|
| Research | Proposed solutions | Key metrics | Edge devices | What is to be offloaded |
|
| Li et al. [15] | A deep learning model coinference framework | Latency, communication size | Devices with cameras | Computer vision algorithms |
| Li et al. [16] | A joint accuracy-and latency-aware execution framework | Latency, accuracy | Devices with cameras | Computer vision algorithms |
| Gao et al. [17] | Edge4Sys system to reduce the computation load of the edge server in a MEC-based UAV delivery scenario | Latency, energy, accuracy | UAV | DNN-based feature extraction and classification |
| Tariq et al. [18] | A fog simulator, covers the network, transmission rage, heterogeneous mobile devices, and mobility feature | Latency, transmission range, mobility | IoT devices | Typical IoT tasks |
| Asad et al. [19] | A fog simulation framework to support latency-sensitive applications | Latency, energy, accuracy | IoT devices | Typical IoT tasks |
| Syed et al. [20] | A fog computing framework to simulate the vehicle-assisted computing environment | Latency, energy consumption, communication size, memory | Vehicles | Compute-intensive tasks |
| The proposed algorithm | MEC-based system to reduce the latency and energy consumption | Latency, energy consumption | Intelligent prosthesis | DNN-based intent recognition tasks |
|
