Buffering strategies examined with synthetic time series

Erlangen. The ongoing project Data buffering for interrupted energy metering examines how local buffering and retransmission differ when metering devices temporarily lose their connection. After such offline phases, measurement series contain gaps, duplicates or misplaced values.

The method note Buffering rules for synthetic measurement series sets out how two buffering strategies are to be compared on synthetic time series. The comparison looks at completeness, duplicate rate, temporal assignment and memory demand under limited storage. Interruption windows are set in a controlled way so that runs remain repeatable.

In the scenario the project is not yet completed. The note describes the planned approach and reports no measurement results. All data are synthetic; this is a document. The approach rests on testing each strategy under identical conditions. Later, it should then be possible to tell which differences stem from the strategy and which from chance.