High fidelity combination of data

A system for operating on data, comprising:

• a processor configured to:
• receive a query comprising an operation to be performed on input data streams;
• parse the query to identify a predicate in the query and to determine candidates based on the query, wherein the predicate and the candidates form at least part of a basis for generating a synthetic time series;
• generate the synthetic time series for each candidate based on the query;
• receive a first input data stream having a first type, wherein the first input data stream includes first input data stream samples associated with first input data stream sample times;
• receive a second input data stream having a second type, wherein the second input data stream includes second input data stream samples associated with second input data stream sample times;
• determine whether at least some points in the second input data stream samples do not have synchronized samples in the first input data stream;
• automatically generate samples for the first input data stream that are synchronized with samples of the second input data stream;
• determine whether at least some points in the first input data stream samples do not have synchronized samples in the second input data stream;
• automatically generate samples for the second input data stream that are synchronized with samples of the first input data stream;
• execute the query by performing the operation on the synthetic time series, the second input data stream samples of the second input data stream and the first input data stream samples, and the automatically generated samples of the first input data stream;
• and
• deliver results of the operation indicating which of the input data streams most closely matches the synthetic time series as an answer to the query;
• and
• a memory coupled to the processor and configured to provide the processor with instructions.

A computer program product for operating on data, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

• receiving a query comprising an operation to be performed on input data streams;
• parsing the query to identify a predicate in the query and to determine candidates based on the query, wherein the predicate and the candidates form at least part of a basis for generating a synthetic time series;
• generating the synthetic time series for each candidate based on the query;
• receiving a first input data stream having a first type, wherein the first input data stream includes first input data stream samples associated with first input data stream sample times;
• receiving a second input data stream having a second type;
• wherein the second input data stream includes second input data stream samples associated with second input data stream sample times;
• determining that at least some points in the second input data stream samples do not have synchronized samples in the first input data stream and that at least some points in the first input data stream samples do not have synchronized samples in the second input data stream;
• automatically generating samples for the first input data stream that are synchronized with samples of the second input data stream and automatically generating samples for the second input data stream that are synchronized with samples of the first input data stream;
• executing the query by performing the operation on the synthetic time series, the second input data stream samples of the second input data stream and the first input data stream samples, and the automatically generated samples of the first input data stream;
• and
• delivering results of the operation indicating which of the input data streams most closely matches the synthetic time series as an answer to the query.