Browsing by Author "Marano, Salvatore"

Now showing 1 - 9 of 9

<> proposal of efficient routing techniques and intelligent vehicular traffic management in the smart cities context, through distributed wireless networks
(2016-02-02) Sottile, Cesare; Pantano, Pietro Salvatore; De Rango, Floriano; Marano, Salvatore
In this thesis, the design of e cient techniques for routing protocol suitable to Vehicular Ad-hoc NETworks (VANETs) has been pro- posed. In particular, the aims of the proposed protocols are to reduce interference issues, due to the data transmissions in wireless environ- ment. The proposed protocols use time series prediction models and also multi-objective metric, based on the evaluation of co-channel in- terference levels, end-to-end delay, and link duration probability along the di erent links from sources towards destinations. These param- eters are modelled through an optimization problem. The key fac- tors are to exploit the advantages available to the Standard 802.11p, based on a dynamic allocation mechanism of the DSRC spectrum, aimed at the reduction of the co-channel interference and the maxi- mization of the link duration probability (two key issues in vehicular environments). Another topic discussed in this thesis is related to the smart vehicular tra c management through VANETs infrastructure and communications(V2I and V2V). A distributed algorithm with the aim to build less congested path for the vehicles in a urban scenario has been developed. It is also considered the problem regarding to enhance air quality around the cities reducing the vehicles CO2 emis- sions. There are di erent causes related to the CO2 emissions such as the average travelled time spent by vehicles inside the city and their average speed. Hence, with a better tra c management the average time spent by the vehicles in the city will be considerably reduced as well as CO2 emissions. These results are demonstrated in a discrete event simulator by using also real tra c data
Bio-inspired techniques applied to the coordination of a swarm of robots involved in multiple tasks
(2017-11-13) Palmieri, Nunzia; Crupi, Felice; Marano, Salvatore; Yang, Xin-She
La tematica di ricerca trattata in questa tesi riguarda il problema di coordinamento di robot attraverso l’utilizzo di algoritmi decentralizzati che usano meccanismi basati sulla Swarm Intelligence. Tali tecniche hanno lo scopo di migliorare le capacità di ogni robot, ciascuno dei quali ha risorse limitate, nel prendere decisioni su dove muoversi o su cosa fare basandosi su semplici regole ed interazioni locali. Negli ultimi anni, infatti, c’è un crescente interesse a risolvere alcuni problemi nell’ambito della robotica attraverso algoritmi che traggono ispirazione da fenomeni naturali e da alcuni animali in natura che esibiscono comportamenti sociali sviluppati e con una notevole capacità di adattamento ambientale. Nel campo della robotica, un aspetto cruciale è la coordinazione dei robot affinché possano compiere dei task in maniera cooperativa. La coordinazione deve essere tale da permettere agli agenti di adattarsi alle condizioni dinamiche dell’ambiente circostante conferendo al sistema caratteristiche di robustezza, flessibilità e affidabilità. Più dettagliatamente, lo scenario di riferimento è un’area nella quale sono disseminati degli oggetti, e dove operano un certo numero di robot che hanno come scopo quello di rilevare gli oggetti stessi e manipolarli. Ciascun robot non conosce la posizione di tali oggetti e non ha conoscenza né dell’ambiente che lo circonda, né della posizione degli altri robot. Il problema è diviso in due sotto-problemi. Un primo problema riguarda l’esplorazione dell’area e l’altro la manipolazione degli oggetti. Essenzialmente, ogni robot esplora in maniera indipendente l’ambiente basandosi sulla propria posizione attuale e sulla posizione degli altri mediante un meccanismo di comunicazione indiretta (stigmergia). Nella fase di manipolazione degli oggetti, invece, è utilizzato un meccanismo di comunicazione diretta attraverso l’uso di una comunicazione wireless. L’algoritmo di esplorazione dell’area trae ispirazione dal comportamento di alcuni tipi di insetti in natura, come le formiche,che utilizzano l’ambiente nel quale vivono come mezzo di comunicazione (stigmergia).Successivamente, quando un robot rileva la presenza di un oggetto, sono stati proposti due approcci. Nel primo caso le informazioni sono diffuse tra i robot secondo un meccanismo di comunicazione“one hop”ed alcune meta-euristiche di derivazione naturale sono state utilizzate come meccanismo decisionale e di coordinamento Il secondo approccio fa riferimento ad una comunicazione “multi-hop” ed è stato proposto un protocollo di coordinamento, anche esso di derivazione biologica. Entrambi gli approcci si basano su meccanismi decentralizzati dove non esiste nessun leader che dia direttive gerarchiche e ciascun robot prende le sue decisioni in maniera autonoma sulla base degli eventi che accadono nell’ambiente. Globalmente si ha un sistema auto organizzato, flessibile ed altamente adattabile. Per testare gli approcci è stato costruito un simulatore sul quale sono stati sviluppati numerosi studi allo scopo di valutare gli algoritmi proposti, la loro efficienza nonché stimare come le principali variabili ed i parametri del modello possono influenzarela soluzione finale.
Evaluation and Optimization of the Energy Behavior of Routing Protocols for Mobile Ad-hoc Networks
(2014-02-28) Fotino, Marco; Palopoli, Luigi; Marano, Salvatore
MANETs (Mobile Ad-hoc NETworks) are networks made up entirely of wireless devices, without the support of any fixed infrastructure. Because of their auto-configuring and self-managing features, and of their many application fields, MANETs have been one of the most investigated research fields in recent years. In particular, the interest of the scientific community is aimed at routing protocols for such networks, which should solve the problem of efficient multihop routing in a distributed environment. My research work is focused on an aspect of MANETs that in recent years has become increasingly important: the evaluation of routing algorithms in terms of energy consumption. My work for the Systems Engineering and Computer Science Ph.D. program is focused on the study of routing algorithms for mobile ad-hoc networks and performance evaluation of such networks, especially trying to highlight and resolve issues related to energy consumption of mobile devices. The study of mobile ad-hoc networks and of the routing algorithms has been carried out mainly through simulations. The algorithms and metrics for routing protocols, obtained by the adaptation, the improvement and the joint application of the solutions known in literature, have been implemented in the ns-2 network simulator software. The methodology used in the context of the Ph.D. program consists in the implementation of the algorithms and in the execution of a large number of simulations, in order to validate the effectiveness of the adopted solutions in the widest possible number of scenarios. Keywords – MANET, Routing, Energy, OLSR, GPSR
Gene expression as a digital communication system
(2018-02-02) Cevallos Vilacrés, Yesenia Elizabeth; Pantano, Pietro; Marano, Salvatore
This PhD thesis extends upon the information theories of digital communication systems to analyse biological communications (nanocommunications) in order to accurately model biological communication as digital communication by providing an essential analysis of the analogies between both systems. As such, this work analysed gene expression from two perspectives: digital communication systems as a general perspective and internetwork systems as a specific perspective (keeping in mind that digital communication networks are a subarea of digital communication systems). First, this work presents a novel layered network model that represents gene expression and the role of the Golgi apparatus as an internetwork router to transmit proteins to a target organ. Second, supported by the aforementioned layered network model, this work presents a digital communication system end-to-end model that represents gene expression with regard to the production of proteinaceous hormones in the endocrine system by using Shannon’s theorem. In addition, each molecular process encoding biological information, from the transcription and translation of deoxyribonucleic acid (DNA) to hormone signalling, is represented by a layered network model. These models apply the general advantages of digital internetworks and systems (i.e., performance and efficiency) to the transmission of biological information in gene expression systems. The proposed models and analysis define the duality between digital and biological communication systems, and the results herein can be used to overcome the disadvantages of both systems. One of the most important applications of the current study is the potential use of the characteristics of both communication systems in the nano/bio-hybrid medical field (i.e., for the treatment of diseases such as cancer). Hence, the analysis presented in this study may prevent side effects by specifically enhancing the transmission of information to a suitable destination (i.e., to specific target organs), thereby facilitating the development of optimal and less expensive treatments.
Protocol Architectures for wireless networks: issues, perspectives and enhancements
(2012-11-09) Loscrì, Valeria; Talia, Domenico; Marano, Salvatore
In this work we have analyzed different types of wireless networks and we have investigated the behavior of different MAC and routing protocols and different protocol architectures. Specifically, we have analyzed a TDMA MAC protocol for Mobile Ad hoc NETworks called Evolutionary-TDMA (E-TDMA). Based on this protocol we have developed a cross-layer approach that uses multiple linkdisjoint paths for a pair source-destination supporting a soft QoS. For this latter point we have developed a Multipath Forward Algorithm to manage and assign the slots. We have developed two different schemes: QoS scheme 1 and QoS scheme 2 and compared these schemes in terms of throughput, delay and overhead with a well-known multipath protocol routing, the AOMDV over 802.11 MAC protocol and AOMDV over E-TDMA. We have obtained good performance exploiting advantages of a TDMA MAC protocol and a Multipath routing protocol. Simulation results have shown as a cross-layer approach permits good performance in terms of throughput, end-to-end data packets delay and overhead to be obtained. In fact, multipath routing protocols for wireless ad hoc networks have been investigated because the use of alternate paths permits a greater fault-tolerance to be obtained. As far as multipath routing protocol is concerned we have developed and analyzed a multipath routing protocol based on geographic positions of the nodes in the network, Geographic Multipath Protocol (GMP). Results, in terms of throughput and delay, show that the multiple paths with minimum interference are better. We have investigated the cross-layer approach also for Wireless Sensor Networks. Although, this kind of networks could be considered an extension of Ad hoc networks, characteristics of WSNs are so different that protocols explicitly developed for MANETs cannot be directly applied for sensors networks. After analyzed some protocols we have proposed some solutions to increase the lifetime of the sensor networks considering other important parameters as throughput and latency. Finally, our analysis of wireless networks finishes with the analysis of Wireless Mesh Networks. We have investigated multi-hop wireless networks based on the IEEE 802.16 technology. Specifically, we have analyzed the Coordinated Distributed Scheme (CDS) of the IEEE Std. 802.16 and we have developed a MAC module 6 Conlcusions 158 supporting the CDS in ns2. Moreover, we have developed two different totally distributed schemes that do not require any changes of the structure of the hardware used in the 802.16. The two schemes proposed manage the assignment of the control slots (Transmit Opportunity XmtOP), in a different fashion. We have developed the two scheme in ns2 in order to compare them with the CDS of the 802.16. Our approaches permit good performance in terms of throughput and delay to be obtained. Our approaches are independent from parameters of the network as density or topology. On the other hand, the CDS is not robust for different network conditions because the behavior of the scheme depends from the setting of particular parameters.
QoS aware multicast routing and meta- heuristics over multi-layered satellite-hap networks
(2011-11-23) Santamaria, Amilcare Francesco; Marano, Salvatore; Palopoli, Luigi
In the last few years, multimedia broadband services and applications are growth. More resources have been requested to the networks and more Quality have to be supplied for the services. Moreover, new kind of networks were born such as wireless network and cellular networks such as 3G/4G and UMTS, also wired connections are evolved and more resources are available to allow new services and applications. In these years, social networks and multimedia applications are increase exponentially making possible new field of applications. With new technologies, in fact, it is possible to supply utility services such as tele-health, wireless health monitoring, public safety services and much more services over the internet, WAN and so on. Moreover, dedicated services and networks can be rapidly installed in case of emergency scenarios such as war or natural disaster cases. For these reasons, new network architectures can be implemented to offer a wide range of services. In particular, it is thinkable to interoperate between different network such as Satellite network, Cellular network and Wired Network to take advantages from all networks reducing weakness. Moreover, it is necessary to increase resources availability making efficient choices and optimizing reservation along links on which network data flows. To make these scenarios possible services and network have to offer a reasonable Quality in terms of Quality of Service (QoS). To do this new protocols and algorithms capable to take advantages from modern network infrastructures have to be designed. A wide range of services can be distributed on the network in a multicast manner, for example multimedia data such as video and audio can take advantages from multicast transmission reducing in a drastic manner the number of control packets and data packets sent into the network. In case of unicast transmission, in fact, a higher number of packets than the multicast are sent into the network to allow each destinations to receive data flows. In order to make more robust, reliable and scalable multicasting on the hybrid network, a multicast algorithm has been associated to the the mul ticast protocol, when a multi-constraints QoS aware multicast have to be addressed. A multicast algorithm that has the main goal to find a multicast tree in an multi-constraints environment belongs to the NP-Complete class of problem. In literature this is known as a Steiner Tree Problem (STP). In order to solve this issue a new class of algorithms has to be addressed. In particular, taking into account limited resources of the routers and time limits a heuristic method has to be designed. In these last years a series of algorithms that derive by the observation of natural evolution and biological process that are also called biologic algorithms, pay out much attention. In this works, the Genetic Algorithm (GA) adapted to Multicast issues has been designed and proposed to provide a better resource management that allow a higher number of multicast sessions and higher number of users to join into the network. This allows a network cost reduction and offers better services with a higher quality to the community. To spread services around wide areas with same quality, a broadband network, which are not sensible to terrestrial overloaded network, have to be addressed to distribute with reliability and scalability multicast services. Satellite network equipped with newest architecture such as DVB-RCS or DVB-S2 that implements two way services represents an optimal solution to distribute broadband services, because it also support a certain realtime interaction between sources and destinations without changing of network. This architecture has the main problem to have a higher round trip delay owing to the distance between satellite and Earth surface, in fact, some applications such as realtime applications do not support satellite communications. For these reasons, and in general to reduce, where possible, round trip delay an intermediate network has been introduced into infrastructure. HAPs meshes are composed of several on air platform that provide wireless broadband communications. They can interact with cellular platforms or with most common wireless networks changing their payload if necessary. An interaction of Satellite and HAP can reduce delays and increase network performances. In order to make this possible an efficient multicast protocol capable to address in an efficient manner data flow and protocols interaction between several entities has been designed and proposed. Moreover, using aforementioned algorithm, efficient routing has been addressed. In this hybrid network, which is based on a DVB-RCS like architecture, local communication can be made exploiting HAPs or wired network. instead, in case of far connections satellite link can be used as bridge among HAPs network, otherwise satellite can be used directly to connect a terrestrial router with another one when HAPs links are overloaded or in case of some areas where HAP coverage is not present.
Resource reservation protocol and predictive algorithms for QoS support in wireless environments
(2008-01) Fazio, Peppino; Talia, Domenico; Marano, Salvatore
Smart homes and smart objects: internet of things for energy-aware monitoring and controlling systems
(2017-02-13) Barletta, Domenico; Pantano, Pietro; Marano, Salvatore; De Rango, Flaviano
In recent years, the design and implementation of Home Automation Systems followed the spread of Internet of Things and its intent to make every home smart for end-users through Internet and Wireless connectivity. Ambient intelligence is an emerging discipline as well that aims to bring intelligence to environments making them sensitive to people through the use of embedded smart objects. Together, Automation Systems, Internet of Things and Ambient Intelligence make great interaction between people and the environments they live in possible. The enhancing of these interactions was one the main goals of this thesis whose intent was to improve the classic concept of Home Automation System through Internet Of Things philosophy. The first research objective was the design and implementation of a basic Internet of Things architecture that can be used for several Home Automation and Monitoring applications such as lighting, heating, conditioning or energy management. In particular, besides controlling features, the system can optimize energy consumptions by increasing awareness of users that have full control of their house and the possibility to save money and reduce the impact of energetic consumption on the earth, matching the new “green” motto requirements. According to IoT, innovative plug-and-play and ready-to-use smart devices were designed and implemented as part of this reactive automation system that offers a user-friendly web application and allows users to control and interact with different plans of their house in order to make life more comfortable and be aware of their energy consumptions. Control and awareness were two important key points for the first stage of research activity. The next challenge was to enhance IoT with intelligence and awareness by exploring various interactions between human beings and the environment they live in. The system was progressively enhanced evolving from a reactive system to a proactive system where the knowledge of human needs, in addition to the current environment status, constitute a new input for the system whose further objective is to find the right automations that can satisfy human needs in real-time. This advanced proactive system can sense and infer the user’s and the environment’s context and consequently decide autonomously how to affect the environment and actuate smart actions. It was experienced by many researchers that, for several reasons, some users might not appreciate these kinds of proactive actuations so, another challenge was the design of a system which is able to learn from user’s context and feedbacks history and, subsequently, to adapt itself to always up-do-date users’ expectations and keep the rules for triggering smart actuations constantly updated. The system was at a later time transformed into a “Inferactive” system. Another important scope of the research was the in-depth examination of the motivations for energy-efficient communications in wireless systems by highlighting emerging trends and identifying the challenges that need to be addressed to enable novel, scalable and energy-efficient wireless communications. One of the most important issues for researchers that must be addressed in designing communication protocols for wireless networks is how to save devices’ energy while meeting the needs of applications. For this reason, the architecture proposed was also enhanced at a communication level with energy-awareness in order to minimize energy consumption by communication devices, protocols, networks, end-user systems and data centers. To do this, some smart devices have been designed with multiple communication interfaces, such as RF and Wi-Fi, by using open-source technology such as Arduino. They have been analyzed under different working conditions and network topology. Communication parameters, data size and device status have been changed dynamically according to different scenarios and specific quality of service required, such as the speed of response, in order to find the best hardware and software configuration that offers the most benefits in terms of energy cost/quality ratio, lowest energy consumption and extended lifetime for battery-powered devices. The last challenge was to improve energy consumption of battery-powered smart devices provided with Wi-Fi that is commonly considered a very energy-expensive communication interface if compared to Bluetooth or ZigBee. Research focused on social IoT solutions for Home Automation Systems where devices are socially connected to Internet and can communicate through social applications like Facebook, Twitter or Google+ with a community of users that usually interact with that device itself. A fuzzy-based solution was proposed to classify the social community interaction with the system in order to implement an adaptive energy saving mode for social Wi-Fi devices according to users’ current context, behavior, habits and feedbacks. Test and performance evaluation regarding energy consumption and efficiency in real world scenario with real hardware devices were performed to prove the efficiency of the solution in terms of electricity consumption, battery lifetime, CPU utilization and increase of comfort and satisfaction levels for community users. The thesis is organized as follows. Chapters 1,2 and 3 contains the basics for a better understanding of the solutions proposed such as Internet of Things, Ambient Intelligence, Smart Homes and Smart Devices energy level issues. Chapter 4 describes the solution proposed, in particular, the system architecture and how the system evolved from reactive, to proactive and finally to “inferactive”. Chapter 5 examines how to deal with energy-aware communication and how Wi-Fi communication was improved in terms of energy consumption through social IoT awareness and fuzzy logic. In closing, Chapter 6 presents the test and the results achieved.
Using multi-layer social networks for opportunistic routing
(2012-10-24) Socievole, Annalisa; Palopoli, Luigi; Marano, Salvatore; De Rango, Floriano