Fabrizio Giuliano

Overgrid: A Fully Distributed Demand Response Architecture Based on Overlay Networks

In this paper, we present Overgrid, a fully distributed peer-to-peer (P2P) architecture designed to automatically control and implement distributed demand response (DR) schemes in a community of smart buildings with energy generation and storage capabilities. As overlay networks in communications establish logical links between peers regardless of the physical topology of the network, the Overgrid is able to apply some power balance criteria to its system of buildings, as they belong to a virtual microgrid, regardless of their physical location. We exploit an innovative distributed algorithm, called flow updating, for monitoring the power consumption of the buildings and the number of nodes in the network, proving its applicability in an Overgrid scenario with realistic power profiles and networks of up to 10 000 buildings. To quantify the energy balance capability of Overgrid, we first study the energy characteristics of several types of buildings in our university campus and in an industrial site to accurately provide some reference buildings models. Then, we classify the amount of “flexible” energy consumption, i.e., the quota that could be potentially exploited for DR programs. Finally, we validate Overgrid emulating a real P2P network of smart buildings behaving according to our reference models. The experimental results prove the feasibility of our approach.

Learning from errors: Detecting ZigBee interference in WiFi networks

In this work we show how to detect ZigBee interference on commodity WiFi cards by monitoring the reception errors, such as synchronization errors, invalid header formats, too long frames, etc., caused by ZigBee transmissions. Indeed, in presence of non-WiFi modulated signals, the occurrence of these types of errors follows statistics that can be easily recognized. Moreover, the duration of the error bursts depends on the transmission interval of the interference source, while the error spacing depends on the receiver implementation. On the basis of these considerations, we propose the adoption of hidden Markov chains for characterizing the behavior of WiFi receivers in presence of controlled interference sources (training phase) and then run-time recognizing the most likely cause of error patterns. Experimental results prove the effectiveness of our approach for detecting ZigBee interference.

MAC design on real 802.11 devices: From exponential to Moderated Backoff

In this paper we describe how a novel backoff mechanism called Moderated Backoff (MB), recently proposed as a standard extension for 802.11 networks, has been prototyped and experimentally validated on a commercial 802.11 card before being ratified. Indeed, for performance reasons, the time critical operations of MAC protocols, such as the backoff mechanism, are implemented into the card hardware/firmware and cannot be arbitrarily changed by third parties or by manufacturers only for experimental reasons. Our validation has been possible thanks to the availability of the so called Wireless MAC Processor (WMP), a prototype of a novel wireless card architecture in which MAC protocols can be programmed by using proper abstractions and a state-machine formal language, which enable easy modifications of legacy operations. Experimental results are in agreement with simulations and prove the effectiveness of Moderated Backoff, as well as the potentialities of the WMP platform.

Exploiting programmable architectures for WiFi/ZigBee inter-technology cooperation

The increasing complexity of wireless standards has shown that protocols cannot be designed once for all possible deployments, especially when unpredictable and mutating interference situations are present due to the coexistence of heterogeneous technologies. As such, flexibility and (re)programmability of wireless devices is crucial in the emerging scenarios of technology proliferation and unpredictable interference conditions.In this paper, we focus on the possibility to improve coexistence performance of WiFi and ZigBee networks by exploiting novel programmable architectures of wireless devices able to support run-time modifications of medium access operations. Differently from software-defined radio (SDR) platforms, in which every function is programmed from scratch, our programmable architectures are based on a clear decoupling between elementary commands (hard-coded into the devices) and programmable protocol logic (injected into the devices) according to which the commands execution is scheduled.Our contribution is two-fold: first, we designed and implemented a cross-technology time division multiple access (TDMA) scheme devised to provide a global synchronization signal and allocate alternating channel intervals to WiFi and ZigBee programmable nodes; second, we used the OMF control framework to define an interference detection and adaptation strategy that in principle could work in independent and autonomous networks. Experimental results prove the benefits of the envisioned solution.

A flexible and reconfigurable 5G networking architecture based on context and content information

The need for massive content delivery is a consolidated trend in mobile communications, and will even increase for next years. Moreover, while 4G maturity and evolution is driven by video contents, next generation (5G) networks will be dominated by heterogeneous data and additional massive diffusion of Internet of Things (IoT). The current network architecture is not sufficient to cope with such traffic, which is heterogeneous in terms of latency and QoS requirements, and variable in space and time. This paper proposes architectural advances to endow the network with the necessary flexibility helping to adapt to these varying traffic needs by providing content and communication services where and when actually needed. Our functional hardware/software (HW/SW) architecture aims at influencing future system standardization and leverage the benefits of some key 5G networking enablers described in the paper. Preliminary results demonstrate the potential of these key technologies to support the evolution toward content-centric and context-aware 5G systems.

MAC learning: enabling automatic combination of elementary protocol components: demo

Cognition as a way to deal with the challenges of future wireless networks has been largely considered by the recent literature, with a main focus on physical layer adaptability and dynamic spectrum access. In this demo, we show how a simple cognition mechanism can be also applied at the MAC layer, by exploiting the emerging paradigm of programmable wireless cards. The idea is using the formal definition of simple MAC protocol components and platform-independent representation of channel events gathered from the wireless node, for emulating the behavior of protocols which are not currently running on the network, learning about their expected performance, and dynamically reconfiguring the wireless node. We demonstrate that programmable nodes, employing our cognition scheme, can find in a distributed way a con-conflicting schedule with other neighbor nodes and can switch from contention-based to scheduled-based protocols as a function of the network load.

ErrorSense: Characterizing WiFi Error Patterns for Detecting ZigBee Interference

Recent years have witnessed the increasing adoption of heterogeneous wireless networks working in unlicensed ISM bands, thus creating serious problems of spectrum overcrowding. Although ZigBee, Bluetooth and WiFi networks have been natively designed for working in presence of interference, it has been observed that several performance impairments may occur because of heterogeneous sensitivity to detect or react to the presence of other technologies. In this paper we focus on the WiFi capability to detect interfering ZigBee links. Despite of the narrowband transmissions performed by ZigBee, in emerging scenarios ZigBee interference can have a significant impact on WiFi performance. Therefore, interference detection is essential for improving coexistence strategies in heterogeneous networks. In our work we show how such a detection can be performed on commodity cards working on time and frequency domain and also analysing data in the error domain. Errors are monitored and classified into error patterns observed in the network in terms of occurrence probability and temporal clustering of different error events. Through statistical analysis we are able to detect the presence of ZigBee transmissions measuring the errors raised by the WiFi card.

Exploring Training Options for RF Sensing Using CSI

This work analyzes human behavior recognition approaches using WiFi channel state information from the perhaps less usual point of view of training and calibration needs. With the help of selected literature examples, as well as with more detailed experimental insights on our own Doppler spectrum-based approach for physical motion/presence/cardinality detection, we first classify the diverse forms of training so far employed into three main categories (trained, trained-once, and training-free). We further discuss under which conditions it is possible to move toward lighter forms of calibration or even succeed in devising fully untrained model-based solutions. Our take home messages are mainly two. First, reduced training might not necessarily kill performance (although, of course, trade-offs will emerge). Second, reduced training must come along with a careful customization of the technical detection approach to the specificities of the behavior recognition application targeted, as it seems very hard to find a one-size-fits-all solution without relying on extensive training.

Demo: Dynamic Adaptations of WiFi Channel Widths Without TX/RX Coordination

Most modern standards for wireless communications support physical layer adaptations, in terms of dynamic selection of channel central frequency, transmission power, modulation format, etc., in order to increase link robustness under time-varying propagation and interference conditions. In this demo, we demonstrate that another powerful solution for extending physical layer flexibility in OFDM-based technologies is the dynamic adaptation of the channel width. Although some standards already define the possibility of utilizing multiple channel widths (e.g. 20MHz, 10MHz, 5MHz for IEEE 802.11a standards), such an utilization is limited to a static configuration of a value defined during the network set-up. Conversely, we demonstrate that channel width adaptations can be performed in real-time during network operation, even on a per-packet basis. To this purpose, we propose an innovative and efficient receiver design, which allows the transmitter to take decisions about the channel width without explicitly informing the receiver.

Making WiFi Work in Multi-hop Topologies: Automatic Negotiation and Allocation of Airtime

We propose a solution for mitigating the performance impairments of CSMA/CA protocols in multi-hop topologies based on the dynamic adaptation of the contention process experienced by nodes in a wireless network. A distributed protocol is used to negotiate the channel airtime for a node as a function of the traffic requirements of its neighbourhood, taking into account bandwidth reserved for the control operations. A mechanism is provided for a node to tune its contention window depending on its allocated airtime. Different from previous schemes, a nodes contention window is fixed in size unless the traffic requirements of its neighbourhood change. The scheme is implemented on legacy commercial 802.11 devices. Extensive experimental results, performed on the CREW European testbed, demonstrate the effectiveness of the approach.