Recognition of unauthorized access of a portable electronic apparatus

Patent metadata

Field	Value
Publication number	US-12468791-B2
Application number	US-202217646981-A
Country	US
Kind code	B2
Filing date	Jan 4, 2022
Priority date	Jan 18, 2021
Publication date	Nov 11, 2025
Grant date	Nov 11, 2025

Abstract

To recognize an authorized user from an unauthorized user of a portable electronic apparatus, the portable electronic apparatus has a plurality of sensors and a user-recognition circuit. The sensors are each configured to generate a signal representative of a respective physical quantity associated with the use, by a user, of the portable electronic apparatus in an operating state. The user-recognition circuit is configured to receive a plurality of electrical signals from the plurality of sensors; determine a plurality of recognition parameters, each associated with a specific mode of use of the portable electronic apparatus by the user; determine a plurality of indicators of use, one for each recognition parameter, wherein each parameter indicates the probability that, at a time instant, the respective recognition parameter is associable with an unauthorized user; determine a probability of intrusion from the plurality of indicators of use; and compare the probability of intrusion with an intrusion threshold.

First claim

What is claimed is:

1 . A method of recognition of an authorized user from an unauthorized user of a portable electronic apparatus, the method comprising:
training a user-recognition circuit in the portable electronic apparatus, after a first start-up of the portable electronic apparatus by the authorized user, the training the user-recognition circuit comprising:
prompting the authorized user to interact with the portable electronic apparatus by conducting pressure tests, performing user activity movement tests, and answering questions about device usage comprising habits of displacement or transportation type,
obtaining information of use of the portable electronic apparatus by the authorized user, at least a portion of the information of use based on the authorized user interacting with the portable electronic apparatus in response to the prompting,
converting the information of use to training data comprising expected values of sets of signal-description parameters, and
performing a supervised training of the user-recognition circuit using the training data as labelled data belonging to respective concordance classes of the authorized user;
performing a verification check at pre-set instants, pre-set situations, or a combination thereof, wherein the pre-set instants correspond to pre-set time intervals to trigger the verification check, and wherein the pre-set situation corresponds to pre-set situations that trigger the verification check, wherein the performing the verification check comprises:
receiving, at the user-recognition circuit, a plurality of electrical signals from a plurality of sensors in the portable electronic apparatus, the plurality of sensors each configured to generate a signal representative of a respective physical quantity associated with a use of the portable electronic apparatus in an operating state,
determining, at the user-recognition circuit, a plurality of recognition parameters from the plurality of electrical signals, each recognition parameter associated with a specific mode of use of the portable electronic apparatus by an external user,
determining a plurality of indicators of use at the user-recognition circuit, one of the indicators of use determined for each of the recognition parameters, each of the indicators of use indicating a probability that, at a time instant, the respective recognition parameter is associable with the unauthorized user or the authorized user,
determining, at the user-recognition circuit, a probability of intrusion starting from the plurality of indicators of use, and
comparing, at the user-recognition circuit, the probability of intrusion with an intrusion threshold and based thereon recognizing the authorized user from the unauthorized user of the portable electronic apparatus; and
configuring the portable electronic apparatus into a permanently blocking state in response to determining an unauthorized user during the verification check.

2 . The method according to claim 1 , wherein the determining the plurality of recognition parameters comprises determining a plurality of sets of signal-description parameters, one for each recognition parameter, each set of signal-description parameters comprising statistical parameters calculated starting from a respective portion of the plurality of electrical signals.

3 . The method according to claim 2 , wherein each set of the signal-description parameters comprises one or more of the following statistical parameters calculated starting from the respective portion of the plurality of electrical signals: auto-correlation, mean, variance, peak-to-peak value, number of peaks, number and position of local maxima and minima and of inflection points, and number of zero-crossings.

4 . The method according to claim 2 , wherein the determining the plurality of indicators of use comprises:
classifying, using a classification algorithm, the recognition parameters, each in a respective discordance class or in a respective concordance class, each recognition parameter classified based on a comparison between current values of the respective set of signal-description parameters and previous values of the respective set of signal-description parameters, the respective concordance class comprising values of the respective set of signal-description parameters that are associable with the authorized user, the respective discordance class comprising values of the respective set of signal-description parameters that are associable with the unauthorized user; and determining a plurality of specific discordance indicators, one for each recognition parameter, the specific discordance indicators each indicating the probability, at the time instant, that the respective recognition parameter belongs to the respective discordance class.

5 . The method according to claim 4 , wherein the classification algorithm includes a clustering algorithm, a decision tree, an artificial neural network, a support-vector machine, a K-means algorithm, or a combination thereof.

6 . The method according to claim 4 , further comprising updating the user-recognition circuit, the updating the user-recognition circuit comprising:
obtaining further information of use of the portable electronic apparatus by the authorized user; converting the further information of use to further training data comprising further expected values of the sets of signal-description parameters; and performing a further supervised training of the user-recognition circuit using the further training data as further labelled data belonging to the respective concordance classes.

7 . The method according to claim 4 , the determining the probability of intrusion starting from the plurality of indicators of use comprising:
determining an overall discordance indicator, starting from the values of the plurality of specific discordance indicators at the time instant, the overall discordance indicator indicating the probability that, at the time instant, the external user is associable with the unauthorized user, and determining an overall discordance indicator over time, starting from values of the overall discordance indicator calculated in a decision time interval; and the comparing the probability of intrusion with the intrusion threshold comprising verifying whether the overall discordance indicator over time is higher than the intrusion threshold.

8 . The method according to claim 7 , wherein the determining the overall discordance indicator at the time instant comprises executing a weighted mean of the plurality of specific discordance indicators.

9 . The method according to claim 7 , wherein the determining the overall discordance indicator at the time instant comprises using an artificial neural network for scheme recognition having at input the plurality of specific discordance indicators.

10 . The method according to claim 7 , wherein, in the decision time interval, the overall discordance indicators at the time instants are calculated with variable frequency.

11 . The method according to claim 1 , further comprising switching, by the portable electronic apparatus, from the permanently blocking state to the operating state upon a reset operation performed by a special authorized user.

12 . The method according to claim 1 , further comprising, by each of a plurality of analysis circuits, one for each recognition parameter, in the user-recognition circuit:
receiving a respective portion of the plurality of electrical signals; determining the respective recognition parameter; and determining the respective indicator of use.

13 . The method according to claim 12 , wherein the plur

Assignees

• St Microelectronics Srl

Inventors

• Alessi Enrico Rosario
• Ferri Simone
• Passaniti Fabio

Classifications

• H04R1/08

  Mouthpieces; {Microphones;} Attachments therefor · CPC title

• G06F2203/04105

  Pressure sensors for measuring the pressure or force exerted on the touch surface without providing the touch position · CPC title

• G06F3/0416

  Control or interface arrangements specially adapted for digitisers · CPC title

• G01P13/00

  Indicating or recording presence, absence, or direction, of movement (electric switches H01H; counting moving objects G06M7/00) · CPC title

• H04W12/12

  Detection or prevention of fraud · CPC title