Telecommunication networks currently follow the reactive approach; that is, network conditions change based on current behaviour of users and state of network resources. However, network efficiency and resource optimization can increase manifold, if the networks had the capability to anticipate behaviour of users and their network utilization. This vision, thus far, has remained largely theoretical. However, considering research in ubiquitous computing and machine learning with respect to: i) understanding user behaviour patterns, ii) establishment of user behaviour models and iii) anticipation of future user behaviour based on the established behaviour model, the vision of anticipatory networks comes closer to reality. The realisation of this vision will be a result of amalgamating human behaviour research in ubiquitous computing that through anticipation of user behaviour will be a key element in network resources optimisation. We call this vision, Human-Aware networking (HAN). Keywords human aware networking; network performance; network optimisation; user behaviour analysis; user behaviour modelling I. INTRODUCTION Human-aware networking is an interdisciplinary research area that lies in the intersection between Ubiquitous computing [1] and Network optimisation. This research area seeks to improve operation quality and efficiency of networks by incorporating a key element in the process: anticipation of user behaviour. Anticipating user behaviour and predicting future user actions, has the potential to establish highly optimised networks that are able to respond to users needs whilst maintaining actual Quality of Service (QoS) and perceived Quality of Experience (QoE). One strand of ubiquitous computing research has focused on understanding human behaviour through the use of unobtrusive monitoring technologies, typically mobile phones. For example, research work in [2] has investigated use of mobile phones in monitoring social interactions of the user. This was done based on non-verbal human behaviour clues, extracted from the sensor data captured through mobile phones, namely distance based on WiFi sensor and mutual orientation based on the phone s compass sensor. In addition, authors managed not only to detect social interaction behaviour, but also to classify it, whether the social interaction was formal or informal.
Realising Anticipatory Networks Through Human Aware Networking
Venet Osmani;Oscar Mayora-Ibarra;Tinku Rasheed;E. Salvadori
2014-01-01
Abstract
Telecommunication networks currently follow the reactive approach; that is, network conditions change based on current behaviour of users and state of network resources. However, network efficiency and resource optimization can increase manifold, if the networks had the capability to anticipate behaviour of users and their network utilization. This vision, thus far, has remained largely theoretical. However, considering research in ubiquitous computing and machine learning with respect to: i) understanding user behaviour patterns, ii) establishment of user behaviour models and iii) anticipation of future user behaviour based on the established behaviour model, the vision of anticipatory networks comes closer to reality. The realisation of this vision will be a result of amalgamating human behaviour research in ubiquitous computing that through anticipation of user behaviour will be a key element in network resources optimisation. We call this vision, Human-Aware networking (HAN). Keywords human aware networking; network performance; network optimisation; user behaviour analysis; user behaviour modelling I. INTRODUCTION Human-aware networking is an interdisciplinary research area that lies in the intersection between Ubiquitous computing [1] and Network optimisation. This research area seeks to improve operation quality and efficiency of networks by incorporating a key element in the process: anticipation of user behaviour. Anticipating user behaviour and predicting future user actions, has the potential to establish highly optimised networks that are able to respond to users needs whilst maintaining actual Quality of Service (QoS) and perceived Quality of Experience (QoE). One strand of ubiquitous computing research has focused on understanding human behaviour through the use of unobtrusive monitoring technologies, typically mobile phones. For example, research work in [2] has investigated use of mobile phones in monitoring social interactions of the user. This was done based on non-verbal human behaviour clues, extracted from the sensor data captured through mobile phones, namely distance based on WiFi sensor and mutual orientation based on the phone s compass sensor. In addition, authors managed not only to detect social interaction behaviour, but also to classify it, whether the social interaction was formal or informal.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
