Abstract: Disclosed herein is a technique for operating a navigation server and a plurality of navigation clients. Each of the navigation clients transmits a model data trained on a local training data set to the navigation server. The navigation client also transmits attribute data characterizing the local training data set to the navigation server. The navigation server uses the attribute data to weigh the contributions of the model data for performing a weighted aggregation into an aggregated set of model parameters for distribution to the navigation client for improving its local model data.

Abstract: According to an aspect of the present inventive concept there is provided a method for eye tracking, comprising: capturing a sequence of digital images of an eye of a user; outputting data including said sequence of images to an image processing unit; processing said data by the image processing unit to determine a sequence of positions of the eye, each position being indicative of a gaze direction, acquiring biosignal data representing an activity of the eye; and in response to detecting closing of the eye based on the acquired biosignal data, pausing at least one of said capturing, said outputting and said processing. A system for implementing the method is also disclosed.

Abstract: The described embodiments relate to systems, methods, and apparatuses for estimating luminaire lifetimes using extrapolated sensor data. During a lifetime of a luminaire (108), a sensor (206) of the luminaire can become inoperable well before the entire luminaire becomes inoperable. In order to estimate when the luminaire will become inoperable, sensor data (504) should be continually tracked even after the sensor fails. In order to continue obtaining sensor data after the sensor fails, the sensor data can be extrapolated based on a correlation between previously received sensor data and other environmental data. The extrapolated sensor data can be tracked after the sensor becomes inoperable so that estimates of luminaire lifetime can be more accurate, compared to not having any sensor data. In this way, maintenance of luminaires can be scheduled according to predictions of luminaire lifetimes, rather than the occurrence of a luminaire failure.

Abstract: A method for monitoring an eye position, comprises: capturing (202) a sequence of digital images of an eye; acquiring (204) a sequence of biosignal data representing eye movements; determining (206) a set of reference eye positions based on the sequence of digital images; and determining (208) a set of intermediate eye positions based on said set of reference eye positions and said sequence of biosignal data, said set of intermediate eye positions representing eye positions relative to said set of reference eye positions, wherein the set of intermediate eye positions represents eye positions between consecutive pairs of images of said sequence of digital images.

Abstract: A system for hand gesture detection is provided, comprising: a wrist wear adapted to be worn about a wrist of a user of the system and including a set of skin electrodes adapted to face the wrist; an impedance measurement circuit adapted to measure at least a first impedance in a first portion of the wrist and a second impedance in a second portion of the wrist which second portion is circumferentially displaced in relation to said first portion, wherein the first impedance is measured via a first electrode group including four skin electrodes of said set of skin electrodes and the second impedance is measured via a second electrode group including four skin electrodes of said set of skin electrodes, and a processing circuit adapted to detect a hand gesture of the user based on the first and the second impedance measured by the impedance measurement circuit.

Abstract: According to an aspect of the present inventive concept there is provided a method for eye tracking, comprising: capturing a sequence of digital images of an eye of a user; outputting data including said sequence of images to an image processing unit; processing said data by the image processing unit to determine a sequence of positions of the eye, each position being indicative of a gaze direction, acquiring biosignal data representing an activity of the eye; and in response to detecting closing of the eye based on the acquired biosignal data, pausing at least one of said capturing, said outputting and said processing. A system for implementing the method is also disclosed.

Abstract: A method for estimating a condition of a person is disclosed. In one aspect, the method includes inputting a set of measurement values to an ensemble of machine learners. Each machine learner in the ensemble of machine learners is trained to make an estimate of the condition of the person based exclusively on features which are extracted from measurements by a single physiological sensor or environment sensor. The method further includes computing, by a machine learner in the ensemble of machine learners for which measurement values corresponding to the features used by the machine learner is available, an individual estimate value of the condition of the person, The method further includes receiving weights to be applied to the individual estimate values. The weights are at least partly adapted to individual characteristics of the person. The method further includes combining individual estimate values based on the received weights to make a final estimate of the condition of the person.

Abstract: An electronic system for estimating a subject's blood pressure, comprising a feature extraction module configured for receiving a subject's photoplethysmogram signal, detecting a plurality of signal characteristic points on the received photoplethysmogram signal, calculating a plurality of distances in both time and amplitude between any two of the detected photoplethysmogram signal characteristic points, and providing a feature information signal comprising information about the calculated distances; and a blood pressure calculation module configured for receiving the photoplethysmogram signal, the feature information signal and anthropometric characteristics of the subject, and calculating systolic, diastolic and continuous mean blood pressure values of the subject.

Abstract: A system for hand gesture detection is provided, comprising: a wrist wear adapted to be worn about a wrist of a user of the system and including a set of skin electrodes adapted to face the wrist; an impedance measurement circuit adapted to measure at least a first impedance in a first portion of the wrist and a second impedance in a second portion of the wrist which second portion is circumferentially displaced in relation to said first portion, wherein the first impedance is measured via a first electrode group including four skin electrodes of said set of skin electrodes and the second impedance is measured via a second electrode group including four skin electrodes of said set of skin electrodes, and a processing circuit adapted to detect a hand gesture of the user based on the first and the second impedance measured by the impedance measurement circuit.

Abstract: An electronic system for estimating a subject's blood pressure, comprising a feature extraction module configured for receiving a subject's photoplethysmogram signal, detecting a plurality of signal characteristic points on the received photoplethysmogram signal, calculating a plurality of distances in both time and amplitude between any two of the detected photoplethysmogram signal characteristic points, and providing a feature information signal comprising information about the calculated distances; and a blood pressure calculation module configured for receiving the photoplethysmogram signal, the feature information signal and anthropometric characteristics of the subject, and calculating systolic, diastolic and continuous mean blood pressure values of the subject.