Biometric information provision device

Abstract

In measuring electrical potential signals, measurement noise such as motion artifacts is removed. [Solution] A biological information providing device is provided, comprising: an adaptive filter having an adaptive filter having an adaptive signal measuring unit that measures an adaptive signal through a pair of electrodes in contact with a living body; an adaptive signal measuring unit that measures the biological impedance generated between the pair of electrodes and outputs a biological impedance signal; a variable coefficient filter unit that generates a control signal based on the biological impedance signal and coefficients; a biological information signal output unit that calculates a biological information signal based on the adaptive signal and the control signal; and a coefficient adjustment unit that adjusts the coefficients so that the signals correlated with the biological impedance signal included in the biological information signal become smaller; and a filter control unit that controls the coefficient adjustment unit so that the convergence speed of the adaptive filter increases when the adaptive signal or the biological impedance signal satisfies predetermined conditions.

Description

[Technical Field]

[0001] This invention relates to a biological information provision device. [Background technology]

[0002] Patent Document 1 describes an electrode system for reducing motion artifacts from potential signals. Patent Document 2 describes an active noise reduction device that reduces residual noise when applied to an in-vehicle environment in which an audio signal is reproduced. Patent Document 3 describes a demodulation device having a section detection unit for detecting a section to be replaced and a replacement unit for replacing the signal in the section to be replaced with a replacement signal. [Prior art document] [Patent] [Patent Document 1] U.S. Patent Application Publication No. 2003 / 0171661 [Patent Document 2] Japanese Unexamined Patent Publication No. 2018-154173 [Patent Document 3] Patent No. 6413023 [Overview of the Initiative] [Problems that the invention aims to solve]

[0003] In measuring electrical potential signals, there is a need to more accurately remove measurement noise such as motion artifacts. [Means for solving the problem]

[0004] In a first embodiment of the present invention, a biological information providing device is provided, comprising: a potential signal measuring unit that measures a potential signal through a pair of electrodes in contact with a living body; a bioimpedance signal measuring unit that measures the bioimpedance generated between the pair of electrodes and outputs a bioimpedance signal corresponding to the bioimpedance; a variable coefficient filter unit that generates a control signal based on the bioimpedance signal and coefficients; a biological information signal output unit that calculates a biological information signal based on the potential signal and the control signal; a coefficient adjustment unit that adjusts the coefficients so that the signals correlated with the bioimpedance signal included in the biological information signal become smaller; and a filter control unit that controls the coefficient adjustment unit so that the convergence speed of the adaptive filter is faster when the potential signal or the bioimpedance signal satisfies predetermined conditions than when the potential signal or the bioimpedance signal does not satisfy the conditions.

[0005] In the biological information providing device, the filter control unit may determine that the biological impedance signal satisfies the above condition when the magnitude of the biological impedance signal exceeds a predetermined first threshold.

[0006] In any of the above-mentioned biological information providing devices, the filter control unit may determine that the biological impedance signal satisfies the above condition if the magnitude of the biological impedance signal exceeds the magnitude of the reference waveform signal obtained by envelope processing of the biological impedance signal.

[0007] Any of the above-mentioned biological information providing devices may further include a phase delay unit that delays the phase of the potential signal input to the biological information signal output unit by the delay time due to the envelope processing.

[0008] In any of the above-mentioned biological information providing devices, the filter control unit may determine that the bioimpedance signal satisfies the above-mentioned conditions when the magnitude of the potential signal exceeds a predetermined first threshold.

[0009] In any of the above-mentioned biological information providing devices, the filter control unit may determine that the potential signal satisfies the above-mentioned condition if the magnitude of the potential signal exceeds the magnitude of the reference waveform signal on which the potential signal has been processed.

[0010] Any biological information providing device may further include a phase delay unit that delays the phase of the potential signal input to the biological information signal output unit by the delay time due to the envelope processing.

[0011] In any biological information providing device, the filter control unit may determine that the potential signal or the bioimpedance signal satisfies the conditions if the correlation between the potential signal and the bioimpedance signal exceeds a predetermined threshold.

[0012] In any biological information providing device, the filter control unit may control the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a first value if the bioimpedance signal or the potential signal satisfies the above conditions. The filter control unit may control the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a second value smaller than the first value if the bioimpedance signal or the potential signal does not satisfy the above conditions.

[0013] In any biological information providing device, when the magnitude of the bio-impedance signal or the magnitude of the potential signal exceeds a predetermined first threshold value, the filter control unit outputs, as a biological information signal, a signal corresponding to the potential signal before it is determined that the bio-impedance signal or the potential signal satisfies the condition. When the magnitude of the bio-impedance signal or the magnitude of the potential signal does not exceed the predetermined first threshold value, the potential signal may be output as a biological information signal. When the magnitude of the bio-impedance signal or the magnitude of the potential signal is less than or equal to the first threshold value and exceeds a second threshold value lower than the first threshold value, the filter control unit may control the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a first value. When the magnitude of the bio-impedance signal or the magnitude of the potential signal is less than or equal to the second threshold value, the filter control unit may control the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a second value smaller than the first value.

[0014] In any biological information providing device, when the magnitude of the bio-impedance signal or the magnitude of the potential signal exceeds a predetermined first threshold value, the filter control unit may control the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a first value. When the magnitude of the bio-impedance signal or the magnitude of the potential signal is less than or equal to the first threshold value and exceeds a second threshold value lower than the first threshold value, the filter control unit may control the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a second value smaller than the first value. When the magnitude of the bio-impedance signal or the magnitude of the potential signal is less than or equal to the second threshold value, the filter control unit may control the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a third value smaller than the second value.

[0015] In any biological information providing device, when the coefficient adjustment unit adjusts the coefficient using the NLMS adaptive algorithm, the filter control unit may control the coefficient adjustment unit to set the stabilization constant of the coefficient adjustment unit to a first value when the bio-impedance signal or the potential signal satisfies the condition. When the bio-impedance signal or the potential signal does not satisfy the condition, the filter control unit may control the coefficient adjustment unit to set the stabilization constant of the adaptive filter to a second value greater than the first value.

[0016] In any biological information providing device, when the bio-impedance signal or the potential signal satisfies the condition, the filter control unit may control the coefficient adjustment unit so that the operation processing speed of the adaptive filter becomes a first operation processing speed. When the bio-impedance signal or the potential signal does not satisfy the condition, the filter control unit may control the coefficient adjustment unit so that the operation processing speed of the adaptive filter becomes a second operation processing speed slower than the first operation processing speed.

[0017] In any biological information providing device, when the bio-impedance signal or the potential signal satisfies the condition, the filter control unit may control the coefficient adjustment unit so that the update interval of the coefficient becomes a first interval. When the bio-impedance signal or the potential signal does not satisfy the condition, the filter control unit may control the coefficient adjustment unit so that the update interval of the coefficient becomes a second interval longer than the first interval.

[0018] In a second aspect of the present invention, there are provided a potential signal measurement unit that measures a potential signal through a pair of electrodes contacting a living body, and a selection unit that outputs a signal corresponding to the potential signal before it is determined that the potential signal satisfies the condition as a biological information signal when the potential signal satisfies a predetermined condition, and selects to output the potential signal as a biological information signal when the potential signal does not satisfy the predetermined condition.

[0019] The biological information providing device may further include: an envelope processing unit that performs envelope processing on the potential signal and outputs a reference waveform signal; a potential signal determination unit that determines that the potential signal satisfies predetermined conditions when the magnitude of the potential signal exceeds the magnitude of the reference waveform signal; a phase delay unit that delays the phase of the potential signal input to the selection unit by the delay time due to the envelope processing; and a noise reduction unit that outputs a signal corresponding to the potential signal before it was determined that the potential signal satisfies predetermined conditions.

[0020] Any biological information providing device may further include: an envelope processing unit that processes the potential signal and outputs a reference waveform signal; a potential signal determination unit that determines that the potential signal satisfies predetermined conditions when the magnitude of the potential signal exceeds the magnitude of the reference waveform signal; a first phase delay unit that delays the phase of the potential signal input to the selection unit by the delay time due to the envelope processing; a noise reduction unit that performs low-pass filtering on the output signal of the first phase delay unit and outputs a signal corresponding to the potential signal before it was determined that the potential signal satisfies the conditions when the potential signal satisfies predetermined conditions; and a second phase delay unit that delays the phase of the potential signal input to the selection unit by the delay time due to the low-pass filtering.

[0021] It should be noted that the above summary of the invention does not enumerate all of its features. Furthermore, subcombinations of these features may also constitute an invention. [Brief explanation of the drawing]

[0022] [Figure 1] This is an example of a schematic diagram illustrating the bioelectrical impedance (BioZ) generated in a biological system (Bioto 500). [Figure 2] An example of a block diagram of the configuration of the biological information providing device 100 related to the comparative example is shown. [Figure 3] An example of a block diagram of an equivalent circuit showing the algorithm of the configuration of the biological information providing device 100 related to the comparative example is shown. [Figure 4A] This figure shows the simulation results of the potential signal d(n) and the bio-information signal e(n) after noise removal when μ=0.2 is set. [Figure 4B] This figure shows the simulation results of the potential signal d(n) and the bioinformation signal e(n) after noise removal when μ=1.0. [Figure 5] An example of a block diagram of the configuration of the biological information providing device 200a according to the first embodiment is shown. [Figure 6] This is an example of a flowchart showing the processing of the biological information providing device 200a according to the first embodiment. [Figure 7] This graph shows an example of switching between multiple step size parameters μ depending on multiple thresholds. [Figure 8] An example of a block diagram of the configuration of the biological information providing device 200b according to the second embodiment is shown. [Figure 9A] This is an example of a flowchart showing the processing of the biological information providing device 200b according to the second embodiment. [Figure 9B] This is an example of a flowchart that further details step S204 in Figure 9A. [Figure 10] An example of a block diagram of the configuration of the biological information providing device 200c according to the third embodiment is shown. [Figure 11] This is an example of a flowchart showing the processing of the biological information providing device 200c according to the third embodiment. [Figure 12] An example of a block diagram of the configuration of the biological information providing device 200d according to the fourth embodiment is shown. [Figure 13A] This is an example of a graph illustrating the function of the filter control unit 60d. [Figure 13B] This is an example of a graph illustrating the function of the filter control unit 60d. [Figure 13C] This is an example of a graph illustrating the function of the filter control unit 60d. [Figure 13D] This is an example of a graph illustrating the function of the filter control unit 60d. [Figure 13E]This is an example of a graph illustrating the function of the filter control unit 60d. [Figure 13F] This is an example of a graph illustrating the function of the filter control unit 60d. [Figure 14] This is an example of a flowchart showing the processing of the biological information providing device 200d according to the fourth embodiment. [Figure 15] An example of a block diagram of the configuration of the biological information providing device 200e according to the fifth embodiment is shown. [Figure 16] This is an example of a flowchart showing the processing of the biological information providing device 200e according to the fifth embodiment. [Figure 17] An example of a block diagram of the configuration of the biological information providing device 200f according to the sixth embodiment is shown. [Figure 18] An example of a block diagram of the configuration of the 200g biological information providing device according to the seventh embodiment is shown. [Figure 19] This flowchart shows an example of the processing performed to remove sudden noise in the 200g biometric information provision device. [Figure 20] An example of a block diagram of the configuration of the biometric information providing device 200h according to the eighth embodiment is shown. [Modes for carrying out the invention]

[0023] The present invention will be described below through embodiments, but these embodiments are not intended to limit the invention as defined in the claims. Furthermore, not all combinations of features described in the embodiments are necessarily essential to the solution of the invention. In the following, embodiments may be distinguished by ordinal numbers for the purpose of describing them, but these ordinal numbers are assigned solely for the convenience of explanation and do not exclude combinations of configurations of embodiments shown with different ordinal numbers. For example, a part of the configuration of the first embodiment may be used by adding, substituting, and / or combining it with a part of the configuration of the third embodiment as needed.

[0024] In this specification, the term "connect" may be used to mean an electrical connection relationship between components. In this case, the term "connect" may mean a direct connection to the component in question via a conductor, but it does not exclude indirect connections involving other components.

[0025] Figure 1 is an example of a schematic diagram illustrating the bioelectrical impedance BioZ generated in the living organism 500. The living organism 500 comprises an epidermal layer 502, a dermis and subcutaneous layer 504, and a muscle layer 506.

[0026] In this embodiment, the living organism 500 is a human. However, the living organism 500 may be another animal. If the living organism 500 is another animal, some skin structures may differ depending on the type of living organism 500.

[0027] In measuring bioimpedance (BioZ), a pair of electrodes 150 are brought into contact with and fixed to the living organism 500, and an alternating current is passed between the electrodes. In this case, the bioimpedance (BioZ) based on the body composition of the living organism 500 can be read by reading the potential difference generated between the pair of electrodes 150. Thus, the alternating signal applied to the living organism 500 via the pair of electrodes 150 may be a differential signal based on the voltage and current applied between the electrodes.

[0028] However, the AC signal applied to the biological tissue 500 may be a signal in which the voltage level applied to one of the pair of electrodes 150 is periodically varied with respect to the ground voltage. In this case, the bioimpedance BioZ can be read via a measuring device connected to the other electrode of the pair of electrodes 150.

[0029] Furthermore, for example, when a living organism 500 activates its muscles, an action potential is generated in the muscle layer 506 of the living organism 500 due to the electrical excitation of cells. In measuring such action potentials, fluctuations in contact impedance have a strong influence, and therefore the impedance originating from the epidermal layer 502, as well as the dermis and subcutaneous layer 504, has a strong influence.

[0030] The epidermal layer 502 is the epidermis of the living body 500. For example, when the living body 500 is a human, the epidermal layer 502 is a part having an average thickness of about 0.2 mm in the skin of parts other than the palm or the sole of the foot.

[0031] In the measurement of the bio-impedance BioZ by the electrode 150, the epidermal layer 502 contributes as a half-cell potential V HC and a variable resistor R HC connected in series to the half-cell potential V ESI and a variable capacitor C ESI . The half-cell potential V HC is an electrostatic potential generated at the part where the electrode 150 and the epidermal layer 502 contact. The component of the half-cell potential V HC contributes as a component of, for example, a frequency of 20 Hz or less in the measurement of the bio-impedance BioZ.

[0032] Here, in the bio-impedance BioZ, the contribution of the contact impedance between the electrode 150 and the skin of the living body 500, that is, between the electrode 150 and the epidermal layer 502 is large. Further, the epidermal layer 502 contributes as a variable resistor R HC connected in series to the half-cell potential V ESI and a variable capacitor C ESI . The variable resistor R ESI and the variable capacitor C ESI vary significantly based on changes in the state of the skin surface and the contact state between the electrode 150 and the skin. This appears as fluctuations in the contact impedance between the electrode 150 and the skin of the living body 500 in the bio-impedance BioZ. When the skin surface is dry, these impedances increase by about 10 times, and thus are expressed as variable resistors and capacitors in the equivalent circuit.

[0033] The dermis and subcutaneous layer 504 are layers such as the dermis layer, subcutaneous tissue, and fascia. In the measurement of bio-impedance, the dermis and subcutaneous layer 504 contribute as a resistor R bodyIt contributes as follows. Of the dermis and subcutaneous layer 504, the dermis is the part of the skin through which capillaries, lymphatic vessels, nerves, etc. pass, and is formed inside the epidermal layer 502. For example, if the living organism 500 is a human, the dermis is the part that has an average thickness of about 2 mm. Of the dermis and subcutaneous layer 504, the subcutaneous tissue is formed further inside the dermis. The subcutaneous tissue is the part that supports the epidermal layer 502 and the dermis, and has an average thickness of about 2 mm to about 9 mm. The subcutaneous layer mainly contains fat cells and large blood vessels, etc. Furthermore, the dermis and subcutaneous layer 504 contain fascia between the subcutaneous tissue and the muscle layer 506. The fascia has a thickness of about 1 mm and is not as large as the skin, but it is the part that generates electrical resistance. Resistance R of the dermis and subcutaneous layer 504 body This is the sum of the electrical resistances of these multiple layers.

[0034] Ultimately, in the impedance measurement of Bio500, the variable resistor R connected in parallel contributes to the bioimpedance BioZ. ESI and variable capacitance C ESI And, resistor R body This is the result.

[0035] The muscle layer 506 is a layer containing multiple muscle fibers 508. Muscle fibers 508 are tissues that are activated by electrical signals transmitted through nerves 510. When muscle fibers 508 move, action potentials are generated within the muscle fibers 508. The muscle layer 506 contains a potential V, which is the sum (combined) of the action potentials of multiple muscle fibers associated with the activity of multiple muscles. EMG (compound action potential V EMG ) occurs. In measuring electromyogram (EMG) signals, such a composite action potential V occurs. EMG Measurements will be taken.

[0036] Such action potential V EMGSince this is a potential generated by the activity of a living organism, it is a potential difference that can occur between the pair of electrodes 150 attached to the living organism 500 even when no external current is applied. However, since the electromyographic signal is a low-potential signal, when measuring such an electromyographic signal, if bioimpedance (BioZ) is also measured, a signal that creates a potential difference between the pair of electrodes 150 is output from outside the living organism 500.

[0037] Electromyography (EMG) signals are an example of "biological information signals" that indicate information about a living organism. EMG signals may have amplitude peaks in a frequency band higher than 20 Hz and below 4 kHz. Other examples of biological information signals may include signals that show potentials generated by the activity of a living organism, such as electrocardiograms (ECG), electroencephalograms (EEG), or electrooculograms (EOG). These biological information signals are merely examples, and biological information signals are not limited to these signals as long as they are based on action potentials of the organism generated by its activity.

[0038] Here, the action potential V EMG In the measurement of the action potential V, motion artifacts (MA) may occur as measurement noise at the contact point between the electrode 150 and the body 500 when the body 500 moves, causing the electrode 150 to shift relative to the body 500 or the body 500 to vibrate. In the measurement of bioimpedance (BioZ), it is known that the fluctuation component of bioimpedance (BioZ) includes a component proportional to MA. Therefore, the action potential V EMG When measuring, by subtracting the fluctuation component of the bioimpedance BioZ, which is proportional to motion artifacts, the action potential V is unaffected by the measurement noise of motion artifacts. EMG It can measure action potentials (V). EMG In measuring the action potential V, if multiple measuring devices are used and the multiple measuring devices are grounded at predetermined intervals around the arm, EMGIt is also possible to identify the three-dimensional location of the electromyogram (EMG) generation. This makes it possible to measure the relationship between EMG signals and arm gestures, and there is a demand for accurate measurement of EMG signals.

[0039] When a living organism (500) moves, the resulting MA (Multi-Auditory Frequency) often has a peak below 20Hz, for example, if the living organism (500) is a human. The effects of such MA can be shielded by a high-pass filter. However, even if the frequency band in which the MA peak appears is below 20Hz, if the MA peak value is large, the effect of the MA, with a tail trailing from the peak, may appear in frequency bands above 20Hz.

[0040] On the other hand, vibrations occurring in the living organism 500 are, for example, vibrations that occur in moving vehicles such as trains, buses, and airplanes on which humans are riding. MA generated when the living organism 500 is riding in a moving vehicle can be directly introduced as noise in the frequency band between 50 Hz and 200 Hz. In this case, MA is known to contribute as fluctuations in capacitive components, and C ESI It contributes as a component proportional to the fluctuating component. Furthermore, the fluctuation of bioimpedance BioZ due to MA may have frequencies greater than a predetermined frequency (e.g., 20 Hz).

[0041] Therefore, variable capacitance C ESI In the measurement, by shielding components in an appropriate frequency range and amplifying the shielded components, it becomes possible to read the fluctuations of capacitive elements proportional to MA, and consequently, the fluctuations of bioimpedance (BioZ) due to MA. Below, such a variable capacitance C proportional to MA is described. ESI This document describes in detail several configurations of a bio-information provider capable of reading fluctuations in biometric signals. In particular, it explains in detail the noise reduction function of the bio-information provider when noise in the potential signal caused by MA (Magnetic Amplifier) ​​suddenly occurs.

[0042] Figure 2 shows an example of a block diagram of the configuration of a biological information providing device 100 according to a comparative example. The biological information providing device 100 is configured to include a potential signal measurement unit 10, a bioimpedance signal measurement unit 20, and an adaptive filter 30.

[0043] The potential signal measurement unit 10 measures potential signals through a pair of electrodes 150 that are in contact with the living body 500. The potential signal measurement unit 10 may amplify the composite action potential measured through the pair of electrodes 150 and output the analog-to-digital converted signal as a potential signal.

[0044] The bioimpedance signal measurement unit 20 measures the bioimpedance generated between a pair of electrodes 150 and outputs a bioimpedance signal corresponding to the bioimpedance BioZ. The bioimpedance signal measurement unit 20 outputs the bioimpedance signal to the adaptive filter 30. The bioimpedance signal measurement unit 20 includes an AC signal output unit 22.

[0045] The AC signal output unit 22 applies an AC signal to the pair of electrodes 150 so that the bioimpedance signal measurement unit 20 can measure the bioimpedance BioZ. For example, the AC signal output unit 22 includes a DC constant voltage source or a constant current source and a mixer for generating a constant AC signal. Another example includes a square wave or sinusoidal AC power supply and a resistor for limiting the amplitude of the signal.

[0046] The AC signal output unit 22 may apply an AC signal as a differential signal to the pair of electrodes 150. In this case, the bioimpedance signal measurement unit 20 measures the bioimpedance BioZ by detecting the impedance to the differential signal. In another example, the AC signal output unit 22 applies a signal with a waveform having a sinusoidal or square wave-like potential difference from a predetermined reference potential to one of the pair of electrodes 150. In this case, the potential signal measurement unit 10 and the bioimpedance signal measurement unit 20 each include an amplifier and a single-ended differential conversion circuit, and may measure the bioimpedance BioZ by extracting a differential signal from the single-ended signal.

[0047] The adaptive filter 30 receives the potential signal V from the composite action potential measured by the potential signal measurement unit 10. EMG This filter removes noise caused by fluctuations in the bioimpedance signal. The adaptive filter 30 comprises a variable coefficient filter section 32, a coefficient adjustment section 34, and a bioinformation signal output section 40.

[0048] The variable coefficient filter unit 32 generates a control signal to remove noise from the potential signal based on the bioimpedance signal and coefficients. The variable coefficient filter unit 32 may generate the control signal by multiplying the bioimpedance signal by the coefficient adjustment unit 34.

[0049] The coefficient adjustment unit 34 adjusts the coefficients of the variable coefficient filter unit 32 based on the bioimpedance signal and the signal output by the bioinformation signal output unit 40. The coefficient adjustment unit 34 adaptively adjusts the coefficients in response to noise mixed into the potential signal in accordance with the feedback based on the output of the bioinformation signal output unit 40.

[0050] The bioinformation signal output unit 40 calculates a bioinformation signal by removing the control signal component from the potential signal based on the potential signal output from the potential signal measurement unit 10 and the control signal output from the variable coefficient filter unit 32, and outputs the calculated bioinformation signal. The bioinformation signal output unit 40 includes a subtractor 42 and has a path for outputting the output from the subtractor 42 to the outside and a path for providing feedback to adjust the variable coefficient filter unit 32. The specific algorithms of the variable coefficient filter unit 32 and the coefficient adjustment unit 34 will be explained in detail with reference to Figure 3.

[0051] Figure 3 shows an example of an equivalent circuit block diagram illustrating the algorithm of the configuration of the biological information providing device 100 according to the comparative example.

[0052] The potential signal measurement unit 10 acquires a biopotential signal from the electrode 150. In this case, the signal from the electrode 150 is acquired by adding together a signal that has noise due to MA, which is correlated with the bioimpedance signal, and the potential signal measurement unit 10 outputs the acquired signal as a potential signal d(n). Here, the action potential V EMG In the measurement, the signal V shows only the "true" action potential, free from noise introduced by bioimpedance (BioZ). E Let's assume that one of the input signals added to the potential signal measurement unit 10 is signal s(n), where s(n) = V E The following conditions are met. Noise x(n) due to bioimpedance BioZ, which is an input signal added to the potential signal measurement unit 10 other than the signal s(n), is introduced via the "unknown" coefficient H(n) shown in the noise intrusion section 12. Here, n is a natural number representing the number of loops in the feedback loop described later, and is a time-dependent variable.

[0053] Therefore, the potential signal d(n) output by the potential signal measurement unit 10 can be considered as a signal obtained by adding H(n) × x(n) to s(n) via the parasitic adder 14. In other words, the potential signal measurement unit 10 outputs a potential signal d(n) that satisfies d(n) = s(n) + H(n) × x(n).

[0054] On the other hand, the bioimpedance signal measurement unit 20 is connected to the electrode 150 in the same way as the potential signal measurement unit 10, but it can extract a signal x(n) based on BioZ by filtering out components of other signals. The bioimpedance signal measurement unit 20 outputs the extracted signal x(n) to the variable coefficient filter unit 32 and the coefficient adjustment unit 34.

[0055] The variable coefficient filter unit 32 outputs a control signal W(n) × x(n) obtained by multiplying x(n) by the coefficient W(n). The coefficient W(n) is a coefficient whose magnitude corresponds to the output of the biological information providing device 100, and whose value is adjusted by a feedback loop via the coefficient adjustment unit 34.

[0056] The bio-information signal output unit 40 outputs a bio-information signal e(n), which becomes the external output of the bio-information providing device 100, based on the potential signal d(n) and the control signal W(n)×x(n). It also outputs this signal to the coefficient adjustment unit 34 as feedback for adjusting the coefficients of the variable coefficient filter unit 32. The bio-information signal output unit 40 includes a subtractor 42.

[0057] The subtractor 42 outputs a bio-information signal e(n) obtained by subtracting the control signal W(n)×x(n) from the potential signal d(n). Therefore, the subtractor 42 outputs a signal e(n) that satisfies the equation e(n) = d(n) - W(n)×x(n) = s(n) + H(n)×x(n) - W(n)×x(n). The bio-information signal output unit 40 outputs the signal e(n) to the outside of the bio-information providing device 100 and to the coefficient adjustment unit 34.

[0058] The coefficient adjustment unit 34 adjusts the coefficient W(n) of the variable coefficient filter unit 32 to adjust the magnitude of the coefficient of the adaptive filter 30. The coefficient adjustment unit 34 then adjusts the coefficient W(n) to adjust the magnitude of the control signal W(n)×x(n) so that the signal (H(n)×x(n)-W(n)×x(n)) which is correlated with the bioimpedance signal x(n) contained in the bioinformation signal e(n)=s(n)+H(n)×x(n)-W(n)×x(n) becomes small. The coefficient adjustment unit 34 may adjust the coefficient W(n) to adjust the magnitude of the control signal W(n)×x(n) so that the error signal (H(n)×x(n)-W(n)×x(n)) which is correlated with the bioimpedance signal x(n) contained in the bioinformation signal e(n)=s(n)+H(n)×x(n)-W(n)×x(n) becomes zero.

[0059] Here, a signal can be represented by the elements of amplitude, frequency, and waveform. In this specification, "signal magnitude" for each signal may mean the amplitude of each signal unless otherwise specified. In particular, in the case of differential AC signals, the signal can take positive or negative values, but "signal magnitude" may mean the amplitude of the signal as expressed in absolute value.

[0060] Thus, the variable coefficient filter section 32 and the coefficient adjustment section 34 implement an adaptive algorithm that, based on feedback of the output signal, adapts so that as the number of loop stages in the feedback loop increases, noise components are removed and the output signal becomes s(n). In other words, the coefficient adjustment section 34 applies feedback such that when the number of loops n in the feedback loop increases, W(n)=H(n) holds true and the signal e(n) converges to asymptotically approach s(n).

[0061] Thus, the variable coefficient filter unit 32 generates a control signal W(n)×x(n) adapted to the current error from signal s(n) contained in signal e(n), such that signal e(n) asymptotically approaches signal s(n) which shows only "true" action potentials. In this respect, the adaptive filter 30 is called an "adaptive" filter. In the end, the variable coefficient filter unit 32 generates a control signal W(n)×x(n) that shows the noise component H(n)×x(n) contained in the potential signal d(n), based on the bioimpedance signal x(n) and coefficient W(n).

[0062] Here, the coefficient adjustment unit 34 may use the LMS (least mean square) algorithm as an adaptive algorithm to adjust the coefficient W(n). In this case, the coefficient adjustment unit 34 uses a parameter μ called the step size parameter to set the coefficient W(n) such that the recurrence relation W(n+1)=W(n)+μ×e(n)×x(n) holds for the coefficient W(n).

[0063] Alternatively, the normalized least mean square (NLMS) algorithm may be used as an adaptive algorithm. In normalized LMS, the coefficients W(n) are set such that the recurrence relation (1) holds.

number

[0064] For example, the step size parameter μ is a parameter whose value is in the range of 0 < μ < 2.0. In the comparative example, the biological information providing device 100 is a constant value within this range.

[0065] The step size parameter μ is a parameter that controls the amount of coefficient update of the variable coefficient filter section 32, i.e., the convergence speed of the adaptive filter 30. If the step size parameter μ is too large, it will attenuate not only the control signal W(n)×x(n) but also the magnitude of the signal output as the main signal e(n). This is because if the step size parameter μ is too large, the coefficients of the variable coefficient filter section 32 change faster than the main signal e(n), and in the feedback via the coefficient adjustment section 34, it becomes impossible to distinguish between the main signal e(n) and the control signal W(n)×x(n), resulting in the output of a coefficient that attenuates the main signal along with W(n) in the next stage of adjustment.

[0066] On the other hand, if the step size parameter μ is small, the convergence accuracy of the adaptive filter 30 improves, but the convergence speed of the adaptive filter 30 slows down. However, when the step size parameter μ is small, if sudden noise is introduced into the bioimpedance signal x(n), the adaptive filter 30 may not be able to follow it at a sufficient speed.

[0067] Figure 4A shows the simulation results of the potential signal d(n) and the bioinformation signal e(n) after noise removal when μ = 0.2. In the figure, line 82 shows the bioinformation signal e(n) after noise removal by the adaptive filter 30 from the potential signal d(n), line 84 shows the potential signal d(n), and regions A and B are also shown. Here, the graph is for explaining the overview and properties, and the units of each axis in the graph are shown in arbitrary units.

[0068] In region A, the difference between the piecewise line 82, which shows the bioinformation signal e(n) after noise removal by the adaptive filter 30, and the piecewise line 84, which shows the potential signal d(n), is not large. In the figure, even in the piecewise line 82 showing the bioinformation signal e(n), a signal of the desired magnitude is obtained.

[0069] In region B, the piecewise line 84 representing the potential signal d(n) contains significant noise. Therefore, the piecewise line 82 representing the bio-information signal e(n) after noise removal by the adaptive filter 30 does not sufficiently remove the effects of the noise. Thus, in region A of the figure, the step size parameter μ is set to an appropriate size at which the potential signal d(n) can be obtained with an appropriate magnitude. However, in region B, the step size parameter μ is too small, and the effects of the noise are not sufficiently removed.

[0070] Figure 4B shows the simulation results of the potential signal d(n) and the bioinformation signal e(n) after noise removal when μ=1.0. In the figure, as in Figure 4A, the piecewise line 84 representing the potential signal d(n) and the piecewise line 82 representing the bioinformation signal e(n) after noise removal by the adaptive filter 30 are shown, along with regions C and D.

[0071] In region C, the piecewise line 82, which represents the bioinformation signal e(n) after noise removal by the adaptive filter 30, is attenuated more significantly than the piecewise line 84, which represents the potential signal d(n), compared to region A in Figure 4A. In the piecewise line 82, which represents the bioinformation signal e(n) after noise removal by the adaptive filter 30, the magnitude of the main signal, the bioinformation signal e(n), itself has also been attenuated, resulting in undesirable behavior.

[0072] In region D, the piecewise line 84, which shows the bioinformation signal e(n) after noise removal by the adaptive filter 30, demonstrates that the noise effect is sufficiently removed. Thus, in region C of the figure, the magnitude of the main signal, the bioinformation signal e(n), is attenuated, indicating that the step size parameter μ is set to an excessively large value. On the other hand, in region D, the noise effect is sufficiently removed, and the step size parameter μ has an appropriate value.

[0073] Motion artifacts can be either steady-state or sudden. If the step size parameter μ is set to a small value to handle steady-state motion artifacts, noise may not be completely removed when sudden motion artifacts are introduced, as shown in the example in region B of Figure 4A. On the other hand, if the step size parameter μ is set to a large value to handle the introduction of sudden motion artifacts, the magnitude of the main signal, the bio-information signal e(n), may be attenuated, as shown in the example in region C of Figure 4B.

[0074] As described above, when the step size parameter μ is set to a constant, the appropriate value of the step size parameter μ differs depending on whether or not there is sudden noise. Therefore, when the step size parameter μ is set to a constant, it may not be possible to set the step size parameter μ to an appropriate value for each time domain.

[0075] Figure 5 shows an example of a block diagram of the configuration of the biological information providing device 200a according to the first embodiment. Below, the configuration of the biological information providing device 200a will be described, focusing on the differences from the biological information providing device 100. In addition to the configuration of the biological information providing device 100, the biological information providing device 200a is equipped with a filter control unit 60a.

[0076] The filter control unit 60a controls the adaptive filter 30 such that when the bioimpedance signal x(n) satisfies predetermined conditions, the convergence speed of the adaptive filter 30 is faster than when the bioimpedance signal x(n) does not satisfy the conditions. The filter control unit 60a includes a bioimpedance signal determination unit 62a and a step size setting unit 64.

[0077] The bioimpedance signal determination unit 62a determines whether the magnitude of the bioimpedance signal x(n) satisfies predetermined conditions. Here, for each signal, "satisfying predetermined conditions" may mean, for example, that the amplitude, expressed as the absolute value of each signal, is greater than a predetermined threshold. In this case, if this threshold is called the first threshold, then "satisfying predetermined conditions" means that the amplitude of the target signal is outside the range of [-first threshold, +first threshold]. In this example, the filter control unit 60a may determine, via the determination of the bioimpedance signal determination unit 62a, that the bioimpedance signal x(n) satisfies predetermined conditions if the magnitude of the bioimpedance signal x(n) exceeds a predetermined first threshold. In response to this determination, the convergence speed of the adaptive filter 30 is adjusted to reduce the control signal W(n) × x(n) corresponding to the noise component appearing in the bioinformation signal e(n).

[0078] The step size setting unit 64 sets the magnitude of the step size parameter μ based on the determination result of the bioimpedance signal determination unit 62a. Specifically, in the example of determination described in the bioimpedance signal determination unit 62a, if the magnitude of the bioimpedance signal x(n) exceeds a predetermined first threshold, the step size setting unit 64 controls the adaptive filter 30 in the coefficient adjustment unit 34 to set the step size parameter μ of the adaptive filter 30 to a predetermined first value. For example, the first value may be μ=1.0 or μ=0.8. On the other hand, if the magnitude of the bioimpedance signal x(n) is less than or equal to the predetermined first threshold, the step size setting unit 64 controls the adaptive filter 30 in the coefficient adjustment unit 34 to set the step size parameter μ of the adaptive filter 30 to a predetermined second value smaller than the first value. For example, the second value may be μ=0.1 or μ=0.2. Thus, in this embodiment, instead of using a constant step size parameter μ, different step size parameters μ are used depending on the conditions.

[0079] In this embodiment, the step size setting unit 64 adjusts the step size parameter μ when adjusting the coefficients of the coefficient adjustment unit 34. However, the step size setting unit 64 may also perform the adjustment by directly controlling the variable coefficient filter unit 32.

[0080] As a result, the filter control unit 60a controls the coefficient adjustment unit 34 to set the step size parameter μ of the coefficient adjustment unit 34 to a predetermined first value when the bioimpedance signal x(n) satisfies predetermined conditions. This allows the filter control unit 60a to adjust the convergence speed of the adaptive filter 30 so as to reduce the influence of the control signal W(n)×x(n) corresponding to the noise component appearing in the bioinformation signal e(n). On the other hand, if the bioimpedance signal x(n) does not satisfy predetermined conditions, the filter control unit 60a may control the coefficient adjustment unit 34 to set the step size parameter μ of the adaptive filter 30 to a predetermined second value smaller than the first value. This allows the filter control unit 60a to adjust the convergence speed of the adaptive filter 30 so as to sufficiently reduce the influence of the control signal W(n)×x(n) corresponding to the noise component appearing in the bioinformation signal e(n), while preventing the magnitude of the signal e(n) itself from becoming too small.

[0081] By performing this control, the bio-information providing device 200a of this embodiment can appropriately adjust the convergence speed of the adaptive filter 30 to remove noise even when sudden noise is mixed into the bio-impedance signal x(n). Furthermore, when no sudden noise is mixed in, the step size parameter is set to a small value, thus preventing the main signal e(n) from being attenuated along with the control signal W(n)×x(n) due to the step parameter μ being set to an excessively large value.

[0082] The filter control unit 60a may set the step size parameter μ to a first or second value, and then make another determination after a predetermined number of cycles in the feedback loop has elapsed, or after a predetermined period of time has elapsed. Depending on this determination, the step size parameter μ may be set to a different value. This allows the adaptive filter 30 to increase the step size parameter μ in response to sudden noise to remove the effect of the noise, and then decrease the step size parameter μ when the sudden noise no longer appears.

[0083] In this embodiment, the filter control unit 60a determines whether the control signal W(n) × x(n) is greater than or equal to a predetermined magnitude, and switches the convergence speed of the adaptive filter 30 in two stages by changing the step size parameter μ in two stages. However, the change in the convergence speed of the adaptive filter 30 performed by the filter control unit 60a is not limited to two stages. That is, in determining whether the control signal W(n) × x(n) is greater than or equal to a predetermined magnitude, multiple thresholds may be provided, and the filter control unit 60a may switch the convergence speed of the adaptive filter 30 in multiple stages according to the magnitude of the noise. A specific example of this will be described later with reference to Figure 7.

[0084] As described above, the filter control unit 60a controls the step size parameter μ when the adaptive algorithm of the adaptive filter 30 is LMS. However, the adaptive algorithm of the adaptive filter 30 is not limited to LMS, and the adaptive filter 30 may employ NLMS, RLS (Recursive Least Squares), or Affine Projection Algorithm (APA) as its adaptive algorithm. In other words, the adaptive filter 30 may be an NLMS adaptive filter, an RLS adaptive filter, or an APA adaptive filter.

[0085] In this case, the filter control unit 60a may, along with or instead of the step size parameter μ, change the stabilization constant ε from a constant to a variable parameter and dynamically change the magnitude of the parameter ε. In equation (1), when the stabilization constant ε is made smaller as a variable parameter, the convergence speed of the adaptive filter 30 increases, and when the stabilization constant ε is made larger as a variable parameter, the convergence speed of the adaptive filter 30 decreases. Therefore, when the bioimpedance signal x(n) satisfies predetermined conditions, the filter control unit 60a may control the adaptive filter 30 in the coefficient adjustment unit 34 to set the stabilization constant ε of the adaptive filter 30 to a predetermined first value. As a result, the filter control unit 60a adjusts the convergence speed of the adaptive filter 30 to reduce the influence of the control signal W(n)×x(n) corresponding to the noise component appearing in the bioinformation signal e(n), similar to when adjusting the step size parameter μ. On the other hand, if the bioimpedance signal x(n) does not satisfy predetermined conditions, the filter control unit 60a may control the adaptive filter 30 in the coefficient adjustment unit 34 to set the stabilization constant ε of the adaptive filter 30 to a predetermined second value that is greater than the first value. In this way, the filter control unit 60a adjusts the convergence speed of the adaptive filter 30 so as to sufficiently reduce the influence of the control signal W(n)×x(n) corresponding to the noise component appearing in the bioinformation signal e(n), while preventing the magnitude of the signal e(n) itself from becoming too small.

[0086] In another embodiment, the filter control unit 60a may control the variable coefficient filter unit 32 so that the operating speed of the adaptive filter 30 becomes a first operating speed when the bioimpedance signal x(n) satisfies predetermined conditions. On the other hand, if the bioimpedance signal x(n) does not satisfy predetermined conditions, the filter control unit 60a may control the variable coefficient filter unit 32 so that the operating speed of the adaptive filter 30 becomes a second operating speed that is slower than the first operating speed. In this way, the filter control unit 60a can adjust the convergence speed of the adaptive filter 30, similar to adjusting the step size parameter μ and the stabilization constant ε.

[0087] For example, if the biometric information providing device 200a is implemented as an IC (Integrated Circuit), each element in the IC operates based on an operating frequency (clock frequency) that references a reference clock. As a specific example of the filter control unit 60a changing the operating processing speed of the adaptive filter 30, the filter control unit 60a may change the operating frequency of the adaptive filter 30.

[0088] In other words, when the filter control unit 60a changes the adaptive filter 30 to a first operating processing speed, it may increase the operating frequency of the adaptive filter 30. For example, the operating frequency corresponding to the first processing speed may be 6kHz or 8kHz. On the other hand, when changing the adaptive filter 30 to a second operating processing speed, it may decrease the operating frequency of the adaptive filter 30. Again, for example, the operating frequency corresponding to the second processing speed may be 3kHz or 2kHz.

[0089] In yet another embodiment, the filter control unit 60a may control the adaptive filter so that the coefficient update interval is a first interval when the bioimpedance signal x(n) satisfies predetermined conditions. On the other hand, if the bioimpedance signal x(n) does not satisfy predetermined conditions, the filter control unit 60a may control the adaptive filter so that the coefficient update interval is a second interval which is longer than the first interval. Here, for example, the second interval may be twice the length of the first interval, or four times the length of the first interval. Alternatively, the filter control unit 60a may control the filter to stop updating the coefficients, or to resume updating the coefficients that have been stopped, depending on the conditions that the bioimpedance signal x(n) satisfies. In this case, the filter control unit 60a may control the filter to stop updating the coefficients for a predetermined period of time and to resume updating the coefficients after the predetermined period has elapsed.

[0090] As in these embodiments, the filter control unit 60a can control the adaptive filter 30 to increase its convergence speed not only by switching the magnitude of the step size parameter μ, but also by switching the operating processing speed of the adaptive filter 30 or the coefficient update interval. Therefore, in the biological information providing device 200a, the operating processing speed of the adaptive filter 30 or the coefficient update interval may be switched instead of, or in combination with, switching the magnitude of the step size parameter μ.

[0091] The biological information providing device 200a can switch the convergence speed of the adaptive algorithm in the adaptive filter 30 by any of the methods described above. The filter control unit 60a of the biological information providing device 200a may switch the convergence speed of the adaptive filter 30 by selecting any of these methods or by combining them.

[0092] As previously mentioned, the filter control unit 60a may switch the convergence speed of the adaptive filter 30 in multiple stages. In this case, the filter control unit 60a may switch the convergence speed of the adaptive filter 30 in multiple stages by combining multiple methods for switching the convergence speed of the adaptive filter 30.

[0093] Figure 6 is an example of a flowchart showing the processing of the biological information providing device 200a according to the first embodiment. The processing of the biological information providing device 200a in this embodiment comprises steps S102 to S116.

[0094] The potential signal measurement unit 10 outputs a potential signal d(n). The bioimpedance signal measurement unit 20 outputs a bioimpedance signal x(n) (S102). The bioimpedance signal determination unit 62a of the filter control unit 60a determines whether the magnitude of the bioimpedance signal x(n) satisfies predetermined conditions (S104).

[0095] The process branches depending on whether the magnitude of the bioimpedance signal x(n) satisfies predetermined conditions (S106). If the magnitude of the bioimpedance signal x(n) satisfies predetermined conditions, the process proceeds to S108; if the magnitude of the bioimpedance signal x(n) does not satisfy predetermined conditions, the process proceeds to S110.

[0096] The step size setting unit 64 of the filter control unit 60a sets the step size parameter μ of the adaptive filter 30 to a first value (for example, μ = 0.8) (S108). After this, the process proceeds to S112.

[0097] The step size setting unit 64 of the filter control unit 60a sets the step size parameter μ of the adaptive filter 30 to a second value smaller than the first value (for example, μ = 0.2) (S110). After this, the process proceeds to S112 and merges.

[0098] The coefficient adjustment unit 34 updates the coefficient W(n) based on the bioimpedance signal x(n) and the bioinformation signal e(n) so that the error signal H(n)×x(n)-W(n)×x(n) is minimized (S112). The adaptive filter 30 multiplies the bioimpedance signal by the coefficient W(n) to generate a control signal W(n)×x(n) that is correlated with the bioimpedance signal (S114).

[0099] The subtractor 42 of the bioinformation signal output unit 40 removes the control signal W(n) × x(n) predicted from the potential signal d(n) superimposed with MA (motion artifact). As a result, the bioinformation signal output unit 40 outputs a bioinformation signal e(n) from which the control signal W(n) × x(n) has been removed (S116). This completes the process.

[0100] This processing allows the adaptive filter 30 to appropriately track the convergence speed of the noise removal filter even when sudden noise is introduced during the measurement of the potential signal d(n), and prevents the main signal e(n) from being attenuated along with the control signal W(n)×x(n) due to an excessive coefficient update frequency.

[0101] Figure 7 shows an example of a graph when switching between multiple step size parameters μ according to multiple thresholds. In the figure, an example is shown in which two thresholds, threshold TH1 and threshold TH2, are provided as thresholds for the magnitude of the bioimpedance signal x(n). In Figure 7, as in Figures 4A and 4B, a piecewise line 84 showing the potential signal d(n) and a piecewise line 82 showing the bioinformation signal e(n) after noise has been removed by the adaptive filter 30 are shown.

[0102] Thresholds TH1 and TH2 are predetermined thresholds, with threshold TH2 being smaller than threshold TH1. In this example, the step size parameter is switched between three levels within a range determined by thresholds TH1 and TH2.

[0103] The bioimpedance signal determination unit 62a determines whether the magnitude of the bioimpedance signal is greater than threshold TH1, greater than threshold TH2 but less than or equal to threshold TH1, or less than or equal to threshold TH2. For example, threshold TH1 is 1mV and threshold TH2 is 0.25mV.

[0104] The step size setting unit 64 controls the coefficient adjustment unit 34 to set the step size parameter μ to 0.5 when the magnitude of the bioimpedance signal is greater than the threshold TH1. The step size setting unit 64 controls the coefficient adjustment unit 34 to set the step size parameter μ to 0.2 when the magnitude of the bioimpedance signal is greater than the threshold TH1.

[0105] As a result, the filter control unit 60a may control the coefficient adjustment unit 34 to set the step size parameter μ of the coefficient adjustment unit 34 to 0.5 if the magnitude of the bioimpedance signal x(n) exceeds TH1. The filter control unit 60a may also control the coefficient adjustment unit 34 to set the step size parameter μ of the coefficient adjustment unit 34 to 0.2 if the magnitude of the bioimpedance signal x(n) is less than or equal to TH1 and exceeds TH2. The filter control unit 60a may also control the coefficient adjustment unit 34 to set the step size parameter of the coefficient adjustment unit 34 to 0.02 if the magnitude of the bioimpedance signal x(n) is less than or equal to TH2.

[0106] In this embodiment, threshold TH1 is an example of a "first threshold," and threshold TH2 is an example of a "second threshold." Furthermore, the step size parameter μ=0.5 is an example of a "first value," the step size parameter μ=0.2 is an example of a "second value," and the step size parameter μ=0.02 is an example of a "third value." Note that more thresholds may be set than in this embodiment, and accordingly, the step size parameter μ may be switched between more steps than in this embodiment.

[0107] In this way, the filter control unit 60a switches the step size parameter μ in three stages, thereby reducing noise mixed into the potential signal d(n) with appropriate accuracy. Furthermore, in the range where the potential signal d(n) is small, the step size parameter μ is also set to a small value, which prevents the main signal e(n) from being attenuated along with the control signal W(n)×x(n).

[0108] Figure 8 shows an example of a block diagram of the configuration of the biological information providing device 200b according to the second embodiment. Below, the configuration of the biological information providing device 200b will be explained, focusing on the differences from the biological information providing device 200a. The biological information providing device 200b has a filter control unit 60b instead of a filter control unit 60a compared to the configuration of the biological information providing device 200a. In addition, the biological information providing device 200b has a phase delay unit 70 between the potential signal measurement unit 10 and the biological information signal output unit 40.

[0109] The filter control unit 60b controls the adaptive filter 30 such that when the bioimpedance signal x(n) satisfies predetermined conditions, the convergence speed of the adaptive filter 30 is faster than when the bioimpedance signal x(n) does not satisfy the conditions. The filter control unit 60b includes an envelope processing unit 66, a bioimpedance signal determination unit 62b, and a step size setting unit 64.

[0110] The envelope processing unit 66 performs envelope processing on the bioimpedance signal x(n) and outputs a signal that shows the shape of the sudden noise. The envelope processing performed by the envelope processing unit 66 includes absolute value conversion of the bioimpedance signal x(n) and low-pass filtering. Here, the absolute value conversion of the signal value may be performed by squaring the signal value, for example. The processing performed by the envelope processing unit 66 on the bioimpedance signal x(n) may include arithmetic processing such as amplification of the signal after low-pass filtering and addition of a predetermined constant to the signal value. As a result, the envelope processing unit 66 outputs a signal that shows the shape of the sudden noise. This signal showing the shape of the waveform is an example of a "reference waveform signal". Through such envelope processing, the filter control unit 60b can easily extract only the noise that has suddenly occurred due to motion artifacts, etc., from the bioimpedance signal x(n). As an example, the envelope processing unit 66 includes a low-pass filter with a cutoff frequency of 10 Hz.

[0111] Here, the envelope processing unit 66 performs envelope processing, including low-pass filtering, on the bioimpedance signal x(n), which may cause a signal delay in the control signal output by the variable coefficient filter unit 32 relative to the potential signal measurement unit 10. For example, if a low-pass filter with a cutoff frequency of 10 Hz is used as the low-pass filter included in the envelope processing unit 66, a signal delay of approximately 30 milliseconds will occur. When such a signal delay occurs, the phase delay unit 70 is added to adjust the phase between the signal output by the potential signal measurement unit 10 and the signal output by the variable coefficient filter unit 32, even when using the envelope processing unit 66.

[0112] The bioimpedance signal determination unit 62b determines whether or not sudden noise is present based on the shape of the bioimpedance signal x(n) that has passed through the envelope processing unit 66. In particular, the bioimpedance signal determination unit 62b determines whether or not the magnitude of the bioimpedance signal x(n) exceeds the magnitude of the reference waveform signal, which is the signal obtained by filtering the bioimpedance signal x(n) by the envelope processing unit 66.

[0113] As a result, the filter control unit 60b determines that the bioimpedance signal x(n) satisfies a predetermined condition if the magnitude of the bioimpedance signal x(n) exceeds the magnitude of the reference waveform signal after the bioimpedance signal x(n) has been filtered by the envelope processing unit 66. This process is equivalent to replacing the process described later for the potential signal d(n) in the fourth embodiment with a process applied to the bioimpedance signal x(n), referring to Figures 13A to 13F.

[0114] Here, the step size setting unit 64 sets the magnitude of the step size parameter μ based on the determination result of the potential signal determination unit 67b. The function of the step size setting unit 64 is the same as that of the step size setting unit 64 included in the filter control unit 60a, except that it is based on the determination result of the bioimpedance signal determination unit 62d.

[0115] The phase delay unit 70 adjusts the phase of the potential signal d(n) output from the potential signal measurement unit 10 in relation to the control signal W(n) × x(n) output from the variable coefficient filter unit 32. As a result, the phase delay unit 70 delays the phase of the potential signal d(n) input to the bio-information signal output unit 40 by the delay time caused by the filtering process by the envelope processing unit 66. In this embodiment of the bio-information providing device 200b, the phase delay unit 70 can be omitted if the phase delay caused by the envelope processing unit 66 is not considered to be significant.

[0116] Figure 9A is an example of a flowchart showing the processing of the biological information providing device 200b according to the second embodiment. The processing of this embodiment comprises steps S202 to S216.

[0117] The potential signal measurement unit 10 outputs a potential signal d(n). The bioimpedance signal measurement unit 20 outputs a bioimpedance signal x(n) (S202).

[0118] The bioimpedance signal determination unit 62b of the filter control unit 60 determines whether or not sudden noise has been introduced into the bioimpedance signal x(n) (S204). This step will be described in detail later with reference to Figure 9B.

[0119] The process branches depending on whether or not it is determined that sudden noise has been introduced into the bioimpedance signal (S206). If it is determined that sudden noise has been introduced into the bioimpedance signal x(n), the process proceeds to S208; if it is not determined that sudden noise has been introduced into the bioimpedance signal x(n), the process proceeds to S210.

[0120] The step size setting unit 64 of the filter control unit 60b sets the step size parameter to the first value (S208). After this, the process proceeds to S212.

[0121] The step size setting unit 64 of the filter control unit 60b sets the step size parameter to a second value, which is smaller than the first value (S210). After this, the process proceeds to S212 and merges.

[0122] The coefficient adjustment unit 34 updates the coefficient W(n) based on the bioimpedance signal x(n) and the bioinformation signal e(n) so that the error signal H(n)×x(n)-W(n)×x(n) is minimized (S212). The variable coefficient filter unit 32 multiplies the bioimpedance signal x(n) by the coefficient W(n) to generate a control signal W(n)×x(n) that is correlated with the bioimpedance signal x(n) (S214).

[0123] The subtractor 42 of the bioinformation signal output unit 40 removes the control signal W(n) × x(n) predicted from the potential signal d(n) superimposed with motion artifacts. As a result, the bioinformation signal output unit 40 outputs a bioinformation signal e(n) from which the control signal W(n) × x(n) has been removed (S216). The process is now complete.

[0124] Figure 9B is an example of a flowchart that further details step S204 in Figure 9A. The process of this embodiment comprises steps S222 to S234.

[0125] The filter control unit 60b converts the potential signal d(n) and the bioimpedance signal x(n) into absolute value signals (S222). For this purpose, the filter control unit 60b may have a modulator, mixer, and / or arithmetic unit, etc. (not shown).

[0126] The envelope processing unit 66 performs envelope processing, including absolute value conversion and low-pass filtering, on the absolute value converted signal to obtain a reference waveform signal (S224). The bioimpedance signal determination unit 62b determines whether the bioimpedance signal exceeds the magnitude of the reference waveform signal after envelope processing, including absolute value conversion and low-pass filtering, performed by the envelope processing unit 66 (S226).

[0127] The process branches depending on whether the bioimpedance signal exceeds the magnitude of the reference waveform signal after envelope processing, including low-pass filtering by the envelope processing unit 66 (S228). If the bioimpedance signal exceeds the magnitude of the reference waveform signal, the process proceeds to S230; if the bioimpedance signal does not exceed the magnitude of the reference waveform signal, the process proceeds to S232.

[0128] If the bioimpedance signal exceeds the magnitude of the reference waveform signal, the bioimpedance signal determination unit 62b determines that there is sudden noise interference (S230). The process then proceeds to S234.

[0129] If the bioimpedance signal does not exceed the magnitude of the reference waveform signal, the bioimpedance signal determination unit 62b determines that there is no sudden noise interference (S232). After this, the process proceeds to S234.

[0130] The bioimpedance signal determination unit 62b outputs the determination result of whether or not there is sudden noise (S234). This completes the process in S204.

[0131] This processing allows the adaptive filter 30 to appropriately track the convergence speed of noise removal even when sudden noise is introduced during the measurement of the biological information signal e(n) in the biological information provision device 200b. Furthermore, it prevents the main signal e(n) from being attenuated along with the control signal W(n)×x(n) due to an excessive coefficient update frequency.

[0132] Figure 10 shows an example of a block diagram of the configuration of the biological information providing device 200c according to the third embodiment. Below, the configuration of the biological information providing device 200c will be explained, focusing on the differences from the biological information providing device 200a. The biological information providing device 200c has a filter control unit 60c instead of a filter control unit 60a compared to the configuration of the biological information providing device 200a. Unlike the filter control unit 60a, the filter control unit 60c is connected to the output of the potential signal measurement unit 10, rather than the output of the bioimpedance signal measurement unit 20.

[0133] The filter control unit 60c controls the coefficient adjustment unit 34 so that when the potential signal d(n) satisfies predetermined conditions, the convergence speed of the adaptive filter 30 is faster than when the potential signal d(n) does not satisfy the conditions. As explained in Figure 3, the x(n) component also appears in the potential signal d(n) = s(n) + H(n) × x(n). Therefore, noise x(n) associated with fluctuations in bioimpedance BioZ can also be detected by measuring the potential signal d(n). The filter control unit 60c includes a potential signal determination unit 67a and a step size setting unit 64.

[0134] The potential signal determination unit 67a determines whether the potential signal d(n) satisfies predetermined conditions. For example, the filter control unit 60c determines that the potential signal d(n) satisfies predetermined conditions if the magnitude of the potential signal d(n) exceeds a predetermined first threshold.

[0135] Furthermore, in the case of the potential signal d(n), multiple thresholds may be provided, similar to the case of the bioimpedance signal x(n) in the first embodiment. As a result, the filter control unit 60 may switch the step size parameter μ in multiple stages.

[0136] Therefore, the filter control unit 60c may control the coefficient adjustment unit 34 to set the step size parameter μ of the coefficient adjustment unit 34 to a predetermined first value (for example, μ = 0.5) if the magnitude of the potential signal d(n) exceeds a first threshold. Furthermore, the filter control unit 60c may control the coefficient adjustment unit 34 to set the step size parameter μ of the coefficient adjustment unit 34 to a second value smaller than the first value (for example, μ = 0.2) if the magnitude of the bioimpedance signal x(n) is less than or equal to TH1 and greater than TH2. Furthermore, the filter control unit 60c may control the coefficient adjustment unit 34 to set the step size parameter of the coefficient adjustment unit 34 to a third value smaller than the second value (for example, μ = 0.02) if the magnitude of the bioimpedance signal x(n) is less than or equal to TH2.

[0137] In this embodiment as well, the adaptive filter 30 may employ NLMS as its adaptive algorithm. That is, the adaptive filter 30 may be an NLMS adaptive filter.

[0138] In this case, the filter control unit 60c may control the stabilization constant ε together with, or instead of, the step size parameter μ. Therefore, the filter control unit 60c may control the coefficient adjustment unit 34 to set the stabilization constant ε of the adaptive filter 30 to a predetermined first value when the potential signal d(n) satisfies predetermined conditions. On the other hand, the filter control unit 60c may control the coefficient adjustment unit 34 to set the stabilization constant ε of the adaptive filter 30 to a predetermined second value that is greater than the first value when the potential signal d(n) does not satisfy predetermined conditions.

[0139] In another embodiment, the filter control unit 60c may control the coefficient adjustment unit 34 so that the operating processing speed of the adaptive filter 30 becomes a first operating processing speed when the potential signal d(n) satisfies predetermined conditions. On the other hand, if the potential signal d(n) does not satisfy the conditions, the filter control unit 60c may control the coefficient adjustment unit 34 so that the operating processing speed of the adaptive filter 30 becomes a second operating processing speed that is slower than the first operating processing speed.

[0140] In yet another embodiment, the filter control unit 60c may control the coefficient adjustment unit 34 so that the coefficient update interval becomes a first interval when the potential signal d(n) satisfies predetermined conditions. On the other hand, the filter control unit 60c may control the coefficient adjustment unit 34 so that the coefficient update interval becomes a second interval which is longer than the first interval when the potential signal d(n) does not satisfy the conditions.

[0141] Figure 11 is an example of a flowchart showing the processing of the biological information providing device 200c according to the third embodiment. The processing of the biological information providing device 200c in this embodiment comprises steps S302 to S316.

[0142] The potential signal measurement unit 10 outputs a potential signal d(n). The bioimpedance signal measurement unit 20 outputs a bioimpedance signal x(n) (S302). The potential signal determination unit 67 of the filter control unit 60c determines whether the magnitude of the potential signal d(n) satisfies predetermined conditions (S304).

[0143] The process branches depending on whether the magnitude of the potential signal d(n) satisfies a predetermined condition (S306). If the magnitude of the potential signal d(n) exceeds a predetermined threshold, the process proceeds to S308; if the magnitude of the potential signal d(n) does not exceed a predetermined threshold, the process proceeds to S310.

[0144] The step size setting unit 64 of the filter control unit 60c sets the step size parameter μ of the adaptive filter 30 to a first value (S308). After this, the process proceeds to S312.

[0145] The step size setting unit 64 of the filter control unit 60c sets the step size parameter μ of the adaptive filter 30 to a second value, which is smaller than the first value (S310). After this, the process proceeds to S312 and merges.

[0146] The coefficient adjustment unit 34 updates the coefficient W(n) based on the bioimpedance signal x(n) and the bioinformation signal e(n) so that the error signal H(n)×x(n)-W(n)×x(n) is minimized (S312). The variable coefficient filter unit 32 multiplies the bioimpedance signal x(n) by the coefficient W(n) to generate a control signal W(n)×x(n) that is correlated with the bioimpedance signal x(n) (S314).

[0147] The subtractor 42 of the bioinformation signal output unit 40 removes the control signal W(n) × x(n) predicted from the potential signal d(n) superimposed with motion artifacts. As a result, the bioinformation signal output unit 40 outputs a bioinformation signal e(n) from which the control signal W(n) × x(n) has been removed (S316). This completes the process.

[0148] This processing allows the adaptive filter 30 to appropriately track the convergence speed of noise removal even when sudden noise is introduced during the measurement of the biological information signal e(n) in the biological information provision device 200c. Furthermore, it prevents the main signal e(n) from being attenuated along with the control signal W(n)×x(n) due to an excessive coefficient update frequency.

[0149] Figure 12 shows an example of a block diagram of the configuration of the biological information providing device 200d according to the fourth embodiment. Below, the configuration of the biological information providing device 200d will be explained, focusing on the differences from the biological information providing device 200a. Unlike the biological information providing device 200a, the biological information providing device 200d includes a filter control unit 60d. Also, similar to the biological information providing device 200b, the biological information providing device 200d includes a phase delay unit 70 between the potential signal measurement unit 10 and the biological information signal output unit 40.

[0150] The filter control unit 60d controls the adaptive filter 30 such that the convergence speed of the adaptive filter 30 is faster than when the potential signal d(n) does not satisfy the condition that the control signal W(n) × x(n) is greater than or equal to a predetermined magnitude, if the potential signal d(n) satisfies the condition. The filter control unit 60d includes an envelope processing unit 66, a potential signal determination unit 67b, and a step size setting unit 64.

[0151] The envelope processing unit 66 filters the potential signal d(n) and outputs a signal that shows the outline of the sudden noise. This makes it easier for the filter control unit 60d to extract only the noise that has suddenly occurred due to motion artifacts, etc., from the potential signal d(n). The envelope processing unit 66 of the filter control unit 60d operates in the same way as the envelope processing unit 66 of the filter control unit 60b, except that it performs envelope processing on the potential signal d(n).

[0152] The potential signal determination unit 67b determines whether or not sudden noise is present based on the shape of the potential signal d(n) that has passed through the envelope processing unit 66. In particular, the potential signal determination unit 67b determines whether or not the magnitude of the potential signal d(n) that has passed through the envelope processing unit 66 exceeds the magnitude of the signal after the potential signal d(n) has been filtered by the envelope processing unit 66. As a result, the filter control unit 60c determines that if the magnitude of the potential signal d(n) exceeds the magnitude of the reference waveform signal after the potential signal d(n) has been envelope-processed by the envelope processing unit 66, the potential signal d(n) satisfies a predetermined condition and the control signal W(n) × x(n) is large in the potential signal d(n). In this embodiment, unlike the second embodiment, the signal after the potential signal d(n) has been envelope-processed by the envelope processing unit 66 corresponds to the "reference waveform signal".

[0153] The phase delay unit 70 delays the phase of the potential signal d(n) output from the potential signal measurement unit 10 with respect to the control signal W(n) × x(n) output from the variable coefficient filter unit 32, thereby adjusting the phase. As a result, the phase delay unit 70 delays the phase of the potential signal d(n) input to the bio-information signal output unit 40 by the delay time caused by the filtering process by the envelope processing unit 66. In this embodiment of the bio-information providing device 200d, the phase delay unit 70 can be omitted if the phase delay caused by the envelope processing unit 66 is not considered to be large.

[0154] Figure 13A is an example of a graph illustrating the function of the filter control unit 60d. The signal shown in the figure is an example of a graph of the potential signal d(n) output by the bio-information signal output unit 40. The line 91 shows the potential signal d(n) input to the filter control unit 60d. Here, as with Figure 4A, the units of each axis in the graph are shown in arbitrary units.

[0155] Figure 13B is an example of a graph illustrating the function of the filter control unit 60d. The signal shown in the figure is a graph obtained by converting the power to absolute values ​​compared to the graph in Figure 13A, and is shown by the line 93. Here, the absolute value conversion may be performed by squaring the signal value. Here, in Figures 13B to 13F, not only are the units of each axis shown in arbitrary units, but graphs with different scales on the vertical axis are shown compared to Figure 13A.

[0156] In the graph of Figure 13A, both positive and negative voltage values ​​are shown. In Figure 9B, these positive and negative values ​​are squared for each time point and shown as absolute values. Therefore, when noise is present, the peak of the noise component appears large, while the contribution of other time components appears small.

[0157] Figure 13C is an example of a graph illustrating the function of the filter control unit 60d. The envelope processing unit 66 acts on the absolute value graph obtained in Figure 13B and performs envelope processing to obtain the outline of the noise. The signal showing the outline of the waveform after envelope processing is shown by the piecewise line 95. The signal shown by the piecewise line 95 corresponds to the "reference waveform signal".

[0158] Figure 13D is an example graph illustrating the function of the filter control unit 60d. The bioimpedance signal determination unit 62b determines whether the signal in Figure 13B is greater than the reference waveform signal.

[0159] Figure 13E is an example of a graph illustrating the function of the filter control unit 60d. In the figure, a bioimpedance signal x(n) similar to that in Figure 13A is shown, along with a rectangular frame 97 indicating the noise intrusion location. This frame indicating the noise intrusion location is shown at a position delayed by the processing time of the filtering process due to the envelope processing shown in Figure 13B. As an example, this delay time is approximately 30 milliseconds. The phase delay unit 70 delays the potential signal d(n) by the amount of delay due to the envelope processing including the low-pass filter, so that the signal in the noise intrusion portion fits well within the frame 97.

[0160] Figure 13F is an example of a graph illustrating the function of the filter control unit 60d. In the figure, the graph of the potential signal d(n) actually delayed by time, as shown by line 99, is shown from the graph in Figure 13D. It can be seen that the noise inclusions in the potential signal d(n) are well contained within the area indicated by the frame.

[0161] Figure 14 is an example of a flowchart showing the processing of the biological information providing device 200d according to the fourth embodiment. The processing of the biological information providing device 200d in this embodiment comprises steps S402 to S416.

[0162] The potential signal measurement unit 10 outputs a potential signal d(n). The bioimpedance signal measurement unit 20 outputs a bioimpedance signal x(n) (S402).

[0163] The potential signal determination unit 67b of the filter control unit 60 determines whether or not sudden noise has been introduced into the bioimpedance signal x(n) (S404). This step may be subdivided as in Figure 9B, but the explanation will be the same if the explanation of the bioimpedance signal x(n) is replaced with the potential signal d(n), so the explanation of the bio-information providing device 200d will be omitted.

[0164] The process branches depending on whether or not it is determined that sudden noise has been introduced into the potential signal d(n) (S406). If it is determined that sudden noise has been introduced into the potential signal d(n), the process proceeds to S408; if it is not determined that sudden noise has been introduced into the potential signal d(n), the process proceeds to S410.

[0165] The step size setting unit 64 of the filter control unit 60d sets the step size parameter to a first value (S208). After this, the process proceeds to S412.

[0166] The step size setting unit 64 of the filter control unit 60b sets the step size parameter to a second value, which is smaller than the first value (S410). After this, the process proceeds to S412 and merges.

[0167] The coefficient adjustment unit 34 updates the coefficient W(n) based on the bioimpedance signal x(n) and the bioinformation signal e(n) so that the error signal H(n)×x(n)-W(n)×x(n) is minimized (S412). The variable coefficient filter unit 32 multiplies the bioimpedance signal x(n) by the coefficient W(n) to generate a control signal W(n)×x(n) that is correlated with the bioimpedance signal x(n) (S414).

[0168] The subtractor 42 of the bioinformation signal output unit 40 removes the control signal W(n) × x(n) predicted from the potential signal d(n) superimposed with motion artifacts. As a result, the bioinformation signal output unit 40 outputs a bioinformation signal e(n) from which the control signal W(n) × x(n) has been removed (S416).

[0169] This processing allows the adaptive filter 30 to appropriately track the convergence speed of noise removal even when sudden noise is introduced during the measurement of the biological information signal e(n) in the biological information provision device 200d. Furthermore, it prevents the main signal e(n) from being attenuated along with the control signal W(n)×x(n) due to an excessive coefficient update frequency.

[0170] Figure 15 shows an example of a block diagram of the configuration of the biological information providing device 200e according to the fifth embodiment. Below, the configuration of the biological information providing device 200e will be explained, focusing on the differences from the biological information providing device 200a. Unlike the biological information providing device 200a, the biological information providing device 200e includes a filter control unit 60e.

[0171] The filter control unit 60e includes a correlation determination unit 69 and a step size setting unit 64. Furthermore, unlike the filter control units 60a to 60d, the filter control unit 60e is connected to both the output of the potential signal measurement unit 10 and the bioimpedance signal measurement unit 20.

[0172] The correlation determination unit 69 compares the potential signal d(n) and the bioimpedance signal x(n). The correlation determination unit 69 may, for example, perform a windowed Fourier transform (FFT, etc.) on the potential signal d(n) and the bioimpedance signal x(n) within a predetermined time window and compare their frequency components. If correlated noise appears in the potential signal d(n) and the bioimpedance signal x(n), and the similarity of the noise exceeds a predetermined threshold, the correlation determination unit 69 may determine that the magnitude of the control signal W(n) × x(n) indicating the noise component contained in the potential signal d(n) satisfies a predetermined condition. As a result, the filter control unit 60e may determine that either the potential signal d(n) or the bioimpedance signal x(n) satisfies a predetermined condition if the correlation between the potential signal d(n) and the bioimpedance signal x(n) exceeds a predetermined threshold.

[0173] Figure 16 is an example of a flowchart showing the processing of the biological information providing device 200e according to the fifth embodiment. The processing of the biological information providing device 200e according to the fifth embodiment comprises steps S502 to S516.

[0174] The potential signal measurement unit 10 outputs a potential signal d(n). The bioimpedance signal measurement unit 20 outputs a bioimpedance signal x(n) (S502).

[0175] The correlation determination unit 69 of the filter control unit 60e compares the potential signal d(n) and the bioimpedance signal x(n) to determine whether or not noise correlated with the noise appearing in the bioimpedance signal x(n) has been mixed into the potential signal d(n) (S504).

[0176] The process branches depending on whether or not it is determined that noise correlated with the potential signal d(n) has been introduced (S506). If it is determined that noise correlated with the potential signal d(n) has been introduced, the process proceeds to S508; if it is not determined that noise correlated with the potential signal d(n) has been introduced, the process proceeds to S510.

[0177] If it is determined that noise correlated with the potential signal d(n) has been introduced, the step size setting unit 64 of the filter control unit 60e sets the step size parameter to the first value (S508). After this, the process proceeds to S512.

[0178] If it is not determined that noise correlated with the potential signal d(n) has been introduced, the step size setting unit 64 of the filter control unit 60e sets the step size parameter to a second value, which is smaller than the first value (S510). After this, the process proceeds to S512 and merges.

[0179] The coefficient adjustment unit 34 updates the coefficient W(n) based on the bioimpedance signal x(n) and the bioinformation signal e(n) so that the error signal H(n)×x(n)-W(n)×x(n) is minimized (S512). The variable coefficient filter unit 32 multiplies the bioimpedance signal x(n) by the coefficient W(n) to generate a control signal W(n)×x(n) that is correlated with the bioimpedance signal x(n) (S514).

[0180] The subtractor 42 of the bioinformation signal output unit 40 removes the control signal W(n) × x(n) predicted from the potential signal d(n) superimposed with motion artifacts. As a result, the bioinformation signal output unit 40 outputs a bioinformation signal e(n) from which the control signal W(n) × x(n) has been removed (S516). The process is now complete.

[0181] This processing allows the adaptive filter 30 to appropriately track the convergence speed of noise removal even when sudden noise is introduced during the measurement of the biological information signal e(n) in the biological information provision device 200e. Furthermore, it prevents the main signal e(n) from being attenuated along with the control signal W(n)×x(n) due to an excessive coefficient update frequency.

[0182] Figure 17 shows an example of a block diagram of the configuration of the biological information providing device 200f according to the sixth embodiment. The differences from the biological information providing device 200a according to the first embodiment will be explained. The biological information providing device 200f includes a filter control unit 60f.

[0183] The filter control unit 60f determines whether the magnitude of the biological information signal e(n) satisfies a predetermined condition based on the control signal output by the variable coefficient filter unit 32 and the biological information signal e(n) output by the biological information signal output unit 40, and controls the convergence speed of the adaptive filter 30 based on the determination. The filter control unit 60f includes a control signal determination unit 71 and a step size setting unit 64.

[0184] The control signal determination unit 71 determines whether the magnitude of the biological information signal e(n) satisfies predetermined conditions based on the control signal W(n) × x(n) and the biological information signal e(n). In this case, the determination may be made based on the correlation between the control signal W(n) × x(n) and the biological information signal e(n) to determine whether or not noise correlated with MA is mixed into the biological information signal e(n). The control signal determination unit 71 is connected to the variable coefficient filter unit 32 and the biological information signal output unit 40.

[0185] The step size setting unit 64 sets the value of the step size parameter μ via the coefficient adjustment unit 34 based on the determination result of the control signal determination unit 71. As a result, the filter control unit 60f controls the convergence speed of the adaptive filter based on the control signal W(n) × x(n) and the biological information signal e(n).

[0186] Through this processing, the convergence speed of the adaptive filter 30 for removing noise in the measurement of the biological information signal e(n) can be appropriately tracked in the biological information providing device 200f. Furthermore, it is possible to prevent the main signal e(n) from being attenuated along with the control signal W(n)×x(n) due to an excessive coefficient update frequency.

[0187] Figure 18 shows an example of a block diagram of the configuration of the biological information providing device 200g according to the seventh embodiment. The biological information providing device 200g includes a potential signal measurement unit 10, a phase delay unit 70, an envelope processing unit 72, a potential signal determination unit 74, a noise reduction unit 76a, and a selection unit 50 connected to a pair of electrodes 150 that come into contact with the living body.

[0188] The potential signal measurement unit 10 outputs a potential signal d(n) based on the measurement results. Sudden noise may be mixed into the potential signal d(n).

[0189] The envelope processing unit 72 performs envelope processing on the potential signal d(n) in a manner similar to that included in the filter control unit 60d of the biological information providing device 200d. The envelope processing unit 72 outputs a reference waveform signal through envelope processing including absolute value conversion and low-pass filtering. This makes it easier for the potential signal determination unit 74 to detect the inclusion of sudden noise based on the reference waveform signal. The envelope processing unit 72 may have a modulator, mixer and / or arithmetic unit, etc., similar to the envelope processing unit 66.

[0190] The potential signal determination unit 74 determines whether the magnitude of the potential signal d(n) exceeds the magnitude of the reference waveform signal. The potential signal determination unit 74 may output signals with different potential levels depending on the determination result. The signals output by the potential signal determination unit 74 may be pulse signals, and may be configured to continuously output a signal of a predetermined level for a predetermined period of time. If the magnitude of the potential signal d(n) exceeds the magnitude of the reference waveform signal, the potential signal determination unit 74 determines that the potential signal d(n) satisfies predetermined conditions. As an example, the potential signal determination unit 74 controls which signal the selection unit 50 will use as the biological information signal output by the biological information providing device 200g based on the different levels of signals output according to the determination result.

[0191] The phase delay unit 70 delays the phase of the potential signal d(n) input to the selection unit 50 and the noise reduction unit 76a by the delay time due to envelope processing by the envelope processing unit 72. As a result, when the potential signal determination unit 74 determines that sudden noise has been introduced into d(n), the phase delay unit 70 can input a signal corresponding to the potential signal from a point in time prior to that point to the selection unit 50 and the noise reduction unit 76a. Here, the phase delay unit 70 may delay the signal by an additional delay time added to the delay time due to envelope processing, so that when the envelope processing unit 72 detects sudden noise, the noise processing performed by the noise reduction unit 76a is carried out according to the potential signal from a point in time sufficiently prior to the time of sudden noise detection.

[0192] The noise reduction unit 76a outputs a signal corresponding to the potential signal d(n) before the potential signal d(n) was determined to satisfy a predetermined condition, when the potential signal determination unit 74 determines that the potential signal d(n) satisfies a predetermined condition.

[0193] As an example, the noise reduction unit 76a may include a flip-flop. In this case, the noise reduction unit 76a latches the potential signal d(n) before it is determined that the potential signal d(n) satisfies the condition using the flip-flop, and outputs a signal corresponding to that potential for a predetermined time. This keeps the potential constant during the intrusion of sudden noise, and allows for the simple removal of sudden noise from the potential signal d(n). As another example, the noise reduction unit 76a may include a timer and a register. In this case, the potential signal d(n) before it is determined that the condition satisfies is stored in the register, and for a predetermined time measured by the timer, the signal of the potential level stored in the register can be used as the noise reduction signal. Thus, the noise reduction unit 76a may have various configurations that allow it to output a signal corresponding to the sudden noise before it is intruded during the occurrence of sudden noise.

[0194] The selection unit 50 selects to output the signal output by the noise reduction unit 76a as a bioinformation signal if the potential signal d(n) satisfies predetermined conditions. On the other hand, the selection unit 50 selects to output the potential signal d(n) as a bioinformation signal if the potential signal d(n) does not satisfy predetermined conditions. As another example, the selection unit 50 may select to replace the electronic signal d(n) with a blank signal for a predetermined period of time if the potential signal d(n) satisfies predetermined conditions.

[0195] As described above, the biological information providing device 200g can detect whether or not sudden noise is mixed into the potential signal d(n), and if it is determined that sudden noise is mixed into the potential signal d(n), it can easily remove the noise. The configuration of the stage after the potential signal measurement unit 10 in the biological information providing device 200g can be included in the configuration of the filter control units 60a to 60f of the biological information providing devices 200a to 200f.

[0196] In this case, if the filter control unit detects a sudden noise in the potential signal d(n) or bioimpedance signal x(n) that is of a predetermined magnitude, each filter control unit 60 may control the bio-information providing device 200g. On the other hand, if no sudden noise is detected, each filter control unit 60 may control the step size parameter μ as described in the bio-information providing devices 200a to 200f.

[0197] In other words, the filter control unit 60, which includes the configuration of the biological information providing device 200g, may output a signal corresponding to the potential signal d(n) before it was determined that the biological impedance signal x(n) or potential signal d(n) satisfies the conditions as a biological information signal e(n) if the magnitude of the biological impedance signal x(n) or the magnitude of the potential signal d(n) exceeds a predetermined first threshold, and may output the potential signal d(n) as a biological information signal e(n) if the magnitude of the biological impedance signal x(n) or the magnitude of the potential signal d(n) does not exceed a predetermined first threshold (e.g., 1mV). Furthermore, if the magnitude of the biological impedance signal x(n) or the magnitude of the potential signal d(n) is less than or equal to the first threshold and exceeds a second threshold lower than the first threshold (e.g., 0.25mV), the filter control unit 60 may control the coefficient adjustment unit 34 to set the step size parameter μ of the coefficient adjustment unit to a first value (e.g., μ=0.5). The filter control unit 60 may control the coefficient adjustment unit 34 to set the step size parameter μ (e.g., μ = 0.2) of the coefficient adjustment unit 34 to a second value smaller than the first value, if the magnitude of the bioimpedance signal x(n) or the magnitude of the potential signal d(n) is less than or equal to the second threshold.

[0198] Thus, the biological information providing device 200 may perform noise reduction processing on the potential signal d(n) when sudden noise is introduced, according to multiple thresholds. On the other hand, when no sudden noise is introduced, noise reduction processing may be performed by changing the convergence speed of the adaptive filter 30.

[0199] Figure 19 is a flowchart showing an example of a process for removing sudden noise performed in the 200g biological information providing device. The process comprises steps S602 to S612.

[0200] The potential signal measurement unit 10 outputs a potential signal d(n) (S602). The potential signal determination unit 74 determines whether or not sudden noise is mixed into the potential signal d(n) (S604). The determination made by the potential signal determination unit 74 may be made according to a reference waveform signal obtained by the envelope processing unit 72 performing envelope processing on the potential signal d(n).

[0201] The process branches depending on whether the potential signal determination unit 74 determines that sudden noise is mixed into the potential signal (S606). If the potential signal determination unit 74 determines that sudden noise is mixed into the potential signal, the process proceeds to S608. On the other hand, if the potential signal determination unit 74 does not determine that sudden noise is mixed into the potential signal, the process proceeds to S612.

[0202] If the potential signal determination unit 74 determines that sudden noise is mixed into the potential signal d(n), the potential signal determination unit 74 outputs a signal that controls the selection unit 50 and the noise removal unit 76a as if sudden noise is mixed into the potential signal. As a result, the noise removal unit 76a outputs a signal corresponding to the potential signal d(n) before it is determined that the potential signal d(n) satisfies the conditions, that is, before the sudden noise was mixed in (S608).

[0203] If the potential signal d(n) satisfies predetermined conditions, the selection unit 50 outputs a signal corresponding to the potential signal d(n) before the sudden noise intrusion as a bioinformation signal (S610). After this, the process ends.

[0204] If the potential signal d(n) does not meet predetermined conditions, the selection unit 50 outputs the potential signal d(n) output by the potential signal measurement unit 10 as a biological information signal (S612). After this, the processing merges with the processing in step S610 and ends.

[0205] Figure 20 shows an example of a block diagram of the configuration of the biological information providing device 200h according to the eighth embodiment. The biological information providing device 200h includes a potential signal measurement unit 10, a phase delay unit 70, an envelope processing unit 72, a potential signal determination unit 74, a pulse stretcher 78, a noise reduction unit 76b, a phase delay unit 79, and a selection unit 50, all connected to a pair of electrodes 150 that come into contact with the living body.

[0206] The following describes the configuration of the biological information provision device 200h, focusing on the differences from the biological information provision device 200g. The biological information provision device 200h differs in the configuration of the stage after the potential signal determination unit 74.

[0207] The pulse stretcher 78 outputs a signal that is a stretched version of the pulse output by the potential signal determination unit 74. This allows the pulse stretcher 78 to cause the selection unit 50 to select the noise-removed signal from the noise removal unit 76b for a sufficiently long period before and after the period corresponding to the period in which sudden noise occurred in the potential signal d(n).

[0208] The phase delay unit 70 delays the phase of the potential signal d(n) input to the selection unit 50 and the noise reduction unit 76a by the delay time due to the envelope processing of the envelope processing unit 72. The phase delay unit 70 corresponds to the "first phase delay unit".

[0209] The noise reduction unit 76b outputs a signal corresponding to the potential signal d(n) before the potential signal d(n) was determined to satisfy a predetermined condition, when the potential signal determination unit 74 determines that the potential signal d(n) satisfies a predetermined condition. Here, if the noise is sudden, the frequency of the sudden noise will be high. Therefore, the noise reduction unit 76b can remove the noise mixed into the potential signal d(n) by performing low-pass filtering. Specifically, the noise reduction unit 76b performs low-pass filtering on the output signal of the phase delay unit 70 and outputs a signal corresponding to the potential signal d(n) before the potential signal d(n) was determined to satisfy a predetermined condition.

[0210] The phase delay unit 79 delays the phase of the potential signal d(n) input to the selection unit 50 by the delay time due to the low-pass filtering of the noise reduction unit 76b. The phase delay unit 79 corresponds to the "second phase delay unit".

[0211] The selection unit 50 selects to output the signal output by the noise reduction unit 76a as the bioinformation signal if the potential signal d(n) satisfies predetermined conditions. On the other hand, the selection unit 50 selects to output the potential signal d(n) as the bioinformation signal if the potential signal d(n) does not satisfy predetermined conditions. As already mentioned, in this embodiment, the period during which the potential signal d(n) satisfies predetermined conditions can be set to a longer period before and after by the pulse stretcher 78. Therefore, the selection unit 50 in this embodiment can reliably remove noise for the period before and after in which sudden noise may be mixed in. The selection unit 50 may also select a blank signal as the biopotential signal if the potential signal d(n) satisfies predetermined conditions.

[0212] As described above, the biological information provider 200h detects whether or not sudden noise is mixed into the potential signal d(n). If the biological information provider 200h determines that sudden noise is mixed into the potential signal d(n), it can remove the noise using a different configuration than that of the biological information provider 200g.

[0213] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications or improvements can be made to the above embodiments. It will be clear from the claims that such modified or improved forms may also be included in the technical scope of the present invention.

[0214] It should be noted that the execution order of operations, procedures, steps, and stages in the apparatus, systems, programs, and methods shown in the claims, specifications, and drawings is not explicitly stated as "before," "prior to," etc., and that these can be implemented in any order unless the output of a previous process is used in a later process. Even if the operation flow in the claims, specifications, and drawings is described using phrases such as "first," "next," etc. for convenience, it does not mean that it is essential to perform the operations in that order. [Explanation of Symbols]

[0215] 10 Potential signal measurement unit 12. Noise-inducing section 14 Adder 20 Bioimpedance signal measurement unit 22 AC signal output section 30 Adaptive Filters 32 Variable coefficient filter section 34 Coefficient adjustment section 40. Biological Information Signal Output Unit 42 Subtractors 50 Selection Section 60 Filter control unit 62 Bioimpedance signal determination unit 64 Step Size Setting Section 66 Envelope Processing Unit 67 Potential signal determination unit 69 Correlation determination unit 70,79 Phase delay section 72 Envelope Processing Unit 74 Potential signal determination unit 76 Noise Reduction Section 78 Pulse Stretcher 82, 84, 91, 93, 95, 99 Linear graph 97th slot 100,200 Biometric Information Provision Devices 150 electrodes 500 living organisms 502 Epidermal layer 504 Dermis and subcutaneous layer 506 Muscle layer 508 muscle fibers 510 nerves

Claims

1. A potential signal measuring unit that measures potential signals through a pair of electrodes that come into contact with the living body, A bioimpedance signal measuring unit that measures the bioimpedance generated between the pair of electrodes and outputs a bioimpedance signal corresponding to the bioimpedance, An adaptive filter having: a variable coefficient filter unit that generates a control signal based on the bioimpedance signal and coefficients; a bioinformation signal output unit that calculates a bioinformation signal based on the potential signal and the control signal; and a coefficient adjustment unit that adjusts the coefficients so that the signals correlated with the bioimpedance signal included in the bioinformation signal become smaller. A filter control unit controls the coefficient adjustment unit so that the convergence speed of the adaptive filter is faster when the potential signal or the bioimpedance signal satisfies predetermined conditions than when the potential signal or the bioimpedance signal does not satisfy the conditions. A biological information providing device equipped with the following features.

2. The bio-information providing device according to claim 1, wherein the filter control unit determines that the bio-impedance signal satisfies the condition when the magnitude of the bio-impedance signal exceeds a predetermined first threshold.

3. The bio-information providing device according to claim 1, wherein the filter control unit determines that the bio-impedance signal satisfies the condition when the magnitude of the bio-impedance signal exceeds the magnitude of a reference waveform signal obtained by envelope processing of the bio-impedance signal.

4. The biological information providing device according to claim 3, further comprising a phase delay unit that delays the phase of the potential signal input to the biological information signal output unit by the delay time due to the envelope processing.

5. The bio-information providing device according to claim 1, wherein the filter control unit determines that the bio-impedance signal satisfies the condition when the magnitude of the potential signal exceeds a predetermined first threshold.

6. The bio-information providing device according to claim 1, wherein the filter control unit determines that the potential signal satisfies the condition when the magnitude of the potential signal exceeds the magnitude of the reference waveform signal obtained by envelope processing of the potential signal.

7. The biological information providing device according to claim 6, further comprising a phase delay unit that delays the phase of the potential signal input to the biological information signal output unit by the delay time due to the envelope processing.

8. The bio-information providing device according to claim 1, wherein the filter control unit determines that the potential signal or the bio-impedance signal satisfies the conditions when the correlation between the potential signal and the bio-impedance signal exceeds a predetermined threshold.

9. The filter control unit controls the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a first value when the bioimpedance signal or the potential signal satisfies the conditions. The biological information providing device according to any one of claims 1 to 8, wherein the filter control unit controls the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a second value smaller than the first value if the bioimpedance signal or the potential signal does not satisfy the conditions.

10. The filter control unit outputs a signal corresponding to the potential signal before it is determined that the bioimpedance signal or the potential signal satisfies the conditions as a bioinformation signal if the magnitude of the bioimpedance signal or the magnitude of the potential signal exceeds a predetermined first threshold, and outputs the potential signal as a bioinformation signal if the magnitude of the bioimpedance signal or the magnitude of the potential signal does not exceed a predetermined first threshold. The filter control unit controls the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a first value if the magnitude of the bioimpedance signal or the magnitude of the potential signal is less than or equal to the first threshold and exceeds a second threshold that is lower than the first threshold. The biological information providing device according to any one of claims 1 to 8, wherein the filter control unit controls the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a second value smaller than the first value when the magnitude of the bioimpedance signal or the magnitude of the potential signal is less than or equal to the second threshold.

11. The filter control unit controls the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a first value when the magnitude of the bioimpedance signal or the magnitude of the potential signal exceeds a predetermined first threshold. The filter control unit controls the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a second value smaller than the first value if the magnitude of the bioimpedance signal or the magnitude of the potential signal is less than or equal to the first threshold and exceeds a second threshold lower than the first threshold. The biological information providing device according to any one of claims 1 to 8, wherein the filter control unit controls the coefficient adjustment unit to set the step size parameter of the coefficient adjustment unit to a third value smaller than the second value when the magnitude of the bioimpedance signal or the magnitude of the potential signal is less than or equal to the second threshold.

12. When the coefficient adjustment unit adjusts the coefficients using the NLMS adaptive algorithm, The filter control unit controls the coefficient adjustment unit to set the stabilization constant of the coefficient adjustment unit to a first value when the bioimpedance signal or the potential signal satisfies the conditions. The biological information providing device according to any one of claims 1 to 8, wherein the filter control unit controls the coefficient adjustment unit to set the stabilization constant of the adaptive filter to a second value greater than the first value when the bioimpedance signal or the potential signal does not satisfy the conditions.

13. The filter control unit controls the coefficient adjustment unit so that the operating processing speed of the adaptive filter becomes the first operating processing speed when the bioimpedance signal or the potential signal satisfies the conditions. The biological information providing device according to any one of claims 1 to 8, wherein the filter control unit controls the coefficient adjustment unit so that the operating processing speed of the adaptive filter becomes a second operating processing speed that is slower than the first operating processing speed when the bioimpedance signal or the potential signal does not satisfy the conditions.

14. The filter control unit controls the coefficient adjustment unit so that the coefficient update interval becomes the first interval when the bioimpedance signal or the potential signal satisfies the conditions. The biological information providing device according to any one of claims 1 to 8, wherein the filter control unit controls the coefficient adjustment unit so that the coefficient update interval becomes a second interval that is longer than the first interval when the bioimpedance signal or the potential signal does not satisfy the above conditions.

15. A potential signal measuring unit that measures potential signals through a pair of electrodes that come into contact with the living body, A selection unit selects to output a signal corresponding to the potential signal before it was determined that the potential signal met the predetermined conditions as a biological information signal if the potential signal met predetermined conditions, and to output the potential signal as a biological information signal if the potential signal did not meet predetermined conditions. A biological information providing device equipped with the following features.

16. An envelope processing unit that performs envelope processing on the aforementioned potential signal to output a reference waveform signal, A potential signal determination unit determines that if the magnitude of the potential signal exceeds the magnitude of the reference waveform signal, the potential signal satisfies predetermined conditions. A phase delay unit that delays the phase of the potential signal input to the selection unit by the delay time due to the envelope processing, When the aforementioned potential signal satisfies predetermined conditions, a noise reduction unit outputs a signal corresponding to the potential signal before it was determined that the potential signal satisfies the conditions, The biological information providing device according to claim 15, further comprising:

17. An envelope processing unit that performs envelope processing on the aforementioned potential signal to output a reference waveform signal, A potential signal determination unit determines that if the magnitude of the potential signal exceeds the magnitude of the reference waveform signal, the potential signal satisfies predetermined conditions. A first phase delay unit delays the phase of the potential signal input to the selection unit by the delay time due to the envelope processing, If the potential signal satisfies predetermined conditions, a noise reduction unit performs low-pass filtering on the output signal of the first phase delay unit and outputs a signal corresponding to the potential signal before it was determined that the potential signal satisfies the conditions. A second phase delay unit delays the phase of the potential signal input to the selection unit by the delay time due to the low-pass filtering, The biological information providing device according to claim 15, further comprising:

Images