Method of processing electrophysiological signals to compute a virtual vehicle key, corresponding device, vehicle and computer program product

A vehicle-key device configured to be equipped onboard a vehicle, the vehicle-key device comprising:

• a PhotoPletysmoGraphy (PPG) sensor configured to collect at least one electrophysiological signal of a human;
• and
• processing circuitry configured to:
• receive the at least one collected electrophysiological signal over a limited time duration;
• discontinue the receiving at an expiry of the limited time duration;
• compute a set of electrophysiological signal features as a function of the at least one collected electrophysiological signal, wherein the computing comprises at least one of:
• i) provide at least one reference electrophysiological signal and apply dynamic time warping processing to the at least one collected electrophysiological signal and to the at least one reference electrophysiological signal;
• ii) apply stacked-auto-encoder artificial neural network (SAE ANN) processing to the at least one collected electrophysiological signal;
• or
• iii) filter the at least one collected electrophysiological signal via joint low-pass and high-pass filtering using a set of filtering parameters including low-pass filter parameters and high-pass filter parameters having a set of low-pass cut-off frequencies and a set of high-pass cut-off frequencies, respectively;
• apply pattern recognition processing to the computed set of electrophysiological signal features, wherein the pattern recognition processing produces a virtual key signal indicative of an identity of the human;
• and
• apply the virtual key signal to a user circuit to switch the user circuit between a first state and a second state as a result of the virtual key signal matching an authorized key signal stored in the user circuit.

A vehicle comprising:

• a vehicle ignition circuit;
• and
• a vehicle-key device coupled to the vehicle ignition circuit, the vehicle-key device comprising:
• a PhotoPletysmoGraphy (PPG) sensor configured to collect at least one electrophysiological signal of a human;
• and
• processing circuitry configured to:
• receive the at least one collected electrophysiological signal over a limited time duration;
• discontinue the receiving at an expiry of the limited time duration;
• compute a set of electrophysiological signal features as a function of the at least one collected electrophysiological signal, wherein the computing comprises at least one of:
• i) provide at least one reference electrophysiological signal and apply dynamic time warping processing to the at least one collected electrophysiological signal and to the at least one reference electrophysiological signal;
• ii) apply stacked-auto-encoder artificial neural network (SAE ANN) processing to the at least one collected electrophysiological signal;
• or
• iii) filter the at least one collected electrophysiological signal via joint low-pass and high-pass filtering using a set of filtering parameters including low-pass filter parameters and high-pass filter parameters having a set of low-pass cut-off frequencies and a set of high-pass cut-off frequencies, respectively;
• apply pattern recognition processing to the computed set of electrophysiological signal features, wherein the pattern recognition processing produces a virtual key signal indicative of an identity of the human;
• and
• apply the virtual key signal to a user circuit to switch the user circuit between a first state and a second state as a result of the virtual key signal matching an authorized key signal stored in the user circuit.

A computer program product loadable stored in a non-transitory memory of at least one processing circuit and comprising software code portions that, when the product is run on the at least one processing circuit, execute the following steps:

• collecting at least one electrophysiological signal of a human over a limited time duration;
• discontinuing the collecting at an expiry of the limited time duration;
• computing a set of electrophysiological signal features as a function of the at least one collected electrophysiological signal, the computing comprising at least one of:
• i) providing at least one reference electrophysiological signal and applying dynamic time warping processing to the at least one collected electrophysiological signal and to the at least one reference electrophysiological signal;
• ii) applying stacked-auto-encoder artificial neural network (SAE ANN) processing to the at least one collected electrophysiological signal;
• or
• iii) filtering the at least one collected electrophysiological signal via joint low-pass and high-pass filtering using a set of filtering parameters including low-pass filter parameters and high-pass filter parameters having a set of low-pass cut-off frequencies and a set of high-pass cut-off frequencies, respectively;
• applying pattern recognition processing to the computed set of electrophysiological signal features, the pattern recognition processing producing a virtual key signal indicative of an identity of the human;
• and
• applying the virtual key signal to a user circuit to switch the user circuit between a first state and a second state as a result of the virtual key signal matching an authorized key signal stored in the user circuit.