EE359 Lecture 12 Outline

Transcription

1 EE359 Lecture 12 Outine Announcements Midterm announcements No HW next week (practice MTs) HW5 posted, due Monday 4pm (no ate HWs) Transmit Diversity Midterm Review Introduction to adaptive moduation Variabe-rate variabe-power MQAM Optima power and rate adaptation

2 Midterm Announcements Midterm: Thursday (11/9), 6-8 pm in (room TBD) Food wi be served after the exam! Review sessions My midterm review wi be durin tomorrow s makeup ecture TA review: Monday 11/6 from 4-6 pm in 364 Packard Midterm oistics: Open book/notes; Brin textbook/cacuators (have extras; adv. notice reqd) Covers Chapters 1-7 (sections covered in ecture and/or HW) Specia OHs next week: Me: Wed 11/8: 9-11am, Thu 11/9: 12-2pm a in 371 Packard Miind: Tues 11/7, 4-6pm, 3rd Foor Packard Kitchen Area + emai Tom: Wed 11/8: 5-7pm, Thu 11/9 2-4pm, 3rd Foor Packard Kitchen Area + emai No HW next week Midterms from past 3 MTs posted: 10 bonus points for takin a practice exam Soutions for a exams iven when you turn in practice exam

3 Review of Last Lecture Array Structure of a Diversity Combiner Performance metrics: Outae probabiity and averae probabiity of error Array and Diversity ain Combinin Techniques Seection Combinin (SC): Path with hihest ain used Maxima Ratio Combinin (MRC): Paths cophased and summed with optima weihts to maximize SNR SC Performance Anaysis Combiner SNR is the maximum of the branch SNRs. CDF easy to obtain (P i p( i < thr )), pdf found by differentiatin. P out obtained from CDF. Averae P s typicay found numericay Diminishin returns with number of antennas. Can et up to about 20 db of ain.

4 Review Continued MRC Performance P With MRC, S =S i for branch SNRs i Optima technique to maximize output SNR Yieds db performance ains Distribution of S hard to obtain Standard averae BER cacuation ò òò... ò = Pb ( S ) p( S) d S = Pb ( S) p( 1) * p( 2) *...* p( M ) d 1d 2... d M sb s s Hard to obtain in cosed form Intera often diveres MGF Approach Cover in HW and ppt, not ecture

5 Transmit Diversity With channe knowede, simiar to receiver diversity, same array/diversity ain Without channe knowede, can obtain diversity ain throuh Aamouti scheme: 2 TX antenna space-time bock code (STBC) Works over 2 consecutive symbos Achieves fu diversity ain, no array ain Part of various wireess standards, incudin LTE Hard to eneraize to more than 2 TX antennas Aamouti code not covered in ecture/exams

6 Midterm Review Overview of Wireess Systems Sina Propaation and Channe Modes Moduation and Performance Metrics Impact of Channe on Performance Fundamenta Capacity Limits Diversity Techniques Main Points

7 Adaptive Moduation Chane moduation reative to fadin Parameters to adapt: Consteation size Transmit power Instantaneous BER Symbo time Codin rate/scheme Ony 1-2 derees of freedom needed for ood performance Optimization criterion: Maximize throuhput Minimize averae power Minimize averae BER

8 Variabe-Rate Variabe-Power MQAM Uncoded Data Bits Deay o 2 M() Bits Point Seector One of the M() Points M()-QAM Moduator Power: P() To Channe (t) (t) BSPK 4-QAM 16-QAM Goa: Optimize P() and M() to maximize R=Eo[M()]

9 Optimization Formuation Adaptive MQAM: Rate for fixed BER M 1.5 P( ) ( ) = 1+ = 1+ - n(5ber) P P( ) K P Rate and Power Optimization ê é P( ) max E o M = E + 2[ ( )] max o2 1 K P( ) P( ) ë P Same maximization as for capacity, except for K=-1.5/n(5BER). ù ú û

10 Optima Adaptive Scheme Power Adaptation Spectra Efficiency î í ì = ³ - = ese 0 ) ( K K K P P K k R B p d K K = æ è ç ö ø ò o ( ). 2 Equas capacity with effective power oss K=-1.5/n(5BER).

11 Spectra Efficiency K 1 K 2 K=-1.5/n(5BER) Can reduce ap by superimposin a treis code

12 Main Points Transmit diversity with channe state information at the TX is same as RX diversity Can obtain diversity ain even without channe information at transmitter via space-time bock codes. Adaptive moduation everaes fast fadin to improve performance (throuhput, BER, etc.) Adaptive MQAM uses capacity-achievin power and rate adaptation, with power penaty K. Comes within 5-6 db of capacity