Protocols for Public Key Cryptosystems - Semantic Scholar

15 downloads 365005 Views 648KB Size Report
SYSterns are briefly considered by exa- mining cryptographic protocols for key distribution and digital signatures us-. This work was partially supported under.
PROTOCOLS

FOR PUBLIC KEY CRYPTOSYSTEMS

Ralph C. Merkle ELXSi

International

Sunnyvale,Ca.

ing both Abstract

New Cryptographic take

full

public

key

and

systems. protocols

The reader is assumed to

which liar

advantage of the unique proare

lic key cryptosystems,

now

for

[1,10]

Several

public key distribution signatures

are

protocols and for

briefly

we

with

each other and with the conventional

as

there

assume

and

ternative.

B

examples

are two communicants, opponent

attempt

A

E

will

to discover the keys, learn the

secrets, times,

E.

will attempt to send secret mes-

sages and sign contracts, while 1. Introduction

in

.

called A and B, and an

al-

fami-

described

For many of the following

digital

compared

be

with the general ideas behind pub-

perties of public key cryptosystems evolving.

conventional

and A

forge

will

contracts.

Some-

attempt to evade a con-

The special strengths of public key tract he signed with B, or SYS terns are

B

will

at-

briefly considered by exatempt

mining cryptographic

for

protocols

to

forge

A’s signature to a new

key contract.

distribution

and digital signatures usA and B will need to apply one functions

to various arguments of vari-

under

ous

so we assume

and much of the

way

This work was partially supported NSF

Grant

work was ISL.

done

10173, at

Stanford

The author would

nowledge much

ENG

the support

of the work

also

University like

of BNR

reported

An extended version has

way

Inc,

fixed

where

[2,9,13,19].

submitted

122 IEEE

we

have

a

one

F which can be applied to any

size output.

discussion of

to CACM.

CH1522-2C/80/OOOO-O122$O0.75@1980

function

arguments of

to ack-

here was done. been

sizes,

one

size

and

produce

a

For a more complete way

functions,

see

The major drawback of this protocol 2. Centralized ~

Distribution is its vulnerability

to both centralized

loss of security and centralized loss of Centralized conventional

key distribution

encryption

of

bribery of personnel at the central site

handling

will compromise all users of the system.

in a multi-user network

Similarlyr destruction

environment before the discovery of pub-

keys

lic key distribution methods.

mechanism

Only con-

ventional encryption functions used,

which

lic

need

(Presently known

cryptographic

systems

Discovery of new public

seems

almost

discovery of more efficient

ones

all

and prob-

key

all

with

but

key

B,

the

single

centers,

Destruction will

center

must

be

improved

center.

The

master

still be stored securely

the

(encrypted)

user

This approach is

This protocol does not fully

the

solve

mon (session) key to A and B

of key distribution method must be

agreed

and

B

further

can

then

on

central keys,

communicate

with

the A no

only

functions.

is nontrivial because no electronic com-

simple

conventional

and

inexpensive

physical methods, e.g., registered mail,

re-

offer only moderate security.

encryption

of

Its use has been defended in

the literature

This problem

munications can be used and is

used

each user and the center to es-

tablish the original keys.

bution center.

quires

key distribution problem: some SOrt

between

assistance from the key distri-

This protocol

used

by IBM [23].

the

prf2ViOUSly

(and

keys can be

key distribution center will send a comusing

if

user keys are encrypted with a

stored anywhere.

a central key distribution center.

If A wishes to communicate with

more

or

suitable provision made for its backup) ,

other system users somehow

deposit a conventional cryptographic

of

key distribution can be in-

master key by the

In centralized key distribution, A, and

reliability

Security can also

able. )

B,

and

not affect the other centers.

key

inevitable,

security

or compromise of a

systems.

central

distribution

key

each with its own keys [1].

Whether or not this will continue is not now known.

the

creased by using two

pub-

the

of

for all users.

centralized

key systems are less efficient than

conventional

destroys

The

be

presently offers a perfor-

mance advantage.

Theft of the central keys, or

function.

was

functions

the only reasonable method key distribution

using

couriers

is

reasonably

although more expensive.

[17,18,20].

123

The

use

secure,

The disadvantage of is 3.

Simple Public ~

this

protocol

that E might actively interfere with

Distribution the exchange of keys.

Worse yet, E

can

force a known k on both A and B. application

This is the most basic of

public key systems

Its

[1,5,6,7,8].

purpose is to allow A and B to agree

on 4. Authenticated

a

Public ~

Distribution

common key k without any prior secret though

arrangements, even all messages.

overhears

keys EA and

then

A computes

E).

‘A “

between A and B is:

It is used to using

to

A

and EB and make them public, while keep-

= k.

A

ing DA and DB secret.

is

al1

the

have

mon

conversation

the

com-

k, then each sends a (session)

others

public

key.

it

with

The two keys

thus agreed on are combined and used

to

encrypt further messages.

great

deal

Firstr no keys and no before

to

simple,

and

recommend

it.

secret

At the end of this protocol, A B

materials

finished.

and

have agreed on a common key, k, which

is both secret and authenticated.

A and B start communicat-

ing, and nothing is retained after have

key

key to the other by encrypting

repeated to agree on a

This protocol is very

exist

enci-

If A and B wish to agree on a

new key k“.

a

The public

ticated access to EX for any user, X.

a conventional

Once A and B

resume

process

encrypt

they both discard k.

talking,

If they later

has

A and B generate EA

a public filer allowing easy and authen-

encryption function.

the

and authenticated communications

The key in future communica-

both

messages

finished

for

phering keys of all users are entered in

tions is k. further

DA(EA(k))

[1]

protocol

EA and DA, and B dis-

discards

cards

secure

DA,

B picks the

sends EA to B (and E).

random key, k, and transmits EA(k) (and

The now classic

A randomly computes enci-

phering and deciphering and

E

This

they

protocol

weaknesses.

It is impossible for E

suffers

from

two

Firstr entries in the pub-

lic file might be altered.

This can

be

to compromise any keys either before the

dealt with both by good physical securi-

conversation

ty, or by using new protocols

takes place, or after it is

over, for the keys exist only during the

tions

conversation.

entries in the public file.

124

5

and

(see

6) for authenticating

secthe

Secondr secret deciphering be

lost.

can be published in newspapers ‘CA magazines, and sent over all avail-

and

keys can

able

This problem must ultimately

communication

channels:

blocking

its correct reception would be very dif-

be solved by good physical security.

ficult. If DCA is compromised, then no 5.

Public ~

Distribution with Certifi-

longer

it

is

possible to authenticate the

users of the system and their public en-

cates ciphering

keys.

The

certificates are

now worthless because the (unauthorized) Kohnfelder entries

them

with

He

DCA.

who

‘as learned ‘CA can produce false certificates at will.

in the public file be authenti-

cated by having a Central Authority sign

person

[3] first suggested that

(CA)

called such

signed entries certificates. 6.

The protocol with

certificates

Public ~

Distribution —— with Tree — Au-

is thentication

the

same as the authenticated protocol, now

except that A and B can entries

in

This

Key distribution with

A

assures

can be compromised, resulting in

and B that each has the

wide loss of authentication.

not

tication

poster.

authen-

[13].

Again, this

The security of this protocol rests

CA system

This prob-

lem can be solved by using tree

enciphering key of some im-

public

certificates

was vulnerable to the criticism that D

proto-

other’s public enciphering key, and the

the

the public file by checking

each other’s certificates. col

check

protocol

attempts

to

on the assumptions that the secret deci-

authenticate entries in the public file.

phering keys of A, B, and

However, instead of signing

been

compromised;

correct copies signed

of

that ‘CA

certificates) ;

not issued deliberately

a

bad

CA A (to

and

have and

it

in

B have

check

entire

trustworthy, or accidentally because

public

entry

file, this protocol ap-

public

file.

Even though H is

applied to the entire public

either

was

the

each

plies a one way hash function, H, to the

the

that CA has

certificate,

because

not

output

un-

long.

it

called

was tricked.

125

of

H

is

only

100

file, or

The (small) output of H

200

will

the bits

be

the root, R, of the public file.

then

If all users of the system know R,

users can authenticate the correct-

all

ness of the (whole) public file by R = H(public file).

puting

to introduce file

will

into

changes imply

R

Where F is a one way function.

com-

Any attempt

enciphering

the

the

public

eliminates

bits

key exists.

to

the

harsh

public

long.

In a similar fashion,

A

really

in the public file is effectively impos-

half of the public file, for

high

degree

has

of

been

published,

assurance

that

it is

selectively authenticate

possible

individual

to know the whole public file

by

Merkle*s “tree authentication,” ‘l’heessence to

is

of tree

authenticate

:

all of the first

the

All A needs to know is the first quarter of

using

[131.

public file (which has YB), and

By applying this concept recursively ,

public

de-

= public file = Yl, Y2, ... Yn,

(so the ith — entry in the public file

the

H(second quarter of public file).

authentication entire

A can confirm YB in the public file R,

log2

Knowing

only

values,

and YB itself.

n

intermediate

The information

needed to authenticate YB, given that

is

has

already

been

authenticated,

denoted Yi, and B’s entry is YB); we can

along the path from R to YB and will

define H(public file) = H(x) as:

called the authentication path. These definitions

H(z)

)

to en-

file by “divide and conquer.” If we fine

know

H(second quarter of public file)

in the public file without having

tries

to

not

F( H(first quarter of public file),

is

it

need

does

H(first half of public file) =

This method is impractical as statFortunately,

100

a

~orrect can be attained.

ed.

only

A can compute H(public file)

scrutiny, and because making alterations

sible after it

filet

file.

of public

glare

the

only knew half the entries in the public

Because the public file will be subjected

of

knowing only this information, and yet A

possibility of compromising DCA be-

cause no secret deciphering

half

public

then A need only know

half of public file) which is

file), an easily detected fact.

the

first

key,

B-s

(which is where YB appears) and Ii(second

# H(altered public

This method effectively

to confirm

If A wishes

= F( H(first half of ~),

are

R

lies be

illustrated

in figure 1, which shows the authentica-

H(second half of ~) )

tion path for Y5.

126

The For a more detailed discussion

only

practical

method

of

the compromising

this

protocol

is

to

reader is referred to [13]. compromise DA or DB. user

Using tree authentication,

is has

A user’s

A

security

thus dependent on himself and no one

an authentication path which can be else.

used to authenticate user A*s public enciphering

key, provided only that R has An “authen-

already been authenticated.

7. Digital Signatures

tication path” is a new form of certificate, with ECA replaced by R. comprom-

This protocol can only be

The use of public key cryptosystems if: DA or D~ is compromised, or if

ised

R is not correctly known by A or

B,

to

provide

digital signatures was sug-

or gested

by

Diffie

and

Hellman

[11.

if there is a false and misleading entry Rivest, Shamir and Adleman

[8] have sug-

in the public file. gested easily

The latter two are

an

attractive

implementation.

detectSignature

techniques

based

on methods

If either A or B has the wrong R,

able.

other than public key cryptosystems have they will be unable to complete the probeen

suggested

by

Lamport

and Diffie

tocol with any other legitimate user who [1,241~ Rabin [151, and Merkle has the correct R, a fact that

will

[13].

be Digital signatures,

whether

based

quickly detected. on conventional encryption functions, on Because the

is

file

public

both public key cryptosystemsr on probabilis-

open to public scrutiny and unalterable, tic computations, or on other techniques false or misleading entries can

be

ra-

a

few

correctness

can

share several pidly

detected.

In

users

concerned

with

practice,

common.

important

These

common

best illustrated by verify

that

the

the

properties

in

properties

are

following

now

public file satisfies classic example.

some

simple

each

user

global

properties

i.e., A wishes to place a purchase

name

appears

order

once and once with

only in the entire public file;

his

stock

broker

B.

A, on the

indiviRiviera, cannot send a written order

dual

users

can

to

then verify that their B

in

New York in time.

All that A can

own entry is correct, and need not bothquickly send to B is information, er

examining

the

rest

of

i.e.,

the public a

sequence

of bits, but B is concerned

file. that A may later disclaim the order.

L27

A

somehow generate a sequence of bits

must

(a digital signature) vince

which

will

disputes

con-

understand

B (and if need be a judge) that A

authorized the order.

be

It must

than

A)

to

generate

it

(to

different.

Failure

to

this point has led to confu-

sion in the literature

easy

We now

for B to validate the digital signature, but impossible for him (or anyone

are

to

turn

[17,20].

specific

digital

signature protocols.

other prevent

charges that B was dabbling in the mark8. ~ Conventional

et illegally with A’s money) .

Signature Protocol

There are digital signature schemes which

do

not

tosystems but it will be convenient tationally

to

let

a

signature

will

an

illegible

If

This

notation

is

generating

and validating signatures can

simply

relying on CA

CA,

to

This protocol weaknesses

of

and

a

by sending it to CA and adjudicate

disputes.

This approach is defended by some

somewhat

because the actual method

that if A and

A and B have a secure method of com-

message

as

misleading

on the observation

municating with CA, then A can “signw

(random

bits) then the signature is rejected invalid.

if

Check-

EA(DA(m))

message

“signature” protocol

B trust some central authority

then be done by

computing m = EA(DA(m)). produces

relies

ro-

A sign message m by

computing the signature, DA(m) . ing

A conventional

involve public key cryp-

is

subject

of centralized

[17’].

to

the

key distribu-

tion (described earlier).

be very different from this model; it is retained

because it is widely known and

because we will not discuss the ences

differ-

among different digital signature

methods, only their common properties. Digital signature protocols are naturally

parts:

a

method of signing messages used by A,

a

method

divided

for

three

authenticating

used by B, and a disputes,

into

method

for

resolving

used by the judge.

It is im-

only

The Basic Digital Signature Protocol ——

a signature

portant to note that two protocols differ

9.

The first public key based signature

that

protocol

having A sign

in the method of resolving

DA (m)

128

and

[1],

message

m

digital

proceeded by

by

computing

giving it to B as the signed

message.

B

EA(DA(m))

(or

=

a

judge)

m,

thus

can

been confused with each other

compute

confirming

the

correctness of the signed message.

for

this

Some criticism of “the” public

reason.

key digital signature protocol has actu-

A is

held responsible for a signed message if

ally

and only if it can be verified by apply-

failed to consider the first protocol at

ing A’s public enciphering key to it.

all.

This

protocol

[16,17,20]

on

public file with.

can

two

have

of

been

public file, discussed previously key

distribution

under

tampered

authenticating

the

and

signed

message.

protocols, solve this

should

be

compromised

and

made

this

the case. The major

be

second A

will

difference

in the second

col

if he can

A’s signing key is compromised.

security for

very

good

physical

A“ The loss to A if DA is compromised can be substantial.

problem

B

will

be left holding the bag if

Clear-

condition is unlikely before he will willing

this

be

to use this protocol.

is to alter the dispute resolu10.

tion protocol so

that

responsible

his

for

secret deciphering

A

is

not

signature

key

is

The Time-Stamp Protocol ——

held if his A protocol that would

compromised

and made public.

report

The fact that altering the

messages signed after the reported

dispute

yet

protocol has not been fully appreciated,

of signatures made before

and

loss

preceding

two

allow

protocols have

129

A

to

loss or theft of DA and disclaim

resolution procedure creates a different

the

proto-

ly, B must be given assurances that this

D

A different method of solving

the

protocol and the first is in the

division of risk:

only

between

agree to this digital signature protocol provide

know

easy to design a system in which this is

held responsible. that

not

then he is unable to disavow his A’ It is signature under this protocol.

can sign any message they desire

seems clear

protocol

D

public.

will

the

is inadequate.

his secret deciphering

with A’s compromised D A, and A

It

disavow

DA

For this reason, some

critics have argued that

under

A second criticism is that A has no

Anyone

effectively

If we assume that A does

recourse

then

the second protocol A can make

public

problem.

key

of this second protocol, and

If we assume khat A knows DA,

criticized

grounds: First, the

might

Methods

be

been

loss

force A to acknowledge the validity

must

the

reported

involve the concept of time.

into

the

following

We introduce

time

protocol

using time-keepers who can

by

digitally time-stamp to

them.

have

We

agreed

The major disadvantage of this pro-

given

tocol ,

assume that both A and B on

a

whose

time-keepers

information

been dealt with fairly.

set

of

tal signature protocol, is the

acceptable

a

accepted in dispute resolution.

then

lost,

answering

‘A

been

he must report this fact to

some agent who will be queries

the

not.

For

this role is played by the thority,

CA.

CA

central sign

will

and decreases relia-

expense

B

is

the

transaction

i.e.,

messages

to

willing

obtain

and the validity-check

time-stamp

au-

both

increases

If

we shall assume

simplicity,

real

network, which

current or

in

of a communications

bility.

status of DA, i.e., has it been lost

presumably

These requirements force the use

for

responsible

about

validity-check,

time. has

require-

ment that B obtain both a time-stamp and

time-stamps will be

If A can report that

as compared with the basic digi-

within

a

the after

has

been

few

days, an off-line

system can be used.

completed,

This modified

pro-

stating that A-s secret deciphering

key

tocol

has

the

fail-soft protocol during communications

not

been

current

compromised These

time.

as

signed

of

messages

outages,

will be called “validity-checks. ”

message

keeper

A

B

time

stamp

then

has

a

more

time-

If

from CA.

by B either as a

or as the standard protocol if

the

DA

A might loss

about

of

it.

have and

‘A If

recently

physical

secret

the signature, otherwise he accepts.

risk should be minimal.

holds

that

a

resolution,

the

applying

for

is good, this

checked

time-stamped

prior

11.

key

This protocol

provides

Witnessed Digital Signatures

to

any reported loss of DA.

assurance

keys

security

judge

A’s public enciphering

AND it has been

deciphering

reported

message has been validly

signed if and only if it can be by

ahd

B would not know

has aiready been reported lost B rejects

In dispute

cheaper

reliable, but it exposes B to some

risk:

the message and ob-

tains a validity-check

used

Off-line operation is

m by computing DA(m) and

sending it to B.

be

communication costs are too high.

In the time stamp protocol, user signs

could

If the value of very

good

high

to all parties that they have

enough,

it

a

transaction

is

might be desirable to

have a witness physically confirm that A

130

signed message m.

The witness, w, would

vious solution is for updates to be

di-

compute DW(”I, W, physically

saw A agree

gitally signed by an appropriate network

to

It would be

administrator,

and

sign message m.”) .

necessary for A and B to

agree

in

ad-

to

is that it reduces B’s risk.

find

pro-

example

leads

application

of

This

The

another

is that it forces A

tures in operating system

a (physically present) witness

code

to

Applications

to

digital signasecurity.

A

the

that

that it is executing today

is not the same that Signature

naturally

ing system is the possibility system

Digital

nodes

major risk to the security of an operat-

to confirm the transaction.

12.

the

ecuting them.

The primary advantage of this

primary disadvantage

for

check the digital signature prior to ex-

vance on acceptable witnesses.

tocol

and

it

was

executing

yesterday: someone might have put a trap

Not

door into the operating system that lets Involvinq Dispute them

do anything they please.

against this possibility, Not

all

applications

of

digital

system

the

To guard operating

could refuse to execute any code

signatures involve contracts between two

in privileged mode unless that code

potentially disputing parties.

been

signatures

Digital

also an ideal method of

are

properly

signed.

had

Carried to its

logical conclusion, the operating system

broadcasting authenticated messages from

would

a central source which must be confirmed

privileged programs each time they

by many separate recipients, or

loaded into central memory If this check

edly

repeat-

confirmed by the same recipient at

different times to insure that the

is

mes-

be

network.

It

application

physically

communications

ly

On the other hand, it

over the network itself.

it

were

would

for any software changes The

incapable

machine

would

be

of executing code was

If privileged programs are digitalsigned

by the programmer who origi-

nally wrote them, as well as by

is very desirable to send updates to the nodes

signature of

signed.

would be clearly undesir-

software.

digital

in privileged mode unless that code

abl~ for any node to start executing the wrong

impossible

to subvert it.

the distribution of network software

to individual nodes of a

the

were implemented in hardware,

sage has not been modified. One example of such an

check

supervisory

The ob-

various

levels, and if the computer

is physically unable to execute unsigned

131

15.

code in privileged mode, then it is poscomplete

sible to have the

privileged

assurance

this

course,

mean

that

secure,

been

1.

modi-

the

but

the

does

W.,

and

Hellman,

in cryptography.

M.

New

IEEE Trans.

on Inform. IT-22, 6(Nov. 1976), 644-654.

programmer.

not necessarily is

2.

Evans A., Kantrowitz, W., and Weiss,

does eliminate a major

E.

A user authentication

system

operating

it

Diffie,

directions

they were given there final

since

checkout and signed by Of

that

programs running on the

computer “right now have not fied

BIBLIOGRAPHY

system not re-

quiring secrecy in the computer.

class of worries.

Comm .

ACM 17, 8(Aug. 1974), 437-442.

3. 13.

Kohnfelder, L.M. Towards a practical

Conclusions

public-key Bachelor’s This paper has briefly described number of cryptographic

protocols.

a 4.

Cer-

Liptonr S.M., and Matyas, S.M.

ing

sible;

safeguarded.

what

however, they are valuable tools

the

digital

signature

Mak-

legal--and

Data Communications

(Feb.

1978), 41-52.

they illustrate

can be achieved and provide feasi-

ble solutions to problems

5.

recurring

of

McEliece, R.J. based

tosystem

interest. Further constructive work

in

theory.

this

A

public-key

on

algebraic

crypcoding

DSN Progress Report, JPL,

(Jan.

and Feb. 1978), 42-44.

area is very much needed. 14.

EE

thesis.

tainly, these are not the only ones pos-

to the system designer:

MIT

cryptosystem.

ACKNOWLEDGEMENTS

6.

over It is a great pleasure for the

au-

Merkle,

R.

Insecure

Secure

Communications

Channels.

COmm. ACM 21,

4 (Apr. 1978), 294-299.

thor to acknowledge the pleasant and informative conversations

he had with

7.

Dov

information

Andelmanr Whitfield Diffie, Martin Hellman,

Raynold

Kahn

Loren

Merkle, R., and Hellman, M.

Kohnfelder,

knapsacks.

and

signatures in trapdoor

IEEE Trans. on

24, 5(Sept. 1978), 525-530.

Frank Olken, and Justin Reyneri.

132

Hiding

Inform.

IT-

8.

man,

L.

16.

Adle-

Rivest, R.L., Shamir, A.t and

Saltzer, J. On Digital

Signatures,

private communication.

A method for obtaining digital

signatures and public-key cryptosystems. 17

Comm. ACM 21, 2(Feb. 1978), 120-126.



Popek G.J. and Kline, C.S.

tion 9.

Wilkes, M.V., Time-Sharing

Enc ryp-

Public Key Algorithms,

Protocolsr

and Digital Signatures in Computer

Computer

Net-

works; in Foundations of Secure Computa-

Systems. Elsevier, New York, 1972.

tion pp. 133-153. 10.

Diffie,

W.,

and

Hellman,

Privacy and authentication:

M.E., 18.

an introduc-

and

Needham R.M.

tion to cryptography, Proceedings of the

Using

IEEE

Large Networks of Computers.

Vol. 67, No. 3, Mar. 1979 pp. 397-

M.D.

for Authentication

in

CACM 21,12

Dec. 1978 pp. 993-999.

427.

11. Squires, J. phones,

Russ

Chicago

monitor

Tribune

Us.

of

pp. 123, June

19.

Merkle, R. Secrecy, authentication,

and

public key systems.

Stanford Elec.

Eng. Ph.D. Thesis, ISL SEL 79-017, 1979.

25, 1975.

12. Davis, R. Remedies sought to

20.

defeat

Microwave Syst., vol. 8, no. 6, pp.

Popek, G.J., and Kline,

cryption

Soviet eavesdropping on microwave links,

and

C.S.

En-

Secure Computer networks.

Computing Surveys

17-

11,4

1979

Dec.

pp.

331-356.

20, June 1978.

13.

Encryption

Schroeder,

Merkle, R.C.

A

certified

21.

digital

Simmons, G.J. Symmetric

metric

signature, to appear, CACM.

Encryption.

and

Computing

AsymSurveys

11,4 Dec. 1979 pp. 305-330. 14.

Kahn,

D.

The

Codebreakers,

New 22.

York: Macmillan. 1967.

Lamport, L. Time, clocks,

and

the

ordering of events in a distributed sys15.

Rabin,

M.O.

,

Digitalized

tem.

signa-

turesr in Foundations of Secure Computation, ed. Demillo,

R.A.,

et.

al .

pp.

155-166.

133

CACM 21,7

Jul

1978 pp. 558-565.

23.

Ehrsam, W.F., Matyas, S.M.,

C.H., and !l?uchman,W.L. key management scheme the

24.

A cryptographic for

implementing

data encryption standard.

Jour. 17,2 1978PP.

Lamport,

IBM

SYS.

106-125.

Constructing

L.,

Meyer,

digital

signatures from a one way function.

SRI

Intl. CSL - 98

Y1

‘2

Y’3

Y4 FIG. I

134