- One-time password
A one-time password (OTP) is a password that is valid for only one login session or transaction. OTPs avoid a number of shortcomings that are associated with traditional (static) passwords. The most important shortcoming that is addressed by OTPs is that, in contrast to static passwords, they are not vulnerable to replay attacks. This means that, if a potential intruder manages to record an OTP that was already used to log into a service or to conduct a transaction, he or she will not be able to abuse it since it will be no longer valid. On the downside, OTPs are difficult for human beings to memorize. Therefore they require additional technology in order to work.
- 1 How OTPs are generated and distributed
- 2 Methods of generating the OTP
- 3 Methods of delivering the OTP
- 4 Comparison of technologies
- 5 Standardization
- 6 See also
- 7 External links
- 8 References
How OTPs are generated and distributed
OTP generation algorithms typically make use of randomness. This is necessary because otherwise it would be easy to predict future OTPs by observing previous ones. Concrete OTP algorithms vary greatly in their details. Various approaches for the generation of OTPs are listed below:
- Based on time-synchronization between the authentication server and the client providing the password (OTPs are valid only for a short period of time)
- Using a mathematical algorithm to generate a new password based on the previous password (OTPs are effectively a chain and must be used in a predefined order).
- Using a mathematical algorithm where the new password is based on a challenge (e.g., a random number chosen by the authentication server or transaction details) and/or a counter.
There are also different ways to make the user aware of the next OTP to use. Some systems use special electronic tokens that the user carries and that generate OTPs and show them using a small display. Other systems consist of software that runs on the user's mobile phone. Yet other systems generate OTPs on the server-side and send them to the user using an out-of-band channel such as SMS messaging. Finally, in some systems, OTPs are printed on paper that the user is required to carry with them.
Methods of generating the OTP
A time-synchronized OTP is usually related to a piece of hardware called a security token (e.g., each user is given a personal token that generates a one-time password). Inside the token is an accurate clock that has been synchronized with the clock on the proprietary authentication server. On these OTP systems, time is an important part of the password algorithm since the generation of new passwords is based on the current time rather than, or in addition to, the previous password or a secret key. This token may be a proprietary device, or a mobile phone or similar mobile device which runs software that is proprietary, freeware, or open-source. An example of time-synchronized OTP standard is TOTP.
All of the methods of delivering the OTP below may use time-synchronization instead of algorithms.
Each new OTP may be created from the past OTPs used. An example of this type of algorithm, credited to Leslie Lamport, uses a one-way function (call it f). The one-time password system works by starting with an initial seed s, then generating passwords
- f(s), f(f(s)), f(f(f(s))), ...
as many times as necessary. If an indefinite series of passwords is wanted, a new seed value can be chosen after the set for s is exhausted. Each password is then dispensed in reverse, with f(f(...f(s))...) first, to f(s). The S/KEY one-time password system (and its confusingly-named derivative, OTP) are based on Lamport's scheme.
If an intruder happens to see a one-time password, he may have access for one time period or login, but it becomes useless once that period expires. To get the next password in the series from the previous passwords, one needs to find a way of calculating the inverse function f−1. Since f was chosen to be one-way, this is extremely difficult to do. If f is a cryptographic hash function, which is generally the case, it is (so far as is known) a computationally infeasible task.
In some mathematical algorithm schemes, it is possible for the user to provide the server with a static key for use as an encryption key, by only sending a one time password.
The use of challenge-response one-time passwords will require a user to provide a response to a challenge. For example, this can be done by inputting the value that the token has generated into the token itself. To avoid duplicates, an additional counter is usually involved, so if one happens to get the same challenge twice, this still results in different one-time passwords. However, the computation does not usually involve the previous one-time password; that is, usually this or another algorithm is used, rather than using both algorithms.
The methods of delivering the OTP which are token-based may use either of these types of algorithm instead of time-synchronization.
Methods of delivering the OTP
A common technology used for the delivery of OTPs is text messaging. Because text messaging is a ubiquitous communication channel, being available in nearly all handsets and with a large customer-base, text messaging has a great potential to reach all consumers with a low total cost to implement. However, the cost of each text messaging often for each OTP might not be suitable for some users. OTP over text messaging may also be encrypted using an A5/x standard which several hacking groups report can be successfully decrypted within minutes or seconds, or the OTP over SMS might not be encrypted by one's service-provider at all. In addition to threats from hackers, the mobile phone operator becomes part of the trust chain. In the case of roaming, more than a single mobile phone operator has to be trusted. Anyone using this information may mount a man-in-the-middle attack. Recently Google has started offering OTP over text messages on all Google accounts. The user can either receive OTP as text message or via an automated call. Incase if the phone isn't accessible, the user can even use backup codes (up to 10) to sign-in to their accounts after signing in with their account password and then using OTP or one of the backup codes.
A mobile phone keeps costs low because a large customer-base already owns a mobile phone for purposes other than generating OTPs. The computing power and storage required for OTPs is usually insignificant compared to that which modern camera-phones and smartphones typically use. Mobile tokens additionally support any number of tokens within one installation of the application, allowing a user the ability to authenticate to multiple resources from one device. This solution also provides model-specific applications to the user's mobile phone. However, a cellphone used as a token can be lost, damaged, or stolen.
EMV is starting to use a challenge-response algorithm (called "Chip Authentication Program") for credit cards in Europe. On the other hand, in access control for computer networks, RSA Security's SecurID is one example of a time-synchronization type of token. Like all tokens, these may be lost, damaged, or stolen; additionally there is an inconvenience as batteries die (typically cannot plug these into a battery-charger, and this is one more battery that must be replaced, or in some cases the whole token must be replaced). A variant of the proprietary token was proposed by RSA in 2006 and was described as "ubiquitous authentication", in which RSA would partner with manufacturers to add physical SecurID chips to devices such as mobile phones.
Recently, it has become possible to take the electronic components associated with regular keyfob OTP tokens and embed them in a credit card form factor. However, because card thickness (.79mm to .84mm) prevents traditional components or batteries from being used, special polymer-based batteries must be used which have a much lower battery life than traditional coin (button) cells. Also, extremely low-powered semiconductor components must be used to conserve the amount of power being used during sleep and/or actual use of the product. Two companies in particular have led in the production of thin "display card OTP" devices. Those being Identita (http://www.identita.com) and NagraID Security (http://www.nidsecurity.com/).
Yubico offers a small USB token with an embedded chip that creates a OTP when a key is pressed and simulates a keyboard to facilitate easily entering a long password.
A new version of this technology has been developed that embeds a keypad into a payment card using the same form and thickness as traditional cards. This card is a credit/debit/ATM/Credit ID card with embedded keypad, display, microprocessor and proximity chip. This new card technology has the potential to leap frog existing card security options.
Authentication-as-a-service providers offer various web-based methods for delivering one-time passwords without the need for tokens. One such method relies on the user’s ability to recognize pre-chosen categories from a randomly-generated grid of pictures. When first registering on a website, the user chooses several secret categories of things; such as dogs, cars, boats and flowers. Each time the user logs into the website they are presented with a randomly-generated grid of pictures. Each picture in the grid has a randomly-generated alphanumeric character overlaid on it. The user looks for the pictures that fit their pre-chosen categories and enters the associated alphanumeric characters to form a one-time access code.
In some countries online banking, the bank sends to the user a numbered list of OTPs that are printed on paper. For every online transaction, the user is required to enter a specific OTP from that list. In Germany (and Austria too), those OTPs are typically called TANs (for 'transaction authentication numbers'). Some banks even dispatch such TANs to the user's mobile phone via SMS, in which case they are called mTANs (for 'mobile TANs').
Comparison of technologies
One OTP implementation versus another
In terms of costs, the cheapest OTP solutions are those that deliver OTPs on paper, and those that generate OTPs on a device that someone already owns. This is because these systems avoid the costs associated with (re-)issuing proprietary electronic tokens and the cost of SMS messaging.
For systems that rely on electronic tokens, algorithm-based OTP generators must cope with the situation where a token drifts out-of-sync with its server if the system requires the OTP to be entered by a deadline. This leads to an additional development cost. Time-synchronized systems, on the other hand, avoid this at the expense of having to maintain a clock in the electronic tokens (and an offset value to account for clock drift). Whether or not OTPs are time-synchronized is basically irrelevant for the degree of vulnerability, but it avoids a need to reenter passwords if the server is expecting the last or next code that the token should be having because the server and token have drifted out-of-sync.
Compared to most proprietary hardware tokens, so long as one already carries a phone or another mobile device in one's pocket, users of mobile devices don't need to carry and protect an extra item (which has no usefulness except that it generates OTPs). In addition to reducing costs considerably, using a phone as a token offers the convenience that it is not necessary to deliver devices to each end-user (who typically already own the device). For many users, a mobile phone may also be trickle-charged to preserve its battery for at least some portion of each day, whereas most proprietary tokens cannot be trickle-charged. However, most proprietary tokens have tamper-proof features.
OTPs versus other methods of securing data
One-time passwords are vulnerable to social engineering attacks in which phishers steal OTPs by tricking customers into providing one or more OTPs that they used in the past. In late 2005 customers of a Swedish bank were tricked into giving up their one-time passwords (The Register article). In 2006 this type of attack was used on customers of a US bank (Washington Post Security Blog). Even time-synchronized OTPs are vulnerable to phishing, by two methods: The password may be used as quickly by the attacker as the legitimate user must use the OTP, if the attacker can get the OTP in plaintext quickly enough. The other type of attack—which may be defeated if one's OTP system implements using the hash chain as discussed above -- is that after the phisher uses this social engineering, the phisher must then use the information gained (past OTP codes which are no longer valid) to predict what OTP codes will be used in the future (e.g. an OTP password-generator that is pseudo-random rather than truly random might or might not be able to be compromised, because pseudo-random numbers are often predictable once one has the past OTP codes (see also main article); the code implemented in the programming of each Direct OTP or token will determine whether one is vulnerable to this type of attack.
Although OTPs are in some ways more secure than a memorized password, users of OTP systems are still vulnerable to man-in-the-middle attacks. OTPs should therefore not be disclosed to any third parties, and using an OTP as one layer in layered security is safer than using OTP alone; one way to implement layered security is to use an OTP in combination with a password that is memorized by the user (and never transmitted to the user, like OTPs often are). An advantage to using layered security is that a single sign-on combined with one master password or password manager becomes safer than using only 1 layer of security during the sign-on, and thus the inconvenience of password fatigue is avoided if one usually has long sessions with many passwords that would need to be entered mid-session (to open different documents, websites, and applications); however, the disadvantage of using many forms of security all at once during a single sign-on is that one has the inconvenience of more security precautions during every login—even if one is logging-in only for a brief usage of the computer to access information or an application that doesn't require as much security as some other top-secret items that computer is used for. See also Related technologies, below.
More often than not, one-time passwords are an embodiment of two-factor authentication (T-FA). T-FA is a form of layered security where it is unlikely that both layers would be disabled by someone using only one type of attack. Some single sign-on solutions make use of one-time passwords. One-time password technology is often used with a security token.
Newer, interactive T-FA approaches, such as Duo Security's Duo Push technology and ENTERSECT Technologies' Interactive Transaction Authentication (ITA) system, attempt to close the loop where attackers could get hold of OTPs, by prompting a user on a paired mobile phone about the transaction taking place. When accepting the transaction, the message is again relayed (over GPRS or SMS technology) to the authentication server. The whole transaction is encrypted using standard Public/Private Key Encryption.
Many OTP technologies are patented. This makes standardization in this area more difficult, as each company tries to push its own technology. Standards do, however, exist -- for example, RFC 1760 (S/KEY), RFC 2289 (OTP), and RFC 4226 (HOTP).
- Initiative For Open Authentication (OATH)
- One-time pad (OTP)
- KYPS (OTP system based on One-time pads)
- Security token
- Two-factor authentication
- RSA Labs OTP standardization proposal
- Cross-platform and HOTP compliant mobile solution
- Open OTP solution for mobile phones
- ^ EOTP - Static Key Transfer
- ^ Barkan, Elad; Eli Biham; Nathan Keller (2003). "Instant Ciphertext-Only Cryptanalysis of GSM Encrypted Communication". Crypto 2003: 600–16. http://cryptome.org/gsm-crack-bbk.pdf.
- ^ Barkan, Elad; Eli Biham; Nathan Keller. "Instant Ciphertext-Only Cryptanalysis of GSM Encrypted Communication by Barkan and Biham of Technion (Full Version)". http://www.cs.technion.ac.il/users/wwwb/cgi-bin/tr-get.cgi/2006/CS/CS-2006-07.pdf.
- ^ Gueneysu, Tim; Timo Kasper; Martin Novotný; Christof Paar; Andy Rupp (2008). "Cryptanalysis with COPACOBANA". Transactions on Computers Nov. 2008: 1498–1513. http://www.sciengines.com/copacobana/paper/TC_COPACOBANA.pdf.
- ^ Nohl, Karsten; Chris Paget (2009-12-27). "GSM: SRSLY?". 26th Chaos Communication Congress (26C3):. http://events.ccc.de/congress/2009/Fahrplan/events/3654.en.html. Retrieved 2009-12-30.
- ^ http://www.encyclopedia.com/doc/1G1-142107014.html
- ^ "Yubico AB". Bloomberg Businessweek. http://investing.businessweek.com/research/stocks/private/snapshot.asp?privcapId=46126480. Retrieved July 13, 2011.
- ^ Ericka Chickowski (2010-11-03). "Images Could Change the Authentication Picture". Dark Reading. http://www.darkreading.com/authentication/security/client/showArticle.jhtml?articleID=228200140.
- ^ "Confident Technologies Delivers Image-Based, Multifactor Authentication to Strengthen Passwords on Public-Facing Websites". 2010-10-28. http://www.marketwire.com/press-release/Confident-Technologies-Delivers-Image-Based-Multifactor-Authentication-Strengthen-Passwords-1342854.htm.
- ^ Duo Push: The next generation of two-factor authentication
Wikimedia Foundation. 2010.