Channel Estimation using Least Mean Square

(LMS) Algorithm for LTE-Advanced

Saqib Saleem, Qamar-ul-Islam


Abstract: For IMT-Advanceds high data rate requirement for the internet and multimedia services, 3GPP has proposed
evolved version of LTE, known as LTE-Advanced. To achieve the targets for next generation mobile communications
systems, the following systems enhancements are proposed in Rel-10: Carrier Aggregation, Co-ordinated Multipoint
Transmission and Reception (CoMP), Relaying Capability, Advanced MIMO techniques and Heterogeneous Networks. In
order to achieve high spectral efficiency and high cell edge throughput, Channel State Information (CSI) is desired to be
known as the both ends of transceiver. Channel can be estimated in time-domain and frequency-domain. For multi-
antenna transmission systems under high mobility conditions when channel is fast fading frequency selective, channel
needs to be estimated at each instance. Under these situations, adaptive algorithms can be used to have knowledge of
channel. In this paper, the behavior of Least Mean Square (LMS) algorithm is determined and the evaluation parameters
used are number of channel taps and CIR samples of the channel. Monte-Carlo Simulations are carried for the
performance and complexity comparison of LMS-based channel estimation for MIMO-OFDM system.

Keywords: MIMO-OFDM, LMS, CIR Samples, Channel Taps, LTE, IMT-Advanced

1 INTRODUCTION
For 4Gs IMTAdvanced systems requirements, e.g.
lessenergyconsumptionperbittransmission,allover
the world service provisioning, flexible usage of
frequencybands, common network architecture,
interworkingwiththeexisting2Gand3Gradioaccess
networks[1]3GPPmadeacorenetworkconsistingof
evolved packet core (EPC), EUTRA and EUTRAN,
whichisgenerallyknownasLTEAdvanced.Byusing
thesystemrequirementsasgiveninTR25.912andTR
25.913,apeakdatarateof326.4Mbpsfor4 4MIMO
and 100 Mbps for SISO system can be achieved by
using 64 QAM modulation technique for DL data
transmission.ButforULcase86.4Mbpsdataratecan
be achieved by using 64 QAM and 57.6 Mbps for 16
QAM and for QPSK, maximum data rate supported
can be 50 Mbps in case of SISO system model [2]. As
compared to HSPA, spectral efficiency of LTE
Advancedis34timesgreaterforDLand23timesfor
UL.InRel10,mobilitysupportfor350500Km/hrand
control plane latency less than 5 ms are under
consideration.[3].

Saqib Saleem is with Institute of Space Technology, Islamabad. He is
currently working as Lecturer in Department of Electrical
Engineering. His areas of interest are Channel Estimation,
Spectrum Sensing, LTE-Advanced.

Dr. Qamar-ul-Islam is also with Institute of Space Technology,
Islamabad. He is currently working as Head of Department of
Electrical Engineering.
In order to make LTEA competitive of IMTA, the
following new techniques are proposed: the
requirementof100MHzbandwidthfor1Gbpscanbe
achieved by carrieraggregation of contiguous and
noncontiguous band, to achieve optimized diversity
gains by using MIMO and beamsteering techniques,
a cell with six sectors, each sector having four
antennasispreferredascomparedtoacellwiththree
sectorsandeachsectorhavingeightantennas,forsoft
handoversthemacrodiversitycanbeachievedbyco
operative MIMO to increase or decrease the system
capacity, coveragearea can be increased with
minimizing interference by using relay which can
decode the data before transmission [4]. To eliminate
the intersymbol interference caused by multipath
fadingchannels,equalizersareusedwhichdependon
thechannelimpulseresponse(CIR)knowledge.Three
kinds of channel estimation techniques are proposed
in [5,6,7]. In first technique, channel can be estimated
by using timedomain statistics of the slow varying
channel under low mobility conditions. Least Square
Error (LSE) and Linear Minimum Mean Square Error
(LMMSE) algorithms can be used for estimating
channel.LSEhaslesscomplexitythanLMMSEdueto
the matrix inversion and dependency on channel
statistics but the performance of LMMSE is better.
DFT method can be used as channel estimator in
frequencydomain. Under noninteger multipath
delays, DCT technique can be used to avoid the high
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 12, DECEMBER 2011, ISSN 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 102
frequency component. The complexity of frequency
domain estimator can be reduced by neglecting the
components having values less than the noise by
using a windowing function on DFTCE. By
considering the channel as first order AR process,
adaptive filtering schemes can be used for channel
estimation. In [7], Kalman filtering based channel
estimator is optimized for different channel filter
lengthsandmultipaths.Inthispaper,LMSalgorithm
isusedfortimevaryingchannelandperformanceand
complexity is analyzed for varying channel impulse
response samples and the multipaths for different
MIMOsystemsaccordingtoLTEAdvanced.
The paper is organized as follows: Physical
Layer of LTEA is explained in Section II and LMS
channel estimation algorithm is given in Section III
with the simulation results given in next section. In
lastsectionconclusionsaredrawn.

2 PHYSICAL LAYER OF LTE-ADVANCED
Physical signals generated by Layer 1of LTEA are
used for following purposes: synchronization, cell
identification, channel state estimation. Reference
Signals (RS), also known as pilot signals, are used to
detect any variation in amplitude and phase of the
received signal [1]. In LTEA, for DL the multiple
access technique used is OFDMA and SCFDMA is
used for UL, both for FDD and TDD modes.
Irrespective of different multiple access techniques
used, LTEA uses same frame structure for UL and
DL. Frame used for FDD mode is of 10 ms duration
containing10subframes.ForULandDLsameframe
structureisfollowed.ForTDDmode,Subframe0and
5 are allocated only for DL transmission while sub
frame 2 is used for UL transmission purposes. Pilot
signals, DwPTS for DL and UpPTS for UL, are
transmittedinsubframe1[8].
ForDLsubframe,OFDMsymbolcanbegeneratedby
[1]

s
I
(t) = o
k,I
c
2]nk(t-N
CP
1
s
)
N1
s
_
N
BW
DL
2
_
_
N
BW
DL
2
_

Where0t<(NCP+N)Ts.WhereTsissymbolduration
andNCPiscyclicprefixlength.o
k,I
isresourceatomfor
l
th
OFDMsymbolandk
th
carrierfrequency.

ForUL,theSCFDMAsignalcanbegeneratedby

s
I
(t) = o
,I
c
-]
2n]
N
TX
. c
]
2n(]+N
stcrt
)
N
d
1
s
(t-N
CP,l
1
s
)
N
TX
-1
=0

N
Tx
2
, [
]=j
N
Tx
2
, [

In LTEA, the data transmission can be by up to four

layers for UL and eight layers for DL case. For two
antenna transmission scheme, 3 bit precoding
codebookisusedfordatamultiplexingforULwhile6
bit precoding codebook is used for four transmit
antennas[8].ToachieveULtransmitdiversity,Single
Antenna Port Mode is defined for PUCCH, PUSCH
and SRS transmissions. For UL control channel with
two transmit antennas, the transmit diversity can be
achieved by Spatial Orthogonal Resource Transmit
Diversity (SORTD) while for PUSCH, OFDM signal
precoded using DFT technique is used. For LTEA,
two reference signal are used for UL : Demodulation
Reference Signal (DRS) and Sounding Reference
Signal (SRS). DRS is precoded but SRS is not pre
coded [2]. UL power control is used to reduce the
interference under slow varying channels and to
lessen the path loss. In DL multiuser MIMO
techniquesaresupportedwithouttheconfigurationof
RRC. The reference signals supported in DL are used
for PDSCH demodulation and Channel State
information(CSI)estimation.

3 LMS BASED CHANNEL ESTIMATION
Toavoidthe matrixinversion,involvedinLSEand
LMMSE [18], LMS algorithm can be used to solve
WienerHolfequation,whichmayor maynotrequire
aprioristatisticalinformationofthechannelanddata.
AsummaryofLMSalgorithmisgivenasfollows

1 LSE method is applied to get the initialized, H

r,t
LS
,
usedforfirstiteration.
2 After finding the filter coefficients, the channel
estimationbecomes

r,t
LMS
|n] = W

H
|n]H

r,t
LS
|n]

Where

,t
LS
|n] = |E

,t
LS
|n] E

,t
LS
|n - 1] E

,t
LS
|n -1 + H]]

WhereLNSfilterhaslengthM.

JOURNAL OF COMPUTING, VOLUME 3, ISSUE 12, DECEMBER 2011, ISSN 2151-9617

https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 103
3 Atn
th
iteration,theerrorisgivenby

E|n] = E

,t
LS
|n] - E

,t
LMS
|n]|n]
4 Fromthiserror,coefficientscanbeupdatedby
w|n +1] = w|n] +pE

,t
LS
|n]E

|n]
Where the value of stepsize parameter p depends on
thecorrelationbetweenthedata.
5 After updating the coefficients. The weighted
errorisgivenby

e|n] = w|n] - w|n]

4 SIMULATION RESULTS
MonteCarlo Simulations for different MIMO
systems employing OFDM as modulation and
demodulation techniques are carried out. FFT size of
128point and channel length of 64 with CP length of
16 is taken in this system. The performance
comparisonofLMSchannelestimatorasafunctionof
CIR samples for different SNR values is shown in
Figure1.FromFigure1,itisclearthatforanychannel
length, the performance is better under low SNR
operating conditions. As we go on increasing CIR
samples, MSE also increases. So for better
performance, less number of CIR samples for low
SNR values are preferred. The effect of channel filter
length on MSE for different LMS estimators is shown
in Figure 2. We note that performance remain same
for LMS and LeakyLMS estimators for all CIR
samples. The complexity of LMS estimator as a
functionofCIRsamplesisgiveninTable4.6.

Fig.1 MSE vs CIR Samples for LMS Estimator

By increasing CIR samples from 5 to 10, the
complexity increases 20%. While further increase of
CIR samples to 20, there is 60% increment in
complexity. The combined effect of SNR and CIR
samplesonperformanceisshowninFigure3.MSEfor
different MIMO schemes is shown in Figure 4. The
performance is better for 2 2 system than S S
and 4 4 systems. Irrespective of CIR samples, low
order MIMO scheme results in better performance.
MSEbehaviorremainssameforCIRsampleslessthan
5,afterthattheperformancedegrades,almostlinearly
forincreasingCIRsamples.


Fig.2 MSE vs CIR Samples for different LMS Estimators



Fig.3 MSE vs SNR vs CIR Samples for LMS Estimator
0 10 20 30 40 50 60 70
0
2
4
6
8
10
12
x 10
4
MSE v/s CIR Samples of LMS Estimator for 2 x 2 System
CIR Samples
M
S
E


SNR=5 dB
SNR=10 dB
SNR=15 dB
SNR=20 dB
SNR=25 dB
10 20 30 40 50 60
1.2
1.4
1.6
1.8
2
2.2
x 10
4
MSE v/s CIR Samples of LMS Estimator for 2 x 2 System
CIR Samples
M
S
E


LMS
Leaky-LMS
0
20
40
60
80
5
10
15
20
25
0
2
4
6
8
10
12
x 10
4
CIR Samples
MSE v/s SNR v/s CIR Samples of LMS EStimator for 2 x 2 System
SNR
M
S
E
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 12, DECEMBER 2011, ISSN 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 104

TABLE 1: COMPLEXITY COMPARISON OF LMS ESTIMATOR FOR DIFFERENT MIMO SCHEMES
CIR Samples 2 2 (pscc)

LS-LMS LMMSE-LMS
S S (pscc)

LS-LMS LMMSE-LMS
4 4 (pscc)

LS-LMS LMMSE-LMS
5 68.65 180.6 147 395 260 705.5
10 80.045 178.83 176.6 440.8 453 753
20 106.23 286.16 238 525 470 1100

TABLE 2: COMPLEXITY OF LMS ESTIMATOR FOR DIFFERENT CHANNEL TAPS FOR 2 2 SYSTEM
Channel Taps Time (pscc)
5 255.2
10 269.6
20 321.4

Figure5showstheperformanceofLMSforthecases
when initially channel estimator is LS and LMMSE.
The performance of LMMSELMS is better than LS
LMSbecauseinfirsttechnique,secondorderchannel
statistics are exploited due to which this method
results in more complexity as given in Table 4.7. By
increasingCIRsamplesfrom5to10inLMMSELMS,
the complexity increases by 8% while in LSLMS this
increment was 20%. While the increment in
complexity is 18.91% when increasing CIR samples
from 5 to 20 but in case of LSLMMSE it was 60%.
Table 1 also demonstrates that for 5 CIR samples the
complexityincreasesby167%incaseofLMMSELMS
as compared to LSLMS. While for 10 CIR samples,
this increment is 140% and this value reduces to 96%
for 20 CIR samples. So the larger the number of CIR
Samples, the increment will be less for LMMSELMS
schemethanthatofLSLMS.Thecomputationaltime
for different MIMO systems for both LMMSELMS
and LSLMS schemes is shown in Table 1. More
computational time results for higher order MIMO
schemes e.g. S S scheme results in 115% more
computational time for both LSLMS and LMMMSE
LMScasesascomparedto2 2.Whilefor4 4,the
increment is almost 290%. The performance as a
functionofchanneltapsforLMSisshowninFigure6.

TABLE3:COMPLEXITYOFLMSESTIMATORFOR
DIFFERENTMIMOSCHEMES
ChannelTaps 2 2(pscc) S S (pscc) 4 4
(pscc)
5 258 260.6 467
10 291 269 526
20 340 470 565

By increasing the channel taps, the

performance also goes on degrading for all SNR
values.Soforbetterperformanceandlesscomplexity,
small number of channel taps are proposed, as for
large number of channel taps not only the
performance degrades but complexity also increases
as given in Table 2. The increment of channel taps
value from 5 to 10 results in 5% more computational
time while 20 channel taps gives 26% more
complexity. . Performance as a function of SNR and
channel taps is shown in Figure 7. The performance
for different MIMO systems as a function of channel
taps is shown in Figure 8. The low order MIMO
systems give better performance for all channel taps.
For less order MIMO scheme, the effect of increasing
the channel taps on performance is not so significant
as for 2 2 case but as we increase the order of
MIMO system, the performance degrades, almost, as
alinearfunctionofincreasingchanneltaps.

Fig.4 MSE vs CIR Samples for different MIMO Schemes

2 4 6 8 10 12 14 16 18 20
5.62
5.64
5.66
5.68
5.7
5.72
5.74
x 10
4
MSE v/s CIR Samples of LMS Estimator for 2 x 2 System
CIR Samples
M
S
E


2 x 2 System
3 x 3 System
4 x 4 System
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 12, DECEMBER 2011, ISSN 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 105
Fig.5 MSE vs CIR Samples for LS-LMS


Fig.6 MSE vs Channel taps for LMS and LMMSE-LMS
Estimator

5 CONCLUSION
Smalllengthofchannelfilterispreferrednotonlyfor
better performance but also for less complexity for
lowSNRvaluesandfor loworderMIMOsystems. If
complexity can be compromised then performance
can be made even better by taking the initially
estimated channel using LMMSE method. Similar to
channelfilter length,lessnumber ofchanneltapsnot
only gives better performance but also less
complexity. To achieve the data rate targets of a
wireless communication system through channel
feedback, LMS channel estimator is optimized for a
systemwithchannelfilterlengthof510CIRSamples
and channel taps less than 10 for both optimized
performanceandcomplexity.


Fig.7 MSE vs SNR vs Channel Taps for LMS Estimator


Fig. 8 MSE vs Channel Taps for different MIMO Schemes


5 10 15 20 25
1.02
1.022
1.024
1.026
1.028
1.03
1.032
1.034
1.036
1.038
x 10
5
MSE v/s CIR Samples of LMS Estimator for 2 x 2 System
CIR Samples
M
S
E


LS-LMS
LMMSE-LMS
0 10 20 30 40 50 60 70
0
2
4
6
8
10
12
14
x 10
6
MSE v/s Channel Taps of LMS Estimator for 2 x 2 System
Channel Taps
M
S
E


SNR=5 dB
SNR=10 dB
SNR=15 dB
SNR=20 dB
SNR=25 dB
0
20
40
60
80
5
10
15
20
25
0
2
4
6
8
10
12
14
x 10
6
Channel Taps
MSE v/s SNR v/s Channel Taps of LMS EStimator
SNR
M
S
E
0 10 20 30 40 50 60 70
2
4
6
8
10
12
14
16
x 10
5
MSE v/s Channel Taps of LMS Estimator for different MIMO Systems
Channel Taps
M
S
E


2 x 2 System
4 x 4 System
3 x 3 System
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 12, DECEMBER 2011, ISSN 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 106
REFERENCES
[1] 3GPP, TS 36.211 V0.1.2. (200611), Physical Channels
andModulation(Release8).Available:www.3gpp.org
[2] 3GPP, TR 36.814 V9.0.0. (201003), EUTRA, Further
advancements for EUTRA Physical Layer Aspects
(Release9).Available:www.3gpp.org
[3] 3GPP,TS36.912V9.2.0(201003),FeasibilityStudyfor
Further Advancements for EUTRA (LTE
Advanced)(Release9).Available:www.3gpp.org
[4] 3GPP, TS 36.913 V9.0.0. (200912), Requirements for
further advancements for EUTRA (LTEAdvanced)
(Release9).Available:www.3gpp.org
[5] Saqib Saleem, QamarulIslam, Optimization of LSE
and LMMSE Channel Estimation Algorithms based on
CIR Samples and Channel Taps, IJCSI International
Journal of Computer Science Issues Vol. 8
Issue.1,pp.437443,January2011
[6] SaqibSaleem,QamarulIslam,LMSandRLSChannel
Estimation Algorithms for LTEAdvanced, Journal of
Computing,Vol.3,Issue.4,pp.155163,April2011.
[7] M.A.Mohammadi,M.Ardabilipour,Performance
Comparison of RLS and LMS Channel Estimation
Techniques with Optimum Training Sequences for
MIMOOFDMSystems,9781424419807,IEEE2008
[8] Seongwook Song, Andrew C.Singer and KoengMo
Sung,Soft Input Channel Estimation for Turbo
Equalization, IEEE Transaction on Signal Processing,
Vol.52,No.10,October2004

JOURNAL OF COMPUTING, VOLUME 3, ISSUE 12, DECEMBER 2011, ISSN 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 107