Resource Blocks for EPoC Considerations. Avi Kliger, BZ Shen, Leo Montreuil Broadcom

Similar documents
Details on Upstream Pilots and Resource Block Configuration for EPoC

Baseline Proposal for EPoC PHY Layer

Baseline Proposal for EPoC PHY Layer IEEE 802.3bn EPoC September 2012 AVI KLIGER, BROADCOM LEO MONTREUIL, BROADCOM ED BOYD, BROADCOM

PROPOSAL FOR PHY SIGNALING PRESENTED BY AVI KLIGER, BROADCOM

Updates to 802.3bn EPoC Upstream Framing Proposal. Avi Kliger, Leo Montreuil Broadcom

EPoC Downstream Pilot Proposal. Christian Pietsch, Qualcomm Avi Kliger, Broadcom

SYMBOL SIZE CONSIDERATIONS FOR EPOC BASED OFDM PHY. Avi Kliger, Leo Montreuil, Tom Kolze Broadcom

PHY High Level Block Diagrams and First Pass Look at PHY Delays. Avi Kliger, Mark Laubach Broadcom

OFDMA PHY for EPoC: a Baseline Proposal. Andrea Garavaglia and Christian Pietsch Qualcomm PAGE 1

Multiple Downstream Profile Implications. Ed Boyd, Broadcom

IEEE p802.3bn EPoC. Channel Model Ad Hoc committee Baseline Channel Model

FORWARD ERROR CORRECTION PROPOSAL FOR EPOC PHY LAYER

Editor: this header only appears here to set number 100 and is not to be included.

EPON over Coax. RF Spectrum Ad Hoc Status Report. Steve Shellhammer (Qualcomm)

IEEE P802.3bn Tutorial E P o C

PXI WiMAX Measurement Suite Data Sheet

Error! No text of specified style in document. Table Error! No text of specified style in document.-1 - CNU transmitter output signal characteristics

EPoC Upstream Modulation Profiles Eugene Dai, PhD, Cox Communications Hal Roberts, Calix Networks

Topics on Channel Architecture

From Control Multiplexer to Gearbox, How Do We Meet MPCP Jitter Requirement? Jin Zhang Marvell

Downstream Bit Loading Procedure

LDPC Code Length Reduction

OFDM TX Shaping for 802.3bn Leo Montreuil

The results in the next section show that OTFS outperforms OFDM and is especially well suited for the high-mobility use case.

PHY Link Channel for EPoC TDD mode. Nicola Varanese, Qualcomm

Channel Model Ad Hoc. Report. Presented by Duane Remein (Huawei)

PXI LTE FDD and LTE TDD Measurement Suites Data Sheet

University of Bristol - Explore Bristol Research. Peer reviewed version. Link to published version (if available): /ICCE.2012.

Selected answers * Problem set 6

Fading & OFDM Implementation Details EECS 562

Outline / Wireless Networks and Applications Lecture 7: Physical Layer OFDM. Frequency-Selective Radio Channel. How Do We Increase Rates?

EPoC Downstream Baseline Proposal (PLC material removed for transfer to PLC baseline)

LDPC FEC PROPOSAL FOR EPOC. Richard S. Prodan Broadcom Corporation

[Insert Document Title Here]

System Impairments Mitigation for NGPON2 via OFDM

Lecture 3 Cellular Systems

802.11ax introduction and measurement solution

OFDMA Networks. By Mohamad Awad

IEEE Working Group on Mobile Broadband Wireless Access <

Page 1. Overview : Wireless Networks Lecture 9: OFDM, WiMAX, LTE

IEEE C802.16a-02/94r1. IEEE Broadband Wireless Access Working Group <

Homeworx Lessons? What can we learn from the first deployment of OFDMA on HFC? Hal Roberts, Calix

Exam 3 is two weeks from today. Today s is the final lecture that will be included on the exam.

EPON over Coax. Channel Bonding Sub-layer

Bit Error Rate Performance Evaluation of Various Modulation Techniques with Forward Error Correction Coding of WiMAX

Coax Resource Allocation & Tone Reordering

MIMO-LTE A relevant Step towards 4G. Prof. Dr.-Ing. Thomas Kaiser CEO mimoon GmbH

50Gb/s technical feasibility analysis. Dekun Liu, Huawei Stanley Shuai, Source Sep, 2017

PERFORMANCE ANALYSIS OF DOWNLINK MIMO IN 2X2 MOBILE WIMAX SYSTEM

Project: IEEE P Working Group for Wireless Personal Area Networks N

3GPP Long Term Evolution LTE

Ten Things You Should Know About MIMO

Are You Ready for DOCSIS 3.1. Presenter: Pete Zarrelli VeEX Field Applications Engineer

AEROHIVE NETWORKS ax DAVID SIMON, SENIOR SYSTEMS ENGINEER Aerohive Networks. All Rights Reserved.

Return Plant Issues SCTE Cascade Range Chapter. Micah Martin January 13, 2008

Rate and Power Adaptation in OFDM with Quantized Feedback

802.16s SOFTWARE PLATFORM

Time division multiplexing The block diagram for TDM is illustrated as shown in the figure

Further Vision on TD-SCDMA Evolution

ECS455: Chapter 4 Multiple Access

Effect of Oscillator Phase Noise and Processing Delay in Full-Duplex OFDM Repeaters

UNIVERSITY OF MICHIGAN DEPARTMENT OF ELECTRICAL ENGINEERING : SYSTEMS EECS 555 DIGITAL COMMUNICATION THEORY

Wireless Physical Layer Concepts: Part III

DSRC using OFDM for roadside-vehicle communication systems

Technical Aspects of LTE Part I: OFDM

Analysis and Improvements of Linear Multi-user user MIMO Precoding Techniques

(Channel Bandwidth: 5 MHz)_MCH_16QAM_1RB#0

DFE Error Performance Under 1000BASE-T1 Noise Environments

Toward SSC Modulation Specs and Link Budget

Efficient Signaling for VoIP in OFDMA

2016 Spring Technical Forum Proceedings

JD7105A Base Station Analyzer

IEEE Working Group on Mobile Broadband Wireless Access < Technical Overview Presentation

Wireless Networks: An Introduction

LTE Direct Overview. Sajith Balraj Qualcomm Research

Project: IEEE P Working Group for Wireless Personal Area Networks N

Basic idea: divide spectrum into several 528 MHz bands.

MACHINE TO MACHINE (M2M) COMMUNICATIONS-PART II

IEEE Broadband Wireless Access Working Group < Title Propose for Uplink Pilot Design in IEEE m

Hotline: Tel: Fax:

CSC344 Wireless and Mobile Computing. Department of Computer Science COMSATS Institute of Information Technology

h max 20 TX Ionosphere d 1649 km Radio and Optical Wave Propagation Prof. L. Luini, July 1 st, 2016 SURNAME AND NAME ID NUMBER SIGNATURE

SourceSync. Exploiting Sender Diversity

Investigation on Multiple Antenna Transmission Techniques in Evolved UTRA. OFDM-Based Radio Access in Downlink. Features of Evolved UTRA and UTRAN

Available online at ScienceDirect. The 4th International Conference on Electrical Engineering and Informatics (ICEEI 2013)

Report ITU-R M (11/2010)

WiMAX Basestation: Software Reuse Using a Resource Pool. Arnon Friedmann SW Product Manager

10GEPON Burst Timing in Coexistence Situation

Project: IEEE P Working Group for Wireless Personal Area Networks N

2 nd Generation OFDM for , Session #11

Transforming MIMO Test

TSTE17 System Design, CDIO. General project hints. Behavioral Model. General project hints, cont. Lecture 5. Required documents Modulation, cont.

Fundamentals of Digital Communication

David Huang Checked By

Performance Analysis of WiMAX Physical Layer Model using Various Techniques

Contents. IEEE family of standards Protocol layering TDD frame structure MAC PDU structure

Calculated Radio Frequency Emissions Report. Cotuit Relo MA 414 Main Street, Cotuit, MA 02635

Understanding Probability of Intercept for Intermittent Signals

Comb type Pilot arrangement based Channel Estimation for Spatial Multiplexing MIMO-OFDM Systems

Synchronizing Transmitter Jitter Testing with Receiver Jitter Tolerance

Transcription:

Resource Blocks for EPoC Considerations Avi Kliger, BZ Shen, Leo Montreuil Broadcom 1

RB Size Current Status in 802.3bn Size in number of symbols (M) Configurable and TBD Size in number of subcarriers (N) Three options are specified: 1,4,8 Configurable Pilot spacing Configurable 1,2,4,8 This presentation proposes values for the number of symbols and pilot patterns Data is written horizontally (subcarrier by subcarrier) and read vertically (symbol by symbol) Time interleaving of codewords IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 2

Number of OFDMA Symbols in a RB Considerations To increase number of symbols Performance with burst noise Improves with longer interleaver Pilots overhead Pilots are transmitted on two symbols in the RB Increasing the number of symbols reduces latency To decrease number of symbols Latency Granularity overhead Granularity overhead is a function of both number of symbols and number of subcarriers For a specific number of subcarriers more symbols means higher overhead IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 3

Performance with Burst Noise Table shows burst noise durations and levels to be considered Assumed to represent worst case conditions in the upstream With simulated burst noise an Interleaver depth of 16 with 20uSec symbols and 11 with 40 usec is required Corresponding Interleaver latency ( CP size = 2.0 usec) is 374 usec and 462 usec for 20u and 40u symbols Burst noise Duration (usec) SNR (db) Interleaver Depth (20uS symbols Interleaver Depth (40uS symbols Upstream 10 10 16 11 1 0 16 8 IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 4

Overhead due to Pilots Assume pilots every 1,2,4 or 8 subcarriers with 20 usec symbols Two pilots in a subcarrier with pilots to protect against burst noise Overhead vs. number of symbols equals 2/L L is the multiplication of the pilot spacing with the number of symbols With a pilots spacing of 8 subcarriers Overhead with 8 symbols is 3.1% Overhead with 12 symbols is 2% Overhead with 16 symbols is 1.5% Below 8 symbols overhead becomes significant, in particular with the more dense pilot patterns overhead (%) 25 20 15 10 5 Piolots overhead pilots spacing = 1 pilots spacing = 2 pilots spacing = 4 pilots spacing = 8 0 2 4 6 8 10 12 14 16 18 20 number od symbols IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 5

Additional Latency due to Interleaver Depth Assume additional latency due to Interleaving is twice the Interleaver depth With 20 usec symbols additional latency is: ~ 350 usec with 8 symbols ~ 530 usec with 12 symbols ~ 750 usec with 16 symbols msec 1.4 1.2 1 0.8 0.6 0.4 Interleaver latency vs. Depth with 20uSec symbols with 40uSec symbols 0.2 0 2 4 6 8 10 12 14 16 18 20 number of symbol IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 6

Proposal for Number of Symbols in a Resource Block Allow three configurable options for the number of symbols (M) in a Resource Block Allow operators to trade-off between latency/overhead and performance with burst noise propose M values with 20 usec symbols M= 8 for low latency, burst noise support is weak M=16 to protect against high level/long burst noise M=12 lower latency, burst noise support is mild Protect well against lower level burst noise All RBs in an OFDMA channel must have the same number of symbols IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 7

Pilot Patterns (1) Pilot patterns are defined for the different RB sizes Pilot spacing (per current decision) Pilot every 1,2,4 or 8 subcarriers Less patterns is possible with cost in overhead or in robustness to frequency response variations Use edge and body pilot patterns To avoid extrapolation every burst starts with a pilot and ends with a pilot Every RB after exclusion starts with a pilot Two pilots used on every subcarrier with pilots to protect against burst noise that hit a symbol with pilots IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 Edge Edge Edge 8

Pilot Patterns (2) Two pilots used on every subcarrier with pilots to protect against burst noise that hit a symbol with pilots Low density pilots are data REs with a lower order modulation Can be used to improve initial frequency and phase correction Edge Edge IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 Edge 9

Pilot Patterns for 8-subcarrier RBs Edge Pattern - 1 Pattern - 2 Pattern - 3 Pattern - 4 P P LD LD P P LD LD P P LD LD P P P P P P P P P P P P P P P P P P P P P P P P P P LD LD P P LD LD P P LD LD P P P P P P P P P P P P P P P P P P P P P P P P LD LD P P LD LD P P LD LD P P M symbols M symbols M symbols M symbols M can equal 8,12 or 16 P Pilots LD Low Density pilots, blank RE for data IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 LD LD LD LD LD LD 10

Pilot Patterns for 4-subcarrier RBs Four pilot patterns are available Edge Pattern - 1 Pattern - 2 Pattern - 3 Pattern - 4 P P LD LD P P LD LD P P LD LD P P P P P P P P P P LD LD P P LD LD P P LD LD P P P P P P P P LD LD P P LD LD P P LD LD P P LD LD M symbols M symbols M symbols M symbols M can equal 8,12 or 16 P Pilots LD Low Density pilots, blank RE for data IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 11

Pilot Patterns for a single subcarrier RBs Two pilot patterns are available Pattern - 1 Pattern - 2 Edge P P LD LD P P LD LD P P LD LD M symbols M symbols M can equal 8,12 or 16 P Pilots LD Low Density pilots, blank RE for data IEEE 802.3bn EPoC Indian Wells, Januaryi 22-24, 2014 12

Proposed Motion (1) Move to: Specify three options for the number of symbols in a Resource Block: 8, 12 and 16 Moved: Seconded: 13

Proposed Motion (2) Move to: Specify edge and body pilot patterns as described in slides 10-12 for EPoC FDD Upstream Moved: Seconded: 14

(this section all new slides) PERFORMANCE ANALYSIS 15

PER Performance with an Interleaver Depth of 16 Symbols 4KQAM (Short Code) 1 usec / 0 db 10 usec / 10 db 16

PER Performance with an Interleaver Depth of 16 Symbols 4KQAM (Short Code) 1 usec / 0 db 10 usec / 10 db 17

PER Performance with an Interleaver Depth of 16 Symbols 4KQAM (Med. Code 1 usec / 0 db 10 usec / 10 db 18

PER Performance with an Interleaver Depth of 16 Symbols 4KQAM (Med. Code 1 usec / 0 db 10 usec / 10 db 19

PER Performance with an Interleaver Depth of 16 Symbols 1KQAM (Long Code 1 usec / 0 db 10 usec / 10 db 20

PER Performance with an Interleaver Depth of 16 Symbols 4KQAM (Long Code) 1 usec / 0 db 10 usec / 10 db 21

Grants to RB Mapping (1) Assumptions Data transmission starts and ends at the boundary of a Resource Block Padding bits ( IDLE are added if required to align to the RB) Markers are mixed with the data (16 REs for each start and end markers) 22

Grants to RB Mapping (2) Gap over fiber Laser on/off + CDR + Dead zone Burst #1 (Data + Pilots + Markers) Burst #2 (Data + Pilots + Markers) Burst #3 (Data + Pilots + Markers) Burst #4 (Data + Pilots + Markers) Burst #5 (Data + Pilots + Markers) Time GAP > 2*RB size + allowed jitter (12 TQs) Markers can either be part of the grant size or additional gap over the fiber Every RB is comprised of N subcarriers in M symbols, total of M*N REs N subcarriers Freq Burst #1 (Data + Pilots + Markers + Padding) Burst #2 (Data + Pilots + Markers + padding) Burst #3 (Data + Pilots + Markers + padding) Burst #4 (Data + Pilots + Markers + padding) M- symbols Time Note: For clarity figure assumes uniform bit loading across frequency, Gaps on fiber Gaps on fiber set by the CLT CLT allocates gap for Laser on/off, CDR+AGC, dead zone, Gap sizes are configurable in the CLT OFDMA TX uses idle padding in the RB to align transmission to RB Markers can either be considered as a fixed overhead added to every grant or as gaps over fiber 23

Grants to RB Mapping (3) Minimum required gap over fiber Avoid two transmitters using same RB Two RBs plus the 12 TQs time for EPON jitter Two Markers are additional overhead Assume: 20 usec symbols and CP size of 2.5 usec BW (MHz) RB size (M) RB size (N) OFDMA symbol size (usec) num of subcarriers in OFDMA frame Subcarrier equivalrnt time duration (nsec) Number of TQs in RB Minimal Gap size (in TQs) 192 16 8 22.5 3840 93.75 46.88 106 192 16 4 22.5 3840 93.75 23.44 59 192 16 1 22.5 3840 93.75 5.86 24 96 16 8 22.5 1920 187.50 93.75 200 96 16 4 22.5 1920 187.50 46.88 106 96 16 1 22.5 1920 187.50 11.72 36 24 16 8 22.5 480 750.00 375.00 762 24 16 4 22.5 480 750.00 187.50 387 24 16 1 22.5 480 750.00 46.88 106 Red are gaps larger than 90 TQs Convert TQs to RBs Conversion ratio depends on available bandwidth One OFDMA frame duration is the symbols size times M Equivalent subcarrier duration equals OFDMA duration divided by the number of subcarriers in the OFDMA frame Calculate RB time as the multiplication of the number of subcarriers in a RB by the subcarrier time duration Minimum required gap over fiber to avoid two transmitters using same RB Two RBs plus maximal allowed jitter (12 TQs) 24