Abstract-The deployment of new network equipment is resulting in increasing energy consumption in mobile broadband networks (MBNs). This contributes to higher CO2 emissions. Over the last 10 years MBNs have grown considerably, and are still growing to meet the evolution in traffic volume carried in wireless networks. To save energy in MBNs, one of the options is to turn off parts of the network equipment in areas where traffic falls below a specific predefined threshold. This paper looks at a methodology for identifying periods of the day when cells or sites carrying low traffic are candidates for being totally or partly switched off, given that the decrease in service quality can be controlled gracefully when the sites are switched off. Based on traffic data from an operational network, potential average energy savings of approximately 30% with some few low traffic cells/sites reaching up to 99% energy savings can be identified.
Facing the huge success of mobile devices, network providers ceaselessly deploy new nodes (cells) to always guarantee a high quality of service. Nevertheless, keeping turned on all the nodes when traffic is low is energy inefficient. This has led to investigations on the possibility to turn off network nodes, fully or partly, in low traffic loads. To accomplish such a dynamic network optimization, it is crucial to predict very accurately low traffic periods. In this paper, we tackle this problem using data mining and propose Spatio-Temporal Ensemble Prediction (STEP). In a nutshell, STEP is based on the following two main ideas: (1) since traffic shows very different behaviors depending on both the temporal and the spatial contexts, several prediction models are built to fit these characteristics; (2) we propose an ensemble prediction technique that accurately predicts low traffic periods. We empirically show on a real dataset that our approach outperforms standard methods on the low traffic prediction task.
With the rapidly increasing deployment of sensor networks, large amounts of time series data are generated. One of the main challenges when dealing with such data is performing accurate predictions in order to address a broad class of application problems, ranging from mobile broadband network (MBN) optimization to preventive maintenance. To this end, time series prediction has been widely addressed by the statistics community. Nevertheless, such approaches fail in performing well when the data are more context-dependent than history-dependent. In this paper, we investigate how latent attributes can be built upon the time series in order to define a spatio-temporal context for predictions. Moreover, such attributes are often hierarchical, leading to multiple potential contexts at different levels of granularity for performing a given prediction. In support of this scenario, we propose the Lattice-Based Spatio-Temporal Ensemble Prediction (LBSTEP) approach, which allows modeling the problem as a multidimensional spatio-temporal prediction. Given an ensemble prediction model, we propose a solution for determining the most appropriate spatio-temporal context that maximizes the global prediction metrics of a set of the time series. LBSTEP is evaluated with a real-world MBN dataset, which exemplifies the intended general application domain of time series data with a strong spatio-temporal component. The experimental results shows that the proposed contextual and multi-granular view of the prediction problem is effective, in terms of both several optimization metrics and the model calculation.
OATAO is an open access repository that collects the work of Toulouse researchers and makes it freely available over the web where possible. This is an author-deposited version published in : http://oatao.univ-toulouse.fr/ Eprints ID : 15178The contribution was presented at DaWaK 2014:http://www.dexa.org/dawak2014 Abstract. Today, huge volumes of sensor data are collected from many different sources. One of the most crucial data mining tasks considering this data is the ability to predict and classify data to anticipate trends or failures and take adequate steps. While the initial data might be of limited interest itself, the use of additional information, e.g., latent attributes, spatio-temporal details, etc., can add significant values and interestingness. In this paper we present a classification approach, called Closed n-set Spatio-Temporal Classification (CnSC), which is based on the use of latent attributes, pattern mining, and classification model construction. As the amount of generated patterns is huge, we employ a scalable NoSQL-based graph database for efficient storage and retrieval. By considering hierarchies in the latent attributes, we define pattern and context similarity scores. The classification model for a specific context is constructed by aggregating the most similar patterns. Presented approach CnSC is evaluated with a real dataset and shows competitive results compared with other prediction strategies.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.