Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
The information disclosure statements (IDS) submitted on January 17, 2024 and July 24, 2024, are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Specification
The specification has not been checked to the extent necessary to determine the presence of all possible minor errors. Applicant’s cooperation is requested in correcting any errors of which applicant may become aware in the specification.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on June 5, 2026 has been entered.
Response to Amendment
The Amendment filed June 5, 2026 has been entered. Claims 1, 3-9, 11-15, & 17-21 remain pending in the application. Claims 1, 3-9, 11-15, & 17-21 have been amended.
Response to Arguments
Applicant’s arguments filed June 5, 2026, see pp. 9-12 of Applicant remarks, have been entered and fully considered but they are not persuasive. In light of the amendments, the rejection(s) have been withdrawn. However, upon further consideration, new grounds of rejections have been made, and Applicant’s arguments are rendered moot.
In response to the Applicant’s argument, with respect to the rejection of claims 1, 7 , & 8, under U.S.C. § 103, that the prior art references Matsumoto (JP 2020-101849 A), in view of Taniguchi (JP 2007-075428 A), and claims 3-6, 9, 11-15, & 17-21, under U.S.C. § 103, that the prior art references Matsumoto, in view of Taniguchi, and further in view of Unuma (US 6941239 B2), fail to disclose, teach, and/or suggest individually or in combination, each and every limitation of the claimed invention, to include the amended features of the invention: “determining that a state of a user wearing the acceleration sensor is running downstairs responsive to determining both: that the amplitudes are less than a first reference value; and that the count value is equal to or greater than a piece count threshold value;”.
The Examiner respectfully disagrees, in light of the amendments, Barfield (US 2014/0122012 A1), Alessi (US 2021-0285773-A1), and Yang (US 2016-0209232 A1), in view of Matsumoto, and further in view of Unuma, further disclose the additional limitations that have been amended, and further identify “determining that a state of a user wearing the acceleration sensor is running downstairs responsive to determining both: that the amplitudes are less than a first reference value; and that the count value is equal to or greater than a piece count threshold value;” and meet these requirements. Therefore, the Applicant’s arguments are unconvincing and the rejection of amended independent claims 1, 9, & 15, and dependent claims 3-8, 11-14, & 17-21, which depend from and incorporate the limitations of amended independent claims 1, 9, & 15, are respectively maintained. Rejections based on the newly cited prior art references follows below.
Claim Objections
Claims 4, 12, & 18 is objected to because of the following informalities:
Claims 4, 12, & 18 recite, “wherein determining that that the state of the user…,” in ll. 1-2, recommend rephrasing to read, “wherein determining that the state of the user…,”.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 3, 5-9, 11, 13-14, & 21 are rejected under 35 U.S.C. 103 as being unpatentable over Barfield (US 2014/0122012 A1, Pub. Date May 1, 2014, hereinafter, Barfield), in view of Alessi et al. (US 2021/0285773 A1, Pub. Date Sep. 16, 2021, hereinafter, Alessi), in view of Yang et al. (US 2016/0209232 A1, Pub. Date Jul. 21, 2016, hereinafter, Yang), and further in view of Matsumoto (JP 2020101849 A, Pub. Date Mar. 3, 2007, hereinafter, Matsumoto).
Regarding independent claim 1, Barfield, teaches:
A measurement device comprising ([Abstract], [0010], [0028], & [0033]: teaches a measurement device (e.g., pedometer) equipped with the hardware components – an acceleration sensor, a memory, and a processor executing instructions):
an acceleration sensor ([Abstract], [0010], [0028], & [0033]);
a memory comprising instructions ([Abstract], [0010], [0028], & [0033]); and
one or more processors configured to execute the instructions to perform ([Abstract], [0010], [0028], [0033]-[0036]):
acquiring time-series acceleration data output by the acceleration sensor ([Abstract], [0010], [0014], [0018], [0025], [0027]-[0034], & [0036]-[0038]: processor continuously gathers sampled data points over time from the accelerometer, which constitutes time-series acceleration data);
determining one or more amplitudes of a waveform corresponding to the time-series acceleration data ([Abstract], [0009]-[0014], [0018], [0021], & [0024]-[0038]: calculating the maximum magnitudes of the sampled signals corresponds to determining the amplitude peaks of the acceleration waveform);
determining a count value of zero-crossing points within a predetermined time ([Abstract], [0010], [0012], [0014]-[0016], [0018], [0020], [0023]-[0025], [0029], [0031]-[0032], & [0036]-[0038]: system processes the waveform to find specific zero-crossing points, tracking how many of the step events occur within a preset time period);
Barfield, is silent in regard to:
determining that a state of a user wearing the acceleration sensor is running downstairs responsive to determining both:
that the amplitudes are less than a first reference value; and
that the count value is equal to or greater than a piece count threshold value; and
However, Alessi, further teaches:
The Examiner is combining Barfield in view of Alessi by implementing that Barfield assesses user states by checking if amplitudes are under a reference a threshold and if step counts meet a piece-count threshold ([0011]-[0015], [0018], [0020]-[0021], [0024], & [0032]).
determining that a state of a user wearing the acceleration sensor is running downstairs responsive to determining both ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]: applies state detection to identify a downstairs/descending state using waveform peak amplitudes):
that the amplitudes are less than a first reference value ([0012]-[0015], [0050]-[0052], [0056]-[0057], [0067], [0073], [0075]-[0077], & [0080]-[0081]); and
that the count value is equal to or greater than a piece count threshold value ([0012]-[0015], [0046]-[0052], [0071], [0075], [0077], [0083], [Claim 6], [Claim 7], [Claim 16], [Claim 17], [Claim 24], [Claim 25], & [Claim 28]); and
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the amplitude and piece count threshold logic of Barfield with the stair-descent state detection parameters of Alessi to arrive at determining that a state of a user wearing the acceleration is running downstairs responsive to determining both: that the amplitudes are less than a first reference value; and that the count value is equal to one or greater than a piece count threshold value. Barfield teaches determining a user’s running state by checking if amplitudes are less than a reference value and count values exceed a threshold (PRE_STEP_COUNT), but does not detail applying this to the specific state of a user running downstairs. Alessi teaches analyzing waveform amplitudes and feature patterns to determine when the state of a user is descending or going downstairs. The problem being solved by this combination is improving user state classification accuracy to enable reliable tracking of elevation changes. This combination represents a predictable variation of known techniques to improve similar devices, as modifying algorithmic threshold parameters to recognize descending activity is a standard practice in wearable sensor programming (KSR).
Barfield, and Alessi, are silent in regard to:
counting one step for a portion of the waveform
However, Yang, further teaches:
The Examiner is combining Barfield and Alessi, in view of Yang, implementing that Barfield measures the time width between crossings to validate steps ([0010]-[0011], [0014], [0029], [0032], & [0036]).
counting one step for a portion of the waveform ([Abstract], [0002]-[0005], [0013]-[0022], [0024]-[0032], [0034]-[0037], [0045], [0047]-[0051], [Claim 1], [Claim 3], [Claim 5], [Claim 8], [Claim 9], [Claim 10], [Claim 12], [Claim 14], [Claim 17], & [Claim 18]: teaches logging exactly one step for a validated waveform event).
It would have been obvious to one of ordinary skill in the art before the effective filing date to integrate the threshold-bounded timing intervals of Yang into the downstairs detection algorithm of Barfield and Alessi to perform counting one step for a portion of the waveform. Barfield tacks total accumulated steps based on periodic thresholds. Yang teaches counting exactly one step per detected waveform event when a specific movement state is verified by the axis signal. Incorporating Yang’s direct one-to-one step counting correlation for a validated state into the prior combination constitutes the substitution of one known counting logic technique for another. The motivation for this modification is to improve the efficiency of the step-counting algorithm by directly assigning step values to isolated waveform portions rather than waiting for cumulative threshold batches. This predictable variation ensures the device accurately logs individual stair impacts in real-time as the user descends (KSR).
Barfield, Alessi, and Yang, are silent in regard to:
from a first timing when acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs.
However, Matsumoto, further teaches:
from a first timing when acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs (Figs. 2-3; [0022]-[0030] & [0035]: teaches in step S8 that if the time width is within the range the waveform is determined to be from walking/running and counted as a step, if not, process loops back, waveform is not counted, Fig. 2, flowchart of step count measurement process, maps to the timing boundaries and time-width range).
PNG
media_image1.png
960
632
media_image1.png
Greyscale
PNG
media_image2.png
741
431
media_image2.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Matsumoto with Barfield, Alessi, and Yang, to validate the step from a first timing crossing above a reference to a second timing crossing below it when the time width falls within a reference range. The primary combination broadly evaluates time periods. The combination of Barfield and Alessi discloses counting steps based on waveform portions and reference ranges, but does not detail counting one step for a portion of the waveform from a first timing where an acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs. Yang teaches validating and counting steps by measuring the interval from when an acceleration signal goes above a predetermined threshold (acceleration reference value) to a subsequent crossing point ([0019] & [0045]). Matsumoto teaches isolating a waveform by identifying a first timing t1 where acceleration changes from negative to positive and a second timing t2 where it changes from positive to negative, and validating the movement if the time width T falls within a predetermined range. Incorporating Matsumoto’s specific timing boundary and time-width range validation into the combined system constitutes a known technique to improve similar devices. A POSITA would be motivated to make this combination to eliminate false positive step counts caused by narrow impulse noise, establishing boundary values that filter out low-level noise, or wide unrelated movements. This adaptation predictably yields results by establishing a time-domain filter customized for the duration of a running downstairs step impact and more accurate activity tracking (KSR).
Regarding dependent claim 3, Barfield, teaches:
The measurement device according to claim 1 ([Abstract], [0010], [0028], & [0033]), wherein the one or more processors are further configured to execute the instructions to perform ([Abstract], [0010], [0014], [0021], [0023], [0025]-[0026], [0028], [0033]-[0036], [0039], & [0056]: discloses processors evaluating sensor data to classify the user’s ambulation state and determine when a walking step is occurring)
that the count value is less than the piece count threshold value ([Abstract], [0010], [0012], [0014]-[0015], [0021], [0023], [0025]-[0026], [0028], [0033]-[0036], [0039], & [0056]: teaches evaluating a total count and that running has less time between impacts, meaning a higher frequency or step count over a predetermined time. Walking has more time between impacts, resulting in a lower frequency where the count value of step crossings falls below the threshold used to trigger the running state).
Barfield, is silent in regard to:
determining that the state of the user is walking responsive to determining both:
that the amplitudes are less than the first reference value; and
However, Matsumoto, further teaches:
The Examiner is combining Barfield in view of Matsumoto by implementing Barfield, who teaches evaluating count thresholds to validate states, evaluating maximum magnitude amplitudes against high/low thresholds to determine walking ([0014]-[0015], [0021], & [0024]).
determining that the state of the user is walking responsive to determining both (Fig. 6a; [Overview], [0006]-[0008], [0015], [0020]-[0028], [0031]-[0033], [0035]-[0038], [Claim 1], [Claim 3] & [Claim 5]: teaches determining the specific sate of walking based on thresholds):
that the amplitudes are less than the first reference value (Fig. 6a; [Overview], [0006]-[0008], [0015], [0020]-[0028], [0031]-[0033], [0035]-[0038], [Claim 1], [Claim 3] & [Claim 5]: limits the walking determination to instances where the maximum amplitude is below a specific reference value)
PNG
media_image3.png
887
729
media_image3.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Barfield in view of Matsumoto to determine that the state of a user is walking responsive to determining both that the amplitudes are less than the first reference value and that the count value is less than the piece count threshold value. Barfield teaches and evaluates determining a walking state based on lower magnitudes and longer time between impacts, evaluating a step count against a piece count threshold (STEPS_THRESHOLD), to validate whether an accumulated series of impacts qualifies as a true walking state, but does not detail a strict upper-bound amplitude requirement for walking. Matsumoto teaches determining that the user’s state is walking when the maximum waveform amplitude (RACC_F_max) is less than a first reference value. Integrating Matsumoto’s strict upper-bound amplitude condition into Barfield’s threshold-based walking verification algorithm constitutes a substitution of one known classification metric for another to improve similar devices. A POSITA would be motivated to make this combination to improve the accuracy of the pedometer, classifying the user’s ambulation state by filtering out high-impact, high-frequency events such as running or stair climbing before calculating walking step counts (KSR).
Regarding dependent claim 5, Barfield, teaches:
The measurement device according to claim 1 ([Abstract], [0010], [0028], & [0033]),
Barfield, is silent in regard to:
wherein determining the reference range includes determining a lower-limit value of the reference range as a first lower-limit value
However, Matsumoto, further teaches:
wherein determining the reference range (Fig. 4; [0024]-[0028] & [0038]: describes determining a “predetermined range” for a time width T (i.e., a reference range), range is defined by a lower limit (first threshold TH_S) is dynamically set to a specific first lower-limit value (Fix 1) and an upper limit (second threshold TH_L)) includes determining a lower-limit value of the reference range ([0024]-[0028]: discloses the first threshold value TH_S functions as a lower-limit value for the valid time width T, acting as a lower boundary of acceptable range) as a first lower-limit value ([0024]-[0028] & [0038]: teaches varying the lower limit of the time width range (TH_S) based on the user’s state, setting a lower value for TH_S (63 ms, the “first fixed value Fix 1”) when the user’s state is estimated to be “running” (RACC_F_max >= Sb).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combination of Barfield and Alessi with Matsumoto to determine the reference range includes determining a lower-limit value as a first lower-limit value. Matsumoto teaches dynamically adjusting a first lower threshold boundary (TH_S) to a specific first fixed value (Fix 1) responsive to identifying a running state (Fig. 6a; [0024]-[0025]). A POSITA would be motivated to apply this dynamic lower-limit adjustment to the downstairs running state to solve the problem of false positive step counts caused by narrow impulse noise during high-impact stair descents. Adapting Matsumoto’s variable lower-limit thresholding logic into the combined pedometer system constitutes a known technique to improve similar devices. This is supported by Matsumoto’s teaching of setting a time width’s lower-limit “first threshold TH_S” to a “first fixed value Fix 1” responsive to the user’s running state, yielding predictable results (KSR).
Barfield, and Matsumoto, are silent in regard to:
responsive to determining that the user is running downstairs
However, Alessi, further teaches:
The Examiner is combining Barfield and Matsumoto in view of Alessi by implementing Matsumoto, who teaches that the lower-limit value is changed responsive to the system determining the user’s running state ([0025]).
responsive to determining that the user is running downstairs ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Barfield with Alessi to arrive to being responsive to a determination that the user is running downstairs. Barfield alters validation thresholds based on differing ambulatory states ([0021]), while Alessi identifies the specific “downstairs” state via signal patterns caused by braking ([0082]-[0083]). The benefit gained by this combination is improved efficiency and precision in activity tracking, allowing the system to distinguish downward stair descent from flat-ground movement. Incorporating Alessi’s downward step state detection into Barfield’s state-based thresholding system is a predictable variation of known activity monitoring algorithms. This motivation is supported by Barfield’s teaching of state-dependent time limits and Alessi’s teaching of downstairs characteristic patterns, yielding predictable results (KSR).
Regarding dependent claim 6, Barfield, teaches:
The measurement device according to claim 3 ([Abstract], [0010], [0028], & [0033]),
Barfield, is silent in regard to:
wherein determining the reference range includes determining a lower-limit value of the reference range as a first lower-limit value
and determining a lower-limit value of the reference range as a second lower-limit value larger than the first lower-limit value responsive to determining that the user is walking.
However, Matsumoto, further teaches:
wherein determining the reference range (Fig. 4; [0024]-[0028] & [0038]: describes determining a “predetermined range” for a time width T (i.e., a reference range), range is defined by a lower limit (first threshold TH_S) is dynamically set to a specific first lower-limit value (Fix 1) and an upper limit (second threshold TH_L)) includes determining a lower-limit value of the reference range ([0024]-[0028]: discloses the first threshold value TH_S functions as a lower-limit value for the valid time width T, acting as a lower boundary of acceptable range) as a first lower-limit value ([0024]-[0028] & [0038]: teaches varying the lower limit of the time width range (TH_S) based on the user’s state, setting a lower value for TH_S (63 ms, the “first fixed value Fix 1”) when the user’s state is estimated to be “running” (RACC_F_max >= Sb)
and determining a lower-limit value of the reference range as a second lower-limit value larger than the first lower-limit value responsive to determining that the user is walking (Fig. 6a; [0024]-[0025]: teaches increasing the lower-limit of the reference range to a larger, second fixed value when the user’s state changes to walking, notes that Fix 2 (125 ms) is larger than Fix 1 (63 ms)).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combination of Barfield and Alessi with Matsumoto to determine the reference range includes determining a lower-limit value responsive to determining that the user is running downstairs, and determining a lower-limit value of the reference range as a second lower-limit value larger than the first lower-limit value responsive to determining that the user is walking. Matsumoto teaches dynamically adjusting a time width’s first threshold lower-limit (TH_S) to a specific first fixed value (Fix 1 at 63 ms) when running, and increasing it to a second, larger fixed value (Fix 2 at 125 ms) responsive to determining the user is walking (Fig. 6a; [0024]-[0025]). A POSITA would be motivated to apply this dynamic lower-limit adjustment to the downstairs running and walking states within the primary combination to solve the problem of false positive step counts caused by narrow impulse noise during high-impact stair descents. Adapting Matsumoto’s variable lower-limit thresholding logic to match expected gait frequencies constitutes a known technique to improve similar devices. This provides the benefit of increased step-counting accuracy across varying intensities and speeds of movement, yielding predictable results (KSR).
Barfield, and Matsumoto, are silent in regard to:
responsive to determining that the user is running downstairs,
However, Alessi, further teaches:
The Examiner is combining Barfield and Matsumoto in view of Alessi by implementing Matsumoto, who teaches that the lower-limit value is changed responsive to the system determining the user’s running state ([0025]).
responsive to determining that the user is running downstairs ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]),
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Barfield with Alessi to arrive to being responsive to a determination that the user is running downstairs. Barfield alters validation thresholds based on differing ambulatory states ([0021]), while Alessi identifies the specific “downstairs” state via signal patterns caused by braking ([0082]-[0083]). The benefit gained by this combination is improved efficiency and precision in activity tracking, allowing the system to distinguish downward stair descent from flat-ground movement. Incorporating Alessi’s downward step state detection into Barfield’s state-based thresholding system is a predictable variation of known activity monitoring algorithms. This motivation is supported by Barfield’s teaching of state-dependent time limits and Alessi’s teaching of downstairs characteristic patterns, improving efficiency and precision when classifying vertical versus horizontal movements, yielding predictable results (KSR).
Regarding dependent claim 7, Barfield, teaches:
The measurement device according to claim 1 ([Abstract], [0010], [0028], & [0033]),
Barfield, is silent in regard to:
wherein the one or more processors further determine whether an amplitude of the portion of the waveform is equal to or more than an amplitude threshold value, and,
in the counting, count one step for the portion of the waveform when the time width is determined as falling within the reference range and amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold value.
However, Matsumoto, further teaches:
wherein the one or more processors further determine whether an amplitude of the portion of the waveform is equal to or more than an amplitude threshold value (Figs. 2 & 5-6; [0011], [0025]-[0029], & [0040]: CPU 11 performs this function, in step S7, determines whether the amplitude of the waveform between the first and second timings (partial waveform) is equal to or more than a predetermined threshold (third threshold TH_A), Fig. 5 depicts the amplitude threshold concept), and,
in the counting, count one step for the portion of the waveform ([0025]-[0029]: establishes that the step counting process first requires the time width of the waveform portion to fall within a defined reference range, step S8 of the flowchart) when the time width is determined as falling within the reference range ([0025]-[0029]: step S6 is a required condition that checks if the first width is within a predetermined range) and amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold value (Figs. 2 & 6b; [0025]-[0029] & [0040]: CPU 11 performs this function, if both the time width check (S6) where time width (T) is determined to fall within the predetermined range, and the amplitude check (S7), where the amplitude of the waveform is determined to be equal to or more than the amplitude threshold TH_A, proceed to the counting step (S8), requires both conditions to be satisfied before counting the step).
PNG
media_image4.png
711
977
media_image4.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the amplitude and time-width validation sequence of Matsumoto into the step-counting measurement device of Barfield, Alessi, and yang. The combination of Barfield, Alessi, and yang teaches a step-counting device that evaluates waveform time widths and amplitudes but do not detail determining whether an amplitude of the portion of the waveform is equal to or more than an amplitude threshold, and counting one step when the time width falls within the reference range and the amplitude is equal to or more than the threshold value. Matsumoto teaches a step-counting processor that determines whether a time width is within a specific range (Step S6) and subsequently determines whether the amplitude of that portion of the waveform is equal to or more than an amplitude threshold value (Step S7, third threshold TH_A). Furthermore, Matsumoto teaches that in the counting phase, the system counts one step for the portion of the waveform only when both the time width is determined as falling within the reference range and the amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold (Step S8, [0028]). The benefit gained by this combination is improved step counting accuracy by reliably filtering out low-amplitude noise or environmental vibrations that match the expected human step time width. This combination represents a substitution of one known waveform validation logic sequence for another, yielding a predictable variation of known techniques to improve similar devices (KSR).
Regarding dependent claim 8, Barfield, teaches:
The measurement device according to claim 7 ([Abstract], [0010], [0028], & [0033]),
Barfield, is silent in regard to:
wherein the one or more processors further determine the amplitude threshold value based on amplitude of the waveform.
However, Matsumoto, further teaches:
wherein the one or more processors further determine the amplitude threshold value (Figs. 1, 2, & 6b; [0011], [0025]-[0029], & [0040]: CPU 11 performs this function, determines and varies the value of the amplitude threshold (third threshold TH_A), based on the actual amplitude variation (maximum and minimum difference) of the recent acceleration waveform) based on amplitude of the waveform (Figs. 6a-6b; [0025]-[0029] & [0040]: figure further illustrates the threshold TH_A being determined based on RACC_F_max-min, which is a measure of the amplitude of the recent acceleration waveform, where CPU 11 varies the amplitude threshold TH_A based on RACC_F_max-min (measure of the amplitude of the acceleration waveform), which is the difference between the maximum and minimum values of the acceleration waveform RACC_F over a recent period (e.g., the last 1 second), Fig. 6a established TH_A as the amplitude threshold value on the Y-axis used to evaluate the partial waveform, Fig. 6b defines the mathematical/logical relationship where TH_A is a function of RACC_F_max-min (amplitude of the acceleration waveform on the X-axis)).
It would have been obvious to one of ordinary skill in the art before the effective filing date to incorporate the dynamic amplitude threshold calculation of Matsumoto into the step-counting measurement device of Barfield, Alessi, and Yang to perform determining the amplitude threshold value based on amplitude of the waveform. The combination of Barfield, Alessi, and Yang discloses a measurement device that counts steps using static amplitude thresholds but does not detail wherein the one or more processors further determine the amplitude threshold value based on amplitude of the waveform. Matsumoto teaches a step-counting processor (CPU 11) that dynamically determines and varies an amplitude threshold value (third threshold TH_A) based on the amplitude of the acceleration waveform (the difference between the maximum and minimum values, RACC_F_max-min) over the most recent one-second interval ([0027]). The benefit gained by this integration is improved efficiency and accuracy in step counting, as an adaptive threshold reduces false positive recognitions of steps by dynamically adjusting to the user’s current movement intensity. This adaptation constitutes a substitution of a dynamically calculated threshold for a static one, which is a predictable variation of known techniques to improve similar devices (KSR).
Regarding independent claim 9, Barfield, teaches:
A method of measurement executed by a measurement device ([Abstract], [0010], [0028], & [0033]: teaches a measurement device (e.g., pedometer) equipped with the hardware components – an acceleration sensor, a memory, and a processor executing instructions) that includes an acceleration sensor ([Abstract], [0010], [0028], & [0033]), the method comprising:
acquiring time-series acceleration data output by the acceleration sensor ([Abstract], [0010], [0014], [0018], [0025], [0027]-[0034], & [0036]-[0038]: processor continuously gathers sampled data points over time from the accelerometer, which constitutes time-series acceleration data);
determining one or more amplitudes of a waveform corresponding to the time-series acceleration data ([Abstract], [0009]-[0014], [0018], [0021], & [0024]-[0038]: calculating the maximum magnitudes of the sampled signals corresponds to determining the amplitude peaks of the acceleration waveform);
determining a count value of zero-crossing points within a predetermined time ([Abstract], [0010], [0012], [0014]-[0016], [0018], [0020], [0023]-[0025], [0029], [0031]-[0032], & [0036]-[0038]: system processes the waveform to find specific zero-crossing points, tracking how many of the step events occur within a preset time period);
Barfield, is silent in regard to:
determining that a state of a user wearing the acceleration sensor is running downstairs responsive to determining both:
that the amplitudes are less than a first reference value; and
that the count value is equal to or greater than a piece count threshold value; and
However, Alessi, further teaches:
The Examiner is combining Barfield in view of Alessi by implementing that Barfield assesses user states by checking if amplitudes are under a reference a threshold and if step counts meet a piece-count threshold ([0011]-[0015], [0018], [0020]-[0021], [0024], & [0032]).
determining that a state of a user wearing the acceleration sensor is running downstairs responsive to determining both ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]: applies state detection to identify a downstairs/descending state using waveform peak amplitudes):
that the amplitudes are less than a first reference value ([0012]-[0015], [0050]-[0052], [0056]-[0057], [0067], [0073], [0075]-[0077], & [0080]-[0081]); and
that the count value is equal to or greater than a piece count threshold value ([0012]-[0015], [0046]-[0052], [0071], [0075], [0077], [0083], [Claim 6], [Claim 7], [Claim 16], [Claim 17], [Claim 24], [Claim 25], & [Claim 28]); and
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the amplitude and piece count threshold logic of Barfield with the stair-descent state detection parameters of Alessi to arrive at determining that a state of a user wearing the acceleration is running downstairs responsive to determining both: that the amplitudes are less than a first reference value; and that the count value is equal to one or greater than a piece count threshold value. Barfield teaches determining a user’s running state by checking if amplitudes are less than a reference value and count values exceed a threshold (PRE_STEP_COUNT), but does not detail applying this to the specific state of a user running downstairs. Alessi teaches analyzing waveform amplitudes and feature patterns to determine when the state of a user is descending or going downstairs. The problem being solved by this combination is improving user state classification accuracy to enable reliable tracking of elevation changes. This combination represents a predictable variation of known techniques to improve similar devices, as modifying algorithmic threshold parameters to recognize descending activity is a standard practice in wearable sensor programming, and yielding an adaptation of dynamic thresholds based on the user’s environmental trajectory (KSR).
However, Yang, further teaches:
The Examiner is combining Barfield and Alessi, in view of Yang, implementing that Barfield measures the time width between crossings to validate steps ([0010]-[0011], [0014], [0029], [0032], & [0036]).
counting one step for a portion of the waveform ([Abstract], [0002]-[0005], [0013]-[0022], [0024]-[0032], [0034]-[0037], [0045], [0047]-[0051], [Claim 1], [Claim 3], [Claim 5], [Claim 8], [Claim 9], [Claim 10], [Claim 12], [Claim 14], [Claim 17], & [Claim 18]: teaches logging exactly one step for a validated waveform event).
It would have been obvious to one of ordinary skill in the art before the effective filing date to integrate the threshold-bounded timing intervals of Yang into the downstairs detection algorithm of Barfield and Alessi to perform counting one step for a portion of the waveform. Barfield tacks total accumulated steps based on periodic thresholds. Yang teaches counting exactly one step per detected waveform event when a specific movement state is verified by the axis signal. Incorporating Yang’s direct one-to-one step counting correlation for a validated state into the prior combination constitutes the substitution of one known counting logic technique for another. The motivation for this modification is to improve the efficiency of the step-counting algorithm by directly assigning step values to isolated waveform portions rather than waiting for cumulative threshold batches. This predictable variation ensures the device accurately logs individual stair impacts in real-time as the user descends (KSR).
Barfield, Alessi, and Yang, are silent in regard to:
from a first timing where acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs.
However, Matsumoto, further teaches:
from a first timing when acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs (Figs. 2-3; [0022]-[0030] & [0035]: teaches in step S8 that if the time width is within the range the waveform is determined to be from walking/running and counted as a step, if not, process loops back, waveform is not counted, Fig. 2, flowchart of step count measurement process, maps to the timing boundaries and time-width range).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Matsumoto with Barfield, Alessi, and Yang, to validate the step from a first timing crossing above a reference to a second timing crossing below it when the time width falls within a reference range. The primary combination broadly evaluates time periods. The combination of Barfield and Alessi discloses counting steps based on waveform portions and reference ranges, but does not detail counting one step for a portion of the waveform from a first timing where an acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs. Yang teaches validating and counting steps by measuring the interval from when an acceleration signal goes above a predetermined threshold (acceleration reference value) to a subsequent crossing point ([0019] & [0045]). Matsumoto teaches isolating a waveform by identifying a first timing t1 where acceleration changes from negative to positive and a second timing t2 where it changes from positive to negative, and validating the movement if the time width T falls within a predetermined range. Incorporating Matsumoto’s specific timing boundary and time-width range validation into the combined system constitutes a known technique to improve similar devices. A POSITA would be motivated to make this combination to eliminate false positive step counts caused by narrow impulse noise, establishing boundary values that filter out low-level noise, or wide unrelated movements that generate improper time widths. This adaptation predictably yields results by establishing a time-domain filter customized for the duration of a running downstairs step impact and more accurate activity tracking, with a predictable variation of known signal thresholding techniques to yield accurate and reliable motion recognition in measurement devices (KSR).
Regarding dependent claim 11, Barfield, teaches:
The method according to claim 9 ([Abstract], [0010], [0028], & [0033]), wherein determining the state of the user includes determining the state of the user to be a walking state ([Abstract], [0010]-[0015], [0017]-[0018], [0020]-[0024], [0026]-[0028], [0031]-[0032], [0036], & [0038]-[0039]: discloses analyzing the acquired sensor data to classify the user’s repetitive ambulatory activity and determine walking states) and the count value is less than the piece count threshold value ([Abstract], [0010], [0012], [0014]-[0015], [0021], [0023], [0025]-[0026], [0028], [0033]-[0036], [0039], & [0056]: teaches maintaining a pre-step count variable and comparing it against a defined step threshold, dictating that the method routes back to data processing if the count is less than (does not equal or exceed) the threshold).
Barfield, is silent in regard to:
when the amplitude of the waveform is less than the first reference value
However, Matsumoto, further teaches:
when the amplitude of the waveform is less than the first reference value ([0024]-[0025]: teaches establishing that the user is in a walking state based on the condition that the maximum acceleration amplitude is less than a designated first reference threshold).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Barfield’s piece count threshold logic with Matsumoto’s specific amplitude thresholding criteria for determining a walking state. Barfield teaches a method of measurement executed by a device that evaluates step counts against piece count threshold values to filter out false step detections. Barfield does not detail determining the state of the user to be a walking state when the amplitude of the waveform is less than a first reference value combined with the count value constraint. Matsumoto discloses a system that determines the user is in a walking state when the maximum amplitude value is less than a defined first reference value. The motivation to combine these references is to improve the efficiency and accuracy of ambulatory motion detection by providing a framework that accommodates steady-state walking while avoiding false positives from non-ambulatory noise. This combination represents a substitution of known adaptive threshold techniques to yield predictable results in motion recognition measurement devices (KSR).
Regarding dependent claim 13, Barfield, teaches:
The method according to claim 9 ([Abstract], [0010], [0028], & [0033]),
Barfield, is silent in regard to:
wherein determining the reference range includes determining a lower-limit value of the reference range as a first lower-limit value
However, Matsumoto, further teaches:
wherein determining the reference range (Fig. 4; [0024]-[0028] & [0038]: describes determining a “predetermined range” for a time width T (i.e., a reference range), range is defined by a lower limit (first threshold TH_S) is dynamically set to a specific first lower-limit value (Fix 1) and an upper limit (second threshold TH_L)) includes determining a lower-limit value of the reference range ([0024]-[0028]: discloses the first threshold value TH_S functions as a lower-limit value for the valid time width T, acting as a lower boundary of acceptable range) as a first lower-limit value ([0024]-[0028] & [0038]: teaches varying the lower limit of the time width range (TH_S) based on the user’s state, setting a lower value for TH_S (63 ms, the “first fixed value Fix 1”) when the user’s state is estimated to be “running” (RACC_F_max >= Sb).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the combination of Barfield and Alessi with Matsumoto to determine the reference range includes determining a lower-limit value as a first lower-limit value. Matsumoto teaches dynamically adjusting a first lower threshold boundary (TH_S) to a specific first fixed value (Fix 1) responsive to identifying a running state (Fig. 6a; [0024]-[0025]). A POSITA would be motivated to apply this dynamic lower-limit adjustment to the downstairs running state to solve the problem of false positive step counts caused by narrow impulse noise during high-impact stair descents. Adapting Matsumoto’s variable lower-limit thresholding logic into the combined pedometer system constitutes a known technique to improve similar devices. This is supported by Matsumoto’s teaching of setting a time width’s lower-limit “first threshold TH_S” to a “first fixed value Fix 1” responsive to the user’s running state, yielding predictable results (KSR).
Barfield, and Matsumoto, are silent in regard to:
when the user is determined to be running downstairs.
However, Alessi, further teaches:
The Examiner is combining Barfield and Matsumoto in view of Alessi by implementing Matsumoto, who teaches that the lower-limit value is changed responsive to the system determining the user’s running state ([0025]).
when the user is determined to be running downstairs ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Barfield with Alessi to arrive to being responsive to a determination that the user is running downstairs. Barfield alters validation thresholds based on differing ambulatory states ([0021]), while Alessi identifies the specific “downstairs” state via signal patterns caused by braking ([0082]-[0083]). The benefit gained by this combination is improved efficiency and precision in activity tracking, allowing the system to distinguish downward stair descent from flat-ground movement. Incorporating Alessi’s downward step state detection into Barfield’s state-based thresholding system is a predictable variation of known activity monitoring algorithms. This motivation is supported by Barfield’s teaching of state-dependent time limits and Alessi’s teaching of downstairs characteristic patterns, yielding predictable results (KSR).
Regarding dependent claim 14, Barfield, teaches:
The method according to claim 9 ([Abstract], [0010], [0028], & [0033]), further comprising:
Barfield, is silent in regard to:
determining whether amplitude of the portion of the waveform is equal to or more than an amplitude threshold value; and
counting one step for the portion of the waveform when the time width is determined as falling within the reference range and amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold value.
However, Matsumoto, further teaches:
determining whether amplitude of the portion of the waveform is equal to or more than an amplitude threshold value (Figs. 2 & 5-6; [0011], [0025]-[0029], & [0040]: CPU 11 performs this function, in step S7, determines whether the amplitude of the waveform between the first and second timings (partial waveform) is equal to or more than a predetermined threshold (third threshold TH_A), Fig. 5 depicts the amplitude threshold concept); and
counting one step for the portion of the waveform ([0025]-[0029]: establishes that the step counting process first requires the time width of the waveform portion to fall within a defined reference range, step S8 of the flowchart) when the time width is determined as falling within the reference range ([0025]-[0029]: step S6 is a required condition that checks if the first width is within a predetermined range) and amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold value (Figs. 2 & 6b; [0025]-[0029] & [0040]: CPU 11 performs this function, if both the time width check (S6) where time width (T) is determined to fall within the predetermined range, and the amplitude check (S7), where the amplitude of the waveform is determined to be equal to or more than the amplitude threshold TH_A, proceed to the counting step (S8), requires both conditions to be satisfied before counting the step).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the amplitude and time-width validation sequence of Matsumoto into the step-counting measurement device of Barfield, Alessi, and Yang. The combination of Barfield, Alessi, and Yang teaches a step-counting device that evaluates waveform time widths and amplitudes but do not detail determining whether an amplitude of the portion of the waveform is equal to or more than an amplitude threshold, and counting one step when the time width falls within the reference range and the amplitude is equal to or more than the threshold value. Matsumoto teaches a sequential process that determines whether a time width is within a specified range (Step S6) and subsequently determines whether the amplitude of that portion of the waveform is equal to or more than an amplitude threshold value (Step S7, third threshold TH_A). Furthermore, Matsumoto teaches that in the counting phase, the system counts one step for the portion of the waveform only when both the time width is determined as falling within the reference range and the amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold (Step S8, [0028]). The benefit gained by this combination is improved step counting accuracy by reliably filtering out low-amplitude noise or environmental vibrations that match the expected human step time width. This combination represents a substitution of one known waveform validation logic sequence for another, yielding a predictable variation of known techniques to improve similar devices to yield accurate motion recognition systems (KSR).
Regarding dependent claim 21, Barfield, teaches:
The measurement device according to claim 3 ([Abstract], [0010], [0028], & [0033]),
when the amplitude of the waveform is less than the first reference value and the count value is less than the piece count threshold value ([0011]-[0012], [0016], [0018]-[0020], [0023], [Claim 1], [Claim 3], [Claim 10], & [Claim 12]: check if the maximum filtered magnitude (amplitude) falls between limits (less than a high threshold/first reference value) and conditionally compares a pre-step count value to a step threshold before advancing the total count)
Barfield, is silent in regard to:
wherein determining the state of the user includes determining the user to be walking
or when the amplitude of the user is less than a second reference value.
However, Matsumoto, further teaches:
The Examiner is combining Barfield in view of Matsumoto by implementing Barfield, who teaches analyzing accelerometer signals to classify and detect walking steps ([0026]).
wherein determining the state of the user (Figs. 3b & 4; ([0006]-[0008], [0011]-[0012], [0015], [0020]-[0028], [0031]-[0033] & [0035]-[0038]) includes determining the user to be walking (Figs. 2 & 6a; [Overview], [0006]-[0008], [0015], [0020]-[0028], [0031]-[0033], [0035]-[0038], [Claim 1], [Claim 3] & [Claim 5]: discloses a device that determines a user’s state (walking/running) to count steps, Fig. 2 illustrates a flow chart for the step count measurement process)
or when the amplitude of the user is less than a second reference value ([0025]: discloses the alternative logic of determining that the user is walking simply when the maximum amplitude values is less than a specific reference threshold).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Barfield’s step counting logic with Matsumoto’s alternative amplitude thresholding criteria for determining a walking state. Barfield teaches a measurement device that determines walking steps based on checking if an amplitude is below a maximum limit and a step count is less than a threshold. Barfield does disclose the alternative logical limitation of determining the user to be walking simply when the amplitude of the waveform is less than a second reference value. Matsumoto discloses a system that determines the user is walking when the maximum amplitude value (RACCC_F_max) is less than a defined reference value (SA). The motivation to combine these references is to improve the efficiency and accuracy of ambulatory motion detection by providing a robust, multi-condition thresholding framework that accommodates both steady-state walking and transitions from slower movements. This combination further represents a substitution of known adaptive threshold techniques to yield predictable results in motion recognition devices (KSR).
Claims 4 & 12 are rejected under 35 U.S.C. 103 as being unpatentable over Barfield, in view of Alessi, in view of Yang, in view of Matsumoto, and further in view of Unuma et al. (US 6941239 B2, Pat. Date Sep. 6, 2005, hereinafter, Unuma).
Regarding dependent claim 4, Barfield, teaches:
The measurement device according to claim 1 ([Abstract], [0010], [0028], & [0033]),
Barfield, is silent in regard to:
wherein determining that that the state of the user is running downstairs
However, Alessi, further teaches:
The Examiner is combining Barfield in view of Alessi by implementing Barfield, who teaches a measurement device determining ambulatory states ([0010]).
wherein determining that that the state of the user is running downstairs ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]: determines the state of going downstairs based on specific signal patterns)
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the amplitude and piece count threshold logic of Barfield with the stair-descent state detection parameters of Alessi to arrive at determining that the state of the user wearing the acceleration sensor is running downstairs. Barfield teaches determining a user’s running state by checking if amplitudes are less than a reference value and count values exceed a threshold (PRE_STEP_COUNT), classifying general ambulator states based on amplitude limits, but does not detail applying this to the specific state of a user running downstairs. Alessi distinguishes the specific state of stepping down, teaches analyzing waveform amplitudes and feature patterns to determine when the state of a user is descending or going downstairs, caused by the user’s natural braking movement against gravity. A POSITA would be motivated to incorporate Alessi’s downward step detection into Barfield’s state determination algorithm as a predictable variation of known classification methods. This combination solves the problem of inaccurate activity tracking during elevation changes, improving user state classification accuracy to enable reliable tracking of elevation changes, yielding the benefit of a robust pedometer capable of isolating stair descent from flat-ground ambulation, as modifying algorithmic threshold parameters to recognize descending activity is a standard practice in wearable sensor programming (KSR).
Barfield, and Alessi, are silent in regard to:
is further responsive to determining that the amplitudes are equal to or greater than a second reference value smaller than the first reference value.
However, Unuma, further teaches:
The Examiner is combining Barfield in view of Unuma by implementing Barfield, who determines states by evaluating if a “maximum magnitude falls between a low threshold and a high limit” (Fig. 1; [0011]-[0015] & [0024]-[0025]).
is further responsive to determining that the amplitudes are equal to or greater than a second reference value smaller than the first reference value ([Col. 18, ll. 45-52], [Col. 31, ll. 28-67], [Col. 32, ll. 1-65], [Col. 33, ll. 6-18], [Col. 35, ll. 66-67] & [Col. 36, ll. 1-31]: defines tracking mobility states bounded by a second reference value (lower threshold) and a first reference value (higher threshold), defines a “running” state using a performance function Wrun (S:0.5,0.8), where parameter a is a lower threshold (0.5G) and b is an upper threshold (0.8G), state of running is recognized when the amplitude S is greater than b (0.8G), with a likelihood of increasing linearly from a to b (running), also teaches concept of using a first reference value (b=0.8G) and a second, smaller reference value (a=0.5G) to define a range for a specific motion state (running) teaches recognizing a state where the power “exceeds the value 0.5G” (second reference) and evaluates it up as the power “increases to 0.8G” (first reference) requiring the amplitude to sit within this bounded range to trigger the specific classification).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify the combination of Barfield and Alessi with Unuma, to teach the state determination is further responsive to determining that the amplitudes are equal to or greater than a second reference value smaller than the first reference value. Barfield teaches evaluating if an amplitude falls between a low threshold and a high limit. Unuma teaches classifying specific ambulatory states by bounding them between a precise lower second reference value and a higher first reference value, such as bounding a state between 0.5 G and 0.8 G, combines characteristics such as amplitude, frequency, and step-derived timing, to recognize “more complicated motions”, allowing the system to determine a user’s state (e.g., running walking) based on the amplitude of an acceleration waveform, and using specific amplitude threshold values for the determination (e.g., 0.5 G and 0.8 G), and also based on step count or gait cycle frequency, from which a step count threshold can be derived. Incorporating Unuma’s strict bounded-range classification logic into the primary combination constitutes the substitution of one known thresholding technique for another to improve similar devices. A POSITA would be motivated to apply this bounded logic to the running downstairs state to solve the problem of false positive classifications, ensuring that overlapping intensity signals from different activities are compartmentalized and yield predictable results (KSR).
Regarding dependent claim 12, Barfield, teaches:
The method according to claim 9 ([Abstract], [0010], [0028], & [0033]),
Barfield, is silent in regard to:
wherein determining that that the state of the user is running downstairs
However, Alessi, further teaches:
The Examiner is combining Barfield in view of Alessi by implementing Barfield, who teaches a measurement device determining ambulatory states ([0010]).
wherein determining that that the state of the user is running downstairs ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]: determines the state of going downstairs based on specific signal patterns)
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the amplitude and piece count threshold logic of Barfield with the stair-descent state detection parameters of Alessi to arrive at determining that the state of the user wearing the acceleration sensor is running downstairs. Barfield teaches determining a user’s running state by checking if amplitudes are less than a reference value and count values exceed a threshold (PRE_STEP_COUNT), classifying general ambulator states based on amplitude limits, but does not detail applying this to the specific state of a user running downstairs. Alessi distinguishes the specific state of stepping down, teaches analyzing waveform amplitudes and feature patterns to determine when the state of a user is descending or going downstairs, caused by the user’s natural braking movement against gravity. A POSITA would be motivated to incorporate Alessi’s downward step detection into Barfield’s state determination algorithm as a predictable variation of known classification methods. This combination solves the problem of inaccurate activity tracking during elevation changes, improving user state classification accuracy to enable reliable tracking of elevation changes, yielding the benefit of a robust pedometer capable of isolating stair descent from flat-ground ambulation, as modifying algorithmic threshold parameters to recognize descending activity is a standard practice in wearable sensor programming (KSR).
Barfield, and Alessi, are silent in regard to:
is further responsive to determining that the amplitudes are equal to or greater than a second reference value smaller than the first reference value.
However, Unuma, further teaches:
The Examiner is combining Barfield in view of Unuma by implementing Barfield, who determines states by evaluating if a “maximum magnitude falls between a low threshold and a high limit” (Fig. 1; [0011]-[0015] & [0024]-[0025]).
is further responsive to determining that the amplitudes are equal to or greater than a second reference value smaller than the first reference value ([Col. 18, ll. 45-52], [Col. 31, ll. 28-67], [Col. 32, ll. 1-65], [Col. 33, ll. 6-18], [Col. 35, ll. 66-67] & [Col. 36, ll. 1-31]: defines tracking mobility states bounded by a second reference value (lower threshold) and a first reference value (higher threshold), defines a “running” state using a performance function Wrun (S:0.5,0.8), where parameter a is a lower threshold (0.5G) and b is an upper threshold (0.8G), state of running is recognized when the amplitude S is greater than b (0.8G), with a likelihood of increasing linearly from a to b (running), also teaches concept of using a first reference value (b=0.8G) and a second, smaller reference value (a=0.5G) to define a range for a specific motion state (running) teaches recognizing a state where the power “exceeds the value 0.5G” (second reference) and evaluates it up as the power “increases to 0.8G” (first reference) requiring the amplitude to sit within this bounded range to trigger the specific classification).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify the combination of Barfield and Alessi with Unuma, to teach the state determination is further responsive to determining that the amplitudes are equal to or greater than a second reference value smaller than the first reference value. Barfield teaches evaluating if an amplitude falls between a low threshold and a high limit. Unuma teaches classifying specific ambulatory states by bounding them between a precise lower second reference value and a higher first reference value, such as bounding a state between 0.5 G and 0.8 G, combines characteristics such as amplitude, frequency, and step-derived timing, to recognize “more complicated motions”, allowing the system to determine a user’s state (e.g., running walking) based on the amplitude of an acceleration waveform, and using specific amplitude threshold values for the determination (e.g., 0.5 G and 0.8 G), and also based on step count or gait cycle frequency, from which a step count threshold can be derived. Incorporating Unuma’s strict bounded-range classification logic into the primary combination constitutes the substitution of one known thresholding technique for another to improve similar devices. A POSITA would be motivated to apply this bounded logic to the running downstairs state to solve the problem of false positive classifications, ensuring that overlapping intensity signals from different activities are compartmentalized and yield predictable results (KSR).
Claims 15, 17, & 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Alessi, in view of Barfield, and further in view of Matsumoto.
Regarding independent claim 15, Alessi, teaches:
A non-transitory computer readable recording medium ([Abstract], [0015], & [0097]: discloses storing the methods as executable instructions on a non-transitory computer-readable medium that configures the device’s processing circuitry) storing instructions executable by one or more processors to cause the processors to perform ([0015], [0097], [Claim 26], [Claim 27], [Claim 28], & [Claim 29])
determining that a state of a user wearing the acceleration sensor is running downstairs responsive to determining both ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]: applies state detection to identify a downstairs/descending state using waveform peak amplitudes):
that the amplitudes are less than a first reference value ([0012]-[0015], [0050]-[0052], [0056]-[0057], [0067], [0073], [0075]-[0077], & [0080]-[0081]); and
that the count value is equal to or greater than a piece count threshold value ([0012]-[0015], [0046]-[0052], [0071], [0075], [0077], [0083], [Claim 6], [Claim 7], [Claim 16], [Claim 17], [Claim 24], [Claim 25], & [Claim 28]); and
Alessi, is silent in regard to:
acquiring time-series acceleration data output by the acceleration sensor;
determining one or more amplitudes of a waveform corresponding to the time-series acceleration data;
determining a count value of zero-crossing points within a predetermined time;
However, Barfield, further teaches:
The Examiner is combining Barfield in view of Alessi by implementing that Alessi distinguishes the environmental state of moving downstairs (Fig. 5; [0082]) and Barfield teaches determining a running state by confirming that amplitudes fall within specific limit thresholds and that step counts exceed piece count thresholds ([0011]-[0015], [0018], [0020]-[0021], [0024], & [0032]).
acquiring time-series acceleration data output by the acceleration sensor ([Abstract], [0010], [0014], [0018], [0025], [0027]-[0034], & [0036]-[0038]: processor continuously gathers sampled data points over time from the accelerometer, which constitutes time-series acceleration data);
determining one or more amplitudes of a waveform corresponding to the time-series acceleration data ([Abstract], [0009]-[0014], [0018], [0021], & [0024]-[0038]: calculating the maximum magnitudes of the sampled signals corresponds to determining the amplitude peaks of the acceleration waveform);
determining a count value of zero-crossing points within a predetermined time ([Abstract], [0010], [0012], [0014]-[0016], [0018], [0020], [0023]-[0025], [0029], [0031]-[0032], & [0036]-[0038]: system processes the waveform to find specific zero-crossing points, tracking how many of the step events occur within a preset time period);
It would have been obvious to one of ordinary skill in the art before the effective filing date to implement Barfield’s step counting methodology as executable instructions stored on Alessi’s downstairs state detection logic. Barfield teaches the measurement methods including amplitude and piece count threshold logic but lacks a non-transitory computer-readable medium storing instructions executable by one or more processors to cause the processors to perform and the detection of moving downstairs. Alessa teaches a non-transitory computer-readable medium containing instructions executable by processors to perform measurement methods, with the stair-descent state detection parameters which detect the state of moving downstairs. Barfield teaches determining a user’s running state by checking if amplitudes are less than a reference value and count values exceed a threshold (PRE_STEP_COUNT), but does not detail applying this to the specific state of a user running downstairs. Alessi teaches analyzing waveform amplitudes and feature patterns to determine when the state of a user is descending or going downstairs. The motivation to combine these teachings is to improve the measurement device’s awareness for vertical descents, while ensuring the method can be stored and executed by modern processors. Further, the problem being solved by this combination is improving user state classification accuracy to enable reliable tracking of elevation changes. This integration represents the application of a known technique to improve similar devices, as modifying algorithmic threshold parameters to recognize descending activity is a standard practice in wearable sensor programming, and yield an adaptation of dynamic thresholds based on the user’s environmental trajectory and software execution environment (KSR).
Alessi, and Barfield, are silent in regard to:
and counting one step for a portion of the waveform from a first timing when acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs.
However, Matsumoto, further teaches:
The Examiner is combining Matsumoto in view of Alessi by implementing that Alessi provides the downstairs context (Fig. 5; [0082]).
and counting one step for a portion of the waveform ([0022]-[0024]: defines a first timing where the waveform becomes positive, a second timing where it becomes negative, and counting a step if the time width between them falls within a specific reference range) from a first timing when acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs (Figs. 2-3; [0022]-[0030] & [0035]: teaches in step S8 that if the time width is within the range the waveform is determined to be from walking/running and counted as a step, if not, process loops back, waveform is not counted, Fig. 2, flowchart of step count measurement process, maps to the timing boundaries and time-width range).
It would have been obvious to one of ordinary skill in the art before the effective filing date to implement Matsumoto’s waveform bounding and time with validation logic into the downstairs step-counting framework of Barfield and Alessi to validate the step from a first timing crossing above a reference to a second timing crossing below it when the time width falls within a reference range. The primary combination broadly evaluates time periods. The combination of Barfield and Alessi identifies downstairs movement using stored instructions, discloses counting steps based on waveform portions and reference ranges, but does not detail counting one step for a portion of the waveform bounded by a first timing where an acceleration goes above an acceleration reference value to a second timing where the acceleration goes below the acceleration reference value when a time width corresponding to the portion of the waveform is determined as falling within a reference range corresponding to running downstairs. Matsumoto teaches isolating a waveform by identifying a first timing t1 where acceleration changes from negative to positive and a second timing t2 where it changes from positive to negative, and validating the movement if the time width T falls within a predetermined range. Incorporating Matsumoto’s specific timing boundary and time-width range validation into the combined system constitutes a known technique to improve similar devices. A POSITA would be motivated to make this combination to eliminate false positive step counts caused by narrow impulse noise, establishing boundary values that filter out low-level noise, or wide unrelated movements, improving the overall accuracy of step detection by filtering out noise or non-ambulatory movements that generate improper time widths. This combination constitutes a predictable variation of known signal thresholding techniques to yield accurate and reliable motion recognition in processor-executed measurement devices, by establishing a time-domain filter customized for the duration of a running downstairs step impact and more accurate activity tracking (KSR).
Regarding dependent claim 17, Alessi, teaches:
The non-transitory computer readable recording medium according to claim 15 ([Abstract], [0015], & [0097]: discloses storing the methods as executable instructions on a non-transitory computer-readable medium that configures the device’s processing circuitry),
Alessi, is silent in regard to:
wherein determining the state of the user includes determining the state of the user to be a walking state and the count value is less than the piece count threshold value.
However, Barfield, further teaches:
wherein determining the state of the user includes determining the state of the user to be a walking state ([Abstract], [0010]-[0015], [0017]-[0018], [0020]-[0024], [0026]-[0028], [0031]-[0032], [0036], & [0038]-[0039]: discloses analyzing the acquired sensor data to classify the user’s repetitive ambulatory activity and determine walking states) and the count value is less than the piece count threshold value ([Abstract], [0010], [0012], [0014]-[0015], [0021], [0023], [0025]-[0026], [0028], [0033]-[0036], [0039], & [0056]: teaches maintaining a pre-step count variable and comparing it against a defined step threshold, dictating that the method routes back to data processing if the count is less than (does not equal or exceed) the threshold).
It would have been obvious to one of ordinary skill in the art before the effective filing date to implement Barfield’s step count threshold logic into the computer-readable medium of Alessi. Alessi teaches a non-transitory computer readable recording medium for evaluating user states, but does not disclose determining the state of the user to be a walking state when the count value is less than the piece count threshold value. Barfield teaches a method of measurement executed by a device that evaluates step counts against piece count threshold values to filter out false step detections, maintaining a pre-step count value and routing the process back if this count value is less than a predetermined piece count threshold value to ensure proper walking step detection. The motivation to combine these references is to improve the efficiency and accuracy of the measurement device by filtering out false positive step registrations caused by short, non-ambulatory movements and a substitution of known adaptive threshold techniques to yield predictable results in motion recognition measurement devices (KSR).
However, Matsumoto, further teaches:
when the amplitude of the waveform is less than the first reference value ([0024]-[0025]: teaches establishing that the user is in a walking state based on the condition that the maximum acceleration amplitude is less than a designated first reference threshold).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the computer-readable medium and piece-count thresholding logic of Alessi and Barfield with Matsumoto’s specific amplitude thresholding criteria for determining a walking state. Alessi and Barfield disclose a computer-readable medium executing measurement methods executed by a device that evaluates step counts against piece count threshold values to filter out false step detections. Alessi and Barfield do not detail determining the state of the user to be a walking state when the amplitude of the waveform is less than a first reference value combined with the count value constraint. Matsumoto discloses a system that determines the user is in a walking state when the maximum amplitude value is less than a defined first reference value. The motivation to combine these references is to improve the efficiency and accuracy of ambulatory motion detection by providing a framework that accommodates and identifies steady-state walking while filtering out false positives or high-impact artifacts from non-ambulatory noise. This combination represents the application of a known technique to improve similar devices by integrating reliable amplitude condition logic to yield predictable results in motion recognition software (KSR).
Regarding dependent claim 19, Alessi, teaches:
The non-transitory computer readable recording medium according to claim 15 ([Abstract], [0015], & [0097]: discloses storing the methods as executable instructions on a non-transitory computer-readable medium that configures the device’s processing circuitry),
The Examiner is combining Alessi in view of Matsumoto by implementing Matsumoto, who teaches modifying the lower-limit threshold value is changed responsive to the system determining the user’s running state ([0025]).
when the user is determined to be running downstairs ([0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]).
Alessi, is silent in regard to:
wherein determining the reference range includes determining a lower-limit value of the reference range as a first lower-limit value
However, Matsumoto, further teaches:
wherein determining the reference range (Fig. 4; [0024]-[0028] & [0038]: describes determining a “predetermined range” for a time width T (i.e., a reference range), range is defined by a lower limit (first threshold TH_S) is dynamically set to a specific first lower-limit value (Fix 1) and an upper limit (second threshold TH_L)) includes determining a lower-limit value of the reference range ([0024]-[0028]: discloses the first threshold value TH_S functions as a lower-limit value for the valid time width T, acting as a lower boundary of acceptable range) as a first lower-limit value ([0024]-[0028] & [0038]: teaches varying the lower limit of the time width range (TH_S) based on the user’s state, setting a lower value for TH_S (63 ms, the “first fixed value Fix 1”) when the user’s state is estimated to be “running” (RACC_F_max >= Sb)
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Matsumoto’s dynamic lower-limit reference range adjustment for running with Alessi’s non-transitory computer-readable medium and downstairs detection context . Alessi teaches a non-transitory computer readable recording-medium configuring a device to detect the state of a user going downstairs, but does not detail determining a reference range by setting a lower-limit value to a first lower-limit value when the user is running downstairs. Matsumoto teaches dynamically adjusting a first lower reference range by determining a lower limit threshold value (TH_S) to be a first specific first fixed value (Fix 1) when it is determined that the user is running (Fig. 6a; [0024]-[0025]). A POSITA would be motivated to apply this dynamic lower-limit adjustment to the downstairs running state to solve the problem of false positive step counts caused by narrow impulse noise during high-impact stair descents. Further, this combination would improve the measurement device’s efficiency and accuracy by adapting signal evaluation thresholds to the high-intensity forces of descending steps at a running pace. By adapting Matsumoto’s variable lower-limit thresholding logic into the combined pedometer system, which constitutes a known technique to improve similar devices. This is supported by Matsumoto’s teaching of setting a time width’s lower-limit “first threshold TH_S” to a “first fixed value Fix 1” responsive to the user’s running state. This combination constitutes the application of a known technique, yielding predictable results in dynamically configuring ambulatory motion recognition software (KSR).
Regarding dependent claim 20, Alessi, teaches:
The non-transitory computer readable recording medium according to claim 15 ([Abstract], [0015], & [0097]: discloses storing the methods as executable instructions on a non-transitory computer-readable medium that configures the device’s processing circuitry), the instructions further causing the one or more processors to perform ([Abstract], [0015], & [0097]):
Alessi, is silent in regard to:
determining whether amplitude of the portion of the waveform is equal to or more than an amplitude threshold value; and
counting one step for the portion of the waveform when the time width is determined as falling within the reference range and amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold value.
However, Matsumoto, further teaches:
determining whether amplitude of the portion of the waveform is equal to or more than an amplitude threshold value (Figs. 2 & 5-6; [0011], [0025]-[0029], & [0040]: CPU 11 performs this function, in step S7, determines whether the amplitude of the waveform between the first and second timings (partial waveform) is equal to or more than a predetermined threshold (third threshold TH_A), Fig. 5 depicts the amplitude threshold concept); and
counting one step for the portion of the waveform ([0025]-[0029]: establishes that the step counting process first requires the time width of the waveform portion to fall within a defined reference range, step S8 of the flowchart) when the time width is determined as falling within the reference range ([0025]-[0029]: step S6 is a required condition that checks if the first width is within a predetermined range) and amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold value (Figs. 2 & 6b; [0025]-[0029] & [0040]: CPU 11 performs this function, if both the time width check (S6) where time width (T) is determined to fall within the predetermined range, and the amplitude check (S7), where the amplitude of the waveform is determined to be equal to or more than the amplitude threshold TH_A, proceed to the counting step (S8), requires both conditions to be satisfied before counting the step).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the non-transitory computer-readable medium of Alessa with the dual-condition thresholding logic for time-width validation sequence and amplitude taught by Matsumoto. Alessi teaches a non-transitory computer-readable medium containing executable instructions for evaluating user movement states, but does not detail the instructions further causing the one or more processors to perform determining whether the amplitude of the portion of the waveform is equal to or more than an amplitude threshold value, and counting one step for the portion of the waveform specifically when the time width is determined as falling within the reference range and the amplitude is determined as being equal to or more than the amplitude threshold value. Matsumoto teaches a sequential process that evaluates whether a time width is within a specified range (Step S6) and subsequently determines whether the amplitude of that portion of the waveform is equal to or more than an amplitude threshold value (Step S7, third threshold TH_A) and that the amplitude simultaneously meets or exceeds this threshold value. Furthermore, Matsumoto teaches that in the counting phase, the system counts one step for the portion of the waveform only when both the time width is determined as falling within the reference range and the amplitude of the portion of the waveform is determined as being equal to or more than the amplitude threshold (Step S8, [0028]). The motivation to combine these teachings is to improve the accuracy and efficiency of the measurement device by filtering out false positive step registrations or low-amplitude noise or environmental vibrations that match the expected human step time width caused by irregular, non-ambulatory motions or ambient noise. This combination represents a substitution of known adaptive threshold techniques into a software-executable format, to yield predictable results in motion recognition systems (KSR).
Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Alessi, in view of Barfield, in view of Matsumoto, and further in view of Unuma.
Regarding dependent claim 18, Alessi, teaches:
The non-transitory computer readable recording medium according to claim 15 ([Abstract], [0015], & [0097]: discloses storing the methods as executable instructions on a non-transitory computer-readable medium that configures the device’s processing circuitry),
The Examiner is combining Alessi in view of Matsumoto by implementing Barfield, who teaches a measurement device determining ambulatory states ([0010]) and Matsumoto, who teaches detecting an elevated pace (running/jogging) based on threshold logic ([Overview], [0024]-[0025], [Claim 1], & [Claim 2]).
wherein determining that that the state of the user is running downstairs (Fig. 5; [0004], [0012]-[0014], [0035], [0048], [0059], [0067], [0069], [0077], [0079], [0082]-[0083], [0086], [0094]-[0095], [0099]-[0100], [Claim 1], [Claim 3], [Claim 4], [Claim 7], [Claim 11], [Claim 17], [Claim 19], [Claim 25], & [Claim 27]: determines the state of going downstairs based on specific signal patterns)
Alessi, is silent in regard to:
is further responsive to determining that the amplitudes are equal to or greater than a second reference value smaller than the first reference value.
However, Matsumoto, further teaches:
is further responsive to determining that the amplitudes are equal to or greater than a second reference value smaller than the first reference value ([0024]-[0025]: discloses bounded threshold logic, checking if the maximum amplitude is equal to or greater than a lower limit (second reference) while being less than an upper limit (first reference)).
It would have been obvious to one of ordinary skill in the art before the effective filing date to incorporate Matsumoto’s dual-bonded amplitude thresholding criteria into the downstairs state detection framework of Alessi. Alessi teaches a non-transitory computer readable recording medium that determines the state of a user moving downstairs, but does not detail determining the running downstairs state responsive to determining that the amplitudes are equal to or greater than a second reference smaller than a first reference value. Matsumoto discloses a method of determining a specific elevated movement state (e.g., walking/running) when an amplitude value (RACC_F_max) falls within a defined range, specifically being equal to or greater than a lower-limit reference value and less than a larger upper-limit reference value. The motivation to combine these teachings is to improve the accuracy and efficiency of the measurement device by categorizing different intensities of movement, such as running versus walking, preventing false positive or false negative state detections during a descent. This combination represents a substitution of known adaptive threshold techniques to yield predictable results in ambulatory motion recognition systems (KSR).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HUGO NAVARRO whose telephone number is (571)272-6122. The examiner can normally be reached Monday-Friday 08:30-5:00 pm EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Eman Alkafawi can be reached at 571-272-4448. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/HUGO NAVARRO/ Examiner, Art Unit 2858 August 15, 2026
/A.A/Primary Examiner, Art Unit 2858
Prosecution Insights