Eliminating Channel Feedback in Next Generation Cellular Networks - - PowerPoint PPT Presentation

▶
eliminating channel feedback in next generation cellular
SMART_READER_LITE
LIVE PREVIEW

Eliminating Channel Feedback in Next Generation Cellular Networks - - PowerPoint PPT Presentation

Apr 06, 2024 597 likes •1.03k views

Eliminating Channel Feedback in Next Generation Cellular Networks Deepak Vasisht Swarun Kumar, Hariharan Rahul, Dina Katabi Cellular Traffic is Increasing Global mobile data traffic will increase 8 fold in 2015-2020 CISCO 30 (Exabytes/month)

slide-1
SLIDE 1

Eliminating Channel Feedback in Next Generation Cellular Networks

Deepak Vasisht Swarun Kumar, Hariharan Rahul, Dina Katabi

slide-2
SLIDE 2

Cellular Traffic is Increasing

Global mobile data traffic will increase 8 fold in 2015-2020 CISCO

10 20 30 2015 2016 2017 2018 2019 2020 Data Demand (Exabytes/month)

Spectrum cannot accommodate this increase

slide-3
SLIDE 3

LTE standard body, 3GPP, is proposing multi-antenna solutions in new releases:

  • Beamforming
  • Coordinated Multi-point
  • Full-Dimensional MIMO

Base station needs to know channels to client

More Antennas

slide-4
SLIDE 4

Channel Acquisition

Use feedback from the client

Feedback overhead is overwhelming

…

slide-5
SLIDE 5

Feedback is Overwhelming

  • Large in current networks, uses lossy compression [3GPP TS

36.211 2010, Irmer et al IEEE Communications 2011]

  • Prohibitive for future deployments with up to 32 antennas
  • According to LTE standard body, 3GPP:

“Identifying the potential issues of CSI acquisition and developing the proper solutions are of great importance”

slide-6
SLIDE 6

R2F2

  • Uses uplink channels to estimate downlink channels
  • Removes feedback overhead
  • Evaluated indoors and outdoors in white spaces

640 660 680 700 720 740

Frequency (MHz) R2F2 testbed Commercial Carriers

slide-7
SLIDE 7

Idea: Use Reciprocity Like in WiFi

In WiFi, Uplink Channel = Downlink Channel

slide-8
SLIDE 8

Idea: Use Reciprocity Like in WiFi

Does not work for cellular networks: Uplink and downlink on different frequencies

In WiFi, Uplink Channel = Downlink Channel

slide-9
SLIDE 9

Problem Statement

How do we estimate channels on one frequency from channels on a different frequency?

slide-10
SLIDE 10

Problem Statement

Uplink Channels at Frequency 1 Downlink Channels at Frequency 2

slide-11
SLIDE 11

Idea: Same Paths on Uplink & Downlink

Uplink Channels at Frequency 1 Downlink Channels at Frequency 2 Paths along which signal is received

slide-12
SLIDE 12

RF-based Localization Systems

−1 −0.5 0.5 1 0.5 1

cos θ Amplitude

600 𝑁𝐼𝑨 User 𝜄 Base Station

slide-13
SLIDE 13

RF-based Localization Systems

−1 −0.5 0.5 1 0.5 1

cos θ Amplitude

−1 −0.5 0.5 1 0.5 1

cos θ Amplitude

600 𝑁𝐼𝑨 650 𝑁𝐼𝑨 User 𝜄 Base Station

Localization systems don’t directly apply

slide-14
SLIDE 14

Idea: Same Paths on Uplink & Downlink

Uplink Channels at Frequency 1 Downlink Channels at Frequency 2 Paths along which signal is received

slide-15
SLIDE 15

Paths to Channels: Ideal Representation

User Base Station 𝜄)

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

𝜚) 𝜚+

slide-16
SLIDE 16

Paths to Channels: Measured Representation

User Base Station 𝜄)

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

Limited number

  • f antennas leads

to convolution with sinc

𝜚) 𝜚+

𝑇-(𝑏), 𝜚), 𝜄)) 𝑇-(𝑏+, 𝜚+, 𝜄+)

slide-17
SLIDE 17

Paths to Channels: Superposition

User Base Station 𝜄)

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

𝑇- 𝑏), 𝜚), 𝜄) + 𝑇-(𝑏+, 𝜚+, 𝜄+)

slide-18
SLIDE 18

Paths to Channels: FFT

User Base Station 𝜄)

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

ℎ) F 𝐺𝐺𝑈(𝑇- 𝑏), 𝜚), 𝜄) + 𝑇-(𝑏+, 𝜚+, 𝜄+))

slide-19
SLIDE 19

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

User

Base Station

ℎ) ℎ+

𝜄)

F F

Uplink to Downlink Channels

Uplink (f) Downlink (f’)

slide-20
SLIDE 20

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

User

Base Station

F F

Uplink to Downlink Channels

Uplink (f) Downlink (f’)

ℎ) ℎ+ ? ?

𝜄)

slide-21
SLIDE 21

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

User

Base Station

F

Channels to Paths

Uplink (f)

ℎ) Goal: To find a set of paths, that can produce channels ℎ) Recall: Each path is represented by (𝑏, 𝜚, 𝜄)

𝜄)

slide-22
SLIDE 22

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

User

Base Station

F

Channels to Paths

Uplink (f)

ℎ) Goal: To find {𝑏7, 𝜚7, 𝜄7}79)

: , that can produce channels ℎ) 𝜄)

Recall: Each path is represented by (𝑏, 𝜚, 𝜄)

slide-23
SLIDE 23

Channels to Paths

Goal: To find {𝑏7, 𝜚7, 𝜄7}79)

: , that can produce channels ℎ)

ℎ;<= = 𝐺𝐺𝑈 ? 𝑇- 𝑏7, 𝜚7, 𝜄7

: 79)

{𝑏7, 𝜚7, 𝜄7}79)

: = 𝑏𝑠𝑕𝑛𝑗𝑜{EF,GF,HF} ℎ) − ℎ;<= +

slide-24
SLIDE 24

Getting Paths from Wireless Channels

  • Optimization is non-linear and constrained
  • Solved using standard interior point method
  • Approximate initialization using RF-localization methods
slide-25
SLIDE 25

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

−1 −0.5 0.5 1 0.2 0.4 0.6 0.8 1

cos θ Amplitude

User

Base Station

F F

Uplink to Downlink Channels

Uplink (f) Downlink (f’)

ℎ) ℎ+

𝜄)

slide-26
SLIDE 26

Evaluation

Goal: To measure the accuracy of R2F2 channel estimates

slide-27
SLIDE 27

Experimental Setup

  • Used USRP N210 software radios as clients and base stations
  • Implemented a 5 antenna LTE base station
  • Located base station close to a commercial base station
slide-28
SLIDE 28

Frequency Separation

  • Used frequencies from 640 to 690 MHz in the White Spaces
  • Evaluation at 30 MHz Uplink-Downlink separation
  • Same as major AT&T and Verizon deployments

640 660 680 700 720 740

Frequency (MHz) R2F2 testbed Commercial Carriers

slide-29
SLIDE 29

100 m 50 m

Indoor Testbed

Base Station Client

slide-30
SLIDE 30

80 m 60 m

Outdoor Testbed

Base Station Client

slide-31
SLIDE 31

Beamforming

slide-32
SLIDE 32

Beamforming

slide-33
SLIDE 33

Beamforming Comparison

0.2 0.4 0.6 0.8 1

  • 5

5 15 25 CDF SNR (dB)

No Beam Ground Truth (Explicit Feedback) R2F2

R2F2 delivers 90% of the MIMO SNR gains, with zero feedback

slide-34
SLIDE 34

Beamforming Comparison: Data Rate

0.2 0.4 0.6 0.8 1 10 20 30 40 50 60 CDF Datarate (Mbps)

No Beam Ground Truth R2F2

R2F2’s achieves 1.7x data rate improvement

slide-35
SLIDE 35

Comparison with RF-localization

0.2 0.4 0.6 0.8 1

  • 5

5 15 25 CDF SNR (dB)

No Beam Ground Truth R2F2 RF-Loc

Delivers only 40% of MIMO SNR gains

slide-36
SLIDE 36

Effect of Frequency Separation

1 2 3 4 5 6 7 8 10 20 30 40 50 SNR Gain (dB) Frequency Separation (MHz)

slide-37
SLIDE 37

Application: Edge Client Nulling

slide-38
SLIDE 38

Application: Edge Client Nulling

BS 1 BS 2 Client 2 Client 1

slide-39
SLIDE 39

Edge Nulling

0.2 0.4 0.6 0.8 1

  • 5

5 10 15 CDF INR(dB) Original After Nulling

  • 5. 3 dB
slide-40
SLIDE 40

Related Work

  • Cellular Networks: Channel feedback compression [Shuang

et al VTC 11, Rao et al 14, Xu et al Access IEEE 14], Statistical channel prediction across frequency bands [Han et al CHINACOM 10, Hugl et al COST 02…]

  • Beyond Cellular Networks: Channel quality prediction [Sen

et al Mobicom 13, Shi et al NSDI 14, Radunovic et al CONEXT 11…], Temporal channel predictions [Cao et al PMRC 04, Wong et al GLOBECOM’05, Dong et al GLOBECOM’01]

slide-41
SLIDE 41

Conclusion

  • R2F2 estimates channels on one frequency from channels on a

different frequency

  • R2F2 accurately estimates downlink LTE channels from uplink

LTE channels

  • R2F2 enables MIMO techniques for FDD systems with zero

channel feedback