Internet was designed specifically as a public network for sharing information.

Correspondence over internet known as e-mail is said to be more convenient than trying

to reach someone by telephone. However, the use of e-mail has s security problem as it

is quit easy for someone to intercept your electronic mail. Security, that is, confidential,

reliable, and known delivery is essential to the success of e-mail. In other words, people

will not use a mail system that they cannot trust to deliver their messages. The

fundamental mechanism for providing security for binary encoded messages in an open

network is encryption. The aim of encryption is to turn an otherwise intelligible message

into gibberish so that anyone who intercepts the message can’t read it. There are two

currently hottest encryption programs available in the market that can provide reliable

protection of electronic mail. These programs are the Pretty Good Privacy (PGP) and the

Privacy Enhanced Mail (PEM).

PGP is an e-mail security program developed a few years ago by Phil Zimmermann.

According to Stalling (1994), PGP combines confidentiality and digital signature

capabilities to provide a powerful, virtually unbreakable and easy-to-use page Free ware

versions are available for windows, the Macintosh, DOS, OS/2, and other platforms.

PGP and PEM provides a variety of security features for electronic mail. These

features are:

• Confidentiality
• Data origin authentication
• Message integrity
• Nonrepudiation of origin

Confidentiality protects the contents of the e-mail message against unauthorized

disclosure. Confidentiality is provided by encrypting messages to be transmitted or to be

stored locally as files.

Authentication permits the authorized receiver of the message to reliably determine

the identity of the sender. In the case of forwarded messages, the authentication feature

identifies the forwarder of the message and not the original sender.

Message integrity gives the authorized receiver of the message assurance that the

message has not been modified in transit. The message received is identical to the

message that the sender sent.

Nonrepudiation is the feature that allows one person to forward message to another

person, who can then verify the identity of the original sender.

PGP is designed to work within existing electronic mail system, primarily those used

on the internet. Although the program is not integrated with any mail program or text

editor, it can be used with all of them.

As Stalling (1994) reports, PGP employs the RSA public key encryption method that

is named after its investors: Riverst, Shamrir, and Adleman and the MD 5 ( Message

Digest Version 5 ) to form a digital signature that assures the receiver that an incoming

message is authentic.

Further more, PGP use the conventional encryption algorithm known as IDEA (

International Data Encryption Algorithm ) for data encryption. IDEA is substantially

faster than RSA, so to reduce encryption time, the IDEA/RSA combination is used.

Besides that, the use of RSA solves the section-key distribution problem because only the

receiver is able to receive the session key that is bound to the message. Thus, to the

extent that RSA is secure, the entire scheme is secure. To this end, PGP provides the user

with several RSA key-size options:

• Casual (384 bits) : known to be breakable, but with much effort.

• Commercial ( 512 bits ) : possible breakable by the NSA and similar organizations.

• Military ( 1024 bits ) : generally believed to be unbreakable.

Sending a PGP message consists of four steps stated as belows:

• Signing

• Compression

• Encryption

• Transmission Encoding

PGP message can be digitally signed by the sender. Signatures are optional, the sender

does not have to sign her message. PGP signatures allow the receiver to verify both the

identity of the sender and that the message has not been tampered with the transit. PGP

also supports detached signatures. A detached signature may be stored and transmitted

separated from the message it signs.

message. Because the compressed message has less redundancy than the original

plaintext, cryptanalysis is more difficult. The compression algorithm used for PGP is

ZIP. Note that the signature is generated before compression. It is preferably to sign an

uncompressed message so that you can store only the uncompressed message together

with the signature for future verification. If you were either to store a compressed

version of the message for later verification or to recompress the message when

verification is required.

PGP can both encrypt and sign a single message. First, a signature is generated for the

message and prepended to the message. Then, the message plus signature is encrypted

using a random encryption key. Finally, the random encryption key is encrypted using

RSA and prepended to the encrypted block. Note that the digital signature is encrypted

along with the message. It is impossible to verify the signature on an encrypted message

without decrypting it first.

PGP converts the raw 8-bits binary stream of ciphertext to a stream of printable ASCII

characters.

Receiving a PGP message involves unraveling everything done by the sender. This

step involves transmission decoding, decrypting, signature verification and message

disposition.

If a message is encoded for transmission, it is decoded from ASCII back into 8-bits

ciphertext.

PGP attempts to decrypt the message which is encrypted. First, the software looks at

the packet that contain the encrypted random encryption key, to see if it can decrypt the

message. Assuming PGP is able to, it decrypts the random encryption key using the

receiver’s private key. Then, it decrypts the message using the random encryption key.

Remember that the encryption is an option step. If the original PGP message was

not encrypted, no decryption is necessary.

PGP determines if the message is signed. If it is, PGP decrypts the hash value using

the sender’s public key. Then, it generated a new hash value for the received message

and compare it to the decrypted hash value. If the two match, the message is accepted as

authentic. This is also an optional step; without a signature, there is nothing to verify.

After the receiver has read the PGP message, he has a number of options. He can

store the message in decrypted form without the signature attached. He can store the in

decrypted form with the signature information. This form of storage is appropriate if the

receiver wants to forward a signed message to a third party, and also provides protection

against modification while the message is being stored. If the receiver wants to protect

the confidentiality of the message, he can choose to save the message in the encrypted

form.

The PEM effort began in 1985 as an activity of the privacy and Security Research

Group (PSRG) under the auspices of the Internet Architecture Board (IAB) (Kent, 1993).

The effort has yielded a series of specifications of which the most recent set, Requests for

Comments (RFCs) are proposed Internet standards. These RFCs are product of the PEM

Working Group within the Internet Engineering Task Force, a subsidiary group of the

IAB.PEM is intended to be compatible with a wide range of key management

approaches. It has mechanisms for using conventional (secret-key) cryptography or

public key cryptography for key management.

PEM provides all essential security features for e-mail users. These features are:

• Confidentiality

• Data origin authentication

• Nonrepudiation of origin

• Key management

The security services for e-mail as provided by the PGP and PEM are almost the same

with the exception that PGP is designed to automatically the essential features such as

confidentiality, authentication and integrity for all messages. It is possible to send a PGP

protected message without providing confidentiality. It is also possible to send a PGP

protected message without providing for authentication integrity. However, all messages

processed by PEM incorporate the authentically, integrity and non-repudiation support

facilities whereas confidentiality is an optional security service.

PEM is designed to work with existing electronic-mail systems primarily the e-mail

systems used in the Internet. As such PEM was designed to fit into existing mail system

architectures. PEM messages can be created with test editors, and most mail system do

not destroy PEM messages in transit.

The major features of PEM are stated as follows:

• PEM is not restricted to particular host or operating system, but rather allows

• PEM is compatible with normal, nonsecure electronic mail.

• PEM is compatible with a range of mail systems.

• PEM operations can be performed on personal computers and large systems.

• PEM support privacy protection of e-mail addressed to mailing lists.

Submission processing in PEM involves four major steps:

• Canonicaliztion

• Message integrity and originality authentication

• Encryption

• Transmission encoding

This is the first step in the PEM submission processing. Canonicalization involves

transforming the message content from the "native" representation for the computer from

which the message is submitted, into a network standard representation. The

canonicalization step must be performed as part of the PEM processing, prior to when the

e-mail system would normally perform the equivalent step. Any change to the message

content after the PEM integrity service has been applied would cause the integrity check

performed by a recipient to fail. Hence, PEM must apply this transformation before the

normal email processing. In addition, if the message content is enciphered for

confidentiality, it would be impossible for the normal email canonicalization to take place

i.e. the message content would be just as unintelligible to this email software as it would

be to a potential dropper.

The second step in PEM message processing begins with calculation of the message

integrity code (MIC). PEM treats the choice of the MIC algorithm as a parameter that

can differ among user communities or evolve as better MIC algorithms are developed.

The MIC is calculated based on the canonicalized version of the message so that it can

be verified in any computing environment. To provide both authentication of the sender

and integrity of the message, the MIC must be protected in somewhat manner that binds

it to the message originator. That is , there must be something about the MIC that shows

that it was created by the same person who created the message, and not by someone else.

This step is optional. PEM only encrypts messages if the type is "ENCRYPTED."

The PEM standards can support multiple encryption algorithms, but currently specify just

one : DES in Cipher Block Chaining mode. The algorithm is specified in the "DEK-

Info" field of the message header along with any data that the algorithm needs. "DES-

CBC" specifies DES in Cipher Block Chaining mode, and the string of the characters

afterward specifies the initialization vector (IV).

The last step involved in sending a PEM message is encoding it for transmission. This

step is also optional. The ENCRYPTED an MIC-ONLY message are encoded. Most e-

mail systems are designed to handle text messages and not binary messages. An

encrypted message is binary and hence may be inadvertently modified in transit. Any

modification will make the message decrypt to nonsense at the receiving end, and so

must be encoded to avoid transmission problems.

When PEM software receive a PEM message, it first scans the message to find the

PEM message boundary, and then looks through the PEM header to find out which

version of PEM was used to process the message. Then it proceeds to deal with the

message.

There are five stages in the process of receiving a PEM message. These stages are

decoding, decrypting, verifying message integrity and authenticity, translation and m

message disposition. The decoding, decrypting and message disposition stages are similar

to those described in receiving a PGP message.

The receiver first checks the PEM header to determine which MIC algorithm and

signature is used for this message. Assuming that the sender used RSA as digital

signature algorithm, the receiver gets the public key of the sender form the originator

certificate field and decrypts the signed MIC value. Finally, the receiver computes the

MIC on the canonical form of the message, and then compares this value with the value

he just decrypted. If they match, then the message has been authenticated.

Finally, after verifying integrity and authenticity the canonical form of the message is

translated into what ever representation is proper for the receiver’s system, and then is

displayed for the receiver to read.

PGP and PEM are both electronic-mail security programs. They both encrypt

messages; they both sign messages. They are both based on public-key crytography.

They have different philosophies, though. PEM is based on the concept of a hierarchical

organization, while PGP is based on a distributed network of individuals. The PEM

might be more suited for applications in the companies, governments and other

organizations. PGP is definitely more suited for people on the Internet