This repo contains the official implementations for the papers of [IEEE TMC 2020] Computation Offloading in Multi-Access Edge Computing: A Multi-Task Learning Approach and [IEEE GlobeCom 2023] A Multi-Head Ensemble Multi-Task Learning Approach for Dynamical Computation Offloading.
The 6 attributes of each MU:
x1 | x2 | x3 | x4 | x5 | x6 |
---|---|---|---|---|---|
the amount of input data necessary to be processed | total number of CPU cycles required to process | local CPU cycle frequency | channel power gain | alpha | beta |
~U(0, 5e5) | =x1*3e3 | ~U(0, 1e9) | ~U(0, 1) | ~U(0, 1) | =1-alpha |
Preset experimental parameters:
Name | Value | Meaning |
---|---|---|
F_t | 2.5e9 | total available computing resource on the server |
kappa | 1e-28 | parameter for local energy consumption |
Pt | 0.3 | transmission power |
PI | 0.1 | execution power |
theta | 1.0 (second) | the maximum tolerable delay |
B | 10e5 | the operational frequency band |
N0(sigma^2) | 7.96159e-13 | for SINR calculation |