Anomalous energy detection for resource-constrained embedded systems using tracing data analysis
Bibliographic record
Abstract
Since the multi-sensor embedded systems are spread all over the urban and suburban areas, there should be a real-time performance analysis framework to monitor the energy consumption of sensors and services. To this end, applying artificial intelligence to model the behavior can play a key role in improving operational efficiency and over-consumption detection. Most of the energy over-consumption detection approaches make use of external devices such as smart plugs or designed for smart buildings rather than computing devices. Through this paper, a new architecture is proposed in which we extract real-time execution traces after identifying the suspicious running applications using the dynamic binary instrumentation technique. The output of this step is fed to a data analytic framework to detect any behavioral anomalies such as over-threshold energy consumption. Finally, depending on the decision made by the anomaly detection step, some sensors might be switched to the standby mode in order to meet the essential expectations for other sensors or services. Since we have a resource-constrained embedded system and a real-time decision-making constraint on sensor control, the amount of computation that could be done in real-time varies a lot. Accordingly, the desired performance may dictate how, when, and where the processing operations might occur (on the embedded system or on a remote server). This work is mainly focused on anomaly detection operations using tracing data analysis. The experimental results show that the over-consumption detection model achieves a high level of accuracy metrics for classifying abnormal consumption behavior.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".