Folie 1 - IEEE Standards Association

Folie 1 - IEEE Standards Association

Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs) Submission Title: The THz Channel Model in Wireless Data Center Date Submitted: 10 Match 2015 Source: Bile Peng Company TU Braunschweig Address Schleinitzstr. 22, D-38102 Braunschweig, Germany Voice:+495313912405, FAX: +495313915192, E-Mail: [email protected] Re: n/a Abstract: This contribution presents some preliminary THz channel modeling results in the future wireless data center scenario. A series of ray tracing simulations are conducted for different channel types. The RMS delay spread and the RMS angular spread are employed as the metric of the multipath richness. A stochastic channel model is developed based on the simulation results and is validated by the ray tracing simulation results. Purpose: Contribution towards developing a wireless data center channel model for use in TG 3d Notice: This document has been prepared to assist the IEEE P802.15. It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein. Release: The contributor acknowledges and accepts that this contribution becomes the property of IEEE and may be made publicly available by P802.15. Submissio n Slide 1 Bile Peng (TU Braunschweig). Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center A Stochastic THz Channel Model in Wireless Data Centers Bile Peng, Thomas Krner TU Braunschweig Submissio n Slide 2 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center

Contents Motivation Ray Tracing Simulation Results Stochastic Channel Model Conclusion Submissio n Slide 3 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Motivation The data center link is responsible for the cooperation between computers and must achieve very high data rates. The data center link is prevailingly wired. However, the wireless link has some significant advantages [1]: More flexibility Less maintenance cost More space for cooling The high data rates of Terahertz (THz) communications makes it a competitive candidate.

This report is a preliminary PHY layer feasibility study of the application of the THz communication in the data center wireless backhaul. Submissio n Slide 4 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Radio Wave Propagation Paths [2,3] Ceiling Reflector Type 1: LoS Type 2: NLoS Type 3: Adjacent casings Submissio n Slide 5 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Selection of Propagation Path Type (1/2) If transmitter and receiver are on the same or adjacent casings, they can be positioned lower than the casing roof to reduce the interference. Reflector Submissio n Slide 6 Bile Peng (TU Braunschweig)

Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Selection of Propagation Path Type (2/2) If transmitter and receiver are close enough, we use the NLoS path with reflection on the ceiling. 1 The short distance compensates for the reflection loss. The AoD/AoA elevations are far from the horizonal direction, which reduces the interference on the LoS paths. Criterion: the elevation () is at ) is at least 2 times Half-Power-Beamwith away from the horizontal direction. Otherwise we select the LoS path. Submissio n Ceiling 2 Slide 7 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Simulation Environment Transmitter

Typical data center (source: http://www.enterprisetech.com/wp-content/uploads/2014/ 11/SIO_DataCenter_Rows1.jpg) Wall Receiver Casing Propagation path Ray tracing simulation Submissio n Slide 8 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Contents Motivation Ray Tracing Simulation Results Stochastic Channel Model Conclusion Submissio n Slide 9 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Statistical Characteristics With Type 1/2 Type1/2: LoS/nLoS channels between 2 nonadjacent casings Multipath richness metric: RMS delay spread with omniantenna

Parity pattern due to reflections on the casing roof RMS delay spread RMS delay spread 10 server in y 2 4 6 8 2 6 4 4 8 5 10 server in x 15 ns 8 6 Tx 6 4 8

2 10 Submissio n 10 server in y Tx 2 10 0 5 Slide 10 10 server in x 15 0 ns Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Impact of Directive Antenna RMS delay spread RMS delay spread 10 4 6 8

2 6 4 4 8 10 5 10 server in x 15 ns server in y server in y 2 10 6 6 2 8 0 10 4 2 5 Omniantenna

8 10 server in x 15 0 ns Directive phased array Antenna: 4x4 phased array The directive antenna reduces the RMS delay spread significantly. Submissio n Slide 11 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Statistical Characteristics With Type 1/2 Type1/2: LoS/nLoS channels between 2 nonadjacent casings Multipath richness metric: RMS angular spread with omniantenna Parity pattern due to reflections on the casing roof RMS angular spread RMS angular spread Tx server in y 40 4 50

2 50 2 server in y 60 60 30 6 40 4 Tx 30 6 20 20 8 8 10 10 10 5 Submissio n 10 server in x 15 10 0

5 Slide 12 10 server in x 15 0 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Impact of Directive Antenna RMS angular spread RMS AoD elevation spread 60 50 2 40 4 30 6 50 2 server in y server in y 60 40 4 30 6

20 20 8 8 10 10 5 10 server in x 15 10 10 0 5 Omniantenna 10 server in x 15 0 Directive phased array Antenna: 4x4 phased array The directive antenna reduces the RMS angular spread significantly as well. Submissio n Slide 13 Bile Peng (TU Braunschweig) Match 2015

doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Statistical Characteristics With Type 3 Type 3: channels between 2 adjacent casings Randomly generated adjacent Tx and Rx The RMS delay spread is lower than the in type 1/2 because of the limited propagation space. 0.7 1 0.6 0.8 Frequency Frequency 0.5 0.4 0.3 0.6 0.4 0.2 0.2 0.1 0 0 1 2 3 4 RMS delay spread (ns) 5 0 0

6 Omniantenna Submissio n 0.05 0.1 0.15 RMS delay spread (ns) 0.2 Directive phased array Slide 14 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Statistical Characteristics With Type 3 Type 3: channels between 2 adjacent casings Randomly generated adjacent Tx and Rx The RMS angular spread is lower than the in type 1/2 because of the limited propagation space. 0.7 0.6 Frequency 0.5 0.4 0.3 0.2 0.1 0 Submissio n 0 10

20 30 RMS angle spread Slide 15 40 50 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Contents Motivation Ray Tracing Simulation Results Stochastic Channel Model Conclusion Submissio n Slide 16 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Stochastic Channel Model 1. 2. 3. 4. 5. 6. 7. Determine number of paths. Determine delay for each path.

Determine pathloss according to delay. Determine angles. Generate uniformly distributed phases. Generate frequency dispersions (Friis law). Generate polarisations. Submissio n Slide 17 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Numbers of Paths Type 1/2, Tx 1 (in corner) LoS Number of paths 1 Probability 100% Reflections Number of paths 17 18 19 20 21 Probability (%) 27 35 22

15 1 Submissio n Slide 18 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Numbers of Paths Type 1/2, Tx 2 (in center) LoS Number of paths 1 Probability 100% Reflections Number of paths 16 17 18 19 20 21 Probability (%) 32 29 12

16 8 3 Submissio n Slide 19 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Numbers of Paths Type 3 (Adjacent casings) LoS Number of paths 1 Probability 100% Reflections Number of paths 3 4 5 6 7 8 9 10 11

Probability (%) 22 13 8 15 8 17 8 6 3 Submissio n Slide 20 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Delay Distribution: type 1/2, Tx 1 LOS Reflection 0.2 0.4 0.35 0.3 Probability Probability 0.15 0.1 0.25

0.2 0.15 0.05 0.1 0.05 0 0 20 40 Delay (ns) 60 0 0 50 100 Relative delay (ns) Path Distribution Parameters LOS Normal distribution =2.26e-8, =8.76e-9 NLOS Negative EXP =4.26e7e7 Submissio n Slide 21 Bile Peng (TU Braunschweig) Match 2015

doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Delay Distribution: type 1/2, Tx 2 LOS Reflection 0.2 0.25 0.2 Probability Probability 0.15 0.1 0.05 0 0 0.15 0.1 0.05 10 20 Delay (ns) 30 0 0 50 100 Relative delay (ns) Path Distribution

Parameters LOS Normal distribution =1.20e-8, =4.56e-9 NLOS Normal distribution =2.98e-8, =1.79e-8 Submissio n Slide 22 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Delay Distribution: type 3 Reflection 0.35 0.3 0.3 0.25 0.25 Probability Probability LOS 0.35 0.2 0.15 0.2 0.15 0.1

0.1 0.05 0.05 0 0 2 4 Delay (ns) 6 0 0 50 100 Relative delay (ns) Path Distribution Parameters LOS Normal distribution =1.80e-8, =8.60e-9 NLOS Negative EXP =4.92e7 Submissio n Slide 23 Bile Peng (TU Braunschweig) Match 2015

doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Delay-Pathloss Correlation: type 1/2, Tx 1 Path Deterministic part Random part (Norm.) LOS p=-20log10(d)-71.52 =0 NLOS pr=-0.294dr-17.44 =4 Submissio n Slide 24 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Delay-Pathloss Correlation: type 1/2, Tx 2 Path Deterministic part Random part (Norm.) LOS p=-20log10(d)-71.52 =0 NLOS pr=-0.385dr-17.95

=4 Submissio n Slide 25 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Delay-Pathloss Correlation: type 3 Path Deterministic part Random part (Norm.) LOS p=-20log10(d)-71.52 =0 NLOS pr=-0.429dr-30.3 =6e7 Submissio n Slide 26 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Pathloss-Angle Correlation Angular difference ( )

Since we want to reduce the multipath effect by highly directive antenna, the propagation paths with low pathloss and similar Angle of Arroval (AoA) to LOS path has a negative impact on the system design. There is no appropriate distribution to describe the relation, therefore we use the correlation matrix. 180 0.45 160 0.4 140 0.35 120 0.3 100 0.25 80 0.2 60 0.15 40 0.1 20 0.05 0 -70 Submissio n Slide 27

-60 -50 -40 -30 -20 Relative Pathloss (dB) -10 0 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Stochastic Channel Example -90 -100 180 LoS path 160 140 Angular difference ( ) Path gain (dB) -110 -120 -130 -140 -150 120 100 80 -160 40 -170 20

-180 0 0.5 1 Time (s) 1.5 0 -160 2 x 10 -8 Channel impulse response Submissio n LoS path 60 -150 -140 -130 -120 Pathloss (dB) -110 -100 Pathloss-angle distribution Slide 28 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center

Validation via RMS Delay Spread 0.35 0.5 0.3 0.4 Frequency Frequency 0.25 0.2 0.15 0.3 0.2 0.1 0.1 0.05 0 0 2 4 6 8 RMS delay spread (ns) 10 12 0 0 Ray Tracing simulation Submissio n 2 4 6 8

RMS delay spread (ns) 10 12 Stochastic channel model Slide 29 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Validation via RMS Angular Spread 0.45 0.4 0.4 0.35 0.35 0.3 Frequency Frequency 0.3 0.25 0.2 0.15 0.25 0.2 0.15 0.1 0.1 0.05 0.05 0

0 10 20 30 40 50 RMS angle spread (ns) 60 70 80 0 0 Ray Tracing simulation 10 20 30 40 50 RMS angle spread (ns) 60 70 80 Stochastic channel model The similar distribution of RMS delay/angle spreads validate the stochastical model. Submissio n Slide 30 Bile Peng (TU Braunschweig)

Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Contents Motivation Ray Tracing Simulation Results Stochastic Channel Model Conclusion Submissio n Slide 31 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center Conclusion The THz communication is a competitive solution for the next generation wireless data center. A ray tracing simulation environment is set up to investigate the channel characteristics. The multipath propagation is a major hurdle of the high speed error free data transmission and the RMS delay/angular spread is used as metric of the multipath richness. A stochastic channel model is developed according to the ray tracing simulation results. Submissio

n Slide 32 Bile Peng (TU Braunschweig) Match 2015 doc.: 15-15-0207-00-003d Stochastic Channel Model for Wireless Data Center List of References 1. 2. 3. T. Krner, Literature review on requirements for wireless data centers doc.: IEEE 802.15-13-0411-00-0thz_Literature Review Zhang W et. al, 3D beamforming for wireless data centers, in Proceedings of the 10th ACM Workshop on Hot Topics in Networks. 2011 K. Ramchadran60 GHz Data-Center Networking: Wireless Worry less?, 2008 Submissio n Slide 33 Bile Peng (TU Braunschweig)

Recently Viewed Presentations

  • NOTETAKING AND STUDY SKILLS The SQRW Method SQRW

    NOTETAKING AND STUDY SKILLS The SQRW Method SQRW

    The SQRW Method SQRW - Survey, Question, Read, Write What exactly is SQRW? Well, it's a study method where you get the best out of textbook study. It's a 4-step strategy and each letter denotes one step so let's go...
  • Partnerships - ODU

    Partnerships - ODU

    Partnerships Introduction to Forms of Business and Formation of Partnerships Operation of Partnerships ... starting a novel, starting a business…it's keeping the thing going that is difficult." Prue Leith, author and executive, quoted in The Adventure Capitalists (Grout and Curry,...
  • UNINETTs SIP-leveranser til sektoren i 2011 sip:Vidar.Faltinsen@uninett.no sip:Jardar.Leira@uninett.no

    UNINETTs SIP-leveranser til sektoren i 2011 sip:[email protected] sip:[email protected]

    Thor-Inge Næsset, IT-sjef, NHH. Tord Tjeldnes, IT-direktør, UiA. Finansiering. Betales over medlemsavgiften (gjelder UH-institusjonene) ... Tore Kristiansen. Eksterne konsulenter fra Norge og Sverige. Førstelinjesupport 24x7 av UNINETT driftsenter.
  • New Jersey Center for Teaching and Learning Progressive

    New Jersey Center for Teaching and Learning Progressive

    Ounces (oz), pounds (lbs), and tons measure weight. Examples of Metric Units. are centimeters (cm), meters (m), and kilometers (km) which measure distance. Grams (g), kilograms (kg), and metric tons (t) measure weight. You can use ratios to convert US...
  • KÖZGAZDASÁGTAN

    KÖZGAZDASÁGTAN

    Meggazdagodás alaplogikája OIKONOMIA Á-(P)-Á' CRHMATISTIKÉ P-(Á-Á')-P' 2. Bármilyen egyéb szempont pl. Szabad/ decentralizált pl. HUNgary 1000 előtt kapitalizmus pl. HUN 1989 után Nem szabad/centralizált feudalizmus pl. HUN középkorban Miért ez no.1. (Ki mondja meg mi a fontos)? mert a meggazdagodás...
  • Construction Surveys - University of Washington

    Construction Surveys - University of Washington

    Construction Surveys List of Figures Readings: Chapter 23 Figures used: 23-1 to 23-10. Material to be considered: 23-7. Stages and Measurements of CS Design stages: topographic Surveying, and site maps Construction stage: Setting-out and positioning works. 60% of surveying hours.
  • Benchmarking - Center for Computation and Technology

    Benchmarking - Center for Computation and Technology

    The term "benchmark" also commonly applies to specially-designed programs used in benchmarking A benchmark should: be domain specific (the more general the benchmark, the less useful it is for anything in particular) be a distillation of the essential attributes of...
  • Access to Environmental Justice: the ongoing Influence of EU Law

    Access to Environmental Justice: the ongoing Influence of EU Law

    Heather Hill Management Company Clg & anor -v- An Bord Pleanála & anor [2019] IEHC 186. Simons J. - 1st judgement on current iteration of s.50B. JR of decision of ABP to grant PP for Strategic Housing Development pursuant to...