Preemptive Mobile Computation Offloading With Hard Deadlines and Concurrent Local Execution

Teymoori, Peyvand; Hekmati, Arvin; Todd, T.D.; Zhao, Dongmei; Karakostas, George

doi:10.1109/tgcn.2021.3061106

Cited by 2 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different optimal offloading strategies are proposed for delay-constrained MCO over wireless transmission channels [14] [15] [16] [17]. Reference [14] studied continuous offloading, which means once the offload starts, it occurs contiguously in one piece without interruption.…”

Section: Introductionmentioning

confidence: 99%

“…This allows the mobile device to adapt to channel conditions at the end of each upload part, provided that the added latency is acceptable and does not result in violating the execution time deadline. Preemptive Offloading, which was introduced in [17], is a generalization of multi-part offloading, where the decisions to continue the upload are made at the start of each time-slot packet transmission.…”

Section: Introductionmentioning

confidence: 99%

“…The wireless communication channels are modelled using a Finite-State Markov Chain (FSMC), which is a frequently used channel model [18] [19]. Although provably optimal offloading algorithms exist for general Markovian wireless channels (e.g., [20] [15] [17]), their computational complexity increases dramatically with the size of the Markov chain, which may prohibit their application in practice. In this paper, we introduce ways of mitigating this complexity by aggregating the state space of the underlying Markov chain into a smaller set of state.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Efficient Mobile Computation Offloading over a Finite-State Markovian Channel using Spectral State Aggregation

Teymoori

Boukerche

2022

2022 IEEE 47th Conference on Local Computer Networks (LCN)

Self Cite

View full text Add to dashboard Cite

This paper considers the problem of mobile computation offloading under stochastic wireless channels while task completion times are subject to deadline constraints. Our objective is to conserve energy for the mobile device by making an optimal decision to execute the task either locally or remotely. In the case of computation offloading, we dynamically vary the data transmission rate, in response to channel conditions. The wireless transmission channel is modelled using a Finite-State Markov Chain (FSMC). We formulate the problem of computation offloading as a constrained optimization problem, and develop an online algorithm to derive the optimal offloading policy. Moreover, to reduce the complexity, we estimate a suboptimal solution of the proposed online algorithm by reducing the size of the FSMC with the help of Markovian aggregation. The numerical results indicate that by applying Markovian aggregation, the running time of the algorithm can be significantly reduced without suffering unreasonable performance degradation.

show abstract