US 12,353,467 B2
Representation and visualization of multivariate sensory time series data
Ismail Uysal, Lakeland, FL (US); and Alla Abdella, West Hartford, CT (US)
Assigned to University of South Florida, Tampa, FL (US)
Filed by UNIVERSITY OF SOUTH FLORIDA, Tampa, FL (US)
Filed on Nov. 6, 2023, as Appl. No. 18/502,620.
Application 18/502,620 is a continuation of application No. 17/550,509, filed on Dec. 14, 2021, granted, now 11,809,475.
Claims priority of provisional application 63/199,199, filed on Dec. 14, 2020.
Claims priority of provisional application 63/159,804, filed on Mar. 11, 2021.
Prior Publication US 2024/0232244 A1, Jul. 11, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/35 (2025.01); G06F 16/353 (2025.01); G06N 20/00 (2019.01)
CPC G06F 16/353 (2019.01) [G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method for cold temperature monitoring using data from multiple and non-uniform temperature sensors, the method comprising:
receiving a first plurality of temperature data sequences from respective temperature sensors in a respective temperature-controlled shipment;
forming a first segment of the representative data sequence that is a first weighted combination of the first plurality of temperature data sequences over a first interval corresponding to a first shortest data sequence of the first plurality of temperature data sequences, wherein each temperature data sequence of the first plurality of temperature data sequences is weighted based on a plurality of first similarity values associated with the first shortest data sequence;
truncating subset of the first plurality of temperature data sequences to exclude the first shortest data sequence corresponding to the first interval to form a second plurality of temperature data sequences;
forming a second segment of the representative data sequence that is a second weighted combination of the subset over a second interval corresponding to a second shortest data sequence of the subset, wherein each of the subset is weighted based on a plurality of second similarity values associated with the second shortest data sequence;
concatenating the first segment of the representative data sequence and the second segment of the representative data sequence; and
outputting a result based on the first segment concatenated with the second segment, the result indicative of a single time series data sequence representation of time-dependent temperatures experienced by the shipment.