International Business Machines Corporation

Gero Dittmann

Cedric Lichtenau

Peter Altevogt

Andreea Anghel

Thomas Pflueger

Andreas Guenther Herkersdorf

Jens Jelitto

Daniel Nikolaus Bauer

Alper Buyuktosunoglu

Jan Leonard Camenisch

Alan Frederic Benner

Patrick Ruch

Francois Abel

Erich M Ruetsche

Laurent Frelechoux

Indira Nair

Emanuel Marc Löertscher

Reinaldo A Bergamaschi

Miguel Angel Prada Delgado

Andreas X Meier

Embodiments are disclosed for a quality control method. The method includes verifying one or more sensors as part of a verification process and assessing a quality of an item or a technical process as part of a quality control process. Two or more measurements values characterizing the item or the technical process from the one or more sensors are obtained. One or more output values based on a computerized process taking the two or more measurements values as inputs are obtained. At least one of the output values obtained is compared to one or more corresponding reference values, to obtain a quality assessment of the item or the technical process.

Quality control based on measurements from verified sensors

Methods are provided for generating a random bit-string from an array of SRAM cells. Such a method includes reading the start-up value of each cell over multiple power-ups of the array, and calculating, from the multiple start-up values of each cell, a bias value indicative of entropy of that cell. The method also includes generating, based on the bias values, an address list in which addresses of cells in the array are listed in order of entropy of the cells. This address list is stored in non-volatile memory. The method further comprises, on a subsequent power-up of the array, generating a random bit-string by reading the start-up values of a set of cells selected in decreasing entropy order of the address list. Hardware random number generators exploiting such methods are also provided.

Random number generation from SRAM cells

A computer-implemented method, a computer system, and a computer program product for managing physical objects. A computer system associates a physical object of a set of physical objects with an object identifier. The computer system obtains a digital fingerprint, from a unique physical property of the physical object, wherein the digital fingerprint is impacted by the unique physical property. The computer system obtains a dataset from the object identifier and the digital fingerprint. The computer system cryptographically signs the dataset to obtain a signature. The computer system instructs to store the object identifier and the signature on a data storage device.

Managing physical objects using crypto-anchors

A system and method interpret optical characterization examinations performed with a set of optical devices. Each device comprises one or more arrays of optical metasurface structures and have arrays tailored to have distinct properties making them differ from one another. First data and second data are accessed that capture a physical fingerprint of each device and an outcome of an optical characterization examination performed with each device, respectively. The outcomes of examinations performed are impacted by the respective, distinct properties of the arrays. Each device is identified based on the first data accessed, which makes it possible to obtain a readout key associated with the identified device. This readout key accounts for the respective one of the distinct properties. Finally, the second data are interpreted according to the readout key obtained to elucidate the outcome of the optical characterization examination. The invention is further directed to related computer program products.

Interpreting optical signals from tailored arrays of metasurfaces

A computer-implemented method for certified sensor readings from a sensor can be provided. The method includes calibrating the sensor, generating a digital certificate, signing, by the sensor, a measurement value with its private key of a public/private key pair, and sending, by the sensor, the signed measurement value such that the sensor is identifiable via its public key and the related generated digital certificate.

Sensor calibration

A computer-implemented method, a computer system, and a computer program product for managing relationships between persons and physical objects based on identifiers. A computer system configures a user device to be capable of interacting with each of a first database and a second database. The computer system verifies, by interacting with the second database, that a given object of the physical objects is a genuine object, based on a given object identifier of the given object and a digital fingerprint detected from a physical fingerprint of the given object. The computer system requests, via the first database, a transfer of a given relationship from a first person to a second person with respect to the given object, based on the given object identifier and a person identifier of the second person.

Managing relationships between persons and physical objects based on physical fingerprints of the physical objects

Embodiments of the invention include a computer-implemented method of processor branch prediction. This method aims at training a machine-learning model of processor branch behavior while a processing unit executes computer instructions. Such instructions include branch instructions, load instructions and store instructions. The load instructions and the store instructions cause a control unit of the processing unit to load data from a memory into processor registers and store data from the processor registers to the memory, respectively. Basically, the training of the model involves, for each of N branch instructions (N&#x3e;2) encountered whilst the processing unit executes said branch instructions: identifying a next branch instruction; and feeding the machine-learning model with carefully chosen inputs.

Computerized branch predictions and decisions

A motion detection device for motion controlled switching of a peripheral device having a switching characteristic is suggested. The motion detection device comprises a motion detector for providing detection signals in response to detected motions and a memory for storing durations between the detection signals. The motion detection device further comprises a signal generator for outputting a switching signal to the peripheral device for switching the peripheral device from a first operation mode to a second operation mode for a activation period. A controller is further included for controlling the signal generator. Therein, the controller is configured to determine the activation period based on at least a selection of the durations between the detection signals stored in the memory and the switching characteristic of the peripheral device.

Motion detection device and system for motion controlled switching of a peripheral device

A method for dynamically selecting a size of a memory access may be provided. The method comprises accessing blocks having a variable number of consecutive cache lines, maintaining a vector with entries of past utilizations for each block size, and adapting said block size before a next access to the blocks.

Selection of variable memory-access size

A method of prefetching data is provided including monitoring sequences of memory addresses of data being accessed by a system, whereby sequences of m+1 memory addresses each are continually identified; and for each identified sequence: converting, upon identifying said each sequence, memory addresses of said each sequence into m relative addresses, whereby each of the m relative addresses is relative to a previous memory address in said each sequence, so as to obtain an auxiliary sequence of m relative addresses; upon converting said memory addresses, feeding said auxiliary sequence of m relative addresses as input to a trained machine learning model for it to predict p relative addresses of next memory accesses by the system, where p&#x2265;1; and prefetching data at memory locations associated with one or more memory addresses that respectively correspond to one or more of the p relative addresses predicted.

Computerized methods for prefetching data based on machine learned sequences of memory addresses

Embodiments of the invention is directed to a method for connecting a device to a network. An example method comprises providing a device assigned with a device identifier and an asymmetric cryptographic key pair that includes a public key and a private key. The device stores the private key on a memory thereof. The device is provided with information as to the assigned device identifier and/or the public key. This information is detectable by a detector so as to be transmissible to a server for it to identify the device identifier and the public key assigned to the device.

Efficient and secure connection of devices to a network without user interfaces

Methods and computerized units grant network access to any one of multiple devices of the same owner. Each of the multiple devices has been previously associated with an owner at an authentication server, whereby device keys for authenticating said multiple devices are stored on the authentication server. Also, said owner has previously been authorized to access the network, such that an owner ID for this owner is stored on the authentication server. In embodiments, present methods comprise, at the authentication server: receiving a network access request for a device to connect to a network, said device being one of the multiple devices; and upon authenticating said device based on a device key associated with this device at the authentication server, confirming that network access can be granted for the device if said owner ID is confirmed to be associated with said device at the authentication server.

Growing community of inventors

Zurich, Switzerland

Gero Dittmann