Joint Task Offloading Scheduling and Transmit Power Allocation for Mobile-Edge Computing Systems
The paper presents a paper on optimizing task offloading scheduling and transmit power allocation in Mobile-Edge Computing (MEC) systems. In contemporary mobile systems, MEC stands as a critical innovation designed to alleviate the computing burden from mobile devices by offloading tasks to nearby MEC servers, improving both execution latency and energy efficiency. This research is oriented towards independent tasks in single-user MEC scenarios with constrained resources, where offloading decisions are paramount.
The authors introduce a low-complexity algorithm to optimize the weighted sum of execution delay and device energy consumption. This algorithm leverages techniques from flow shop scheduling and convex optimization, improving the efficiency of MEC systems significantly. The primary challenges addressed include the task offloading scheduling and optimizing transmit power allocation—two dimensions crucial for ensuring the effectiveness of MEC operations.
The paper adopts a unique perspective by treating the task scheduling issue as a two-machine flow shop scheduling problem and employs Johnson's algorithm to achieve an optimal sorting process for task offloading. The tasks are categorized into two sets based on their transmission and execution times, and optimal permutation scheduling is achieved to minimize the computational workload. The incorporation of convex optimization techniques allows for calculating the optimal power allocation vector, which aligns with real-world radio resource limitations.
Numerical results, derived through simulation, demonstrate that task scheduling becomes increasingly significant when the computational and radio resources are evenly balanced. The proposed algorithm achieves near-optimal delay performance while enabling substantial energy savings, offering practical applicability by optimizing task offloading across various MEC scenarios.
The implications of this research extend towards both theoretical and practical advancements in MEC systems. The incorporation of flow shop scheduling in task scheduling provides a structured methodology that can be further explored for multi-user and multi-server environments. Practically, the results indicate how MEC systems can maintain energy efficiency without compromising execution performance. The paper opens pathways for future exploration of hybrid MEC systems, wherein mobile devices have some computational capabilities, thus requiring joint optimization of task offloading decisions, scheduling, and resource allocation.
In conclusion, this paper provides a methodical approach to jointly optimizing task offloading scheduling and transmit power allocation in MEC systems, leading to enhanced execution efficiencies and presenting a foundation for future research developments in dynamic edge computing environments.