Now, that we have acknowledged that the need of a “Unique ID System” exists (Refer my previous post), we must further acknowledge the purpose of deploying such a system i.e. the application areas and the usage environments.
A) Application Areas:

Civil applications
 
• Availing social benefits (RSBY, NREGS, etc) and
• Exercising citizen rights (Voting, etc)  OR

Law Enforcement

Identifying a person from prints left behind at crime scene

Difference between Civil Applications and Law Enforcement

The difference is in the nature of participation of the subject, which is voluntary in the Civil Applications and involuntary in the law enforcement. Also, in the case of the law enforcement, the print left behind, also known as latent prints can be partial or be from the palm area; thereby making the identification very difficult and complex.

Further, the implications of a “False Acceptance” or “False Reject” are very different among these systems. While a “False Acceptance” in a civil application would result in multiple benefits to an individual under a scheme (which can be rectified subsequently), it would allow a criminal prosecution against a wrong individual, and vice-versa for “False Reject”. It would be imperative to note that the system requirements for Law Enforcement system would be more rigid and complex.

B) Usage:

Unique ID Alone Vs Unique ID

It is also important to asses if

• The output desired from such a system is to be able to identify a new record submitted as already existing in the historical database built over time, alone
OR
• To authorize a transaction request with biometric based authorization data.

The primary difference between the two environments being system uptime and availability requirement; while the “Unique ID” can be established online as a transaction, they are generally done in batches for optimization purposes and service is merely delayed, not denied in case of downtime. On the other hand, for transaction environment needs, the system uptime and efficient response are mandatory for continual service else it is perceived as service denied.

Provision for a Transaction Environment

Further, the transaction environment described above envisages an online environment and assumes communication channel availability at the point of transaction. However, if the need is that transaction be allowed offline as well, then there is a requirement of secure token issuance and lifecycle management – which will require a collection of systems like KMS, SCMS, etc.

FINO has successfully implemented transaction environments around a fingerprints based “Unique ID System” for various civil applications.

A National Identification Number (NIN) or National Identity Card number as currently being used by the governments of many countries as a means of tracking their citizens, permanent residents, and temporary residents for the purposes of work, taxation, government benefits, health care, and other governmentally-related functions. Sometimes, the number appears on an identity card issued by such a country. This unique identification number is issued by a central repository after analyzing the details of the applicant’s data. However, the system would allow issuing two different Unique IDs based on alternate credentials presented by the same individual with malicious intent or just by plain mistake.

If the need is to be able to identify an individual with a alphanumeric string of a certain length alone then all that is required is an intelligent composition and sequencing of this alphanumeric string which can convey information about the domicile, gender, age, etc. and is easy to remember for the individual.

There are many such compositions that can be invented, FINO also has invented one after a considerable deliberation and successfully put to use. However, if the need is also to be able to identify an individual based on more than just demographic details, etc. in order to control intentional fraud or mistakes then the solution needs to go beyond the above mechanism and thus the need of a “Unique ID System”.
At this stage it is of utmost importance to understand and acknowledge the underlying need for such a system. A “Unique ID System” should be capable of addressing a person uniquely in a system and not by the data presented by the person in a system, even though supported by collaterals as there are chances of fraud in generating/ procuring and/ or validating these documents at ground level.

In order to do this, the system should be capable of capturing and identifying one or more natural or biometric characteristics of the individual. These characteristics should be quantifiable using efficient algorithms to search and compare against a historic collection of such characteristics obtained from other individuals.
The concern here is Biometrics technologies collectively are relatively a new realm of technology at various stages of development and have different success ratios.

Keep watching this space for more action on Unique Identity System. In my next post I will try to address the question on “Why do we need a “Unique ID System”?”