Daniel Zucchetto

Data-driven QoE optimization techniques for multi-user wireless networks

The proliferation of heterogeneous data services and applications, with different communication requirements, has led to the design of Quality of Service (QoS) mechanisms to provide service differentiation and, possibly, performance guarantee to a range of classes of applications. In this paper, we propose a data-driven media delivery framework for the optimization of multi-user wireless networks that differs from the classic approaches in the following aspects. First, it goes beyond the QoS paradigm to embrace the Quality of Experience (QoE) approach, which discriminates data streams based on their actual content rather than just their class. Second, it applies cutting-edge cognitive science techniques to automatically learn data models and discover optimization strategies. To substantiate our argumentation, we discuss a couple of use cases regarding the transmission of multimedia content over a wireless link shared by users belonging to different QoE classes of service.

Uncoordinated Access Schemes for the IoT: Approaches, Regulations, and Performance

Internet of Things devices communicate using a variety of protocols, differing in many aspects, with the channel access method being one of the most important. Most of the transmission technologies explicitly designed for IoT and machine-to-machine communication use either an ALOHA-based channel access or some type of Listen Before Talk strategy, based on carrier sensing. In this article, we provide a comparative overview of the uncoordinated channel access methods for Internet of Things technologies, namely ALOHA-based and Listen Before Talk schemes, in relation to the ETSI and FCC regulatory frameworks. Furthermore, we provide a performance comparison of these access schemes, in terms of both successful transmissions and energy efficiency, in a typical Internet of Things deployment. Results show that Listen Before Talk is effective in reducing inter-node interference even for long-range transmissions, although the energy efficiency can be lower than that provided by ALOHA methods. Furthermore, the adoption of rate adaptation schemes lowers the energy consumption while improving the fairness among nodes at different distances from the receiver. Coexistence issues are also investigated, showing that in massive deployments Listen Before Talk is severely affected by the presence of ALOHA devices in the same area.

State Modulated Traffic Models for Machine Type Communications

Machine-to-machine (M2M) traffic is variegate and finding a traffic model which can cover a wide range of M2M sources is challenging. In this paper we address this challenge by proposing an extension of legacy renewal processes for modeling of M2M traffic sources. To this end, we first describe the model and derive some performance parameters, as the overall packet arrival distribution and its moments. We then discuss the packetgeneration process and consider the counting variable in atime interval, and give the mean and the Laplace transformof the z-transform for this variable. Successively, we present the asymptotic expansion for the variance and the Index of Dispersion of Counts (IDC). We derive the expression of the two first coefficients of this expansion in the general case, while more explicit expressions are provided for some special cases. More specifically, for the special case of a source model with two states, geometric distribution of the numbers of arrivals in each state, and exponential inter-arrival times, we solve for the modelparameters in terms of mean, variance and two IDCs. The model is then applied to real M2M traces obtained from an operational network. Albeit the match is not perfect, yet the proposed model captures the main features of the traces, in particular the large burstiness in the packet arrival process.

Cell traffic prediction using joint spatio-temporal information

In future cellular networks, the ability to predict network parameters such as cell load will be a key enabler of several proposed adaptation and resource allocation techniques. In this study, we consider a joint exploitation of spatio-temporal data to improve the prediction accuracy of standard regression methods. We test several such methods from the literature on a publicly available dataset and document the advantages of the proposed approach.