-The unprecedented growth in content demand on smartphones has significantly increased the energy consumption of current cellular and backbone networks. Apart from achieving stringent carbon footprint targets, provisioning high data rates to city vehicular users while maintaining quality of service (QoS) remains a serious challenge. In previous work, to support content delivery at high data rates, the number and locations of caching points (CPs) within a content distribution network (CDN) were optimized while reducing the operational energy consumption compared to typical cellular networks. Further reduction in energy consumption may be possible through sleep cycles, which reduces transmission energy consumption. However, sleep cycles degrade the quality of service. Therefore, in this paper, we propose a novel load adaptation technique for a CP which not only enhances content download rate but also reduces transmission energy consumption through random sleep cycles. Unlike a non-load adaptive (deterministic) CP, the performance results reveal that the load adaptive CP achieves considerably lower average piece delay (approximately 60% on average during the day), leveraging the introduction of random sleep cycles to save transmission energy. The proposed CP saves up to 84% transmission energy during off-peak hours and 33% during the whole day while fulfilling content demand in a city vehicular environment.
In this paper, we propose energy efficient Information Piece Delivery (IPD) through Nano Servers (NSs) in a vehicular network. Information pieces may contain any data that needs to be communicated to a vehicle. The available power (renewable or non-renewable) for a NS is variable. As a result, the service rate of a NS varies linearly with the available energy within a given range. Our proposed system therefore exhibits energy aware rate adaptation (RA), which uses variable transmission energy. We have also developed another transmission energy saving method for comparison, where sleep cycles (SC) are employed. Both methods are compared against an acceptable download time. To reduce the operational energy, we first optimise the locations of the NSs by developing a mixed integer linear programming (MILP) model, which takes into account the hourly variation of the traffic. The model is validated through a Genetic Algorithm (GA1). Furthermore, to reduce the gross delay over the entire vehicular network, the available renewable energy (wind farm) is optimally allocated to each NS according to piece demand. This, in turn, also reduces the network carbon footprint. A Genetic Algorithm (GA2) is also developed to validate the MILP results associated with this system. Through transmission energy savings, RA and SC further reduce the NSs energy consumption by 19% and 18% respectively, however at the expense of higher download time. MILP model 4 (with RA) and model 5 (with SC) reduced the delay by 81% and 83% respectively, while minimising the carbon footprint by 96% and 98% respectively, compared to the initial MILP model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.