Advances, Systems and Applications
From: Task offloading optimization mechanism based on deep neural network in edge-cloud environment
Notation | Description |
---|---|
\(C=\{c_1,c_2,c_3,\ldots ,c_n\}\) | Crowd task set |
\(l=\{l_{c_1},l_{c_2},l_{c_3},\ldots l_{c_n}\}\) | Crowd tasks posted location set |
\(T_{satrt}\) | Crowd task release time |
\(T_{end}\) | Crowd task deadline |
\(T=\{t_{c_1},t_{c_2},t_{c_3},\ldots ,t_{c_n}\}\) | Maximum allowable delay set for crowd tasks |
\(D=\{d_{c_1},d_{c_2},d_{c_1},\ldots ,d_{c_n}\}\) | The amount of data contained set in the crowd task |
\(W=\{w_1,w_2,w_3,\ldots ,w_m\}\) | Crowd worker set |
\(W_{id}\) | Crowd worker id |
\(S_{{c_i}{w_j}}\) | Task offload policy |
\(r_{local}\) | Local data processing rate |
\(q_{local}\) | Consumption per bit of data processed locally |
\(x_{edge}\) | Edge server data transfer rate |
\(r_{edge}\) | The rate at which edge server data is processed |
\(y_{edge}\) | Transmission energy consumption per bit data of edge server |
\(q_{edge}\) | Energy consumption per bit of data processed by edge servers |
\(T_{local}(c_i,w_j)\) | Local time consumption of worker \(w_j\) task \(c_i\) |
\(E_{local}(c_i,w_j)\) | Local energy consumption of worker \(w_j\) task \(c_i\) |
\(T_{edge}(c_i,w_j)\) | The marginal time consumption of worker \(w_j\) task \(c_i\) |
\(E_{edge}(c_i,w_j)\) | The marginal energy consumption of worker \(w_j\) task \(c_i\) |
\(W_{local}\) | The total consumption of local processing computing tasks |
\(W_{edge}\) | The total consumption of edge servers processing computing tasks |