UE Computation Offloading Based on Task and Channel Prediction of Single User

This paper established a partial computation offloading paradigm to improve energy consumption and delay performance in single user scenario. This algorithm utilizes task and channel prediction to obtain future state to optimize objective function. A unique advantage of this algorithm is task and channel prediction utilizing markov property. The proposed algorithm utilizing task and channel prediction achieved more than 90% energy consumption reduction compared with no prediction. Finally, simulation achieved reduce in energy consumption are presented to corroborate the theoretical analysis as well as validate the effectiveness of the proposed algorithm. The algorithm proposed in this paper greatly improves the performance of energy-sensitive and delay-sensitive devices.