Scribd
Upload
38 views5 pages

Example: Estimation / Filtering: EE263 Autumn 2015 S. Boyd and S. Lall

The document describes an estimation and filtering problem. A signal u passes through a second-order system H(s) and is sampled at 10Hz by an A/D converter with 3-bit quantization, producing the signal y. Ignoring the quantization and using an equalizer G(s) to invert H(s) results in poor estimation of u. Formulating it as an estimation problem using techniques from the course yields a much better estimate of u, with RMS error well below the quantization error level.

Uploaded by

Mouli
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
38 views5 pages

Example: Estimation / Filtering: EE263 Autumn 2015 S. Boyd and S. Lall

The document describes an estimation and filtering problem. A signal u passes through a second-order system H(s) and is sampled at 10Hz by an A/D converter with 3-bit quantization, producing the signal y. Ignoring the quantization and using an equalizer G(s) to invert H(s) results in poor estimation of u. Formulating it as an estimation problem using techniques from the course yields a much better estimate of u, with RMS error well below the quantization error level.

Uploaded by

Mouli
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

EE263 Autumn 2015 S. Boyd and S.

Lall

Example: Estimation / Filtering

1
Estimation / filtering

u w y
H(s) A/D

I signal u is piecewise constant (period 1 sec)

I filtered by 2nd-order system H(s), step response s(t)

I A/D runs at 10Hz, with 3-bit quantizer

2
Typical behavior

u(t)
0

−1
0 1 2 3 4 5 6 7 8 9 10
1.5
s(t)

0.5

0
0 1 2 3 4 5 6 7 8 9 10
1
w(t)

−1
0 1 2 3 4 5 6 7 8 9 10
1
y(t)

−1
0 1 2 3 4 5 6 7 8 9 10

problem: estimate original signal u, given quantized, filtered signal y


3
Simple approach

one simple approach:

I ignore quantization

I design equalizer G(s) for H(s) (i.e., GH ≈ 1)

I approximate u as G(s)y

. . . yields terrible results

4
Better approach

formulate as estimation problem (EE263) . . .


1

u(t) (solid) and û(t) (dotted)


0.8

0.6

0.4

0.2

−0.2

−0.4

−0.6

−0.8

−1
0 1 2 3 4 5 6 7 8 9 10

RMS error 0.03, well below quantization error (!)

You might also like