Summary Abstract DescriptionWearable Device
Field of the InventionThe present invention relates to wearable devices, in particular to wearable devices capable of indicating whether or not a current wearer of the device habitually wears the device.
Background to the Invention
Biometric identification systems are typically expensive.
There is a need for relatively low cost alternatives which remain relatively difficult to compromise.
*:*::* Summary of the Invention *... * * S...
*:*. According to an aspect of the invention there is provided * a wearable device capable of indicating whether or not a *.* current wearer of the device habitually wears the device, ::::. the device comprising: a sensor unit operable to gather environmental information relevant to the device; a memory unit operable to store information gathered by the sensor unit; a processor unit operable to process information from the sensor unit and/or information stored in the memory unit and to generate an output based on the information; and an indicator unit operable to indicate whether or not a current wearer habitually wears the device according to the output of the processor unit.
Suitably, the wearable device comprises a strap to allow the device to be strapped to a wearer. Suitably, the strap is a strap for an arm, preferably a wrist strap.
Suitably, the sensor unit is operable to gather environmental information of one or more of the following types: temperature information, humidity information, stress/strain information relating to tension of the strap, vascular pulse information relating to the wearer, respiratory information relating to the wearer, position information, and information relating to whether the device is being worn.
Suitably, the device comprises a timer operable to time stamp information gathered by the sensor unit.
Suitably, the sensor unit is operable to gather vascular pulse information including one or more of pulse shape, pulse frequency and pulse amplitude.
::. Suiably, the processor unit is operable to generate an output based on the correlation of current information gathered by the sensor unit with information associated with the habitual wearer of the device. Suitably, the processor unit is operable to generate an output based on behavioural patterns in information that has been stored in the memory unit since the sensor unit indicated that the device started to be worn. Suitably the processor unit is operable to generate an output based on correlation of current information gathered by the sensor unit with information associated with the habitual wearer of the device, in combination with behavioural patterns in information that has been stored in the memory unit since the sensor unit indicated that the device started to be Suitably, the processor unit is operable to generate a private key which is used authenticate the current wearer as the habitual wearer or not by comparison of information gathered by the sensor unit since the device last started to be worn to the private key.
Suitably, the memory unit is operable to store the private key of the habitual wearer in a private key database. The private key database suitably holds information related to time-stamped behavioural states of the owner. Suitably, the memory unit is arranged to store information that the processor unit identifies as different to that used in *:::* generating the private key. S... * S ***.
Suitably, the memory unit is operable to buffer information gathered by the sensor unit for a S..
predetermined period, and to store data from the buffered : ., information in the memory to provide an indication of the behavioural state in the period immediately prior to a removal of the wearable device from a current wearer.
Suitably, the processor is arranged to use the stored buffered information in generating its output so that the output is based upon the stored buffered information as well as data clustered and classified over time.
Suitably, the indicator unit comprises a visual display unit, operable to give a visual indication in response to a signal received from the processor unit. Suitably, the indicator unit comprises a wired or wireless output unit operable to give an indication to a device compatible with the wearable device.
Brief Introduction to the Drawings
For a better understanding of the invention, and to show how embodiments of the same may be carried into effect, reference will now be made, by way of example, to the accompanying diagrammatic drawings in which: Figure 1 shows a generic graphical example of how the output from the sensor into is used to update the fuzzy rules and membership functions in example embodiments of the present invention; *..... Figure 2 shows an example of a degree of membership associated with three behavioural patterns varying over time, as evaluated by the processor unit in an example embodiment of the present invention; S..
: .. Figure 3 shows a graphical representation for an example membership function for use in example embodiments of the present invention.
Description of Example Embodiments
A device according to an example embodiment of the present invention is envisioned to be continually worn, or substantially continually worn by its owner. This individual is referred to as the habitual wearer. The device comprises a sensor unit operable to gather environmental information relevant to the device; a memory unit operable to store information gathered by the sensor unit; a processor unit operable to process information from the sensor unit and/or information stored in the memory unit and to generate an output based on the information; and an indicator unit operable to indicate whether or not a current wearer habitually wears the device according to the output of the processor unit.
That output may be provided on a visual display which forms part of the device, and/or may be provided to a compatible device such as in response to an identity challenge to the weaver of the wearable device. The indicator unit may conveniently comprise short-range wireless connection functionality.
An indication of whether or not a current wearer is a is habitual wearer may be used as part of an identity * ** verification system, for example a system in which a **** unique characteristic of the device is associated with the identity of a known habitual wearer of that device. ***
While being worn and optionally further while not being : *. worn the sensor unit abstracts information regarding the **** wearer and the environment in which the device is located.
In the example embodiments described below the sensor unit comprises a vascular pulse sensor, a stress/strain sensor associated with a strap by which the device is worn, and the device further comprises a timer operable to time stamp information gathered by the sensor unit. The timer may conveniently comprise part of the sensor unit, or the memory unit.
Patterns in the information gathered by the sensor unit and built up in the memory unit are analysed by the processor unit. The processor unit is arranged to apply fuzzy rules based on membership functions to generate its output, thereby allowing the indicator unit to make an indication concerning the current wearer of the device.
A generic graphical example of how the information from the sensor unit is used by the processor unit to update fuzzy rules and membership functions is illustrated schematically in Figure 1. For simplicity, within this example, classification of a "behaviour" is described in relation to specific physical behaviours. An example of a behaviour is an 8 hour long interval where the device has been removed, such as from 11pm to 7am, has a strong association with an overnight sleep behaviour. Behaviours such as this are classified through clustering of gathered data. Hence, no such physical behaviours need be known a * ** priori to the processor unit. Furthermore there is no S...
limit to the behaviours which are established by the *, , processor unit based on abstracted data which is clustered and then classified, other than those imposed by the S..
limitations in storage capacity and processing power. * S. * S * S...
In an example embodiment the sensor unit is able to detect when the device is being worn. For example, the device may comprise a wrist strap, and the sensor unit may recognise the doing and/or undoing of the strap. As mentioned above, in this embodiment the sensor unit comprises a vascular pulse sensor, the output of which can also be used by the processor unit to determine whether or not the device is being worn based on the presence or absence of a detected vascular pulse.
After removal of the device and upon subsequent reattachment to a wearer, private key authentication algorithms are undertaken by the processor. The result of the private key authentication algorithms is a confidence value associated with the match observed between information abstracted from the device wearer, to probabilities associated with membership functions identifying behavioural patterns and their respective signatures stored within the private key associated with the habitual wearer.
In other embodiments of the present invention the authentication procedure is periodically initiated as well as, or in the alternative to being initiated upon reattachment to a wearer.
Upon initiation of the authentication procedure the *.... following steps are undertaken by the processor unit: * S S...
Step 1 -Probabilistic identification of envisaged S. : behavioural patterns: *..
: .. Based upon information, such as elapsed time, date, etc. a *...
probability (degree of membership) is associated with each envisaged behavioural pattern. A graphical example of three different behaviours is illustrated in Figure 2.
From the probabilities, weights may be applied to a set of potential pre-classif led behaviours.
Step 2 -Weights are accumulated and applied to their respective associated membership functions, providing a soft degree of association of the current device wearer to the habitual wearer.
Each behaviour has an associated membership function, which identifies a degree of membership to the habitual wearer, based upon input parameters (time elapsed, time, date, etc) and correlation to the private key of the habitual wearer. An output is provided by the processor unit indicating conformity to the norm with respect to the habitual wearer. The indicator unit is operable to indicate whether or not a current wearer is the habitual wearer of the device according to this output received from the processor unit. The indicator unit may provide a constant indication while the device is being worn, or alternatively may provide an indication at a particular point in time, or in response to an indication request detected by the sensor unit. * **
A graphical representation for an example membership * .** S...,, function is shown in Figure 3. In this example the *:*. behaviour considered is an overnight sleep. *.*
Step 4 -The output of the indicator unit can be provided : .. as an input to a third party application which uses this S...
to trigger warnings, limit device performance, initiate additional authentication procedures, etc. Alternatively the output of the indicator unit can be manually checked.
Step 5 -Additional authentication may be required and may take the form of authentication over a prolonged time interval. For instance, if the current wearer has a low conformity to the habitual wearer's private key, authentication is repeated at a later time and the trend relative to the habitual wearer's private key is analysed by the processor unit. If the trend is determined by the processor unit as returning to the habitual wearer's norm, this is acknowledged, and a new classification of behaviour is formed. If the trend is determined by the processor unit as not returning to the habitual wearer's norm, this may be used to further validate that the current wearer is not the habitual wearer.
A number of specific behavioural scenarios may result from the classification and the clustering of information from the sensor unit. Examples of some specific behavioural scenarios are given below.
1. Device taken off at night, re-worn in the morning.
Utilising the time of day and the elapsed may be used to identify whether the device has been taken off over night.
Changes in vascular pulse shape associated with the time * ** ****** spent asleep, whether or not the owner showers before wearing the device etc. may be identified and applied within membership functions associating vascular pulse ** * characteristics observed upon the re-wearing of the *** * 20 device, and behavioural characteristics. * S. * S S S...
* 2. Device is taken off at a time of the day when the owner typically has a shower, and is then re-worn after 15 minutes have elapsed.
Similar to 1, by utilising the behavioural pattern, probabilities appropriate to specific behaviour may be identified, and changes envisaged in for example vascular pulse shape may be validated 3. Device is taken off, after a short amount of time (e.g. <1 minute) has elapsed, then the device is re-worn.
Within such a short time period little change is to be expected and hence no sudden changes in either the strain on the wearable strap, or the vascular pulse shape is expected.
The device as described offers a low cost means of identifying a current wearer of the device as the habitual wearer. In addition the information gathered by the sensor unit may also be useful in monitoring the health of the habitual wearer. In embodiments of this type the processor unit generates a health-related output to the indicator unit, the indicator unit operable to distinguish between a health-related input and an identification input and to provide a suitable indicated output.
Attention is directed to all papers and documents which e..
are filed concurrently with or previous to this * specification in connection with this application and * which are open to public inspection with this * 20 specification, and the contents of all such papers and : *, documents are incorporated herein by reference. ****
S S..
All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive.
Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.
The invention is not restricted to the details of the foregoing embodiment (s). The invention extends to any novel one, or any novel combination, of the features
disclosed in this specification (including any
accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed. * * *1 S * S. **S. * * ** * S * S. * S. S..
S S *S * S S S...
I S..
S
Claims Illustrations
