Boos$ng’the’Spectral’Eﬃciency’’ of’Wireless’Networks...

Boos$ng the Spectral Efficiency of Wireless Networks via Spa$al Mul$plexing

with pCell Technology

BWRC, University of California, Berkeley April 1st, 2016

Patents, Patents Pending April 1, 2016

Antonio Forenza, PhD Co-‐founder & CTO, Artemis Networks

Outline

•  Background •  SINR analysis of volumes of coherent signal •  Descrip$on of pCell SDR wireless plaRorm •  Experimental results •  Conclusions

April 1, 2016 Patents, Patents Pending

Forecast of Mobile Data Traffic


Global Mobile Data Traffic (Actual/Projected)

ExaB

ytes/m

onth

[Cisco’16] “Cisco visual network index: global mobile data traffic forecast update 2015-‐2020” February 2016

CAGR = 40-‐70%

Solu$ons to Spectrum Crunch


•  3 solu$ons [Qualcomm, SKT, Nokia] •  More cell density •  More spectrum •  More spectral efficiency

“New forecasts show much more spectrum is needed.” “Operators will con$nue to deploy new cells, including small cells, and will con$nue to upgrade networks to state-‐of-‐the-‐art 4G LTE technology to take advantage of improved spectral efficiency, but that is not enough.” [CTIA’15] CTIA report to FCC, June 22, 2015

Spectrum

Cell density

Spectral efficiency

Current performance

Future performance

[3GPP] K. Mallinson, “2020 vision for LTE”, June 2012

More Cell Density

•  Small cells perform poorly •  Intercell interference [Andrews’15] •  Handoff overhead [3GPP-‐TR36.839]

•  Solu$ons •  SON and ICIC [AT&T’12][Mao’08] •  CoMP [3GPP-‐TR36.819][ChinaMob’11]

•  High cost, difficult deployment •  Zoning permits •  Backhaul


“AT&T has dropped plans to deploy 40,000 small cells on its network by the end of 2015” Fierce Wireless, March 2015

[Andrews’15]

[BS/Km3]

[bps/H

z/m

2 ]

0

BS-‐to-‐BS distance of 50m

More Spectrum

Patents, Patents Pending April 1, 2016 [Rappaport’14]

•  Na$onal Broadband Plan [FCC’10] •  547 MHz available in 2010 for mobile broadband •  Allocate 500 MHz by 2020

•  LTE-‐U [Forum’15][Qualcomm’15] •  Operate LTE in ISM unlicensed 5GHz band •  Use 200 MHz of spectrum (+300 MHz for future use)

•  Millimeter-‐wave [Heath’16] •  Pros: large bandwidth

•  BW=4GHz @ 28 and 38 GHz •  BW=7GHz @ 60 GHz

•  Cons: pathloss, body absorp$on (blocking) •  Short-‐range links up to 200m

[Qualcomm’15]

More Spectral Efficiency

•  Massive-‐MIMO •  Theory: asympto$cally noise/interference-‐free links [Marzeqa’10][Larsson’14] •  Requires upgrade to 4G LTE standard, proposed for Pre-‐5G [3GPP-‐Rel.13] •  Prac$cal only at mmWave, complex RF design •  Centralized cellular architecture

•  Pilot contamina$on •  Coordina$on between cells

•  4-‐8x mux gain with 64-‐128 antennas [ZTE’16][Intel’16]

•  pCell •  Theory: dirty paper coding [Caire’03][Viswanath’03][Yu’04] •  Compliant with standard LTE Rel.8 devices, readily deployable •  Works at any frequency band, simple frequency-‐agile RRHs (pWaves) •  Distributed architecture, no cells •  16x mux gain with 24 antennas, can scale to higher orders


4 users, SNR=10dB

pCell C-‐RAN

[Samsung’12] [Larsson’14]

Downlink Spectral Efficiency 5 MHz, 2-‐antenna LTE devices

pCell Performance Today


5 10 15 20 25 30 35 40 45 50 55 60

Spectral Efficien

cy (b

ps/H

z)

pCell LTE HSPA+

[Rysavy’14] Rysavy Research/4G Americas, August 2014

1.2 bps/Hz per sector

59.3 bps/Hz per area ≥1 m2

1.7 bps/Hz per sector

Presented results

Cellular vs. pCell Coverage


Base Sta$on User Equipment Power distribu$on

Cellular Coverage

pCell Coverage

Cellular vs. pCell SINR


Base Sta$on User Equipment SINR (darker is higher)

pCells

Real-‐world Cellular SINR

Real-‐world pCell SINR

Cellular Layout, Sparse Users


Cellular Layout, Clustered Users


Serendipitous Layout, Sparse Users


Data Rate Performance

April 1, 2016

200mW Max 64-‐QAM

500mW Max 256-‐QAM

Outline



System and Channel Models

•  System model: N transmiqers, U users •  Channel model [Svantesson’00]

•  Volume of coherent signal


SINR at Displaced Loca$on


ZF precoding

Spherical Bessel func$on of the first kind and order Legendre polynomial of degree

•  With linear precoder •  Mul$pole expansion [Poon’05]

•  Exact expression

Dimension of Volume of Coherent Signal


•  Approxima$on for •  LOS components only and

•  Radius of volume of coherent signal

kr <<1

N =U

Envelope SINR

April 1, 2016

Centralized transmiqers (ULA with λ/2 spacing)

Distributed transmiqers N = 10, U = 8

Patents, Patents Pending

Volumes of Coherent Signal


N = 16, U = 10, SINRo = 5dB

Outline



Hardware Architecture


LTE UEs

Internet

pCell Data Center

Sowware Architecture


Internet

pCell Processing

IP Routing

0 1

2

3

4

5

6

7

A

E F

B C

D

H

G

pCell Data Center

Fronthaul

pCell antenna

LTE UE

pCell

eNB

VRI 0

eNB

VRI 1

eNB

VRI 2

eNB

VRI 3

eNB

VRI 4

eNB

VRI 5

eNB

VRI 6

eNB

VRI 7

TDD LTE Frame


SF #0 SF #1 SF #2 SF #3 SF #4 SF #5 SF #6 SF #7 SF #8 SF #9 SRS 1 SRS 2 SRS 1 SRS 2

1 msec

Outline



Indoor Trials


16 iPhone 6 Pluses in 1 Square Meter


1m x 1m Plexiglass table

Measured LTE MAC-‐level SE


Uplink (Uniform Peak: 27.5 bps/Hz)

18 meters

Downlink (Avg. 59.3, Peak 59.8 bps/Hz)

14 m

eters

Outdoor Trials


0.5 to 5W PA

Power (if Fiber)

•  Roowops in downtown SF •  600 roowops available •  58 roowops for first trial •  About 500 pWaves for beta-‐tes$ng

pWave 1.0: 600-‐6000 MHz

Patents, Patents Pending

•  0.5 to 5W PA •  GPS or Ethernet sync •  PoE or Fiber

April 1, 2016

pWave Installa$ons


Power (if Fiber)

LTE Antennas

Fiber, power

pWaves

Frequency-‐agile RRH

Outline



Conclusions

•  Higher SE is the prac$cal solu$on to “spectrum crunch”

•  Distributed antennas enable confined volumes with peak SINR

•  Volumes for mul$ple users yield large spa$al mul$plexing gain

•  Large mul$plexing gain with a deployable SDR wireless plaRorm


References


•  [Artemis’15] S.G. Perlman and A. Forenza, “An introduc$on to pCell”, Feb. 2015. hqp://www.rearden.com/artemis/An-‐Introduc$on-‐to-‐pCell-‐White-‐Paper-‐150224.pdf •  [Cisco’16] Cisco, “Cisco visual network index: global mobile data traffic forecast update, 2015-‐2020”, White paper, Feb. 3, 2016

hqp://www.cisco.com/c/en/us/solu$ons/collateral/service-‐provider/visual-‐networking-‐index-‐vni/mobile-‐white-‐paper-‐c11-‐520862.html •  [Qualcomm] “The 1000x mobile data challenge”, Nov. 2013 hqps://www.qualcomm.com/inven$on/1000x •  [SKT] H. Park, “Efficient spectrum resource usage for next-‐genera$on N/W”, SK Telecom, 3GPP workshop on Rel.12 and onwards, Jun. 2012

hqp://www.3gpp.org/wp/workshop/2012-‐06-‐11_12_RAN_REL12/Docs/RWS-‐120020.zip •  [Nokia] “Support up to 1000 $mes more capacity”, hqp://networks.nokia.com/innova$on/technology-‐vision/1000x-‐capacity •  [3GPP] K. Mallinson, “2020 vision for LTE”, 3GPP, June 2012 hqp://www.3gpp.org/news-‐events/press-‐clippings/1261-‐2020-‐vision-‐for-‐lte •  [CTIA’15] Sawanobori, T., Roche, R., “Mobile Data Demand: Growth Forecasts Met”, June 22, 2015. CTIA.

hqp://www.c$a.org/docs/default-‐source/default-‐document-‐library/062115mobile-‐data-‐demands-‐white-‐paper.pdf •  [Andrews’15] A.K. Gupta, X. Zhang, J.G. Andrews, “Poten$al throughput in 3D ultradense cellular networks,” IEEE Asilomar Conf. on Sign. Syst. and Comp., Nov. 2015

[3GPP-‐TR36.839] 3GPP, “Mobility Enhancements in Heterogeneous Networks (Release 11),” TR 36.839, v11.0.0, Sep. 2012 hqp://www.3gpp.org/dynareport/36839.htm •  [AT&T’12] “AT&T deploying SON technology as part of latest network upgrades”, Feb. 2012 hqp://www.cellular-‐news.com/story/Operators/53297.php •  [Mao’08] X. Mao, A. Maaref, K. H. Teo “Adap$ve Sow Frequency Reuse for Inter-‐Cell Interference Coordina$on in SC-‐FDMA Based 3GPP LTE Uplinks,” Globecom, 2008 •  [3GPP-‐TR36.819] 3GPP TR 36.819, “Coordinated mul$-‐point opera$on for LTE physical layer aspects (Rel. 11)”, Dec. 2011, hqp://www.qtc.jp/3GPP/Specs/36819-‐b10.pdf •  [ChinaMob’11] “C-‐RAN, the road towards green RAN”, Oct. 2011 hqp://labs.chinamobile.com/cran/wp-‐content/uploads/CRAN_white_paper_v2_5_EN.pdf, p, 22 •  [AT&T’15] Fierce Wireless, “AT&T drops goal of deploying 40,000 small cells on its network by end of 2015,” Mar. 2015

hqp://www.fiercewireless.com/story/aq-‐drops-‐goal-‐deploying-‐40000-‐small-‐cells-‐end-‐2015-‐ci$ng-‐benefits-‐leap-‐de/2015-‐03-‐05 •  [FCC’10] FCC, “Connec$ng America: the na$onal broadband plan”, p.84 hqp://download.broadband.gov/plan/na$onal-‐broadband-‐plan.pdf •  [Forum’15] LTE-‐U Forum, “Coexistence study for LTE-‐U SDL,” LTE-‐U technical report, Feb. 2015

hqp://www.lteuforum.org/uploads/3/5/6/8/3568127/lte-‐u_forum_lte-‐u_technical_report_v1.0.pdf •  [Qualcomm’15] “LTE-‐U/LAA, MuLTEfire and Wi-‐Fi, making best use of unlicensed spectrum,” Sep. 2015

hqps://www.qualcomm.com/media/documents/files/making-‐the-‐best-‐use-‐of-‐unlicensed-‐spectrum-‐presenta$on.pdf

References


•  [Heath’16] R.W. Heath, “Millimeter wave for 5G finding a new spectrum mother load,” Invited talk, Jan. 2016 hqp://www.ece.utexas.edu/~rheath/presenta$ons/2016/MillimeterWave5GFindingNewSpectrumMotherLoad.pdf

•  [Rappaport’14] T. Rappaport, “Millimeter wave cellular communica$ons,” IEEE Comm. Theory Workshop, May 2014 hqp://www.ieee-‐ctw.org/2014/slides/session1/Ted_Rappaport_CTW2014.pdf

•  [Marzeqa’10] T. L. Marzeqa, “Noncoopera$ve cellular wireless with unlimited numbers of base sta$on antennas,” IEEE Trans. on Wireless Commun., vol. 9, no. 11, pp. 3590–3600, Nov. 2010.

•  [Larsson’14] E. Larsson, O. Edfors, and T.L. Marzeqa, “Massive MIMO for next genera$on wireless systems”, IEEE Comm. Magazine, pp.186-‐195, Feb. 2014. •  [3GPP-‐Rel.13] D. Flore, “Ini$al priori$es for the evolu$on of the evolu$on of LTE in Release-‐13”, 3GPP RAN, Sep. 20, 2014.

hqp://www.3gpp.org/news-‐events/3gpp-‐news/1628-‐rel13 •  [ZTE’16] “ZTE wins two awards at MWC”, Feb. 2016 hqps://www.youtube.com/watch?v=tO1a6ESsO5w •  [Intel’16] Intel “How will Massive MIMO Improve 5G Data Transfers? Intel Explains”, Feb. 2016 hqps://www.youtube.com/watch?v=k3ZqaxikXdk •  [Samsung’12] Y. Kim, “Full dimension MIMO and beamforming technologies for B4G,” Samsung, Oct. 2012 •  [Caire’03] G. Caire and S. Shamai (Shitz), “On the achievable throughput of a mul$antenna Gaussian broadcast channel,” IEEE Trans. on Inform. Theory, vol. 49, pp. 1691–

1706, July 2003. •  [Viswanath’03] S. Vishwanath, N. Jindal, and A. Goldsmith, “Duality, achievable rates, and sum-‐rate capacity of Gaussian MIMO broadcast channels,” IEEE Trans. on Inform.

Theory, vol. 49, pp. 2658–2668, Oct. 2003. •  [Yu’04] W. Yu and J.M. Cioffi, “Sum capacity of Gaussian vector broadcast channels,” IEEE Trans. on Inform. Theory, vol. 50, pp. 1875–1892, Sept. 2004. •  [Svantesson’00] T. Svantesson, “A study of polariza$on diversity using an electromagne$c spa$o-‐temporal channel model,” Proc. IEEE Veh. Tech. Conf., vol. 1, Sep. 2000. •  [Poon’05] A.S.Y. Poon, R.W. Brodersen, and D.N.C. Tse, “Degrees of freedom in mul$ple-‐antenna channels: A signal space approach,” IEEE Trans. Info. Th., Feb. 2005.

Thank you


pCell Videos at www.artemis.com


Boos$ng’the’Spectral’Eﬃciency’’ of’Wireless’Networks...

Documents

Transcript of Boos$ng’the’Spectral’Eﬃciency’’ of’Wireless’Networks...