DETAILED ACTION
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 .
Claim Rejections - 35 USC § 103
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 1-15, 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over Angell et al. (US 2012/0063270) in view of Patton et al. (US 2020/0387560).
Regarding claim 1, Angell et al. (US 2012/0063270) teaches a computing system for event detection and localization (i.e., An event detection and localization system is provided that employs a plurality of smartphones [abstr.], [0006]), comprising: a computing device (smartphones or any other communication devices) processor (typically present in communication devices); and a memory device including instructions that, when executed by the computing device processor (typically present in communication devices), enables the computing system to: obtain sensor data from a plurality of devices, each device including at least one sensor (i.e., Each smartphone comprises a memory for storing an event detection process; and at least one hardware device to implement the event detection process. The hardware device is operative to detect an event based on a signal obtained using a sensor in the vicinity of the smartphone; obtain a time of arrival of the signal; obtain a location of the smartphone at the time of arrival [0006]-[0007]); from at least a subset of the plurality of devices, detect a potential event (i.e., the detected events may comprise one or more of a gunshot, a biological threat, a chemical threat and a radiological threat [0006], [0019]); dynamically form a network of devices from the subset of the plurality of devices based on a detection of the potential event (i.e., a user can optionally be alerted that an event is detected and the user can validate the detected event. The time of arrival can be obtained, for example, by starting a timer process that determines an absolute arrival time in a reference time frame. The location of the smartphone at the time of arrival can be obtained, for example, by enabling a location awareness capability of the smartphone [0006], [0026], [0031]); synchronize and correlate the sensor data across the network of devices to validate an occurrence of the potential event (i.e., the primary requirement is to begin a timer process that can then use a highly accurate and synchronized time from either a cell tower 140 or GPS time to determine the absolute arrival time in a reference frame that is the same for all the smartphones 110 in the network 100 [0029]); apply geolocation techniques to correlated data to determine a location of the potential event (i.e., when the gunshot is detected during step 320, the GPS or location awareness capability is enabled during step 340 to determine the location of the smartphone [0030]); and generate an event detection and geolocation result (i.e., Upon detection of a gunshot waveform during step 320, the smartphone gunshot detection and localization process 300 records the time of arrival of the sound wave during step 330 and records the location of the smartphone at the time of detection during step 340 [0027]-[0028]).
Angell does not specifically teach detect a potential event using a trained model applied to the sensor data, the trained model being operable to identify events based on predefined criteria.
However, the preceding limitation is known in the art of communications. Patton teaches the use machine learning to detect event wherein: each event is detected from one or more normalized signals and includes an event location, an event category, an event an event truthfulness, an event severity, and an event time. Entity notification preferences defining events relevant to an entity are accessed. The entity notification preferences include category preferences, location preferences, distance preferences, truth preferences, severity preferences, and time preferences ([0011]). A single source classifier can be configured to compute a single source probability for a single event type or to compute a single source probability for each of a plurality of different event types. A single source classifier can compute a single source probability using artificial intelligence, machine learning, neural networks, logic, heuristics, etc ([0042], [0239]-[0240]). Artificial intelligence and/or machine learning is used to determine if an event is relevant to an entity based on current location ([0279]-[0281]). Event identification module 1118 can (possibly using artificial intelligence and/or machine learning) determine that event 1324 is within distance range 1343 of entity 1321 (and otherwise satisfies preferences 1326) [0305]-[0306]). Therefore, it would have been obvious to one of ordinary skill in the art, at the time to have implemented the technique of Angell within the system of Patton in order to provide a single system that assesses the validity and severity of diverse events, reducing maintenance costs and overall complexity.
Regarding claim 2, Angell in view of Patton teaches all the limitations above. Angell further teaches wherein the plurality of devices comprises at least one of smartphones, wearables, drones, and autonomous vehicles (i.e., apparatus are provided for event detection and localization using a plurality of smartphones [0006]).
Regarding claim 3, Angell in view of Patton teaches all the limitations above. Patton further teaches wherein the trained model is a machine learning model trained on historical event data associated with events of interest, and wherein events of interest include at least auditory, environmental, or physical phenomena ([0073], [0281], [0297]-[0298]).
Regarding claim 4, Angell in view of Patton teaches all the limitations above. Patton further teaches the trained model is operable to identify events using multimodal data inputs, including audio data, accelerometer data, and barometric pressure data ([0058], [0281], [0308]).
Regarding claim 5, Angell in view of Patton teaches all the limitations above. Patton further teaches dynamically forming the network includes prioritizing devices based on at least one of proximity to the potential event, resource availability, and signal strength (i.e., Distance measurer 705 can determine a signal location of a normalized signal and measure a distance between the signal location of the normalized signal and the signal location of a previously received normalized signal (when a location of a previously received normalized signal is available) [0172]).
Regarding claim 6, Angell in view of Patton teaches all the limitations above. Angell further teaches wherein geolocation techniques include at least one of Time Difference of Arrival (TDOA) or Angle of Arrival (AoA) (i.e., obtain a time of arrival of the signal [0006], [0032]-[0033]).
Regarding claim 7, Angell in view of Patton teaches all the limitations above. Angell further teaches wherein synchronizing sensor data includes time-aligning sensor outputs using at least one synchronization mechanism (i.e., use a highly accurate and synchronized time from either a cell tower 140 or GPS time to determine the absolute arrival time in a reference frame that is the same for all the smartphones 110 in the network 100 [0022], [0029]).
Regarding claim 8, Angell in view of Patton teaches all the limitations above. Patton further teaches wherein dynamically forming the network includes selecting devices based on compatibility with event-specific data processing requirements ([0249]-]0250]).
Regarding claim 9, Angell in view of Patton teaches all the limitations above. Patton further teaches wherein the event detection and geolocation result includes at least one of event type, timestamp, geolocation coordinates, and event confidence level (i.e., The entity may desire some level of confidence in the veracity of a detected event before taking an action [0104]-[0105], [0156], [0257]).
Regarding claim 10, Angell in view of Patton teaches all the limitations above. Patton further teaches wherein correlating sensor data includes dynamically adjusting correlation thresholds based on environmental conditions, including at least one of noise levels, signal interference, or device density ([0189]).
Regarding claim 11, Angell et al. (US 2012/0063270) teaches a computer-implemented method for event detection and localization (i.e., An event detection and localization system is provided that employs a plurality of smartphones [abstr.], [0006]), comprising obtaining sensor data from a plurality of devices, each device including at least one sensor (i.e., Each smartphone comprises a memory for storing an event detection process; and at least one hardware device to implement the event detection process. The hardware device is operative to detect an event based on a signal obtained using a sensor in the vicinity of the smartphone; obtain a time of arrival of the signal; obtaining a location of the smartphone at the time of arrival [0006]-[0007]); from at least a subset of the plurality of devices, detect a potential event (i.e., the detected events may comprise one or more of a gunshot, a biological threat, a chemical threat and a radiological threat [0006], [0019]); dynamically form a network of devices from the subset of the plurality of devices based on a detection of the potential event (i.e., a user can optionally be alerted that an event is detected and the user can validate the detected event. The time of arrival can be obtained, for example, by starting a timer process that determines an absolute arrival time in a reference time frame. The location of the smartphone at the time of arrival can be obtained, for example, by enabling a location awareness capability of the smartphone [0006], [0026], [0031]); synchronizing and correlate the sensor data across the network of devices to validate an occurrence of the potential event (i.e., the primary requirement is to begin a timer process that can then use a highly accurate and synchronized time from either a cell tower 140 or GPS time to determine the absolute arrival time in a reference frame that is the same for all the smartphones 110 in the network 100 [0029]); applying geolocation techniques to correlated data to determine a location of the potential event (i.e., when the gunshot is detected during step 320, the GPS or location awareness capability is enabled during step 340 to determine the location of the smartphone [0030]); and generating an event detection and geolocation result (i.e., Upon detection of a gunshot waveform during step 320, the smartphone gunshot detection and localization process 300 records the time of arrival of the sound wave during step 330 and records the location of the smartphone at the time of detection during step 340 [0027]-[0028]).
Angell does not specifically teach detect a potential event using a trained model applied to the sensor data, the trained model being operable to identify events based on predefined criteria.
However, the preceding limitation is known in the art of communications. Patton teaches the use machine learning to detect event wherein: each event is detected from one or more normalized signals and includes an event location, an event category, an event an event truthfulness, an event severity, and an event time. Entity notification preferences defining events relevant to an entity are accessed. The entity notification preferences include category preferences, location preferences, distance preferences, truth preferences, severity preferences, and time preferences ([0011]). A single source classifier can be configured to compute a single source probability for a single event type or to compute a single source probability for each of a plurality of different event types. A single source classifier can compute a single source probability using artificial intelligence, machine learning, neural networks, logic, heuristics, etc ([0042], [0239]-[0240]). Artificial intelligence and/or machine learning is used to determine if an event is relevant to an entity based on current location ([0279]-[0281]). Event identification module 1118 can (possibly using artificial intelligence and/or machine learning) determine that event 1324 is within distance range 1343 of entity 1321 (and otherwise satisfies preferences 1326) [0305]-[0306]). Therefore, it would have been obvious to one of ordinary skill in the art, at the time to have implemented the technique of Angell within the system of Patton in order to provide a single system that assesses the validity and severity of diverse events, reducing maintenance costs and overall complexity.
Regarding claim 12, Angell in view of Patton teaches all the limitations above. Angel further teaches synchronizing sensor data across the network of devices using at least one synchronization mechanism, the synchronization mechanism including GPS-disciplined clocks or dynamic time warping (i.e., use a highly accurate and synchronized time from either a cell tower 140 or GPS time to determine the absolute arrival time in a reference frame that is the same for all the smartphones 110 in the network 100 [0029]).
Regarding claim 13, Angell in view of Patton teaches all the limitations above. Patton further teaches detecting the potential event comprises applying multimodal data inputs to a trained model, the multimodal data inputs including at least audio signals, accelerometer readings, and barometric pressure variations ([0058], [0281], [0308]).
Regarding claim 14, Angell in view of Patton teaches all the limitations above. Patton further teaches iteratively refining geolocation results by integrating additional sensor data from dynamically added devices to the network after an initial geolocation estimate (i.e., In some cases (gunshot detection), only the addition of specific smartphone applications and possibly server processing provide the complete sensor array [0018], [0022]).
Regarding claim 15, Angell in view of Patton teaches all the limitations above. Angell further teaches wherein applying geolocation techniques includes: analyzing time-stamped audio data to compute relative time delays across devices using Time Difference of Arrival (TDOA) (i.e., obtain a time of arrival of the signal [0006], [0032]-[0033]); and refining TDOA estimates using cross-correlation techniques to enhance alignment of signal waveforms ([0016], [0018]-[0019]).
Regarding claim 18, Angell et al. (US 2012/0063270) teaches a non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor of a computing system, causes the computing system to: obtain sensor data from a plurality of devices, each device including at least one sensor (i.e., Each smartphone comprises a memory for storing an event detection process; and at least one hardware device to implement the event detection process. The hardware device is operative to detect an event based on a signal obtained using a sensor in the vicinity of the smartphone; obtain a time of arrival of the signal; obtain a location of the smartphone at the time of arrival [0006]-[0007]); from at least a subset of the plurality of devices, detect a potential event (i.e., the detected events may comprise one or more of a gunshot, a biological threat, a chemical threat and a radiological threat [0006], [0019]); dynamically form a network of devices from the subset of the plurality of devices based on a detection of the potential event (i.e., a user can optionally be alerted that an event is detected and the user can validate the detected event. The time of arrival can be obtained, for example, by starting a timer process that determines an absolute arrival time in a reference time frame. The location of the smartphone at the time of arrival can be obtained, for example, by enabling a location awareness capability of the smartphone [0006], [0026], [0031]); synchronize and correlate the sensor data across the network of devices to validate an occurrence of the potential event (i.e., the primary requirement is to begin a timer process that can then use a highly accurate and synchronized time from either a cell tower 140 or GPS time to determine the absolute arrival time in a reference frame that is the same for all the smartphones 110 in the network 100 [0029]); apply geolocation techniques to correlated data to determine a location of the potential event (i.e., when the gunshot is detected during step 320, the GPS or location awareness capability is enabled during step 340 to determine the location of the smartphone [0030]); and generate an event detection and geolocation result (i.e., Upon detection of a gunshot waveform during step 320, the smartphone gunshot detection and localization process 300 records the time of arrival of the sound wave during step 330 and records the location of the smartphone at the time of detection during step 340 [0027]-[0028]).
Angell does not specifically teach detect a potential event using a trained model applied to the sensor data, the trained model being operable to identify events based on predefined criteria.
However, the preceding limitation is known in the art of communications. Patton teaches the use machine learning to detect event wherein: each event is detected from one or more normalized signals and includes an event location, an event category, an event an event truthfulness, an event severity, and an event time. Entity notification preferences defining events relevant to an entity are accessed. The entity notification preferences include category preferences, location preferences, distance preferences, truth preferences, severity preferences, and time preferences ([0011]). A single source classifier can be configured to compute a single source probability for a single event type or to compute a single source probability for each of a plurality of different event types. A single source classifier can compute a single source probability using artificial intelligence, machine learning, neural networks, logic, heuristics, etc ([0042], [0239]-[0240]). Artificial intelligence and/or machine learning is used to determine if an event is relevant to an entity based on current location ([0279]-[0281]). Event identification module 1118 can (possibly using artificial intelligence and/or machine learning) determine that event 1324 is within distance range 1343 of entity 1321 (and otherwise satisfies preferences 1326) [0305]-[0306]). Therefore, it would have been obvious to one of ordinary skill in the art, at the time to have implemented the technique of Angell within the system of Patton in order to provide a single system that assesses the validity and severity of diverse events, reducing maintenance costs and overall complexity.
Regarding claim 19, Angell in view of Patton teaches all the limitations above. Angell further teaches wherein the instructions are configured to prioritize devices for network formation based on proximity to the potential event, available resources, or compatibility with event-specific data processing requirements (i.e., detect an event based on a signal obtained using a sensor in the vicinity of said smartphone [0033], [claim 5]).
Regarding claim 20, Angell in view of Patton teaches all the limitations above. Patton further teaches wherein the instructions are configured to share geolocation processing tasks among devices in the network, reducing resource consumption on individual devices ([0045], [0063], [0089]).
Allowable Subject Matter
Claims 16-17 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JEAN ALLAND GELIN whose telephone number is (571)272-7842. The examiner can normally be reached MON-FR 9-6 PM.
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, JINSONG HU can be reached at 571-272-3965. 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.
/JEAN A GELIN/Primary Examiner, Art Unit 2643
