With a lot of material from Rich Nicholls, CTL/RCL and Kurt Sundstrom, of unknown whereabouts

Similar documents
Practical issue: Group definition. TSTE17 System Design, CDIO. Quadrature Amplitude Modulation (QAM) Components of a digital communication system

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

Lecture 13. Introduction to OFDM

Orthogonal frequency division multiplexing (OFDM)

EC 551 Telecommunication System Engineering. Mohamed Khedr

Orthogonal Frequency Division Multiplexing (OFDM)

Multi-carrier Modulation and OFDM

Receiver Designs for the Radio Channel

Design and Simulation of COFDM for High Speed Wireless Communication and Performance Analysis

S.D.M COLLEGE OF ENGINEERING AND TECHNOLOGY

Fundamentals of OFDM Communication Technology

Experimenting with Orthogonal Frequency-Division Multiplexing OFDM Modulation

Fading & OFDM Implementation Details EECS 562

Basic idea: divide spectrum into several 528 MHz bands.

CHAPTER 3 ADAPTIVE MODULATION TECHNIQUE WITH CFO CORRECTION FOR OFDM SYSTEMS

SHIV SHAKTI International Journal of in Multidisciplinary and Academic Research (SSIJMAR) Vol. 3, No. 4, August-September (ISSN )

OFDMA and MIMO Notes

OFDM AS AN ACCESS TECHNIQUE FOR NEXT GENERATION NETWORK

Orthogonal Frequency Division Multiplexing & Measurement of its Performance

ICI Mitigation for Mobile OFDM with Application to DVB-H

OFDM system: Discrete model Spectral efficiency Characteristics. OFDM based multiple access schemes. OFDM sensitivity to synchronization errors

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

Modern Quadrature Amplitude Modulation Principles and Applications for Fixed and Wireless Channels

SC - Single carrier systems One carrier carries data stream

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

Baseline Proposal for EPoC PHY Layer

University of Bristol - Explore Bristol Research. Link to publication record in Explore Bristol Research PDF-document.

Multipath can be described in two domains: time and frequency

OFDM (Orthogonal Frequency Division Multiplexing) SIMULATION USING MATLAB Neha Pathak MTech Scholar, Shri am Institute of Technology

Lecture 3 Cellular Systems

Performance Evaluation of OFDM System with Rayleigh, Rician and AWGN Channels

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

Broadband OFDM-FDMA System for the Uplink of a Wireless LAN

Orthogonal Frequency Division Multiplexing (OFDM) based Uplink Multiple Access Method over AWGN and Fading Channels

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

Underwater communication implementation with OFDM

Channel Estimation in Multipath fading Environment using Combined Equalizer and Diversity Techniques

Chapter 5 OFDM. Office Hours: BKD Tuesday 14:00-16:00 Thursday 9:30-11:30

Optimal Number of Pilots for OFDM Systems

Wireless Networks (PHY)

Comparative Study of OFDM & MC-CDMA in WiMAX System

Technical Aspects of LTE Part I: OFDM

UNIFIED DIGITAL AUDIO AND DIGITAL VIDEO BROADCASTING SYSTEM USING ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING (OFDM) SYSTEM

Decrease Interference Using Adaptive Modulation and Coding

Performance Evaluation of Wireless Communication System Employing DWT-OFDM using Simulink Model

MITIGATING CARRIER FREQUENCY OFFSET USING NULL SUBCARRIERS

Bit Loading and Peak Average Power Reduction Techniques for Adaptive Orthogonal Frequency Division Multiplexing Systems

Performance Analysis of OFDM System with QPSK for Wireless Communication

ANALYSIS AND STUDY OF MULTI-SYMBOL ENCAPSULATED ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING

ENHANCING BER PERFORMANCE FOR OFDM

Wireless Communication Systems: Implementation perspective

S PG Course in Radio Communications. Orthogonal Frequency Division Multiplexing Yu, Chia-Hao. Yu, Chia-Hao 7.2.

Multi-Carrier Systems

Wireless Communication: Concepts, Techniques, and Models. Hongwei Zhang

Principles and Experiments of Communications

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

SPARSE CHANNEL ESTIMATION BY PILOT ALLOCATION IN MIMO-OFDM SYSTEMS

Mobile & Wireless Networking. Lecture 2: Wireless Transmission (2/2)

Techniques for Mitigating the Effect of Carrier Frequency Offset in OFDM

Revision of Wireless Channel

Lecture 3: Wireless Physical Layer: Modulation Techniques. Mythili Vutukuru CS 653 Spring 2014 Jan 13, Monday

BER Analysis for MC-CDMA

ATSC 3.0 Physical Layer Overview

IEEE pc-00/11. IEEE Broadband Wireless Access Working Group <

An OFDM Transmitter and Receiver using NI USRP with LabVIEW

PEAK TO AVERAGE POWER RATIO REDUCTION USING BANDWIDTH EFFICIENCY INCREASING METHOD IN OFDM SYSTEM

Analytical Review of Orthogonal Frequency Division Multiplexing (OFDM)

Wireless Communication

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

UTILIZATION OF AN IEEE 1588 TIMING REFERENCE SOURCE IN THE inet RF TRANSCEIVER

Chapter 0 Outline. NCCU Wireless Comm. Lab

INTERNATIONAL JOURNAL OF PURE AND APPLIED RESEARCH IN ENGINEERING AND TECHNOLOGY

Outline Chapter 4: Orthogonal Frequency Division Multiplexing

Anju 1, Amit Ahlawat 2

Pilot-Assisted DFT Window Timing/ Frequency Offset Synchronization and Subcarrier Recovery 5.1 Introduction

Clipping and Filtering Technique for reducing PAPR In OFDM

Implementation and Comparative analysis of Orthogonal Frequency Division Multiplexing (OFDM) Signaling Rashmi Choudhary

Page 1. Outline : Wireless Networks Lecture 6: Final Physical Layer. Direct Sequence Spread Spectrum (DSSS) Spread Spectrum

[Insert Document Title Here]

Part 3. Multiple Access Methods. p. 1 ELEC6040 Mobile Radio Communications, Dept. of E.E.E., HKU

Chapter 3 Introduction to OFDM-Based Systems

Simulative Investigations for Robust Frequency Estimation Technique in OFDM System

OFDM Systems For Different Modulation Technique

Performance Evaluation of STBC-OFDM System for Wireless Communication

Wireless Medium Access Control and CDMA-based Communication Lesson 16 Orthogonal Frequency Division Medium Access (OFDM)

Wireless Networks: An Introduction

An Equalization Technique for Orthogonal Frequency-Division Multiplexing Systems in Time-Variant Multipath Channels

Lecture LTE (4G) -Technologies used in 4G and 5G. Spread Spectrum Communications

EE359 Lecture 18 Outline

Comparison of MIMO OFDM System with BPSK and QPSK Modulation

A Research Concept on Bit Rate Detection using Carrier offset through Analysis of MC-CDMA SYSTEM

ANALYSIS OF BER AND SEP OF QPSK SIGNAL FOR MULTIPLE ANENNAS

Chapter 2 Overview - 1 -

EE678-Application Assignment Discrete Wavelet Multitone Modulation

Communication Theory

Algorithm to Improve the Performance of OFDM based WLAN Systems

2: Diversity. 2. Diversity. Some Concepts of Wireless Communication

CHAPTER - 6. Higher MIMO performance OFDM Multiplexing

Optimized BPSK and QAM Techniques for OFDM Systems

Survey on Effective OFDM Technology for 4G

Transcription:

Signal Processing for OFDM Communication Systems Eric Jacobsen Minister of Algorithms, Intel Labs Communication Technology Laboratory/ Radio Communications Laboratory July 29, 2004 With a lot of material from Rich Nicholls, CTL/RCL and Kurt Sundstrom, of unknown whereabouts

Outline OFDM What and Why Subcarrier Orthogonality and Spectral Effects Time Domain Comparison Equalization Signal Flow PAPR management Cool Tricks 2

Digital Modulation Schemes Single Carrier PSK, QAM, PAM, MSK, etc. Demodulate with matched filter, PLLs Common Standards: DVB-S, Intelsat, GSM, Ethernet, DOCSIS Multi-Carrier OFDM, DMT Demodulate with FFT, DSP Common Standards: DVB-T, 802.11a, DAB, DSL-DMT 3

......... Communication and Interconnect Technology Lab What is OFDM? Orthogonal Frequency Division Multiplexing Split a high symbol rate data stream into N lower rate streams Transmit the N low rate data streams using N subcarriers Frequency Division Multiplexing (FDM) & Multi-Carrier Modulation (MCM) N subcarriers must be mutually orthogonal N exp j 2 f t 2 Stream -N/2 Subcarrier spacing = f Partition available bandwidth into N orthogonal subchannels High Rate Complex Symbol Stream Serial to Parallel Hold (T hold = 1/ f sec) Stream 1 exp j 2 f t N exp j 2 1 f t 2 Complex Baseband OFDM Signal s(t) -N( f)/2 0 f (N-1)( f)/2 Stream N/2-1 OFDM Conceptual Block Diagram 4

Why OFDM? Reduces symbol rate by more than N, the number of subcarriers Fading per subcarrier is flat, so single coefficient equalization Reduces equalizer complexity O(N) instead of O(N 2 ) Fully Captures Multipath Energy For Large Channel Coherence Time, OFDM/DMT can Approach Water Pouring Channel Capacity Narrowband interference will corrupt small number of subcarriers Effect mitigated by coding/interleaving across subcarriers Increases Diversity Opportunity Frequency Diversity Increases Adaptation Opportunities, Flexibility Adaptive Bit Loading OFDMA PAPR largely independent of modulation order Helpful for systems with adaptive modulation 5

Downsides of OFDM Complexity FFT for modulation, demodulation Must be compared to complexity of equalizer Synchronization Overhead Cyclic Extension Pilot Tones PAPR Increases the length of the symbol for no increase in capacity Simplify equalization and tracking for no increase in capacity Depending on the configuration, the PAPR can be ~3dB-6dB worse than a single-carrier system Phase noise sensitivity The subcarriers are N-times narrower than a comparable single-carrier system Doppler Spread sensitivity Synchronization and EQ tracking can be problematic in high doppler environments 6

Subcarrier Orthogonality Orthogonality simplifies recovery of the N data streams Orthogonal subcarriers = No inter-carrier-interference (ICI) Time Domain Orthogonality: Every subcarrier has an integer number of cycles within T OFDM Satisfies precise mathematical definition of orthogonality for complex exponential (and sinusoidal) functions over the interval [0, T OFDM ] Frequency Domain Orthogonality: ICI = 0 at f = nf 0 f f Some FDM systems achieve orthogonality through zero spectral overlap BW inefficient! OFDM systems have overlapped spectra with each subcarrier spectrum having a Nyquist zero ISI pulse shape (really zero ICI in this case). BW efficient! 7

OFDM Signal (Time & Frequency) 1.5 TIME DOMAIN: 2 OFDM subcarriers (BPSK) 1 0.5 0-0.5-1 -1.5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (Normalized by Tofdm) FREQUENCY DOMAIN: OFDM Subcarriers 2 through 10 1 0.8 0.6 0.4 0.2 0-0.2-0.4 0 2 4 6 8 10 12 Frequency (Normalized by 1/Tofdm) 8

Practical Signal Spectra Magnitude 0 10 Single carrier signals require filtering for spectral containment. This signal has narrow rolloff regions which requires long filters. 20 30 0 500 1000 1500 2000 Frequency OFDM spectra have naturally steep sides, especially with large N. The PAPR is often higher, which may result in more spectral regrowth. The blue trace is an unfiltered OFDM signal with 216 subcarriers. The red trace includes the effects of a non-linear Power Amplifier. 9

Time-Domain Comparisons By greatly increasing the symbol period the fading per subcarrier becomes flat, so that it can be equalized with a single coefficient per subcarrier. The addition of the cyclic prefix eliminates Inter- Symbol Interference (ISI) due to multipath. 10

Frequency Domain Equalization Design System Such That T Delay Spread < T Guard and B Coherence > B Subcarrier Subcarriers are perfectly orthogonal (no ISI or ICI) Each Subcarrier experiences an AWGN channel Equalizer Complexity : Serial Data Rate = 1/T, OFDM Symbol Rate = 1/(NT) FEQ performs N complex multiplies in time NT (or 1 complex mult per time T) Time domain EQ must perform MT complex multiplies in time T where M is the number of equalizer coefficients Channel Frequency Response (at time t) Subcarrier n Frequency 11

802.11a PHY Block Diagram I DAC Guard Interval Insertion Window I & Q HPA /2 BPF Duplexer Q DAC Data Scrambler FEC Encoder Interleaver QAM Mapping Pilot Insertion S/P IFFT (TX) P/S Data Descrambler FEC Decoder Deinterleaver QAM Demap Channel Estimation & Correction P/S FFT(RX) S/P Guard Interval Removal To MAC Sublayer RSSI I ADC LPF Digital LPF Symbol Timing Frequency Correction Signal Detect AGC /2 LNA BPF Frequency Offset Estimation Q ADC LPF 12

802.11a Processing 802.11a is a TDD contention-based, bursty protocol Full acquisition, synchronization, and EQ training can be performed for each burst or frame The short training symbols provide timing, AGC, diversity selection, and initial carrier offset The long training symbols provide fine synchronization and channel estimation Two FFT periods allow 3dB increase in channel estimation SNR by combining (averaging) the estimates Tracking is facilitated by the four pilot tones 13

802.11a Time/Frequency Signal Structure 8.125 MHz Short Training Symbols Long Training Symbols DATA FRAME Data Symbols FREQUENCY 53 Subcarriers (48 data, 4 pilot, 0 @ DC) 0-8.125 MHz 800 ns 4 s Indicates Pilot Tone Location TIME 14

DVB-T Time/Frequency Signal Structure Since DVB-T is a continuous transmit signal, channel estimation is facilitated easily by rotating pilots across the subcarrier indices. Interpolation provides channel estimation for every subcarrier. This figure is from reference [4] 15

Peak to Average Power Ratio Single Carrier Systems PAPR affected by modulation scheme, order, and filtering Constant-envelope schemes have inherently low PAPR For example: MSK, OQPSK PAPR increases with modulation order e.g., 64-QAM PAPR is higher than QPSK As Raised Cosine excess bandwidth decreases, PAPR increases Squeezing the occupied spectrum increases PAPR Multi-Carrier Systems PAPR affected by subcarrier quantity and filtering PAPR is only very weakly connected to modulation order PAPR increases with the number of subcarriers Rate of increase slows after ~64 subcarriers The Central Limit Theorem is still your friend Whitening is very effective at reducing PAPR Symbol shaping decreases PAPR 16

PAPR with 240 subcarriers P(PAPR < Abscissa) 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 64-QAM 20% RRC PAPR Cumulative Distribution Function 64-QAM OFDM-48 64-QAM 802.11a OFDM-240 N = 240 requires no more than 1dB additional backoff compared to 802.11a, and about 3.5dB more than a single-carrier system. The results shown use only data whitening for PAPR reduction. Additional improvements may be possible with other techniques. 0 3 4 5 6 7 8 9 10 11 12 PAPR (db) 17

PAPR Mitigation in OFDM Scrambling (whitening) decreases the probability of subcarrier alignment Subcarriers with common phase increase PAPR Symbol weighting reduces the effects of phase discontinuities at the symbol boundaries Raised Cosine Pulse weighting Works well, requires buffering Signal filtering Easier to implement 18

Time-Domain Weighting The phase discontinuities between symbols increase the size of the spectral sidelobes. Weighting the symbol transitions smooths them out and reduces the sidelobe amplitudes. Typically Raised- Cosine weighting Is applied. Tapered Regions This figure is informative content from the IEEE 802.11a specification. The two-fft period case applies only to preambles for synchronization and channel estimation. 19

Effect of Symbol Weighting With no RC weighting With 1% RC weighting Applying a tiny bit of symbol weighting in the time domain has a significant effect on PAPR. In this case only 1% of the symbol time is used for tapering. The blue trace is prior to the PA, the red trace after. Application of the 1% RC window meets the green transmit mask. 20

Cool and Interesting Tricks OFDMA Different users on different subcarriers Adaptive Bit Loading Seeking water filling capacity Adaptation to Channel Fading Adaptation to Interference 21

OFDMA Subcarrier Division Pilot Tones Data Subcarriers... Control User #1 User #2 User #3 User #N-1 User #N Redundant Control The 802.16 standard describes multiple means to implement OFDMA. In one mode each user s signal occupies contiguous subcarriers which can be independently modulated. Another mode permutes each user s subcarriers across the band in a spreading scheme so that all user s subcarriers are interlaced with other user s subcarriers. The first method allows for adaptive modulation and the second method increases frequency diversity. 22

Subcarrier Division with TDM Each color is for a distinct terminal. Control Subcarriers Subcarriers Redundant Control Subcarriers OFDM Symbols 23

Channel Frequency Response Multipath Frequency Selective Fading 5 Frequency (MHz) -5-4 -3-2 -1 0 1 2 3 4 5 0 Response (db) -5-10 -15-20 -25-30 Shannon s Law applies in each flat subinterval v = 100 km/hr f = 2 GHz t = 0.5 m sec 24

Adaptive Bit Loading Communication and Interconnect Technology Lab Response (db) 5 0-5 -10-15 -20-25 -30-5 -4-3 -2-1 0 1 2 3 4 5 High SNR At Receiver Deep Fade (Bad) Frequency (MHz) 6 bps/hz Low SNR At Receiver 4 bps/hz 2 bps/hz 0 bps/hz 64 QAM Channel Bandwidth 16 QAM QPSK OFDM Symbol Sub Carriers 25

Per-Subcarrier Adaptive Modulation Signal level Frequency 26

References [1] IEEE Std 802.11a-1999 [2] Robert Heath, UT at A, http://www.ece.utexas.edu/~bevans/courses/realtime/lectures/20_ofdm/346,22,ofdm and MIMO Systems [3] Hutter, et al, http://www.lis.ei.tum.de/research/lm/papers/vtc99b.pdf [4] Zabalegui, et al, http://www.scit.wlv.ac.uk/~in8189/csndsp2002/papers/g1/g1.2.pdf 27

Backup No! Go forward! 28

Cyclic Prefix (Guard Interval) Delay Spread Causes Inter-Symbol-Interference (ISI) and Inter-Carrier-Interference (ICI) Non-linear phase implies different subcarriers experience different delay (virtually all real channels are non-linear phase) Adding a guard interval between OFDM symbols mitigates this problem Zero valued guard interval will eliminate ISI but causes ICI Better to use cyclic extension of the symbol Symbol #1 T OFDM Symbol #2 T OFDM T G T FFT Subcarrier #2 ICI Subcarrier #1 (delayed relative to #2 ) Guard interval eliminates ISI from symbol #1 to symbol #2 Cyclic extension removes ISI and ICI! 3.5 cycles of subcarrier #1 inside the FFT integration period ICI! 29

DVB-T Time/Frequency Signal Structure Since DVB-T is a continuous transmit signal, channel estimation is facilitated easily by rotating pilots across the subcarrier indices. Interpolation provides channel estimation for every subcarrier. This figure is from reference [3] 30

Advantages SCM Sensitivity (margin) Complexity Memory Phase noise sensitivity Frequency registration Reduced PA Backoff Less Overhead (no cyclic prefix) OFDM Single Frequency Networks Simple EQ Flexibility Statistical Mux OFDMA BW, TDMA LOW SNR, avoid DFE PAPR not affected by modulation order. Automatically integrates multipath. IEEE Politics 31