/h, QPSK 1/2, tDRU and pDRU, uncorrelated channel)
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
2.00
- 1.00 - 0.50 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00SNR[ dB]
Go
od
ut
SFBC+AH,tDRU
SFBC+AH,pDRU
SFBC+PC, tDRU
SFBC+PC, pDRU
(mVehA 30km/hrate 1 SFBC schemes (mPedB 3km/h, QPSK 1/2, tDRU and pDRU, uncorrelated channel)
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
2.00
- 1.00 - 0.50 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00SNR[ dB]
Go
od
ut
SFBC+AH,tDRU
SFBC+AH,pDRU
SFBC+PC, tDRU
SFBC+PC, pDRU
(mVehA 30km/h
Results of rate-1 schemesQPSK, SFBC+PC, SFBC+AH (6/6)
rate 1 SFBC schemes (mVehA 120km/ h, QPSK 1/ 2, tDRU and pDRU, uncorrelated channel)
1.0E- 02
2.1E- 01
4.1E- 01
6.1E- 01
8.1E- 01
1.0E+00
1.2E+00
1.4E+00
1.6E+00
1.8E+00
2.0E+00
- 1.00 - 0.50 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00
SNR[dB]
Go
od
pu
t
SFBC AH tDRU
SFBC AH pDRU
SFBC PC tDRU
SFBC PC pDRU
Rate 3
SFBC+SM+AH, SM+AH, and SM+PC
Rate 3 - Detailed of Simulated Schemes
– SM+AH: cf Appendix– SM+PC (case I): Precoders chosen as first three
columns of 16e’s V(4,4,6). Precoder changed every PRU.
– SM+PC (case II): 4 precoders with largest chordal distance from 4-bit DFT-based codebook (cf Appendix). Precoder changed every PRU.
Results of rate-3 schemesQPSK ½, SM+AH vs. SM+PC, MMSE, mPedB 3km/h, tone-based DRU (1/6)
rate 3 schemes (mPedB 3km/h, QPSK 1/2, tDRU, uncorrelated channel)
1.0E-02
1.0E-01
1.0E+00
2.00 2.50 3.00 3.50 4.00 4.50 5.00SNR[dB]
BLE
R
SM+AH.1RU CE
SM+PC(case I),1RU CE
SM+PC(case II), 1RUCE
Results of rate-3 schemesQPSK ½, SM+AH vs. SM+PC, MMSE, mPedB 3km/h, PRU-based DRU (1/6)
rate 3 schemes (mPedB 3km/h, QPSK 1/2, pDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00SNR[ dB]
BL
ER
SM+AH.1RU CE
SM+PC(case I),tDRU,1RU CE
SM+PC(case II),tDRU,1RU CE
Results of rate-3 schemesQPSK ½, SM+AH vs. SM+PC, MMSE, mVehA 30km/h, tone-based DRU (3/6)
rate 3 schemes (mVehA 30 km/h, QPSK 1/2, tDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
2.00 2.50 3.00 3.50 4.00 4.50 5.00SNR[ dB]
BL
ER
SM+AH.1RU CE
SM+PC(case I),1RU CE
SM+PC(case II ), 1RU CE
Results of rate-3 schemesQPSK ½, SM+AH vs. SM+PC, MMSE, mVehA 30km/h, PRU-based DRU (4/6)
rate 3 schemes (mVehA 30 km/h, QPSK 1/2, pDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00SNR[ dB]
BL
ER
SM+AH .1RU CE
SM+PC(case I),tDRU, 1RU CE
SM+PC(case II),tDRU, 1RU CE
Results of rate-3 schemesQPSK ½, SM+AH vs. SM+PC, MMSE, mVehA 120km/h, tone-based DRU (5/6)
rate 3 schemes (mVehA 120 km/h, QPSK 1/2, tDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00SNR[ dB]
BL
ER
SM+AH .1RU CE
SM+PC(case I),1RU CE
SM+PC(case II), 1RU CE
Results of rate-3 schemesQPSK ½, SM+AH vs. SM+PC, MMSE, mVehA 120km/h, PRU-based DRU (6/6)
rate 3 schemes (mVehA 120km/h, QPSK 1/2, pDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00SNR[ dB]
BL
ER
SM+AH .1RU CE
SM+PC(case I),tDRU,1RU CE
SM+PC(case II),tDRU,1RU CE
Results of rate-3 schemes16 QAM ½, SM+AH vs. SM+PC, MMSE, mPedB 3km/h, tone-based DRU (1/6)
rate 3 schemes (mPedB 3km/h, 16 QAM 1/2, tDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
8.00 8.50 9.00 9.50 10.00 10.50 11.00 11.50 12.00 12.50SNR[ dB]
BL
ER
SM+AH.1RU CE
SM+PC(case I),1RU CE
SM+PC(case II), 1RUCE
Results of rate-3 schemes16 QAM ½, SM+AH vs. SM+PC, MMSE, mPedB 3km/h, PRU-based DRU (2/6)
rate 3 schemes (mPedB 3km/h, 16QAM 1/2, pDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
8.00 9.00 10.00 11.00 12.00 13.00 14.00SNR[ dB]
BL
ER
SM+AH.1RU CE
SM+PC(case I),tDRU,1RU CE
SM+PC(case II),tDRU,1RU CE
Results of rate-3 schemes16 QAM ½, SM+AH vs. SM+PC, MMSE, mVehA 30km/h, tone-based DRU (3/6)
rate 3 schemes (mVehA 30 km/h, 16 QAM 1/2, tDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
8.00 8.50 9.00 9.50 10.00 10.50 11.00 11.50 12.00 12.50SNR[ dB]
BL
ER
SM+AH.1RU CE
SM+PC(case I),1RU CE
SM+PC(case II), 1RU CE
Results of rate-3 schemes16 QAM ½, SM+AH vs. SM+PC, MMSE, mVehA 30km/h, PRU-based DRU (4/6)
rate 3 schemes (mVehA 30 km/h, 16 QAM 1/2, pDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
8.00 8.50 9.00 9.50 10.00 10.50 11.00 11.50 12.00 12.50
SNR[dB]
BLE
R
SM+AH.1RU CE
SM+PC( case I),tDRU, 1RU CE
SM+PC(case II),tDRU, 1RU CE
Results of rate-3 schemes16 QAM ½, SM+AH vs. SM+PC, MMSE, mVehA 120km/h, tone-based DRU (5/6)
rate 3 schemes (mVehA 120 km/ h, 16 QAM 1/ 2, tDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
8.00 8.50 9.00 9.50 10.00 10.50 11.00 11.50 12.00 12.50
S NR[ dB ]
BL
ER
SM+AH.1RU CE
SM+PC(case I),1RU CE
SM+PC(case II), 1RU CE
Results of rate-3 schemes16 QAM ½, SM+AH vs. SM+PC, MMSE, mVehA 120km/h, PRU-based DRU (6/6)
rate 3 schemes (mVehA 120km/h,16 QAM 1/2, pDRU, uncorrelated channel)
1.0E- 02
1.0E- 01
1.0E+00
8.00 9.00 10.00 11.00 12.00 13.00 14.00
SNR[dB]
BLE
R SM+AH.1RU CE
SM+PC(case I),tDRU,1RU CESM+PC(case II),tDRU,1RU CE
Summary• Precoder cycling offers better or as good goodput than Antenna Hopping.• Complexity analysis shows an advantage to precoder cycling schemes. • DSFBC with MMSE receiver does not offer a performance advantage over
SM with MLD, but it requires a more complex receiver.• Hybrid SFBC+SM does not offer a performance advantage over SM with
MMSE receiver, but it requires a more complex receiver.
• For 4x2 – Rank 1– Precoder Cycling + precoded pilots is preferred to SFBC + AH +
common pilots• For 4x2 – Rank 2
– Precoder Cycling + SM + precoded pilots is preferred to SM + AH + common pilots and to DSFBC + AH + common pilots
• For 4x4 – Rank 3– Precoder Cycling + SM + precoded pilots is preferred to SM + AH +
common pilots and to SFBC + SM + AH + common pilots
Proposed SDD Text Changes
• Delete the bracketed rows in Table 4.
• Delete line 17 on page 100.
• Delete lines 8 to 11 on page 101.
• Apply the same changes in section 11.12.2.1.1 on page 119 in the UL MIMO section
Appendix – detailed simulation assumptions
OFDM parameters 10 MHz (1024 subcarriers)
Number of OFDM symbols
per subframe 6
Data burst size
The burst size is assumed to be:
PRU-based LDRU
- 4 RUs distributed over 48 PRUs
Paired-tone-based LDRU
- 4 RUs: subcarriers distributed over 24 distributed PRUs
Permutation Tone based LDRU and PRU based LDRU
Number of total RU in one
subframe 48
AoD No constraint in uncorrelated channel
Number of Antennas 4 transmitter, 2 receiver [4Tx, 2Rx] (for rates 1 and 2)
4 transmitter, 4 receiver [4Tx, 4Rx] (for rate 3)
Antenna configuration ULA: 4 lambda spacing
Angular spread 3 degree
Modulation/Coding QPSK 1/2, 16QAM 1/2 with LTE FEC
8 Turbo decoding iterations
Appendix – detailed simulation assumptions
Channel model
1. Uncorrelated (mandatory with first priority)
Modified Veh A 30 km/h
Modified Ped B 3 km/h
Modified Veh A 120 km/h
Modified Ped B and modified Veh A are defined in section 3.2.9 of IEEE
802.16m-08/004r3.
Channel estimation
MMSE channel estimator (Wiener filter) is assumed, and the r.m.s. delay of channel model is assumed to be known. There’s no unbiased operation. No pilot boosting. 1. Dedicated pilot (i.e. precoded by W)
Single-PRU level CE (channel estimation shall be done within 1PRU)
Two-PRU level CE (channel estimation shall be done within 2PRU)
2. Common pilot (i.e. non-precoded by W)
Single-PRU level CE (channel estimation shall be done within 1PRU)
Two PRU level CE (channel estimation shall be done within 2PRU)
MIMO detector MRC for rate 1 schemes LMMSE (for all ranks) MLD for SM with precoding with rank 2
Appendix – detailed simulation assumptions
SFBC + AH Rate 1 *
1 2*
2 1
s s
s s
z ,
1 0 1 0 1 0 0 0 0 0 0 0
0 1 0 0 0 0 1 0 1 0 0 0, , , , ,
0 0 0 1 0 0 0 1 0 0 1 0
0 0 0 0 0 1 0 0 0 1 0 1
W
D-SFBC + AH Rate 2
*34
*43
*12
*21
ss
ss
ss
ss
z ,
1 0 0 0 1 0 0 0 1 0 0 0
0 1 0 0 0 0 1 0 0 0 1 0, ,
0 0 1 0 0 1 0 0 0 0 0 1
0 0 0 1 0 0 0 1 0 1 0 0
W
SM + AH Rate 2
1
2
s
s
z ,
1 0 0 0 1 0 0 0 1 0 0 0
0 1 0 0 0 0 1 0 0 0 1 0, , , , ,
0 0 1 0 0 1 0 0 0 0 0 1
0 0 0 1 0 0 0 1 0 1 0 0
W
Appendix – detailed simulation assumptions
SM + AH Rate 3 1
2
3
s
s
s
z ,
1 0 0 1 0 0 1 0 0 0 0 0
0 1 0 0 1 0 0 0 0 1 0 0, , ,
0 0 1 0 0 0 0 1 0 0 1 0
0 0 0 0 0 1 0 0 1 0 0 1
W
SFBC + SM + AH Rate 3
*1 2
*2 1
3 5
4 6
s s
s s
s s
s s
z ,
0010
0001
1000
0100
,
1000
0100
0010
0001
W
SM + PC Rate 2 (precoder cycling)
4 precoders with large chordal distance from 4-bit DFT-based codebook
1 1 3 3:, 2 3 , :, 2 4 , :, 1 3 , :, 1 4
1 1 1 1 1 1 1 1
1 1 1 1 1 11 1 1 1, , ,
1 1 1 1 1 1 1 12 2 2 2
1 1 1 1 1 1
W W W W W
j jW
j j
Appendix – detailed simulation assumptions
Pilot type in tone-based DRU Pilot type in PRU-based DRU
SFBC + AHD-SFBC + AHSFBC + SM + AH
Non-precoded pilots- common
Non-precoded pilots- common
Precoder cycling Precoded pilots- shared
Precoded pilots- dedicated