DOC PREVIEW
Berkeley ELENG 290C - Lecture 8: Equalization: FIR II and DFE

This preview shows page 1-2-3-4 out of 12 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

EE290C – Spring 2011Lecture 8: Equalization: FIR II and DFEElad AlonDept. of EECSEE290C Lecture 8 2VM TX + Equalization• First implementation (K.-L. J. Wong, JSSC 2004):EE290C Lecture 8 3Constant Current VM TX Design• D. Dettloff, ISSCC 2010EE290C Lecture 8 4RX FIR Equalizers• Currently not very popular – why?EE290C Lecture 8 5Coefficient Shuffling Architecture• Lee, Razavi, CICC 2001EE290C Lecture 8 6Shuffling Implementation• Can shuffle digital tap coefficients or analog inputsEE290C Lecture 8 7Interleaving• Often combined with tap shuffling• Ex: J. Jaussi, ISSCC 2004EE290C Lecture 8 8DFE: ImplementationRX_in• Design driven by feedback latency constraint• New summer output needs to settle (at comparator input) within one bit timeEE290C Lecture 8 9DFE Summer Design• Key constraint:tdig_fb+ tana_settle= Tbit• Need some Nτof settling:τsum= (Tbit–tdig_fb)/Nτ• Typically want ~4τ for ~98% settlingEE290C Lecture 8 10A ReminderEE290C Lecture 8 11Self-Loading 0 0.2 0.4 0.6 0.8 100.10.20.30.40.50.6GBW/(ωT/γ)GmCL = 20fFCL = 40fFCL = 80fF• Creates bandwidth limit• No matter how much power you spend• DFE has increased “self-loading” due to taps…EE290C Lecture 8 12DFE Summer DesignEE290C Lecture 8 13Maximum Cancelled ISI0 2 4 6 8 1000.010.020.030.040.05MDFEGm!0Gb/s DFEtdig = 30pstdig = 40pstdig = 50ps• There is a maximum MDFEyou can achieve• Don’t forget that every tap has to handle worst-case possible ISI• Implies max. number of taps (in given tech.)EE290C Lecture 8 14TX FIR RX DFE Duality• Can use same techniques we talked about in TX for RX DFE• But still need to watch out for digital power…d0clkVin0 – Imax/20 – Imax/20 – Imax/4z-1z-1d1d2d1d0d2d1d2EE290C Lecture 8 15Closing the First TapEE290C Lecture 8 16Split Summation• B. Leibowitz, ISSCC 2007EE290C Lecture 8 17All CML Design• H. Wang, VLSI 2009• Over 20Gb/s in 65nm CMOSEE290C Lecture 8 18Loop Unrolling• K. Parhi, ISCAS 1990+1+α-1+α+α+1-α-1-α-αDQ1−nddClk1|1=−nndd0|1=−nndddClknx+α-αEE290C Lecture 8 19Extended Loop UnrollingEE290C Lecture 8 20LE + DFEEE290C Lecture 8 21Practical DFE Design IssuesEE290C Lecture 8 22Practical DFE Design IssuesEE290C Lecture 8 23Practical DFE Design


View Full Document
Download Lecture 8: Equalization: FIR II and DFE
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Lecture 8: Equalization: FIR II and DFE and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Lecture 8: Equalization: FIR II and DFE 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?