Apparatus and method for low-power electrogram devices for human machine interfaces

Abstract

In an embodiment of the invention, a human-machine interface system includes a wirelessly powered electrogram sensor including a set of signal electrodes, an analog front end (AFE) configured to receive a biopotential signal from the set of signal electrodes and provide an amplified signal, an analog to digital converter (ADC) configured to receive the amplified signal and digitize the amplified signal, and a transmitter configured to receive the digitized signal, modulate the digitized signal into an output signal and transmit the output signal.

Description

APPARATUS AND METHOD FOR LOW-POWER ELECTROGRAM DEVICES FORHUMAN MACHINE INTERFACESCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The current application claims the benefit of and priority under 35 U.S.C. §119(e) to U.S. Provisional Patent Application No. 63 / 654,682, entitled “Apparatus and Method for Low- Power EMG and EEG Devices for HMI”, filed May 31, 2024 the disclosure of which is incorporated herein by reference in its entirety for all purposes.FIELD OF THE INVENTION

[0002] The present invention relates generally to biopotential sensors and more specifically to wirelessly powered electrogram sensors that can be configured for different types of electrogram.BACKGROUND OF THE INVENTION

[0003] In recent years, there has been a growing need for recording various biopotential signals, such as electrocardiograms (ECGs), electromyograms (EMGs), and electrooculograms (EOGs). Concurrent recording of EOG and EMG is particularly useful in soft robotics to aid individuals in performing basic movements with ease. From a point-of care application perspective, a low-power, multi-channel device that can be wirelessly configured to record and transmit these biosignals can be useful. Designing the analog front-end (AFE) for a neural sensing chip presents several challenges, including accommodating a wide range of biopotential signals (pV to mV) necessitating programmable gain, bandwidth (BW), and sampling.

[0004] Addition applications include emerging human-machine interface (HMI) devices such as augmented reality and virtual reality (AR / VR) glasses that need to capture a human’s intention and use it to control objects on a computer screen. Some existing devices capture human intention using the following methods:

[0005] 1) Capturing real-time movement of the eyeball using a camera embedded in an AR / VR headset and running eye tracking algorithms to derive the intention of the person. This typically requires high power consumption due to the use of a camera, large number of pixels, and processing algorithms.

[0006] 2) Using a hat equipped with an array of electroencephalogram (EEG) sensors to decode brain activities. This approach is also bulky and requires high power consumption.

[0007] 3) Wearing an electromyography (EMG) band with multiple sensors to decode human intention by measuring an EMG signal on the wrist. This is also shown to be power hungry and bulky due to the high-power consumption of EMG devices.SUMMARY OF THE INVENTION

[0008] In an embodiment of the invention, a human-machine interface system includes a wirelessly powered electrogram sensor including a set of signal electrodes, an analog front end (AFE) configured to receive a biopotential signal from the set of signal electrodes and provide an amplified signal, an analog to digital converter (ADC) configured to receive the amplified signal and digitize the amplified signal, and a transmitter configured to receive the digitized signal, modulate the digitized signal into an output signal and transmit the output signal.

[0009] Additional embodiments of the invention include an external computing device configured to receive the transmitted output signal and determine a user intention from the output signal.

[0010] In further embodiments of the invention, the analog front-end includes a programmable gain amplifier and a tunable bandwidth filter controlled by a wireless command signal from the external computing device.

[0011] In some embodiments of the invention, the analog front end comprises at least one filter to reduce noise.

[0012] In more embodiments of the invention, the analog to digital converter is a SAR ADC.

[0013] In still more embodiments of the invention, the wirelessly powered electrogram sensor also includes a plurality of pairs of AFEs and ADCs, each pair of AFE and ADC associated with a pair of signal electrodes of the set of signal electrodes, a serializer configured to combine the outputs of the plurality of analog to digital converters into the digitized signal for the transmitter.

[0014] In additional embodiments of the invention, pairs of signal electrodes are configured to each capture a different type of biopotential signal.

[0015] In further embodiments of the invention, the different type of biopotential signals include at least electromyogram (EMG) and electrooculogram (EOG).

[0016] In some embodiments of the invention, the wirelessly powered electrogram sensor is mounted near a user’s eye and configured to detect oculomotor EMG signals indicative of gaze direction or eye movement.

[0017] In more embodiments of the invention, the different type of biopotential signals include at least electromyogram (EMG) and electroencephalogram (EEG).

[0018] In still more embodiments of the invention, the wirelessly powered electrogram sensor is mounted on a hand and configured to detect EMG signals corresponding to finger gestures or hand movement for use in gesture-based control of an external computing device.

[0019] In additional embodiments of the invention, the different type of biopotential signals include at least electromyogram (EMG) and electrocardiogram (ECG).

[0020] In further embodiments of the invention, the wireless transmitter uses ultra-wideband (UWB) modulation to transmit the digital signals with reduced power consumption.

[0021] In some embodiments of the invention, a method for detecting finger movement includes emitting radar waves from a radar sensor attached to a finger over a period of time, receiving reflected radar waves at the radar sensor over the period of time, determining, based upon the received reflected radar waves over the period of time, at least one of: distance, direction, and radial velocity of at least one object.

[0022] In more embodiments of the invention, the method also includes determining that the at least one object is another finger.

[0023] In still more embodiments of the invention, a wearable HMI system includes a ringshaped device configured to be worn on a finger and comprising a radar sensor, wherein the radar sensor is configured to detect motion of adjacent fingers using reflected radar waves and to transmit motion data to an external computing device for gesture recognition.

[0024] In additional embodiments of the invention, the radar sensor operates in the millimeterwave frequency range and detects sub-millimeter movements with a resolution of 100 micrometers or better.

[0025] In further embodiments of the invention, a method for user input in an augmented or virtual reality environment includes detecting a user's eye muscle signals using EMG electrodes placed near the eye, determining gaze direction or blink patterns from the EMG signals, and using the determined eye activity as a control input to select or manipulate objects on a head mounted display.

[0026] In additional embodiments of the invention, a multi-modal HMI system includes a plurality of wirelessly powered electrogram sensors configured to capture signals of different biopotential signal types including EMG and EOG, amplify, digitize, and transmit the capturedsignals to a central processing system, a radar-based motion detection unit configured to detect fine motor movement and transmit detected movement as radar information, and a central processing system configured to integrate the captured biopotential signals and radar information into integrated user intent information and infer user intent based on a combination of neural and physical cues derived from the integrated user intent information.DESCRIPTION OF THE FIGURES

[0027] Fig. 1 illustrates a circuit diagram of an electrogram sensor in accordance with several embodiments of the invention.

[0028] Fig. 2 illustrates an analog front end (AFE) according to several embodiments of the invention.

[0029] Fig. 3A illustrates a clock recovery section according to some embodiments of the invention.

[0030] Fig. 3B illustrates a clock recovery schematic and its measured operational range within various carrier frequencies and modulation indexes according to some embodiments.

[0031] Fig. 3C illustrates a schematic of a transmitter and the measured phase noise in free- running mode in accordance with several embodiments.

[0032] Fig. 3D illustrates measured phase noise of a transmitter in accordance with several embodiments of the invention.

[0033] Fig. 4 illustrates a wirelessly powered electrogram sensor system having multiple channels in accordance with several embodiments of the invention.

[0034] Fig. 5 illustrates potential placements of electrodes in accordance with several embodiments of the invention.

[0035] Fig. 6A illustrates a power harvesting section of a sensor system in accordance with an embodiment of the invention.

[0036] Fig. 6B illustrates a data and clock recovery section of a sensor system in accordance with an embodiment of the invention.

[0037] Fig. 6C shows a chart with minimum required modulation index (m) in some embodiments of the invention.

[0038] Fig. 6D shows a chart with a recovered clock’s period jitter for clock recovery in some embodiments of the invention.

[0039] Fig. 7A illustrates an AFE with a two-stage capacitive-coupled amplifier in accordance with several embodiments of the invention.

[0040] Fig. 7B illustrates programmable gain and a low comer bandwidth in accordance with several embodiments of the invention.

[0041] Fig. 8A illustrates a UWB Class-E power oscillator and an SMDL as an antenna for transmitting in accordance with several embodiments of the invention.

[0042] Fig. 9 illustrates an eye tracking human machine interface (HMI) in accordance with several embodiments of the invention.

[0043] Fig. 10 illustrates a hand gesture HMI in accordance with several embodiments of the invention.

[0044] Fig. 11 illustrates a process for obtaining a biopotential signal and wirelessly transmitting data indicative of the biopotential signal in accordance with several embodiments of the invention.DETAILED DISCLOSURE OF THE INVENTION

[0045] Apparatuses and methods for low-power electrogram, e.g., EMG and EEG, devices for human-machine interfaces (HMI) are disclosed. Many embodiments of the invention implement low-power and optionally wirelessly powered electrogram based sensors that can wirelessly transmit their sensed data.

[0046] Embodiments may utilize electrogram sensors to capture neural signals from various locations in human body. Electrogram sensors can include, but are not limited to, action potential (AP), local field potential (LFP), electromyogram (EMG), electrocorti cogram (ECOG), electrocardiogram (ECG), electroencephalogram (EEG), electromyogram (EMG), and / or electrooculogram (EOG).

[0047] Neural signals include AP and LFP. AP signals have a bandwidth of up to 10kHz and offer high spatial resolution (down to the level of single neurons) but are susceptible to instability and deterioration over time. On the other hand, LFP signals occupy a narrower bandwidth (up to 500Hz) and provide stable but lower-resolution complementary data, which is valuable for understanding the brain's network-level dynamics. LFP and AP signals may be recorded from the same electrode implanted in the brain.

[0048] In various embodiments of the invention, an electrogram sensor may be implemented as a neural sensing implant or within a wearable device on a face or hand.

[0049] Several embodiments of the invention involve placing EMG sensors around the eye to measure the movement of eyeball using the EMG signals produced around the eye.

[0050] Additional embodiments include placing EMG sensors around the fingers (e.g., as a ring) to capture muscle movements of the fingers using the EMG signals generated at the fingers.

[0051] Further embodiments include placing EMG sensors at various locations on the hand or wrist.

[0052] EMG sensors in accordance with embodiments of the invention can amplify the EMG signals, digitize them, and wirelessly transmit them to an AR / VR glass or an external unit for further processing. The wireless channel may be Bluetooth or Wi-Fi. If the signal is processed in an external unit, the processed data can be sent to an AR / VR headset and different objects on the screen can be moved based on the signal measured by the EMG sensor.

[0053] Electrogram sensors in accordance with embodiments of the invention may be wirelessly powered. The source of the wireless power can be an external unit or on an AR / VR headset itself. In many embodiments, clock recovery by the sensor allows external controller to set or control the sampling rate of the sensor using the embedded clock signal.

[0054] Electrogram sensors in accordance with embodiments of the invention may be lightweight and cheaper to produce by not having an onboard power source. The simpler circuitry and economic costs can enable such sensors to be disposable.Electrogram Sensors

[0055] In many embodiments of the invention, a wirelessly powered electrogram sensor includes a power harvesting section for recovering energy from a received wireless power transfer (WPT) signal, a clock recovery section (which can include an envelope detector, a comparator, a Schmitt trigger) for recovering a clock signal from the received wireless signal, an analog front end (AFE) that interfaces with signal electrodes for connecting to the human body, and a transmitter section for transmitting data obtained by the AFE. In some embodiments, the received signal is in the 40 MHz band (e.g., 40.68MHz) and the signal transmitted back is a different rate (e.g., 433 MHz), although different frequencies may be used as appropriate to a particular application. A circuit diagram of an electrogram sensor in accordance with several embodiments of the invention is illustrated in Fig. 1.

[0056] A power harvesting section may include a 4-stage fully differential rectifier that can harvest energy from a received signal and generate, for example, a nominal 1.3V. The energy can be provided to a Low Dropout Regulator (LDO) that then provides, for example, a regulated 1 ,2V supply for the rest of the system using the harvested voltage.

[0057] Many embodiments of the invention utilize an analog front end (AFE) that includes a two-stage amplifier and successive approximation register (SAR) ADC for each set of signal electrodes. An AFE according to several embodiments is illustrated in Fig. 2. The AFE(s) may have gain and / or bandwidth tuning for different biological signals (using a 20-bit passcode for example) and may be controllable by external controller. In some embodiments, an AFE can provide a gain that is variable from 35dB to 55dB to the signal from the electrodes.

[0058] In some embodiments of the invention, an AFE includes one or more filters that can reduce noise. Noise filters can include, but are not limited to, a high pass filter to remove DC noise / low-frequency noise and a notch filter to remove 60Hz signal.

[0059] A two-stage design for an AFE can have higher input impedance and lower area compared to a single stage design. Many embodiments of the invention utilize a two-stage AFE. In some embodiments, the two stages use thick oxide low-leakage pseudo resistors to determine the high-pass corner. The first stage may employ a high-gain folded cascode design with a PMOS input pair. The gain of each amplifier in several embodiments can be independently controlled with digital codes that change the feedback capacitances of capacitors C2 and / or C4. The second stage may have high pass corner frequency controlled using a two-bit current DAC that adjusts the voltage 'EBB71on the gate of pseudo resistors.

[0060] In several embodiments of the invention, the power consumption of the first and second stages of the amplifier, including bias generation and common-mode feedback (CMFB), amount to 3pW and 10.7pW, respectively. Since the PMOS input pair in the AFE is biased with the same DC voltage as the output nodes, a VDD latch at the output nodes can turn off the input pair’s tail current. In some embodiments of the invention, common-mode feedback (CMFB) can be applied to stabilize the common mode of the amplifier using two transistors that provide linear gmwith dynamic degeneration based on the DC mismatch between the reference level and output common mode level. As a result, in certain embodiments the linear input common mode range for the CMFB block may increase from 530mV-670mV to 420mV-780mV.

[0061] In further embodiments, a digitizer, such as a 10-bit I pW (@10kHz clock) successive approximation register (SAR) ADC, can be used to digitize the amplified signal from its associated AFE.

[0062] A clock recovery section eliminates the need for power-hungry phase-locked loops (PLLs) and offers external sampling rate control. A clock recovery section according to some embodiments of the invention is illustrated in Fig. 3A. In some embodiments of the invention, a clock signal is amplitude modulated (AM) onto the WPT signal. Upon receipt by the electrogram sensor, the WPT signal can be self-mixed and low pass filtered to isolate the baseband and DC components. The envelope of the signal can be recovered with an open-loop comparator. To enhance the clock's robustness and resistance to noise, a Schmitt trigger with a window ranging from 0.35V to 0.95V (for example) may be applied to the envelope. For optimal WPT efficiency, the AM index can be selected as low as 10%. The clock recovery schematic and its measured operational range within various carrier frequencies and modulation indexes according to some embodiments are shown in Fig. 3B. Carrier frequencies of 2GHz and above may be out of the selfmixer bandwidth and below 10MHz can experience attenuation from decoupling capacitors. The data from the SAR-ADC can be converted into a Retum-to-Zero (RZ) format before transmission.

[0063] The digitized signal can be transmitted using a transmitter section of the electrogram sensor. In various embodiments, the transmitter can be configured to operate based on either On- Off Keying (OOK) or UWB schemes on frequencies that are, for example, within the 433MHz ISM band. Several embodiments can operate across a wide range of data rates (e.g., 10kbps- 400kbps). The AFE may support gain and bandwidth tuning, which can allow reconstruction of various biological signals with the same wireless setup. A class-E power oscillator, without a power amplifier, may be employed as a transmitter to minimize power consumption. Fig. 3C shows a schematic of a transmitter and the measured phase noise in free-running mode in accordance with several embodiments.

[0064] Depending on the power budget and operational distance of the implant, the transmitter can operate in either OOK or UWB mode. In the UWB mode, power consumption decreases by more than 20 times, but the detection of the transmitted signal becomes more challenging as it occupies a larger bandwidth. To enable easier detection, the pulse width in the UWB mode can be configured as either 250ns or 550ns, with corresponding energy efficiencies of 280pJ / bit and 580pJ / bit. The oscillator's asymmetric driving (EN1 and EN2) can be used to achieve a rapid start-up time of lOOps. In many embodiments, the Tx with SMDL tank has a low phase noise (PN) of - 80.8dB measured at 300kHz offset in the free running mode. Fig. 3D illustrates measured phase noise of a transmitter in accordance with several embodiments of the invention.

[0065] As will be discussed in systems further below, an external computing device may be used as a receiver to receive transmitted signals from a wirelessly powered electrogram sensor. The receiver may be a display, such as AR or VR glasses / goggles. In some embodiments, the receiver may also provide a wireless signal to power and / or control the electrogram sensor. In other embodiments, the control device that provides a wireless power and control signal is a separate device.

[0066] In further embodiments, an external high-quality (Q=54@900MHz) wire-wound 0805 Surface Mount Device Inductor (SMDL) with an inductance of 56nH in a receiver can be used to receive the transmitted signal from the electrogram sensor by serving both as the inductance for the LC tank of the oscillator and as an antenna. The orientation of the SMDL coil may be orthogonal to the circular powering coil, reducing unwanted near-field coupling between the power signal and the transmitter signal. In some embodiments, an SMDL may be used as an antenna in biomedical applications for distances of up to 2 meters.

[0067] In some embodiments of the invention, when the output channels are turned off, they continuously send ‘ 1’s to align the data rate of the system with the recovered clock.

[0068] Although specific architectures of electrogram sensors are described above, one skilled in the art will recognize that any of a variety of architectures may be utilized in accordance with embodiments of the invention.Multi-Channel Sensors

[0069] In additional embodiments of the invention, the sensor system may capture and transmit a number of channels of sensor data concurrently using multiple AFEs and combining their outputs with, for example, a serializer. Power consumption may be lowered by disabling unused channels or lowering the sampling rate. Some embodiments of the invention can enable simultaneous ECG, EMG, and EOG recording in a fully wireless setup.

[0070] A wirelessly powered electrogram sensor system having multiple channels in accordance with several embodiments of the invention is illustrated in Fig. 4. The illustrated system is similar to that of Fig. 1 but with 8 signal electrodes. Fig. 5 illustrates potential placementsof the electrodes. Pairs of electrodes may be taken by individual AFEs to produce biopotential signals. Furthermore, different pairs of electrodes may provide different types of biopotential signals. For example, a first pair can include first electrode and a second electrode that captures an EMG signal for a first AFE. A second pair can include a third electrode and a fourth electrode that captures an EEG signal for a second AFE. In some embodiments, an electrode may serve as a common ground or reference. One skilled in the art will recognize that other numbers of electrodes and combinations to provide biopotential signals are possible.

[0071] The system's clock can be derived from the envelope of a 2W 40.68MHz power link, and its frequency can be externally configured by a user. Fig. 6A illustrates a power harvesting section of the sensor system in accordance with an embodiment of the invention. Fig. 6B illustrates a data and clock recovery section of the sensor system in accordance with an embodiment of the invention. Fig. 6C shows a chart with minimum required modulation index (m) and Fig. 6D shows a chart with a recovered clock’s period jitter for clock recovery. Since the modulation index can be as low as 3.6% (at 400kHz), the clock transmission does not significantly impact power transfer. In several embodiments, the AFE’s configuration and passcode are sent at 10kbps with pulse width modulated amplitude shift keying (PWM-ASK) on the same inductive link. The data transmission can be a one-time operation that takes 2.1ms.

[0072] The AFE in certain embodiments employs a distributed architecture with active electrodes for interference suppression. As depicted in Fig. 7A, the AFE may utilize a two-stage capacitive-coupled amplifier structure to improve input impedance and reduce area compared to a single-stage design. It can offer programmable gain (38dB to 56dB) and a low corner bandwidth (0.5Hz to 5Hz), with an input impedance of 1GQ at 10Hz, as shown in Fig. 7B. Similar to the sensor of Fig. 1, the AFE may filter, amplify, and / or digitize a captured biopotential signal.

[0073] In several embodiments of the invention, when channels are turned off, they continuously send ‘ 1’ s, aligning the data rate of the system with the recovered clock. The amplified signals may undergo digitization, serialization, and conversion to return to zero (RZ) format. In many embodiments, the serialized data can be wirelessly transmitted at 433MHz using a UWB Class-E power oscillator and an SMDL as an antenna as shown in Fig. 8A. In the scenario where all channels are off, the serializer can transmit the recovered clock.

[0074] As will be discussed in systems further below, an external computing device may be used as a receiver to receive transmitted signals from a wirelessly powered electrogram sensor.The receiver may be a display, such as AR or VR glasses / goggles. In some embodiments, the receiver may also provide a wireless signal to power and / or control the electrogram sensor. In other embodiments, the control device that provides a wireless power and control signal is a separate device.

[0075] Although specific architectures for electrogram sensors and multi-channel sensors are discussed above, one skilled in the art will recognize that additional variations are possible within embodiments of the invention.Human-Machine Interface using Electrogram Sensors

[0076] Human-machine interface (HMI) devices may be used to interact with displays such as augmented reality and virtual reality (AR / VR) glasses. HMI systems generally operate to capture a user’s intention and use it to control objects on a screen or other aspects of the display device.

[0077] In additional embodiments of the invention, an HMI system can incorporate one or more wirelessly powered electrogram sensors that can capture one or more biopotential signals from a user. Sensor(s) can be placed near a user’s eye to capture signals indicative of the eye’s movement and / or contraction of muscles around the eye. Sensor(s) can be placed on or near a user’s fingers to capture signals indicative of movement of one or more fingers.

[0078] An example of an HMI for eye tracking in accordance with an embodiment of the invention is illustrated in Fig. 9. The HMI system 900 includes electrodes 902, 904, and 906 that can be placed around a user’s eye 907. The electrodes 902, 904, and 906 may be attached to the surface of the user’s skin, e.g., using adhesive or glue. The electrodes 902, 904, and 906 can be connected to a sensor system 908, which may be similar to a multi-channel wirelessly powered electrogram sensor system as those discussed further above with respect to Fig. 4. The sensor system 908 may be configured to obtain one or more biopotential signals from the electrodes 902, 904, 906. In further embodiments, the sensor system 908 obtains different types of biopotential signals simultaneously, e.g., using different pairs of electrodes. For example, one pair of electrodes 904 and 902 may capture an electromyogram (EMG) signal, while another pair of electrodes 906 and 902 may capture an electrooculogram (EOG) signal. These different types of signals can provide diverse potential sensing.

[0079] The sensor system 908 may include a transmitter section that can amplify, digitize, and transmit the signals as an output signal to an external receiver 910. The external receiver 910 maycommunicate wirelessly with the sensor system 908 to receive the transmitted signal. The external receiver may be a display, such as but not limited to, a head-mounted display (HMD), augmented reality (AR) glasses / goggles, or virtual reality (VR) glasses / goggles worn by the user. In some embodiments, the receiver may also provide a wireless signal to power and / or control the electrogram sensor. In other embodiments, the control device that provides a wireless power and control signal is a separate device.

[0080] An example of an HMI for hand gesture tracking in accordance with an embodiment of the invention is illustrated in Fig. 10. The HMI system 1000 includes electrodes 1002, 1004, and 1006 that can be placed around a user’s hand or fingers 1007. The electrodes 1002, 1004, and 1006 may be attached to the surface of the user’s skin, e.g., using adhesive or glue. The electrodes 1002, 1004, and 1006 can be connected to a sensor system 1008, which may be similar to a multichannel wirelessly powered electrogram sensor system as those discussed further above with respect to Fig. 4. The sensor system 1008 may be configured to obtain one or more biopotential signals from the electrodes 1002, 1004, 1006. In further embodiments, the sensor system 908 obtains different types of biopotential signals simultaneously, e.g., using different pairs of electrodes. For example, one pair of electrodes 1004 and 1002 may capture an electromyogram (EMG) signal, while another pair of electrodes 1006 and 1002 may capture an electrocardiogram (ECG) signal. These different types of signals can provide diverse potential sensing.

[0081] The sensor system 1008 may include a transmitter section that can amplify, digitize, and transmit the signals as an output signal to an external receiver 1010. The external receiver 1010 may communicate wirelessly with the sensor system 908 to receive the transmitted signal. The external receiver may be a display, such as but not limited to, a head-mounted display (HMD), augmented reality (AR) glasses / goggles, or virtual reality (VR) glasses / goggles worn by the user. In some embodiments, the receiver may also provide a wireless signal to power and / or control the electrogram sensor. In other embodiments, the control device that provides a wireless power and control signal is a separate device.

[0082] Although specific HMI systems are discussed above with respect to Figs. 9 and 10, one skilled in the art will recognize that any of a variety of HMI systems may be utilized in accordance with embodiments of the invention. For example, HMI systems in other embodiments may be adapted to other parts of the body and / or may utilize different number of electrodes.Processes for Capturing Biopotential Signals using a Wirelessly Powered Electrogram Sensor

[0083] A process for obtaining a biopotential signal and wirelessly transmitting data indicative of the biopotential signal is illustrated in Fig. 11. The process may use electrogram sensors such as those described further above.

[0084] The process 1100 includes electrogram sensor receiving (1102) a wireless power transfer (WPT) signal that includes an embedded clock signal on a receive antenna. The electrogram sensor recovers (1104) power and the clock signal from the WPT signal. The power may be used to power other functions on the sensor and the clock signal may be used in generating a transmit signal.

[0085] A biopotential signal is captured (1106) using a set of signal electrodes on the electrogram sensor. In further embodiments, multiple biopotential signals can be obtained by different pairs of signal electrodes. The multiple biopotential signals may be different types.

[0086] The biopotential signal is provided to an analog front-end (AFE) that amplifies (1108) the biopotential signal. In some embodiments, the process includes tuning the gain and / or bandwidth of the AFE. When there are multiple biopotential signals, each may be passed to a different AFE.

[0087] The amplified biopotential signal is digitized (1108) using, for example, an SAR ADC. When there are multiple signals, each may be passed to a different ADC. The digitized signal is transmitted (1110) by a transmit antenna on the electrogram sensor.

[0088] Although a specific process is discussed above with respect to Fig. 11, one skilled in the art will recognize that any of a variety of processes may be utilized in accordance with embodiments of the invention as may be appropriate to any particular application.Low-power Radar-based Sensors:

[0089] In a radar-based approach, a low-power radar operating above 1GHz and below ITHz can be used on one finger as a ring. The radar can be used to capture the relative distance of the finger that has the radar sensor from other fingers. This information can be used to capture smallest movements of the finger, which can eventually be transmitter to an AR / VR for further processing. The radar data can be used to transfer human intention to a computer to control the movement of different objects.

[0090] In embodiments of the invention, the radar sensor can operate in near-field or far field. The depth resolution of a far field radar sensor is determined by the carrier frequency, phase resolution, and bandwidth. The accuracy of a radar sensor can be as small as lOOum, which is sufficient to capture small movement of the fingers. Object localization and motion detection using radars is discussed further in International Patent Publication No. WO 2024 / 064907 (Application No. PCT / US2023 / 074927), the relevant portions of which are hereby incorporated by reference in their entirety. In many embodiments of the invention, a radar signal can be captured and utilized in combination with one or more biopotential signals in sensor systems such as in Figs. 1 and 4 and processes such as in Fig. 11.Conclusion

[0091] While the above description contains many specific embodiments of the invention, these should not be construed as limitations on the scope of the invention, but rather as an example of one embodiment thereof. It is therefore to be understood that the present invention may be practice otherwise than specifically described, without departing from the scope and spirit of the present invention. Thus, embodiments of the present invention should be considered in all respects as illustrative and not restrictive.

Claims

WHAT IS CLAIMED IS:

1. A human-machine interface system comprising: a wirelessly powered electrogram sensor comprising: a set of signal electrodes; an analog front end (AFE) configured to receive a biopotential signal from the set of signal electrodes and provide an amplified signal; an analog to digital converter (ADC) configured to receive the amplified signal and digitize the amplified signal; and a transmitter configured to receive the digitized signal, modulate the digitized signal into an output signal and transmit the output signal.

2. The human-machine interface system of claim 1, further comprising an external computing device configured to receive the transmitted output signal and determine a user intention from the output signal.

3. The human-machine interface system of claim 2, wherein the analog front-end includes a programmable gain amplifier and a tunable bandwidth filter controlled by a wireless command signal from the external computing device.

4. The human-machine interface system of claim 1, wherein the analog front end comprises at least one filter to reduce noise.

5. The human-machine interface system of claim 1, wherein the analog to digital converter is a SAR ADC.

6. The human-machine interface system of claim 1, wherein the wirelessly powered electrogram sensor further comprises: a plurality of pairs of AFEs and ADCs, each pair of AFE and ADC associated with a pair of signal electrodes of the set of signal electrodes;a serializer configured to combine the outputs of the plurality of analog to digital converters into the digitized signal for the transmitter.

7. The human-machine interface system of claim 6, wherein a plurality of the pairs of signal electrodes are configured to each capture a different type of biopotential signal.

8. The human-machine interface system of claim 7, wherein the different type of biopotential signals include at least electromyogram (EMG) and electrooculogram (EOG).

9. The human-machine interface system of claim 8, wherein the wirelessly powered electrogram sensor is mounted near a user’s eye and configured to detect oculomotor EMG signals indicative of gaze direction or eye movement.

10. The human-machine interface system of claim 7, wherein the different type of biopotential signals include at least electromyogram (EMG) and electroencephalogram (EEG).

11. The human-machine interface system of claim 10, wherein the wirelessly powered electrogram sensor is mounted on a hand and configured to detect EMG signals corresponding to finger gestures or hand movement for use in gesture-based control of an external computing device.

12. The human-machine interface system of claim 7, wherein the different type of biopotential signals include at least electromyogram (EMG) and electrocardiogram (ECG).

13. The human-machine interface system of claim 1 , wherein the wireless transmitter uses ultra-wideband (UWB) modulation to transmit the digital signals with reduced power consumption.

14. A method for detecting finger movement, the method comprising: emitting radar waves from a radar sensor attached to a finger over a period of time;receiving reflected radar waves at the radar sensor over the period of time; determining, based upon the received reflected radar waves over the period of time, at least one of: distance, direction, and radial velocity of at least one object.

15. The method of claim 6, further comprising: determining that the at least one object is another finger.

16. A wearable HMI system comprising: a ring-shaped device configured to be worn on a finger and comprising a radar sensor, wherein the radar sensor is configured to detect motion of adjacent fingers using reflected radar waves and to transmit motion data to an external computing device for gesture recognition.

17. The wearable HMI system of claim 16, wherein the radar sensor operates in the millimeter-wave frequency range and detects sub-millimeter movements with a resolution of 100 micrometers or better.

18. A method for user input in an augmented or virtual reality environment comprising: detecting a user's eye muscle signals using EMG electrodes placed near the eye; determining gaze direction or blink patterns from the EMG signals; and using the determined eye activity as a control input to select or manipulate objects on a head mounted display.

19. A multi-modal HMI system comprising: a plurality of wirelessly powered electrogram sensors configured to: capture signals of different biopotential signal types including EMG and EOG; amplify, digitize, and transmit the captured signals to a central processing system; a radar-based motion detection unit configured to detect fine motor movement and transmit detected movement as radar information; anda central processing system configured to integrate the captured biopotential signals and radar information into integrated user intent information and infer user intent based on a combination of neural and physical cues derived from the integrated user intent information.

Images