Scribd
Upload
82 views7 pages

LFSRs

Linear feedback shift registers (LFSRs) are efficient designs used for test pattern generation and output response analysis. They use flip-flops and XOR gates rather than counters. There are two types: external feedback LFSRs and internal feedback LFSRs. An LFSR generates periodic sequences defined by its characteristic polynomial. Maximum-length LFSR sequences have a length of 2n - 1 and are generated using primitive characteristic polynomials.

Uploaded by

Satyakiran Kunapareddy
Copyright
© Attribution Non-Commercial (BY-NC)
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
82 views7 pages

LFSRs

Linear feedback shift registers (LFSRs) are efficient designs used for test pattern generation and output response analysis. They use flip-flops and XOR gates rather than counters. There are two types: external feedback LFSRs and internal feedback LFSRs. An LFSR generates periodic sequences defined by its characteristic polynomial. Maximum-length LFSR sequences have a length of 2n - 1 and are generated using primitive characteristic polynomials.

Uploaded by

Satyakiran Kunapareddy
Copyright
© Attribution Non-Commercial (BY-NC)
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/ 7

Linear Feedback Shift Registers (LFSRs)

Efficient design for Test Pattern Generators & Output Response Analyzers (also used in CRC)
FFs plus a few XOR gates better than counter
fewer gates higher clock frequency External Feedback LFSR

D Q CK

D Q CK

D Q CK

D Q CK

Two types of LFSRs


External Feedback Internal Feedback
higher clock frequency

Internal Feedback LFSR


D Q CK D Q CK D Q CK D Q CK

Characteristic polynomial

defined by XOR positions P(x) = x4 + x3 + x + 1 in both examples


C. Stroud, Dept. of ECE, Auburn Univ. 10/04

LFSRs (cont)
Characteristic polynomial of LFSR n = # of FFs = degree of polynomial XOR feedback connection to FF i coefficient of xi
coefficient = 0 if no connection coefficient = 1 if connection coefficients always included in characteristic polynomial:
xn (degree of polynomial & primary feedback) x0 = 1 (principle input to shift register)

Note: state of the LFSR polynomial of degree n-1 1x 0x 1x Example: P(x) = x3 + x + 1 1x


0 1 2 3

D Q
1

D Q
2

D Q
3

CK C. Stroud, Dept. of ECE, Auburn Univ. 10/04

CK

CK

LSFRs (cont)
An LFSR generates periodic sequence
must start in a non-zero state,

The maximum-length of an LFSR sequence is 2n -1


does not generate all 0s pattern (gets stuck in that state)

The characteristic polynomial of an LFSR generating a maximum-length sequence is a primitive polynomial A maximum-length sequence is pseudo-random:
number of 1s = number of 0s + 1 same number of runs of consectuive 0s and 1s 1/2 of the runs have length 1 1/4 of the runs have length 2 (as long as fractions result in integral numbers of runs)
C. Stroud, Dept. of ECE, Auburn Univ. 10/04

Example: Characteristic polynomial is P(x) = x3 + x + 1 Beginning at all 1s state


7 clock cycles to repeat maximal length = 2n-1 polynomial is primitive
1x0
D Q
1

LFSRs (cont)

1x1
D Q
2

0x2
3

1x3
D Q CK

CK

CK

Properties:
four 1s and three 0s 4 runs:
2 runs of length 1 (one 0 & one 1) 1 run of length 2 (0s) 1 run of length 3 (1s)

1 1 1 0 0 1 0 1

1 0 0 1 0 1 1 1

1 1 0 0 1 0 1 1

1 2 3 4 5 6 7

Note: external & internal LFSRs with same primitive polynomial do not generate same sequence (only same length)
C. Stroud, Dept. of ECE, Auburn Univ. 10/04

LFSRs (cont)
Reciprocal polynomial, P*(x)
P*(x) = xn P(1/x)
example: P(x) = x3 + x + 1 then: P*(x) = x3 (x-3 + x-1 +1) = 1 + x2 + x3 = x3 + x2 +1

if P(x) is primitive, P*(x) is also primitive


same for non-primitive polynomials

Polynomial arithmetic
modulo-2 (xn + xn = xn - xn = 0)
Addition/Subtraction (x5 + x2 + 1) + (x4 + x2) x5 x2 1 + x4 x2 x5 x4 1 = x5 + x4 + 1 Multiplication (x2 + x + 1) (x2 + 1) x2 + x + 1 x2 + 1 x2 + x + 1 x4 + x3 + x2 = x4 + x3 + x + 1
C. Stroud, Dept. of ECE, Auburn Univ. 10/04

Division x2 + x + 1 x2 + 1 x4 + x3 + x + 1 x4 + x2 x3 + x2 + x + 1 x3 +x x2 + 1 x2 + 1 0

LFSRs (cont)
Non-primitive polynomials produce sequences < 2n-1
Typically primitive polys desired for TPGs & ORAs

Example of non-primitive polynomial


P(x) = x3 + x2 + x + 1
External Feedback LFSR Internal Feedback LFSR

DQ DQ CK 0 0 0 0 0 0 1 1 1 1 1 1 DQ CK 1 1 0 0 1 0 1 1 0 0 0 0 1 1 0 DQ CK 0 1 0 1 0 1 0 1 0 0 0 0 0 0 0 1 1 0 0 1 1 0 1 0 1 1 0 0 1 1 CK

DQ CK

DQ CK

1 1 0 0 1 1 1 1 0

1 0 1 1 0 1

C. Stroud, Dept. of ECE, Auburn Univ. 10/04

LFSRs (cont)
Primitive polynomials with minimum # of XORs
Degree (n) 2,3,4,6,7,15,22 5,11,21,29 8,19 9 10,17,20,25,28 12 13,24 14 16 18 23 26,27 30 Polynomial xn + x + 1 xn + x2 + 1 xn + x6 + x5 + x + 1 xn + x4 + 1 xn + x3 + 1 xn + x7 + x4 + x3 + 1 xn + x4 + x3 + x + 1 xn + x12 + x11 + x + 1 xn + x5 + x3 + x2 + 1 xn + x7 + 1 xn + x5 + 1 xn + x8 + x7 + x + 1 xn + x16 + x15 + x + 1

C. Stroud, Dept. of ECE, Auburn Univ. 10/04

You might also like