Signal PRocessing and INnovative Transmissions


  • AMBIENT BACKSCATTERING COMMUNICATIONS: Ambient backscatter is an intriguing wireless communication paradigm that allows small devices to compute and communicate by using only the power they harvest from far-field radio-frequency (RF) signals in the air. Ambient backscattering devices reflect RF signals emitted by existing or legacy communications systems, such as digital TV broadcasting, cellular, or Wi-Fi ones, which are designed for transporting information and are not intended for RF energy transfer. We deal with mathematical modeling, practical design, and performance analysis of wireless broadband networks operating over fading channels with ambient backscatter devices.
  • COGNITIVE COMMUNICATIONS: Due to the explosive growth in wireless data services, mainly driven by video communications, together with the emerging "Internet of Things" (IoT) paradigm and the diffusion of machine-to-machine communications, forthcoming wireless systems must be able also to support an enormous number of low-rate devices, which will require new approaches and policies for spectrum allocation and management, including new forms of spectrum sharing. In this context, cognitive radio (CR) approaches are expected to play a major role. In this research activity, we study spectrum-sharing paradigms for CR networks, where a secondary user (SU) can maintain or even improve the performance of a primary user (PU) transmission, while also obtaining a low-data rate channel for its own communication. In particular, we propose a new scheme that allows the SU to convolve its block-precoded symbols with the received PU signal in the time domain, which gives rise to the term convolutive superposition. 
  • COOPERATIVE COMMUNICATIONS: Multiple-input multiple-output (MIMO) technology is a well-established approach to achieve throughput and energy gains in a wireless network. Multiple antennas, separated by at least half the wavelength, are used to increase the throughput by means of multiplexing or to reduce the energy consumption through diversity. However, application of MIMO techniques to wireless networks poses several implementation challenges in many cases, e.g., the limited physical size of nodes. A viable solution to the latter problem is represented by cooperative MIMO communication, whereby data-exchange between a MIMO source-destination pair can be assisted by various relays, each equipped with one or more antennas, which cooperate with the source to emulate a multi-antenna node. In this context, we consider the design and analysis of both centralized and decentralized MIMO cooperative techniques, by accounting for different amounts of channel state information.
  • MODELING AND OPTIMIZATION OF WIND AND SOLAR POWER SYSTEMS: Global warming, dangerous climate changes, and progressive growth of energy-hungry appliances have led to great emphasis on clean renewable energy resources. The inherent variability and uncertainty of wind and solar powers pose major technical and economical challenges to their large-scale integration into the electricity grid. Existing tools for the purpose of wind/solar power prediction rely on "top-down models", which do not attempt to model the statistical properties of the underlying random signals, but instead exploit the statistical relation between predicted and actual values of the power to tune model parameters. Such prediction models fail to reliably and accurately predict ramp events, i.e., sudden and large increases or decreases in wind and solar powers. The basic objective of the proposed research is to completely reverse such a viewpoint by developing "bottom-up models", which include a detailed characterization of the nonstationarity of wind and solar energy sources.
  • SOFTWARE-DEFINED NETWORKS: The unceasing increase in processing capability, storage capacity, and optical bandwidth, along with miniaturization and cost reduction of devices, is gradually changing traditional network architectures by allowing "intelligence" to migrate from the network’s core part to the edge. In a few years, the edge of future 5G (fifth generation) networks will be composed of a very large number of heterogeneous geo-distributed (i.e., not located within the same local area network) "smart" nodes and devices interconnected by means of ultra-broadband (fixed and radio) links; a flat core optical network with a limited number of "dumb" and fast nodes will ensure wide area connectivity between 5G edge domains. In the 5G edge network and services infrastructure, software will play the most important role. Indeed, relying on recent advances in network functions virtualization (NFV), software defined networking (SDN), cloud and edge-fog computing, a large amount of networks services and functions (up to applications) will be implemented in software running on general-purpose hardware. In this edge scenario, network resource allocation problems are enriched by new degrees of freedom, i.e., the feasibility of dynamically instantiating, moving, and orchestrating over wide areas ensembles of virtual machines (VMs), which can run network services and functions. We aim at studying management of virtual resources in software-centric 5G networks, by understating how and where the top-down governance of the software-centric network can meet the bottom-up self-organizing capabilities to be deployed in the edge network devices and terminals.




Contacts and direction

SPRINT GROUP - Via Claudio 21, 80125 Naples, Italy




COORDINATOR: Prof. Francesco Verde [f (dot) verde (at) unina (dot) it]

Please publish modules in offcanvas position.