PKI REQUIREMENTS FOR PAKISTAN
By Dr. Syed Irfan Hyder
July 30 - Aug 05, 2001
This article briefly describes the role of public key infrastructure, certification authorities and public and private key encryption in dealing with security issues of electronic transactions, especially in the business environment of Pakistan.
Traditionally systems have been using symmetric key encryption i.e use of a single key for encryption and decryption to ensure privacy, authentication and integrity of electronic messages. For example, when ATM machines are used with a pin number and an ATM card for banking transactions or when a teller logs in the computer in a bank branch using password authentication, symmetric encryption is employed.
There are certain attributes of these transactions which are not present when a similar transaction is happening through the Internet. (i) The employee or the customer was present in person and was physically authenticated when the password or pin-number was issued by the bank. (ii) The employee or the customer is physically reachable and could be asked for any authorisation or authentication using other forms of identities. (iii) The machine and the place from where it is used for accessing the transaction is physically known or managed by the bank or can be identified using caller line identification, (iv) there is a well controlled and manageable number of clients or customers for specific transactions.
Presence of these four attributes provide security and comfort to the bank and enables it to provide electronic services even though there is no e-commerce legislation in the country. Similarly, Central Depository Company (CDC) allows its members to make share transactions electronically. Some banks have already given computer terminals to its trusted corporate and preferred customers from which on leased lines they are routinely making bank transactions. Even Customs has started receiving shipping manifests on diskettes and is conducting limited type of transactions with preferred customers.
The aforementioned four attributes are typically not present in Internet based transactions. Internet digitally separates the clients from the bank. In many cases, it may not be physically possible for the bank to authenticate the client as for instance if one opens a merchant account in some Internet bank in USA. The place and machine of transaction are neither managed nor known in advance to the bank. Anyone could potentially initiate the transaction from anywhere or any machine.
It is here in these conditions that the symmetric key system employing an identical key for encryption and decryption is no longer helpful. It also poses some problems. First, if the secrecy of key is compromised by any of the parties then the system becomes open. The keys have to be periodically changed to avoid this problem. Therefore it is vital to keep the key secret. Second, if some distance separates the parties who wish to exchange messages, there is a need to transport the key from one party to another prior to any message exchange. For Key transportation some secure channel would have to be used, other than the one that is used to communicate encrypted messages, otherwise, the key itself will need to be encrypted by another key, and this would regress infinitely. This problem was overcome in the past by exchanging beforehand pads of one-time-use only keys, or by installing software manually at the two ends or employing other similar methods. These methods work well if the number of parties involved is not very large and the distance separating the parties is manageable, however in case of large number of parties that may be widely separated some trusted third party provider of keys must be involved.
These problems are solved using asymmetric key encryption in which a pair of keys for every individual. One of these keys is made public and the other one is kept private. A message encrypted by one of these two keys can only be decrypted by the other key of the pair. If A wishes to send a message to B, he will encrypt the message with B's public key. This will assure A that only B will be able to read the message because only she has the other key (private key) of the pair. This means that the key for locking the message is different from the key used for unlocking. The public-key algorithm has made the digital signature application possible. But the authenticity of digital signatures (dignature) depends upon the confidence that the public-key used in the dignature is indeed owned by the "signatory". So to make dignatures worthwhile we need a trusted third party to certify that owner of the public-keys are who they claim to be. This certification can be carried out electronically by issuing digital certificates.
Prerequisites for the use of Digital Signatures
Paper signature usually have an intrinsic association with a particular person because they are that person's own handwriting. However, public-private key pairs used to create digital signatures have no intrinsic association with anyone in particular — they are nothing more than large numbers. When a recipient obtains the public key actually for a digitally signed communication, how can he or she verify that the public key actually belongs to the purported sender? An impostor could have generated the public-private key pair and entered that public key in a public database under the purported sender's name.
The solution to this problem is to enlist a third party, trusted by both the sender and recipient, to perform the tasks necessary to associate a person or entity on one end of the transaction with the key pair used to create the digital signature on the other. Such a trusted third party is called a certification authority."
Role of Certification Authorities
The trusted third party that issues these digital certificates is known as a Certification Authority. A CA can be a government agency or a commercial body. It is only a matter of trust. An "electronic identity", issued by a CA, is proof that the user is known by the CA. Before issuing a digital certificate, the CA performs identity verification on the user or business entity. The CA acts like a trusted electronic notary, telling everyone who the valid users are and what their digital signatures should look like. Therefore, through third-party trust, anyone trusting the CA can have confidence in the user's identity.
Need for Cross-certification
Cross-certification is simply an extended form of third-party trust. It is a process in which two CAs securely exchange cryptographic keying information, so that each can effectively certify the trustworthiness of the other's keys. From a technical perspective, the process involves the creation of "cross certificates" between two CAs. Therefore, users in one CA domain can implicitly trust users in the other CA domain.
Since cross-certification extends third-party trust, it is important that each CA domain, in addition to exchanging cryptographic keying information, be completely comfortable with the other domain's security policies and practices which it employs in issuing certificates and in carrying out its operations.
Need for Public Key Infrastructure
It is reasonable to expect that with growth in e-commerce, the number of CAs will also grow, just like the number of ISPs has grown. With multiple CAs operating there will be a need to establish chains or hierarchies of CAs to achieve the goal of cross certification and certification of CAs themselves.
This infrastructure which binds all CAs together is known as a PKI or Public-Key Infrastructure. It is the whole system of digital certificates, certificate servers and CAs. PKI manages the generation and distribution of public/private key pairs, as well as the certificates used to provide confidence in the validity of the keys. Although in principle public keys are available to anyone, it is important that their authenticity and ownership be verified by a PKI.
A PKI provides: confidentiality; access control; integrity; authentication; and non-repudiation services for electronic commerce transactions, and for their supporting information technology applications, and publishes the public key (along with the user's identification) as "certificates" on open bulletin boards (such as X.500 directories).
A Public Key Infrastructure (PKI) comprises a number of services, which may or may not be provided by a single organization. A PKI consists of the following: Certification Authority, Registration Authority, Certificate Directories, Certificate Revocation System, Management of Key Histories, Timestamping, Root Cas, Other services may include: Key Backup and Recovery System, Support for Non-Repudiation, Automatic Key Update, and Cross-certification.
PKI is extremely beneficial where there are large number of partners who are unknown to each other or where the number is so great that it is difficult to manage their relationships as for instance in e-Government correspondence among Federal Government, Provincial Government and Local Government employees.
Need for Certification Authorities and PKI in Pakistan
As seen from the above discussion, transactions on the Internet require an elaborate public key infrastructure consisting of legislation, certification authorities, repositories and regulation. However, as argued in an earlier article "B2B and/or B2C Infrastructure" in Pakistan & Gulf Economist, March 12-15, 2001, the potential for primarily internet based transactions with consumers that are complete strangers to business firms is very slim. There are not just enough internet enabled "C" in Pakistan and internet enabled "B" retailers to make B2C transactions feasible in Pakistan.
What is feasible in Pakistan is electronic linkages of established communities of trading partners and converting their existing paper based transactions into electronic form. This automation effort is technically, economically and cost-wise justifiable in our current state of economy. These well-known, well-established communities exist in various industries and sectors. For, example linkages of consumer goods manufacturers with their established community of distributors can be justified on business grounds. Similarly, linkages of oil companies with their dealers and linkages of pharmaceuticals with their retailers is also justifiable. Also linkages students fee transactions done by parents with schools is another case with business feasibility. Pakistan Customs has a list of established authenticated traders known as Gold Customers. In total there may be about 800 volume exporters and importers whose authenticity and other credentials have been verified by Customs over the years who also contribute to more than 80% of the transactions and with which Customs may be very interested in establishing electronic linkages.
Transacting with these parties does not require a national PKI but may require some trusted third party like a value added service provider. Similarly, when an Oil company has to link up with its dealer network; or where parents have to pay fees to a school; or consumer goods manufacturers have to transact with its distributors then in each of these cases the parties are known to each other and are physically accessible for purposes of verification, registration and authentication. These transactions can easily happen through a trusted third party value added service providers which are popularly known as B2B (Business to Business) exchanges. In such cases the trusted service provider performs many of the functions of the PKI and in fact may also use some kind of restricted PKI service or PKI-like security mechanisms.
For many of these communities of trading partners an involved national pubic key infrastructure may be an over-kill given our current level of development. It may be more feasible initially to use the traditional systems for physical authentication and password protections to enable e-commerce using proprietary, leased line or virtual private networks further enhanced by hardware authentication. Later as the confidence grows and as the electronic transactions become more widespread, and banks shed their apprehension, the time would be ripe to introduce public-key infrastructure at a national scale. There would then be commercial reasons for asking prominent service providers like VeriSign or Baltimore to develop national PKIs. For the moment, PKI solutions can also be implemented on a restricted scale by many of these trading communities. They can themselves evaluate the costs and the feasibility and can select the technology most suited for their transactions.
The phenomenon of e-commerce is still in its early stages, where it is trying to take a concrete shape. Merchants are concerned about the integrity, authenticity and non-repudiation of electronic messages especially those traversing the internet. Consumers ask how can they trust a web site offering to deliver goods to their doorsteps provided they submit their credit card number now. Business transactions may involve legal contracts, or exchange of confidential plans. How will parties obtain redress in cases of dispute or fraud? Similarly banks see additional risks in supporting e-commerce over the open channel of Internet. All these issues are manageable on restricted networks or virtual private networks. These are also manageable when we analyse the attributes of trading communities with the business reasons to do e-commerce.
The author is Vice President and Dean, PAF-Karachi Institute of Economics and Technology. He is also Head of E-Commerce Task Force of Ministry of Science and Technology.