Reconfigurable sensor unit for electronic device

[0001] This application is a continuation in part of U.S. patent application Ser. No. 14/749,118 which was filed on Jun. 24, 2015, now pending, which claims the benefit of U.S. Provisional Application No. 62/121,104 which was filed on Feb. 26, 2015. The content of each patent application identified above is incorporated by reference in its entirety to the maximum extent allowable by law.

[0002] This disclosure relates to the field of sensors, and more particularly, to reconfigurable sensors for electronic devices. This disclosure also relates to techniques for configuring those sensors.

[0003] Portable electronic devices such as smartphones, smartwatches, other wearables, and tablets are ever more popular in the world today. Certain functions of these devices depend on the device having a knowledge of the device's orientation, or of conditions of the environment in which the device currently resides. For example, a smartphone may rotate its user interface from a portrait view to a landscape view based upon the orientation in which the smartphone is held. As another example, a smartwatch may activate its display or alter the brightness of its display based upon the orientation in which the smartwatch is held, or based upon the light in the environment in which the smartwatch resides, respectively. Such portable electronic devices may also log the physical activity of a user. For example, a smartphone or smartwatch may count the number of steps taken by a user.

[0004] In order to determine the orientation of the device or conditions of the environment in which the device resides, sensors such as accelerometers and gyroscopes are employed. Typically, the electronic device includes a system on chip (SOC) that receives raw data from the sensors, and then determines the device orientation or conditions of the environment.

[0005] While this approach is effective and enables the electronic device to perform commercially desirable functions, further features may yet be desired. Therefore, further development is needed.

[0006] This summary is provided to introduce a selection of concepts that are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in limiting the scope of the claimed subject matter.

[0007] An electronic device disclosed herein includes a printed circuit board (PCB) having at least one conductive trace thereon and a system on chip (SoC) mounted on the PCB and electrically coupled to the at least one conductive trace. A sensor chip is mounted on the PCB in a spaced apart relation with the SoC and is electrically coupled to the at least one conductive trace such that the sensor chip and SoC are electrically coupled. The sensor chip itself includes at least one sensing device and a control circuit. The control circuit is configured to receive configuration data as input, the configuration data including sense channel configuration, clock configuration, feature configuration, and classification configuration. Data is acquired from the at least one sensing device in accordance with the sense channel configuration, and triggers for feature computation are generated in accordance with the clock configuration or events relating to physical phenomena. Features of the data are computed in response to the triggers and in accordance with the feature configuration. A context of the electronic device relative to its surroundings is determined based on the computed features and in accordance with the classification configuration, and the context is output.

[0008] Also disclosed herein is a sensor chip mounted on a printed circuit board (PCB) and electrically coupled to a system on chip (SoC) mounted on the PCB via at least one conductive trace. The sensor chip includes at least one sensing device, and a control circuit. The control circuit is configured to receive configuration data as input, and acquire data from the at least one sensing device in accordance with the configuration data. The control circuit classifies a context of the at least one sensing device relative to its surroundings based on analysis of the acquired data in accordance with the configuration data.

[0023] The present description is made with reference to the accompanying drawings, in which example embodiments are shown. However, many different embodiments may be used, and thus the description should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. Like numbers refer to like elements throughout.

[0024] With initial reference to FIG. 1, an electronic device 100 is now described. The electronic device 100 may be a smartphone, tablet, smartwatch, or other wearable device. The electronic device 100 includes a printed circuit board (PCB) 99 having various components mounted thereon. Conductive traces 97 printed on the PCB 99 serve to electrically couple the various components together in a desired fashion.

[0025] A system on chip (SoC) 116, which comprises a central processing unit (CPU) 117 coupled to a graphics processing unit (GPU) 119, is mounted on the PCB 99. Coupled to the SoC 116 are a transceiver 101 via which the SoC 116 can communicate with remote servers over the internet, and a touch sensitive display 103 via which the SoC 116 may display output and receive input. A variety of sensors are coupled to the SoC 116, including a light sensor 102 for determining the level of ambient light in the environment in which the electronic device 100 resides, a magnetometer 104 used to determine the orientation of the electronic device 100 with respect to the magnetic field of the Earth, a barometer 106 used to determine the air pressure in the environment (and thus, the altitude of the electronic device 100), a microphone 114 used to detect audible noises in the environment, and a proximity sensor 118 used to determine proximity of the user with respect to the electronic device 100.

[0026] A configurable sensor unit 107 is mounted on the PCB 99 spaced apart from the SoC 116, and coupled thereto by the conductive traces 97. For illustrative purposes only, for example, the configurable sensor unit 107 includes an accelerometer 108 and/or gyroscope 110 coupled to a control circuit 112. The accelerometer 108 is used for determining accelerations experienced by the electronic device 100, and the gyroscope 110 is used to determining an orientation of the electronic device 100 with respect to the environment. It should be understood that any now existing or future developed type of sensors could be used with this disclosure, the scope of which is not limited by the type of configurable sensors used. The configurable sensor unit 107 may be formed from discrete components and/or integrated components and/or a combination of discrete components and integrated components, and may be formed as a package.

[0027] As illustrated in FIG. 1 as an example, configurable sensor unit 107 is not a portion of the SoC 116, and is a separate and distinct component from the SoC 116. The sensor unit 107 and the SoC 116 are separate, distinct, mutually exclusive chips mounted on the PCB 99 at different locations and coupled together via the conductive traces 97. It should be understood that this example configuration is just one optional configuration and SOC 116 and/or the features/function thereof could be integrated with the sensor unit 107, which is also included in the scope of this disclosure.

[0028] In operation, the SoC 116 may acquire data from the various sensors 102, 103, 104, 106, 114, and 118 at an acquisition rate, and may process the data so as to determine a context of the electronic device 100 relative to its environment. The acquisition rate for some of these various sensors, in some operating scenarios, may be relatively low due to the nature of the contexts to be determined from these sensors not necessitating a higher acquisition rate. Therefore, power consumption as a result of acquisition of data from these various sensors may be relatively low. Contexts will be explained in detail.

[0029] On the other hand, the contexts to be determined from some other sensor, for example, in some operating scenarios, the accelerometer 108 and/or the gyroscope 100 may involve the use of a higher acquisition rate than the other sensors. Therefore, to save power, a control circuit 112 of the configurable sensor unit 107 that consumes less power than the SoC 116 may be used to perform acquisition of data from the accelerometer 108 and/or the gyroscope 110, and may perform some processing of that data. So as to provide the same level of configurability as would be present were the SoC 116 performing the acquisition and processing of the data from the accelerometer 108 and/or the gyroscope 110, the sensor unit 107 may receive configuration data as input.

[0030] In operation, the control circuit 112, as stated, receives the configuration data as input. The control circuit 112 acquires data from the accelerometer 108 and/or the gyroscope 110, and processes the data so as to generate a context of the electronic device 100 relative to its surroundings. This processing is performed by the control circuit 112 using a processing technique operating in accordance with the configuration data received as input. The processed data is then output by the control circuit 112 to the SoC 116 for use thereby.

[0031] With additional reference to FIG. 2, sample details of an example control circuit 112 are now given. Although specifics will be described, it should be understood that the functions of the control circuit 112 may be achieved through different circuit implementations with different circuit topologies, elements and/or units, and made from suitable components, and that any and all such circuit implementations are within the scope of this disclosure. It should also be understood that while certain functions and operations are described as being performed by certain components of the control circuit 112, the performance of those functions and operations by any suitable hardware is contemplated and within the scope of this disclosure.

[0032] The control circuit 112 includes a multiplexer 190 that receives three dimensional data in the X, Y, and Z directions from the accelerometer 108 and gyroscope 110. The output of the multiplexer 190 is fed to a charge pump 200, which in turn feeds its output to a filter 204. The filter feeds its output to a programmable logic device such as an arithmetic logic unit (ALU) 206, which feeds its output to computed feature registers 208. A classification unit 210 reads data from the computer feature registers 210, and outputs contexts to data registers 212. The data registers 112 are coupled to an I2C bus interface 214. In addition, an interrupt generator 202 is coupled to the output of the charge pump 200, and the output thereof may be read by the SoC 116. Configuration registers 216 are coupled to the filter 204, ALU 206, computed feature registers 208, and classification unit 210. In addition to the connections described above, the outputs of the filter 204, ALU 206, and computed feature registers 208 are also each coupled to the data registers 212.

[0033] In operation, the accelerometer 108 and gyroscope 110 each output data for the three spatial dimensions X, Y, Z. This data is received by the multiplexer 190, which selectively multiplexes that data to the charge pump 200. The charge pump 200 outputs that data to the filter 204 which may be an analog or digital filter. The filter 204 filters the data and outputs it to an arithmetic logic unit (ALU) 206.

[0034] The ALU 206 serves to extract features of the data itself, such as a mean acceleration over a given period of time, variance of acceleration, radian acceleration, number of acceleration peaks, number of zero crossings in the filtered acceleration signal, peak acceleration values, linear acceleration values, energy in band, mean of roll, pitch, and yaw, variance of roll, pitch, and yaw, mean of linear acceleration, variance of linear acceleration, etc. This list of features that may be extracted from the data is not intended to be exhaustive. Indeed, the ALU may extract any useful features from accelerometer 108 or gyroscope 110 data, and all such features are within the scope of this disclosure.

[0035] The computed feature registers 208 store the features extracted by the ALU 206. The classification unit 210 may be a separate ALU, and classifies the context of the electronic device 100 based upon the features extracted by the ALU and stored in the computed feature registers 208. For example, the context of the electronic device 100 may be where on the user's body it is carried (i.e. in pocket, in hand, in holster), a current method of locomotion of the user (i.e. running, walking, driving, bicycling, climbing stairs), or an orientation of the electronic device 100 with respect to gravity. Another example context may be movement of the electronic device 100 in a gesture, such as a user raising a smartwatch in to a position to view the screen thereof, shaking the electronic device 100, double tapping the touch sensitive display 103 of the electronic device 100, rotating the electronic device 100 either clockwise or counterclockwise, and swiping the touch sensitive display 103 to the left, right, top, or bottom.

[0036] The contexts classified by the classification unit 210 are then stored in the data registers 212, and may in turn be read out by the I2C bus 214 to the SoC 116. The SoC 116 then uses the classified contexts in its operations. The interrupt generator 202 serves to generate interrupts for the SoC 116 so as to facilitate communication between the configurable sensing unit 107 and the SoC 116.

[0037] In some applications, the multiplexer 190 may receive additional data from the SoC 116. For example, the SoC 116 may send additional data from the light sensor 102, magnetometer 104, barometer 106, microphone 114, proximity sensor 118, or touch sensitive display 103 to the multiplexer 190, such that this data may be used by the other components of the sensor unit 107.

[0038] The described functionalities of the filter 204, ALU 206, and classification unit 210 operate in accordance with configuration data stored in the configuration registers 216. For example, the configuration data may be changeable parameters of the processing technique performed by the control circuit 112. Thus, the configuration data may change the pre-processing performed by the control circuit 112 by either replacing the filtering technique performed by the filter 204, or by replacing the values of the coefficients of the filter 204.

[0039] The configuration data may control the performance level of the sensor unit 107 by altering the balance between latency, accuracy, and power consumption. This balance may be changed by altering the acquisition rate of the data, and thus the rate at which the multiplexer 190 switches from among its various inputs. This balance may also be changed by altering the range of accepted values of the data, for example by the ALU 206 ignoring data outside of the range of accepted values when extracting or computing features of the data.

[0040] The configuration data may change the techniques used by the ALU 206 to extract and compute the features from the data. The ALU 206 may be programmed to extract and compute a plurality of features from the data, and the configuration data may serve to select a subset of that plurality of features for the ALU 206 to actually extract and compute. In some cases, the configuration data may program or reprogram the ALU 206 for extracting and computing different features. The configuration data may also program or reprogram the way in which the ALU 206 extracts and computes any of the features.

[0041] The balance between latency, accuracy, and power consumption may also be altered by the configuration data switching operation of the sensor unit 107 between a lower power consumption mode and a higher power consumption mode. In the lower power consumption mode, the sensor output data rate (ODR) is lower and ALU 206 extracts and computes a smaller subset of features of the plurality of features which the ALU 206 is programmed to extract and compute, whereas in the higher accuracy mode, the ALU 206 extracts and computes a larger subset (or all) of the features of the plurality of features which the ALU 206 is programmed to extract and compute. When these lower power consumption and higher accuracy modes are used, the sensor unit 107 operates in the lower power consumption mode until features of a threshold value or values are extracted by the ALU 206, at which point the sensor unit 107 switches to operation in the higher power consumption mode.

[0042] The configuration data may control the techniques used by the classification unit 210 for classifying the features extracted and computed by the ALU 206 and stored in the computed feature registers 208, for example by using a decision tree. The classification unit 210 may be programmed to classify the features into a plurality of contexts, and the configuration data may select a subset of that plurality of contexts for the classification unit 210 to classify. In addition, the configuration data may program or reprogram the classification unit 210 for classifying and determining different contexts. The configuration data may also program or reprogram the way in which the classification unit 210 classifies and determines the different contexts.

[0043] The SoC 116 may receive additional or new configuration data from the transceiver 101, and it may then pass this additional or new configuration data to the configuration registers 216. In some instances, the SoC 116 may generate the additional or new configuration data itself, for example based upon user input, and then pass the additional or new configuration data to the configuration registers 216.

[0044] It should be understood that the configuration data stored in the configuration registers 216 may operate on the various components in a variety of ways. For example, the configuration data may directly provide the techniques or algorithms used by the various components, may select which pre-programmed techniques or algorithms are to be used, may provide the coefficients used by the techniques or algorithms, or may replace standard techniques, algorithms, or coefficients. Thus, the various components may operate in a first fashion in the absence of the configuration data, and in a second fashion in the presence of the configuration data. The configuration registers 216 are reprogrammable on the fly by the SoC 116, and thus the alteration of the operation of the sensor unit 107 may not involve the changing of firmware. Therefore, the various components may be considered to be fully programmed at a first point in time, and that programming may be changed by the configuration data at a second point in time. For example, the execution of different applications by the SoC 116 may cause reprogramming of the configuration registers 216 on the fly. An example is when the electronic device 100 is acting as an activity tracker, and the configuration registers 216 are loaded with different configurations as a function of user input to indicate which activity they are going to undertake (e.g. walking, cycling, weight lifting, etc).

[0045] In addition to the power savings provided by this sensor unit 107, the sensor unit 107 facilitates easier manufacture of electronic devices 100. Due to the configuration data being easily changeable in the configuration registers 116, different sensor units 107 need not be stocked by a manufacturer for different electronic devices 100. The same design of sensor unit 107 may be used by a manufacturer for different electronic devices 100, with differences in the configuration data stored in the configuration registers 116 of each, thereby easing manufacturing supply chain constraints.

[0046] With reference to FIG. 3, a sample implementation of the electronic device 100 and techniques for configuring the electronic device 100 is now described. Here, a series of configuration messages are sent to the configuration registers 216. The information contained within the configuration messages is in turn potentially used by the math function logic gates 203, filter 204, ALU 206, computed feature registers 208, and classification unit 210. The configuration messages may be a sense channel configuration message, filtering configuration message, clock configuration message, feature configuration message, and classification configuration message.

[0047] The sense channel configuration message serves to define which sensors or combinations of sensors (i.e. accelerometer 108, gyroscope 110, etc) data is acquired from, and/or which channels of the selected sensors the data is acquired from, and/or which processed versions of that data (e.g. derived sensor data) is to be used. As shown in FIG. 4, the sense channel configuration message contains a message ID field which holds an identifier of what type of message it is (i.e. here, that it is a sense channel configuration message), a sub ID field which holds a further identifier of what type of message it is, a field containing a number of channels along which to acquire data, and a configuration field for each channel. Shown in FIG. 5 are the different ways the contents of the configuration field defines each channel. For example, a given channel may be a X, Y, or Z axis of the accelerometer 108 or gyroscope 110 or magnetometer 104 (shown in FIG. 1), Euclidean norms or magnitudes of acceleration along any axis, Euclidean norms or magnitudes of angular rate along any axis, Euclidean norms or magnitudes of magnetic field along any axis, and/or normalized values of any of the preceding.

[0048] The filtering configuration message defines the filter applied to acquired data by the filter 204. The filter 204, as explained above, is a digital filter with a transfer function defined by:

[0049] $H\left(z\right)=\frac{\sum _{l=0}^N{a}_l{z}^{-1}}{\sum _{k=0}^M{b}_k{z}^{-k}}$

[0050] In particular, the contents of the filter configuration message define the constants of the numerator and denominator of the digital filter applied by the filter 204. Using the numerator and denominator as described above, almost any filter such as low pass, high pass, band pass can be described.

[0051] The format of the filtering message is shown in FIG. 6. In particular, the filtering message includes a message ID field which holds an identifier of what type of message it is (i.e. here, that it is a filtering configuration message), a sub ID field which holds a further identifier of what type of message it is, input and output channel fields, and numerator and denominator coefficient fields. Identifier names correlated to output channels are shown in FIG. 7.

[0052] The clock configuration message defines triggers to be used by the ALU 206 in computing features of the data. The triggers may be defined as being generated at periodic intervals (ie. one per second), or may be defined as being generated at detectable events within the data. For example, a detectable event may be a zero amplitude crossing of the data (in the positive direction, in the negative direction, or either), or a peak of the data (positive, negative, or either).

[0053] The format of the clock configuration message is shown in FIG. 8. In particular, the filtering message includes a message ID field which holds an identifier of what type of message it is (i.e. here, that it is a clock configuration message), a sub ID field which holds a further identifier of what type of message it is, a clock index field used to define a clock, a reset trigger index used to generate an index for resetting the trigger, and a clock duration field used to define a clock duration after which a trigger is generated. Sample reset trigger names and indexes are shown in FIG. 9.

[0054] The feature configuration message defines feature of the data to be computed by the ALU 206, and these features may be computed at the triggers. For example, a condition of the data may be a zero amplitude crossing of the data (e.g. in a positive direction, in a negative direction, or either), and the ALU 206 may count the number of occurrences of the zero amplitude crossing between each trigger. A different condition of the data may be a characteristic of the data computed at the triggers. For example, the condition may be a data maxima, a data minima, a data mean, a data variance, and the energy in the band at a trigger, or between two triggers. Other sample conditions may be an amplitude peak (positive, negative, or either), a peak to peak value, a total energy between triggers, or a total time elapsed during the feature.

[0055] The format of the feature configuration message is shown in FIG. 10. In particular, the feature configuration message includes a message ID field which holds an identifier of what type of message it is (i.e. here, that it is a feature configuration message), a sub ID field which holds a further identifier of what type of message it is, a number of features to be configured, a feature index, and definitions for feature numbers, feature input channels, feature output channels, and input reset triggers. Shown in FIG. 11 are indices for the feature buffers that may be used on the feature outputs, and shown in FIG. 12 are the various features that may be computed at the feature channels. It should be noted that some entries of the table of FIG. 12 also denote triggers for use with the clock configuration message, and these entries are so labeled.

[0056] The classification configuration message defines a decision tree to be used in classifying and determining context of the electronic device 100 relative to its surroundings. The format of the classification configuration message is shown in FIG. 13. The classification configuration message includes message ID and sub ID fields as with the other configuration messages, a field for the output destination, and then data fields defining the nodes, node indices, and results of the decision tree. Shown in FIG. 14 are the operations that may be performed at the nodes, and their respective indices.

[0057] It should be appreciated that via the various configuration messages described above, the control circuit 112 may be configured in a variety of ways to determine a variety of contexts of the electronic device 100. For example, the control circuit 112 may be determined to determine each time a user carrying the electronic device 100 has taken a step. In such a case, the configuration messages may be as follows: the sense channel configuration message may set the ADC 201 to convert the X, Y, and Z channels of the accelerometer, together with a Euclidean norm, for input; the filtering configuration message may set the filter 204 to be a low pass filter; the clock configuration message may set the triggers as amplitude threshold crossings of the data in the positive direction; the feature configuration message may set the ALU 206 to determine the peak to peak value between two triggers, the energy in band for a two second window, and the duration between two triggers; and the classification configuration message may set the classification unit 210 to determine that a user has taken a step as a function of a duration since a last amplitude threshold crossing of the data in the positive direction, a peak to peak amplitude between such amplitude threshold crossings, and the energy in band for two seconds.

[0058] A sample of the decision tree for determining the steps, as defined by the classification configuration message, is as follows:

[0059]
This example decision tree shows the nodes as configured, the operators and operands of the nodes as configured, and the steps to be taken as a function of the operator and operands.

[0060] Many modifications and other embodiments will come to the mind of one skilled in the art having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is understood that various modifications and embodiments are intended to be included within the scope of the appended claims.