Leveraging machine learning for efficient mobility management and data transmission in fog computing
Bibliographic record
Abstract
Fog computing is a new proposed architecture that complements the existing cloud computing through one or more layers of intermediate computing servers.These servers, called fogs, are deployed at the edge of the network and bridge the gap between end-user devices and the cloud.They provide compute and storage resources to devices in a similar way to the cloud but in a more distributed fashion.This new computing paradigm offers a set of new challenges that we aim to address in this research.We leverage the use of machine learning, particularly deep learning algorithms capable of taking advantage of the large volumes of data generated in smart city scenarios, to improve the efficiency of the fog computing middleware JAMScript.The first part of our work focuses on optimizing the handover procedure for mobile devices in a fog computing environment using a set of fog and cost predictors.These predictors are used to reduce the service interruptions experienced while transitioning from one fog node to another.We simulate a city level fog network with real-world data derived from taxi traces in Shanghai city.We then model the fog associations for vehicles using a feedforward neural network as well as the cost (latency) of interacting with a particular fog server using an recurrent neural network (RNN) with long short-term memory (LSTM) cells.We present a system architecture that describes the components of this predictive system as well as a smarter request routing scheme that can be implemented using it.The second part of our research introduces a learning logger architecture that utilizes an ensemble of LSTMs to model data streams derived from devices at the fog servers.We show how predictions from the learning model can be used to partially replace actual data, thereby saving on valuable bandwidth without compromising the integrity and usability of the data.Finally, we conduct a series of experiments that showcase the performance of our predictive systems and quantify their benefit in a fog computing environment.In these experiments, we use a Docker-container based emulator for a vehicular fog network created using JAMScript to evaluate the fog and cost predictors.For the learning logger experiments, we use a set of environmental sensor data streams.Experimental results show that these systems can yield a considerable reduction in resource usage and gain in transmission efficiency in a fog computing environment.7 Conclusion and Future Work 7.1 Future Work . . . . . . . .
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".