Joint allocation of radio and computational resources in wireless application offloading

Abstract

Providing femto-access points (FAPs) with computational capabilities will allow (either totally or partially) offloading of highly demanding applications from smart-phones to the so called femto-cloud. Such offloading promises to be beneficial in terms of battery saving at the mobile terminal (MT) and/or latency reduction in executing applications, if the energy and/or time required for the communication process are compensated by the energy and/or time savings that result from remote computation. For this scenario, we provide a general framework that allows the joint allocation of radio and computational resources resulting in an optimized trade-off between energy consumption and latency. Such energy ansumption includes contributions from both communication and computation segments. Complex communication models (e.g. multi-input multi-output (MIMO) transmission) can be encompassed within the proposed framework. Moreover, practical constraints can be included, such as energy constraints, application constraints, as well as radio and computational constraints. As a result of the optimization, the optimal communication strategy (transmission power, rate, etc.) between the handset and the serving FAP is obtained, as well as the optimal distribution of computation tasks between the handset and the FAPs in the femto-cloud. A particular case is defined: when all the computation tasks are forced to be done in either the handset or the femto-cloud. In such a case, the proposed framework will provide the necessary information to decide whether offloading the application is beneficial