Browse Prior Art Database

System and method dynamically learning and checking user unlocking speed

IP.com Disclosure Number: IPCOM000246578D
Publication Date: 2016-Jun-20
Document File: 3 page(s) / 59K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method to reinforce device locking/unlocking mechanisms, not requiring specific hardware into the device itself and perceived as natural movement for how recent devices are worn.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

System and method dynamically learning and checking user unlocking speed

Locking/unlocking devices is important to protect personal information. There are different methods especially in mobile devices but users are always looking for more usable, amazing and secure given the level of security their hardware is able to provide. The usual mechanism implemented in smartphone operating systems displaying 9 points with some of them to be joined is widely spread and perceived as amazing. It is not secure as the fingerprint sensor but most of devices have not that hardware available. Moreover with new devices (such as smart watch) there could be some other movements that the user perceives more natural and amazing such as moving a save lock through values or rotating the wrist and so on.

This approach can be reinforces as described below, it applies to wrist movements that can be done forward or back or also to the already existing 9 points grid mentioned above for instance if we look at how quick the user will reproduce the sequence. In fact we expect, as in other things, that after a first initial phase the user

will do the unlock sequence, whatever it is, in a very similar way. The device needs however to learn and decide when the steady state is reached to enable this aspect and double check the sequence and the time needed to do it. If the device is stolen it is not unlocked if the sequence is not correct and done with the expected cadence.

There is an unlocking code definition and a later associated learning phase related. The length of this learning phase is automatically defined through the mechanism described below. The point is that the unlocking sequence when done is recorded and translated into a diagram like the following:

Fig.1

This is just an instance of the unlocking valid sequence: 7 3 -2 5 -8 1 8. The trend gives also an idea of how quick it has been done. At definition time the sequence is defined and the device can be unlocked just doing the sequence itself. However the learning phase is started and it will lead to a more and more secure unlocking mechanism. This is how it happens: all the unlocking correct sequence time trends are recorded and overlapped and a new diagram is progressively built. The sequence points can be imagined to happen in different discrete time values and crowd around a more and more evident time value while going on.

1


Page 02 of 3

Fig.2

In Fig.2 three additional collected samples for the unlocking sequence value 3 are reported: one 0,1 sec before, one 0,1 sec later and one at the same time of the already existing one. Some points are expected to reach really high value of occurrences while going on collecting these information and clusters of collected samples mass around the right sequence values. These clusters are important because they are used to determine the...