Filter Bank Multi-Carrier (FBMC) for Future Wireless Systems ...FBMC OFDM (LTE,WLAN) Current...

Filter Bank Multi-Carrier (FBMC)for Future Wireless Systems

CD Laboratory Workshop

Ronald NisselNovember 15, 2016

Dependable Wireless Connectivity for the Society in Motion

Motivation

Slide 2 / 27


Multicarrier Modulation

xl,k...Transmitted data symbol

00

13

17

Time index, k

Freq

uenc

yin

dex,

l

−2 −1 0 1 2

0

0.5

1

1.5

p(t)

tT0

−5 0 5

−50

0

P(f

)

f

F0= f T0

FBMCOFDM(LTE,WLAN)

Current generation of wireless systems use OFDM.Can we do better?

Slide 3 / 27


5G Design Paradigm: Flexible Time-Frequency Allocation

Time

Freq

uenc

y

Robust communcation(reduced spectral efficiency)

Low latency, high velocity(high efficiency if L >> K)

Large area(high efficiency if K >> L)

Required time-frequencyresources for one symbol

L ... Number of subcarriers per blockK ... Number of time-symbols per blockF ... Frequency-spacingT ... Time-spacing

F

T ∝ 1F

Slide 4 / 27


How does FBMC work?

Slide 5 / 27


FBMC: Time-Frequency Localized Prototype Filter

(Complex) orthogonal for TF = 2

Slide 6 / 27


FBMC Offset-QAM: Transmit Real-Valued Symbols

Orthogonal only in the real domain: TF = 0.5

Slide 7 / 27


Comparison of Multicarrier Schemes

MaximumSpectral

EfficiencyTF = 1

Time-Localization

Frequency-Localization

(Bi)-Orthogonal

IndependentTransmitSymbols

PureOFDM yes yes no yes yes

Windowed/FilteredOFDM

no yes yes yes yes

FBMC-OQAM yes yes yes realonly yes

CodedFBMC-OQAM yes yes yes

yes,after de-

spreadingno

Slide 8 / 27


Comparison of Multicarrier Schemes

−50 −40 −30 −20 −10 0 10 20 30 40 50

−100

−80

−60

−40

−20

0

Normalized Frequency, f/F

Power

SpectralDensity

(dB)

−50 −40 −30 −20 −10 0 10 20 30 40 50

−100

−80

−60

−40

−20

0

Normalized Frequency, f/F

Power

SpectralDensity

(dB)

LTE like:TF = 1.07

No OOBreduction

TF = 1.3

TF = 1.3

CP-OFDM

WindowedOFDM

FilteredOFDM

FBMCOQAMTF = 1(complex)

21 sub-carriers

Slide 9 / 27


Real World Testbed Measurements

Slide 10 / 27


Vienna Wireless Testbed

TX antenna

RX

Transmitter:

Receiver:

Slide 11 / 27


Channel Estimation in FBMC1

1“On Pilot-Symbol Aided Channel Estimation in FBMC-OQAM”,R. Nissel,et.al.,2016Slide 12 / 27


Channel Estimation in FBMC

24

68

1012

14

24

68

1012

0

0.5

1

1.5

Channel

Time-positio

n k

Subcarrier-position l

Data symbolsPilot symbols

Time, t

Frequency, f LS estimation,

Interpolation

Slide 13 / 27


Channel Estimation in FBMC

Cancel the imaginary interference at the pilot positions:

Time

Frequency

1 Auxiliary 2 Auxiliary Coding

Novel Contribution:I We suggest the usage of two auxiliary symbols.I We extended the coding approach to more than 8 symbolsI Downloadable MATLAB code: www.nt.tuwien.ac.at/downloads/I Validated by real world measurements

Slide 14 / 27

www.nt.tuwien.ac.at/downloads/


Achievable Rate Improvement: FBMC vs. OFDM (LTE)

−10 −5 0 5 10 15 20 25 30

0

5

10

15

20

25

30

SNROFDMD

(dB)

Achieva

bleRate

Improvem

ent(%

)

1 Auxiliary symbol(low complexity)

2 Auxiliary symbols(low complexity)

Coding(high complexity)

OFDM no cyclic prefix

Improvement due tobetter spectrum utilization

Improvement due tono cyclic prefix

FBMC

Slide 15 / 27


Real World Throughput Measurements

−10 −5 0 5 10 15 20 25 30

0

5

10

15

20

25

30

SNROFDMD

(dB)

ThroughputIm

provem

ent(%

)

1 Auxiliary symbol(low complexity)

2 Auxiliary symbols(low complexity)

Coding(high complexity)

FBMC

95 % confidenceinterval obtainedby bootstrapping

mean

Slide 16 / 27


MIMO in FBMC2

2“Enabling Low-Complexity MIMO in FBMC-OQAM”, R. Nissel and M. Rupp, 2016Slide 17 / 27


Enabling MIMO in FBMC By Hadamard Spreading (+1,-1)

Coded FBMC-OQAM:

Time

Code

Frequency

Novel Contribution:I Matrix formulation: new analytical insightsI We investigate the block interference and the influence of time-varying channelsI Downloadable MATLAB code: www.nt.tuwien.ac.at/downloads/I Validated by real world measurements

Slide 18 / 27



Real World MIMO Measurements

−5 0 5 10 15 20 25 30

10−4

10−3

10−2

10−1

Estimated Signal-to-Noise Ratio for OFDM (dB)

BitErrorRatio

OFDMFBMC

−5 0 5 10 15 20 25 30

10−4

10−3

10−2

10−1

Estimated Signal-to-Noise Ratio for OFDM (dB)

BitErrorRatio

95 % confidenceinterval for FBMCobtained bybootstrapping 2×1 MIMO

Alamouti’s Space-TimeBlock Code

2×2 MIMOMaximum-LikelihoodDetection (2×bit rate)

2×2 MIMO (2×bit rate)Zero-Forcing Equalizer

16-QAM

Slide 19 / 27


Bit Error Probability for FBMC in Doubly SelectiveChannels3

3“OFDM and FBMC-OQAM in Doubly-Selective Channels: Calculating the Bit ErrorProbability”, R. Nissel and M. Rupp, 2016, submitted

Slide 20 / 27


Closed-Form BEP Expression

−5 0 5 10 15 20 25 30 35

10−3

10−2

10−1

100

BitErrorProbability,BitErrorRatio

TheoryGauss approx.Simulation

−5 0 5 10 15 20 25 30 35

10−3

10−2

10−1

100

SSiiggnnaall--ttoo--NoNoiissee RRaattiioo [[ddBB]],, sseeee ((1188))

BitErrorProbability,BitErrorRatio

OFDM (noCP)

CP-OFDM

FBMC

Vehicular A, 500km/h

OFDM (noCP)

CP-OFDM ≈FBMC

Doubly- flat

Pedestrian A, 10km/h64-QAM

Slide 21 / 27


Outlook

Slide 22 / 27


Doubly-Selective Channels in FBMC

Questions:I Are one-tap equalizers good enough?I How good is FBMC compared to OFDM?I Which prototype filter

I HermiteI PHYDYAS

Fair comparison:

Sign

al-t

o-In

terf

eren

ceR

atio

Subcarrier Spacing

Optimal Subcarrier Spacing

Limiting factor:Doppler spread

(time-variant channel)

Limiting factor:delay spread

(multipath delays)

Slide 23 / 27


Doubly-Selective Channels in FBMC

510

15

24

68

1012

0

0.5

1

1.5

OFDM Symbol

LS estimation of the piecewise mean channel

Subcarrier

|ˆ hLS

l,k|

510

15

24

68

1012

0

0.5

1

1.5

OFDM Symbol

LS estimation of the piecewise mean channel

Subcarrier

|ˆ hLS

l,k|

50100

150200

24

68

1012

0

0.5

1

1.5

time index

MMSE channel estimation

Subcarrier

|H[l,m]|

50100

150200

24

68

1012

0

0.5

1

1.5

time index

MMSE channel estimation

Subcarrier

|H[l,m]|AMMSE

I Channel estimation using time-frequency correlation

I Channel equalization

Slide 24 / 27


Further Questions in FBMC

Bit error probability, including channel estimationI Similar to our submitted paper (which assumes perfect channel knowledge),I But now also include channel estimation.

mmWave transmissionsI TestbedI Jointly with Erich and Martin

Further investigate real world hardware effectsI QuantizationI Non-linear power amplifier

Slide 25 / 27


Conclusions

Slide 26 / 27


Conclusions

In contrast to most other research groups:I Downloadable MATLAB code: www.nt.tuwien.ac.at/downloads/I Theory is validated by real world testbed measurements

Main contribution in 2016:I Channel estimation in FBMCI MIMO in FBMCI Bit error probability for FBMC in doubly-selective channels

Outlook: A lot of open questions for the next years!

Slide 27 / 27



Backup Slides

Slide 1 / 9


Ambiguity Function OFDM

Slide 2 / 9


Ambiguity Function CP-OFDM

Slide 3 / 9


Block InterferenceNo guard-symbols:

KT KT KT

Block 1 Block 2 Block 3

Time

Tran

smit

Pow

er

Guard time-slot:

KT T KT T KT

Block 1 Block 2 Block 3

Set time slotto zero

Time

Tran

smit

Pow

er

Slide 4 / 9


Block Interference

0 50 100 150 200 250

0

10

20

30

40

50

Spreading Length (Symbols)

Signal-to-InterferneceRatio(dB)

0 50 100 150 200 250

0

10

20

30

40

50


Signal-to-InterferneceRatio(dB) One guard slot per block

(reduced efficiency)

No guard symbols(maximum spectral efficiency)

1 ms for 15 kHz subcarrier spacing, (125µs for 120 kHz)

Slide 5 / 9


Block Interference

0 50 100 150 200 250

0

5

10

15

20


Efficiency

Loss

(%)

0 50 100 150 200 250

0

5

10

15

20


Efficiency

Loss

(%)

One guard slot per block

LTE efficiency loss due to the CP

1ms for 15 kHz subcarrier spacing, (125µs for 120 kHz)

Slide 6 / 9


Time-Variant Channels

0 10 20 30 40 50

0

10

20

30

40

Signal-to-Interference

Ratio(dB)

Velocity (km/h) for F=15 kHz and fc=2.5 GHz

TheorySimulation

0 50 100 150 200 250 300 350 400


0 1 2 3 4 5 6 7

x 10−3Normalized maximum Doppler shift νmax/F

0 10 20 30 40 50

0

10

20

30

40


Ratio(dB)


Spreading Length K = 8

Spreading Length K = 32

One guard time slot per block

Slide 7 / 9


Flexible Subcarrier Spacing, TF = 1 (no CP)

Channel Model: VehicularA, 250 km/h

5 10 15 20 25 30 35 40 45

0

5

10

15

20

25

30

35

Subcarrier Spacing (kHz)


Ratio(dB)

TheorySimulation

5 10 15 20 25 30 35 40 45

0

5

10

15

20

25

30

35

Subcarrier Spacing (kHz)


Ratio(dB)

OFDM

FBMC-OQAMHermite

FBMC-OQAMPhydyas

15 kHz(LTE)

Slide 8 / 9


Peak-To-Power Average Ratio and Signal Power Over Time

Pr(P

APR

≤PA

PR0)

6 8 10 12

10−4

10−3

10−2

10−1

100

PAPR0 (dB)0 0.1 0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Sign

alPo

wer

Time (ms)

1Aux.1Aux.flipped

2Aux.Cod.

6TFBMC

1Aux.

1Aux.flipped

2Aux.

Cod.

OFDM

Slide 9 / 9

Date post:	26-Feb-2021
Category:	Documents
Upload:	others
View:	4 times
Download:	0 times

Filter Bank Multi-Carrier (FBMC) for Future Wireless Systems ...FBMC OFDM (LTE,WLAN) Current...

Documents