The objective of this dissertation is to propose the cross-layer spectrum management architecture for white space applications that improves the performance the main functions of the spectrum management. In the proposed cross-layer architecture, white space network devices such as white space devices and the spectrum map server cooperate to support the extended spectrum map, the inter-cell transmit power adaptation (ITPA), and the frequency-domain coexistence beacon (FCB). Upon the cross-layer architecture, firstly, white space devices (WSDs) achieve a faster search and higher accuracy in spectrum sensing with the extended spectrum map, the extended DHCP (Dynamic Host Configuration Protocol), and the FCB. Secondly, WSDs achieve the precise selection of their operating channel in spectrum decision with the extended spectrum map and the ITPA. In spectrum sharing, thirdly, the collaboration- based spectrum sharing with the ITPA achieve more accommodation of WSDs by increasing the number of channels shared between WSDs. Finally, WSDs with the FCB and the extended spectrum achieve effective spectrum mobility by obtaining the occupancy-status of channels precisely.

The pervasiveness and operational autonomy of mesh-based wireless sensor networks (WSNs) make them an ideal candidate in offering sustained monitoring functions at reasonable cost over a wide area. To extend the functional lifetime of battery-operated sensor nodes, stringent sleep scheduling strategies with communication duty cycles running at sub-1% range are expected to be adopted. Although ultra-low communication duty cycles can cast a detrimental impact on sensing coverage and network connectivity, its effects can be mitigated with adaptive sleep scheduling, node deployment redundancy and multipath routing within the mesh WSN topology. This work proposes a cross-layer organizational approach based on sleep scheduling, called Sense-Sleep Trees (SS-Trees), that aims to harmonize the various engineering issues and provides a method to extend monitoring capabilities and operational lifetime of mesh-based WSNs engaged in wide-area surveillance applications. Various practical considerations such as sensing coverage requirements, duty cycling, transmission range assignment, data messaging, and protocol signalling are incorporated to demonstrate and evaluate the feasibility of the proposed design approach.

Recently, wireless access to Internet applications and services has attracted a lot of attention. However, there is no single wireless network that can meet all mobile users’ requirements. Con-sequently, integrated heterogeneous wireless networks are introduced to meet diverse wireless Internet applications and services requirements. On the other hand, integrated heterogeneous wireless networks pose new challenges to the design and development of reliable transport mechanisms. Wireless Application Protocol version 2 (WAP 2.0) is one of the promising trans-port mechanisms. It uses wireless profiled TCP (WP-TCP), which is fully compatible with TCP, as one of the reliable transport protocols to cope with the wireless link impairments. For WAP 2.0 to continue providing reliable and efficient transport services in the future, one of the key is-sues is to thoroughly study, understand, and improve its performance in integrated heterogeneous wireless networks. In this thesis, we develop analytical frameworks and propose a solution to respectively study and improve the performance of WP-TCP in integrated heterogeneous wireless networks. Spe-cifically, we consider WP-TCP short- and long-lived flows over integrated wireless local area network (WLAN) and wireless wide area network (WWAN), where WLAN can be static or mo-bile. In order to facilitate the analysis of WP-TCP performance in integrated WLAN and WWAN, we first construct a novel WLAN link model, which captures the impact of both uncor-related and correlated transmission errors, and derive mathematical expressions that describe packet loss probability and packet loss burst length over WWAN-WLAN link. Then, we develop analytical frameworks for studying the performance of WP-TCP short- and long-lived flows. Differently from those reported in the literature, our analytical framework for WP-TCP short-lived flows takes into account both correlated and uncorrelated packet losses. Furthermore, our analytical framework for long-lived flow can be used to study the short-term (during vertical handover) and long-term performances of WP-TCP and it captures the effects of vertical handover, such as excessive packet losses and sudden change in network characteristics, which are commonly experienced in integrated static WLAN and WWAN. By using the devel-oped analytical frameworks, we extensively analyze the performance of WP-TCP flows and in-vestigate the optimal protocol design parameters over a wide range of network conditions. Finally, based on our analytical studies, we propose a receiver-centric loosely coupled cross-layer design along with two proactive schemes, which significantly improve the vertical hand-over performance. The proposed solution is easy to implement and deploy, compatible with tra-ditional TCP, and robust in the absence of cross-layer information. Extensive simulations have been conducted to confirm the effectiveness and practicability of our schemes.

The pervasiveness and operational autonomy of mesh-based wireless sensor networks (WSNs) make them an ideal candidate in offering sustained monitoring functions at reasonable cost over a wide area. To extend the functional lifetime of battery-operated sensor nodes, stringent sleep scheduling strategies with communication duty cycles running at sub-1% range are expected to be adopted. Although ultra-low communication duty cycles can cast a detrimental impact on sensing coverage and network connectivity, its effects can be mitigated with adaptive sleep scheduling, node deployment redundancy and multipath routing within the mesh WSN topology. This work proposes a cross-layer organizational approach based on sleep scheduling, called Sense-Sleep Trees (SS-Trees), that aims to harmonize the various engineering issues and provides a method to extend monitoring capabilities and operational lifetime of mesh-based WSNs engaged in wide-area surveillance applications. Various practical considerations such as sensing coverage requirements, duty cycling, transmission range assignment, data messaging, and protocol signalling are incorporated to demonstrate and evaluate the feasibility of the proposed design approach.

Recently, wireless access to Internet applications and services has attracted a lot of attention. However, there is no single wireless network that can meet all mobile users’ requirements. Con-sequently, integrated heterogeneous wireless networks are introduced to meet diverse wireless Internet applications and services requirements. On the other hand, integrated heterogeneous wireless networks pose new challenges to the design and development of reliable transport mechanisms. Wireless Application Protocol version 2 (WAP 2.0) is one of the promising trans-port mechanisms. It uses wireless profiled TCP (WP-TCP), which is fully compatible with TCP, as one of the reliable transport protocols to cope with the wireless link impairments. For WAP 2.0 to continue providing reliable and efficient transport services in the future, one of the key is-sues is to thoroughly study, understand, and improve its performance in integrated heterogeneous wireless networks. In this thesis, we develop analytical frameworks and propose a solution to respectively study and improve the performance of WP-TCP in integrated heterogeneous wireless networks. Spe-cifically, we consider WP-TCP short- and long-lived flows over integrated wireless local area network (WLAN) and wireless wide area network (WWAN), where WLAN can be static or mo-bile. In order to facilitate the analysis of WP-TCP performance in integrated WLAN and WWAN, we first construct a novel WLAN link model, which captures the impact of both uncor-related and correlated transmission errors, and derive mathematical expressions that describe packet loss probability and packet loss burst length over WWAN-WLAN link. Then, we develop analytical frameworks for studying the performance of WP-TCP short- and long-lived flows. Differently from those reported in the literature, our analytical framework for WP-TCP short-lived flows takes into account both correlated and uncorrelated packet losses. Furthermore, our analytical framework for long-lived flow can be used to study the short-term (during vertical handover) and long-term performances of WP-TCP and it captures the effects of vertical handover, such as excessive packet losses and sudden change in network characteristics, which are commonly experienced in integrated static WLAN and WWAN. By using the devel-oped analytical frameworks, we extensively analyze the performance of WP-TCP flows and in-vestigate the optimal protocol design parameters over a wide range of network conditions. Finally, based on our analytical studies, we propose a receiver-centric loosely coupled cross-layer design along with two proactive schemes, which significantly improve the vertical hand-over performance. The proposed solution is easy to implement and deploy, compatible with tra-ditional TCP, and robust in the absence of cross-layer information. Extensive simulations have been conducted to confirm the effectiveness and practicability of our schemes.

As a result of rapid progress in research and development, today's wireless world exhibits several heterogeneous communication networks, such as cellular networks, satellite networks, wireless local area networks (WLAN), mobile ad hoc networks (MANET), and sensor networks. These networks are complementary to each other. Hence, their integration can realize a unified wireless system that has the best features of the individual networks. This has spurred much research interest in designing integrated next-generation of wireless systems (NGWS). While existing wireless networks have been extensively studied individually, the integrated wireless system brings new challenges in architecture design, system management, and protocol design. The different wireless networks use different communication technologies and are based on different networking paradigms. Therefore, it is challenging to integrate these networks such that their heterogeneities are hidden from each other and a harmonious inter-operation among them is achieved. The objective of this research is to design a scalable, secure, and robust architecture and to develop seamless mobility management protocols for NGWS. More specifically, an architecture that integrates the heterogeneous wireless systems is first proposed for NGWS. Next, a cross-layer (Layer 2 + 3) handoff management protocol is developed for NGWS. Afterward, analytical modeling is developed to investigate the handoff performance of the existing mobility management protocols for different types of applications. Finally, a framework for multi-layer mobility management is developed to support the seamless handoff support to all types of applications in NGWS.

The future success of communication networks hinges on the ability to overcome the mismatch between requested quality of service (QoS) and limited network resources. Spectrum is a natural resource that cannot be replenished and therefore must be used efficiently. On the other hand, energy efficiency (EE) is also becoming increasingly important as battery technology has not kept up with the growing requirements stemming from ubiquitous multimedia applications. The qualities of wireless channels vary with both time and user. We use channel state information (CSI) to dynamically assign wireless resources to users to improve spectral and energy efficiency. We first investigate a series of general treatments of exploiting CSI in a distributed way to control the medium access to maximize spectral efficiency for networks with arbitrary topologies and traffic distributions. As the first step, we propose decentralized optimization for multichannel random access (DOMRA), which uses local CSI and two-hop static neighborhood information to achieve performance comparable with the global optimal channel-aware Aloha. The generic framework developed in DOMRA proved to be very useful in improving cellular networks as well. We develop cochannel interference avoidance (CIA) medium access control (MAC), which is optimized by DOMRA, to mitigate the downlink severe cochannel interference that is usually experienced by cell-edge users. Aloha-based schemes have low channel utilization efficiency because of the collision of entire data frames. We further develop channel-aware distributed MAC (CAD-MAC), which avoids collision through signaling negotiation ahead of data transmission. CAD-MAC completely resolves the contention of networks with arbitrary topologies, achieves throughput close to centralized schedulers, and is robust to any channel uncertainty. Then we address energy-efficient wireless communications while emphasizing orthogonal frequency multiple access (OFDMA) systems. We first discover the global optimal energy-efficient link adaptation in frequency-selective channels using the strict quasiconcavity of energy efficiency functions. This link adaptation optimally balances the power consumption of electronic circuits and that of data transmission on each subchannel. The global optimal energy-efficient transmission can be obtained using iterative operations, which may be complex to be implemented in a practical system. Besides, running iterative algorithms consumes additional energy. Hence, we further develop a closed-form link adaptation scheme, which performs close to the global optimum. Besides, since subchannel allocation in OFDMA systems determines the energy efficiency of all users, we develop closed-form resource allocation approaches that achieve near-optimal performance too. In an interference-free environment, a tradeoff between EE and spectral efficiency (SE) exists, as increasing transmit power always improves SE but not necessarily EE. We continue the investigation in interference-limited scenarios and show that since increased transmit power also brings higher interference to the network, SE is not necessarily higher and the tradeoff is improved. Especially, in interference-dominated regimes, e.g., local area networks, both spectral- and energy-efficient communications desire optimized time-division protocols and the proposed DOMRA, CIA-MAC, and CAD-MAC can be used to improve both spectral and energy efficiency.

Monitoring and controlling smart grid assets in a timely and reliable manner is highly desired for emerging smart grid applications. Wireless Sensor Networks (WSNs) are anticipated to be widely utilized in a broad range of smart grid applications due to their numerous advantages along with their successful adoption in various critical areas including military and health care. Despite these advantages, the use of WSNs in such critical applications has brought forward a new challenge of ful lling the Quality of Service (QoS) requirements of these applications. Providing QoS support is a challenging issue due to highly resource constrained nature of sensor nodes, unreliable wireless links and harsh operation environments. In this thesis we critically investigate the problem of QoS provisioning in WSNs. We identify challenges, limitations and requirements for applying QoS provisioning for WSNs in smart grid applications. We nd that the topic of data prioritization techniques at the MAC layer to provide delay bounds in condition monitoring applications is not well developed. We develop six novel QoS schemes that provide data di erentiation and reduce the latency of high priority tra c in a smart grid context. These schemes are namely; Delay-Responsive Cross layer (DRX), Fair and Delay-aware Cross layer (FDRX), Delay-Responsive Cross layer with Linear backo (LDRX), Adaptive Realistic and Stable Model (ARSM), Adaptive Inter-cluster head Delay Control (AIDC) and QoS-aware GTS Allocation (QGA). Furthermore, we propose a new Markov-based model for IEEE 802.15.4 MAC namely, Realistic and Stable Markovbased (RSM). RSM considers actual network conditions and enhances the stability of the WSNs. We show through analytical and simulation results that all of the presented schemes reduce the end-to-end delay while maintaining good energy consumption and data delivery values.

Οι εφαρμογές πολυμέσων έχουν αποκτήσει τα τελευταία χρόνια μία αυξανόμενη ζήτηση από τους χρήστες γενικά του Διαδικτύου καθώς προσφέρουν νέες και ποικιλόμορφες δυνατότητες ανταλλαγής πληροφοριών εικόνας και ήχου. Όμως οι εφαρμογές αυτές υπόκεινται σε περιορισμούς που έχουν να κάνουν κυρίως με τη φύση τους και χαρακτηρίζονται από τις υψηλές απαιτήσεις σε ρυθμό μετάδοσης δεδομένων (bandwidth-consuming applications) και την ευαισθησία τους στις καθυστερήσεις που παρουσιάζονται κατά τη μετάδοση των πακέτων από τον αποστολέα στο παραλήπτη (delay-sensitive applications). Από την άλλη μεριά οι εφαρμογές αυτές φέρεται ότι είναι λιγότερο ευαίσθητες στις απώλειες των πακέτων (packet-loss tolerant applications). Το ζητούμενο όμως με τις εφαρμογές πολυμέσων, πέρα από το εύρος των υπηρεσιών τις οποίες προσφέρουν, είναι και η παρεχόμενη ποιότητα των υπηρεσιών (Quality of Service, QOS) στο τελικό χρήστη. Η ποιότητα αυτή των υπηρεσιών συνδέεται άμεσα με τα προαναφερόμενα χαρακτηριστικά των εφαρμογών πολυμέσων. Η μέχρι τώρα προσέγγιση από την ερευνητική κοινότητα αλλά και τις εταιρείες παροχής υπηρεσιών Διαδικτύου (Internet Service Providers), σε ότι αφορά την εξασφάλιση της ποιότητας υπηρεσιών, έχει εστιασθεί είτε στην επιμέρους βελτιστοποίηση της απόδοσης των πρωτοκόλλων μετάδοσης, είτε στην εγκατάσταση επιπλέον εξοπλισμού για τη δημιουργία δικτύων διανομής πολυμέσων (Content Distribution Networks, CDNs) που τοποθετούνται συνήθως κοντά στον τελικό χρήστη. Επιπρόσθετα η αυξανόμενη προσπάθεια της ερευνητικής κοινότητας με σκοπό την αύξηση της ποιότητας υπηρεσιών προσέφερε νέες καινοτόμες λύσεις με την μορφή των υπηρεσιών-αρχιτεκτονικών όπως οι Ολοκληρωμένες Υπηρεσίες (Integrated services, Intserv) και οι Διαφοροποιημένες Υπηρεσίες (Differentiated Services, Diffserv) οι οποίες φιλοδοξούν να προσφέρουν εγγυήσεις ποιότητας υπηρεσιών σε συγκεκριμένες ομάδες χρηστών. Όμως και οι δύο αυτές αρχιτεκτονικές δεν κατάφεραν μέχρι τώρα να αποτελέσουν μια ολοκληρωμένη λύση για τη παροχή εγγυήσεων ποιότητας υπηρεσιών στο χρήστη λόγω των μεγάλων δυσκολιών στην εφαρμογή τους που έχουν να κάνουν τόσο με χρηματοοικονομικά κριτήρια όσο και με τη ίδια τη δομή του Διαδικτύου. Έτσι βλέπουμε ότι παρόλη τη πρόοδο που έχει γίνει μέχρι σήμερα στη τεχνολογία των δικτύων η παροχή ποιότητας υπηρεσίας στο Διαδίκτυο από άκρο σε άκρο δεν είναι ακόμη στις μέρες μας εφικτή με αποτέλεσμα οι υπηρεσίες μετάδοσης πολυμέσων στο Διαδίκτυο – π.χ “youtube” – να επηρεάζονται σημαντικά από τις όποιες μεταβολές στη κατάσταση του δικτύου. Προς το σκοπό αυτό η ερευνητική κοινότητα έχει στραφεί στη μελέτη μηχανισμών οι οποίοι θα είναι να θέση να προσαρμόζουν το ρυθμό μετάδοσης της πολυμεσικής πληροφορίας, ανάλογα με τις εκάστοτε συνθήκες του δικτύου, έτσι ώστε να προσφέρουν τη μέγιστη δυνατή ποιότητα υπηρεσίας στο τελικό χρήστη. Η προσπάθεια αυτή μπορεί να κατηγοριοποιηθεί σε δύο μεγάλες κατηγορίες ανάλογα με το τρόπο δρομολόγησης της πολυμεσικής πληροφορίας, όπως παρακάτω: • Μηχανισμοί προσαρμογής για unicast μετάδοση: Σε αυτή τη περίπτωση οι μηχανισμοί προσαρμογής προσαρμόζουν το ρυθμό μετάδοσης της πληροφορίας από ένα σημείο (αποστολέας) προς ένα σημείο (παραλήπτης). • Μηχανισμοί προσαρμογής εκπομπής πολλαπλής διανομής (multicast): Στη περίπτωση αυτή οι μηχανισμοί προσαρμογής προσαρμόζουν το ρυθμό μετάδοσης της πληροφορίας που λαμβάνει χώρα από ένα σημείο (αποστολέας) προς πολλά σημεία (παραλήπτες). Σε ότι αφορά τη unicast μετάδοση την επικρατέστερη πρόταση αποτελεί ο μηχανισμός ελέγχου συμφόρησης με την ονομασία TCP Friendly Rate Control (TFRC) που έχει γίνει αποδεκτός ως διεθνές πρότυπο από τη Internet Engineering Task Force (IETF). Στη περιοχή της multicast εκπομπής ο μηχανισμός TCP-friendly Multicast Congestion Control (TFMCC) έχει γίνει επίσης αποδεκτός ως πειραματικό πρότυπο από την IETF. Παρόλα αυτά όμως εργαστηριακές μελέτες και πειράματα έχουν δείξει ότι τόσο ο μηχανισμός TFRC όσο και ο TFMCC δεν είναι και οι πλέον κατάλληλοι μηχανισμοί προσαρμογής για τη μετάδοση πολυμέσων. Τα κυριότερα προβλήματα αφορούν στη “φιλικότητα” των μηχανισμών αυτών προς το πρωτόκολλο Transmission Control Protocol (TCP) καθώς και στις απότομες διακυμάνσεις του ρυθμού μετάδοσης. Ιδιαίτερα οι απότομες διακυμάνσεις του ρυθμού μετάδοσης είναι ένα στοιχείο μη επιθυμητό από τις εφαρμογές πολυμέσων και ιδιαίτερα από τις εφαρμογές πολυμέσων πραγματικού χρόνου. Στη περιοχή των ασυρμάτων δικτύων τα προβλήματα τα οποία αντιμετωπίζονται κατά τη μετάδοση των πολυμέσων δεν έχουν τόσο άμεση σχέση με τη συμφόρηση του δικτύου (αυτή παρατηρείται κυρίως στα ενσύρματα δίκτυα), όσο με τις απώλειες των πακέτων που είναι ένα άμεσο αποτέλεσμα του μέσου διάδοσης. Η μέχρι τώρα προσέγγιση αφορά στην επιμέρους βελτιστοποίηση των διαφόρων πρωτοκόλλων των στρωμάτων του OSI έτσι ώστε να μειωθούν τα προβλήματα διάδοσης και να ελαχιστοποιηθούν οι απώλειες πακέτων και οι καθυστερήσεις από τον αποστολέα στο παραλήπτη. Κατά τα τελευταία χρόνια όμως κερδίζει όλο και περισσότερο έδαφος μια διαφορετική προσέγγιση η οποία έχει επικρατήσει να ονομάζεται διεθνώς ως “cross-layer optimization-adaptation”. Κατά τη προσέγγιση αυτή διαφαίνεται ότι θα μπορούσαμε να πετύχουμε τη βελτιστοποίηση της παρεχόμενης ποιότητας στις εφαρμογές πολυμέσων μέσω κάποιων μηχανισμών προσαρμογής, που θα εμπλέκουν περισσότερα του ενός εκ των στρωμάτων του OSI στις τρέχουσες συνθήκες του δικτύου. Η μεθοδολογία, οι προκλήσεις, οι περιορισμοί καθώς και οι εφαρμογές της διαστρωματικής (cross layer) προσαρμογής αποτελούν ένα ανοικτό ερευνητικό πεδίο το οποίο βρίσκεται αυτή τη στιγμή σε εξέλιξη. Σκοπός της παρούσας Διδακτορικής Διατριβής είναι καταρχήν η μελέτη των υπαρχόντων μηχανισμών ελέγχου συμφόρησης που αφορούν τα ενσύρματα δίκτυα μη εγγυημένης ποιότητας, όπως το Διαδίκτυο. Προς τη κατεύθυνση αυτή αξιολογούνται αρχικά οι υπάρχοντες μηχανισμοί ελέγχου συμφόρησης και διαπιστώνονται τα κυριότερα προβλήματα τα οποία σχετίζονται με τη ποιότητα υπηρεσιών. Η αξιολόγηση γίνεται με βάση τα κριτήρια που αφορούν τόσο στη φιλικότητα των μηχανισμών αυτών προς το TCP όσο και στα κριτήρια που αφορούν τη ποιότητα υπηρεσιών των εφαρμογών πολυμέσων. Η αξιολόγηση αυτή μας οδηγεί στο σχεδιασμό νέων πρωτοκόλλων τα οποία υπόσχονται υψηλότερη φιλικότητα προς το TCP και καλύτερη ποιότητα υπηρεσιών. Ένα σημαντικό στοιχείο που διαφοροποιεί τα πρωτόκολλα αυτά από τις άλλες προσεγγίσεις είναι η ομαλή (smooth) συμπεριφορά κατά την οποία ελαχιστοποιούνται οι απότομες μεταβολές του ρυθμού μετάδοσης, που είναι μη επιθυμητές από τις εφαρμογές πολυμέσων, διατηρώντας παράλληλα ένα υψηλό ρυθμό απόκρισης στις απότομες μεταβολές των συνθηκών του δικτύου. Ένα δεύτερο σημαντικό στοιχείο της εργασίας αυτής είναι οι προσθήκες στις βιβλιοθήκες του προσομοιωτή ns-2 οι οποίες είναι ήδη αντικείμενο εκμετάλλευσης από άλλους ερευνητές. Για το σκοπό αυτό τα νέα πρωτόκολλα καθορίζονται πλήρως και ενσωματώνονται στις βιβλιοθήκες του προσομοιωτή ns-2 έτσι ώστε να είναι διαθέσιμα στην ερευνητική κοινότητα ως μέρος του προσομοιωτή, για περαιτέρω μελέτη και αξιολόγηση. Επεκτείνονται παράλληλα υπάρχοντα ερευνητικά εργαλεία προσομοιώσεων τα οποία επιτρέπουν την ανάλυση και την αξιολόγηση υπαρχόντων και μελλοντικών μηχανισμών προσαρμογής με βάση κριτήρια ποιότητας που αφορούν ειδικά τις εφαρμογές πολυμέσων, πλέον των “κλασσικών” κριτηρίων που σχετίζονται με μετρικά δικτύων. Σε ότι αφορά στα ασύρματα δίκτυα μελετάται η διαστρωματική προσαρμογή και εάν και κατά πόσο είναι δυνατό να επιτευχθεί η αύξηση της ποιότητας της παρεχόμενης υπηρεσίας μέσα από την εφαρμογή μια τέτοιας σχεδίασης. Μελετώνται οι διάφοροι τρόποι και μεθοδολογίες σχεδίασης μιας διαστρωματικής προσαρμογής και προτείνεται ένα νέο πλαίσιο με το οποίο είναι δυνατό να αυξήσουμε τη ποιότητα υπηρεσίας σε υβριδικά δίκτυα, που αποτελούνται τόσο από ενσύρματους όσο και από ασύρματους χρήστες. / Multimedia applications have gained in recent years an increasing demand from Internet users as they offer new opportunities and diverse multimedia services. These applications, however, are subject to restrictions which mainly have to do with its nature and are characterized by high requirements of the transmission rates (bandwidth-consuming applications) and their sensitivity to delays in the transmission of packets by the consignor to consignee (delay-sensitive applications). On the other hand, allegedly these applications are less sensitive to packet losses (packet-loss tolerant applications). The issue, however, with multimedia applications, except for the scope of the services which offer, is the Quality of Services (QoS) that is offered to the end user. This quality of services is directly linked to the above characteristics of multimedia applications. The approach so far by the research community and also the Internet Service Providers (ISPs), as regards ensuring the quality of service, has focused either to individually optimizing the efficiency of transmission protocols, or in the installation of additional equipment (servers) for the establishment of distribution networks (Content Distribution Networks, CDNS) which are normally positioned close to the final user. In addition, the growing effort of the research community with a view to increasing the quality of service offered new innovative solutions in the form of services-architectures like the Integrated Services (Intserv) and Differentiated Services (Diffserv ) which aspire to offer guarantees of quality of services in specific user groups. But these two architectures failed until now to become the solution for the provision of guarantees of quality of services to the end user due to difficulties in applying them which have to do with financial criteria and the structure of the Internet itself. Therefore, we can see that despite the progress made so far in networks technology the provision of QoS across the Internet is not still feasible with the result that multimedia services via the Internet (for example “YouTube”) are significantly affected by the changes of the network conditions. To this course, the research community has directed to the study of mechanisms which will be able to adjust the transmission rate of multimedia data, according to the conditions of the network, so as to offer the best possible quality of service to the end user. This effort could be classified into two broad categories, according to the way the multimedia information is routed, as follows: • Adaptation mechanisms for unicast transmission: In this case the adaptation mechanisms regulate the transmission rate between the sender and the receiver in a unicast connection. • Adaptation mechanisms for multicast transmission: In this case the adaptation mechanisms regulate the transmission rate between the sender and a group of receivers. Regarding the unicast transmission the predominant proposal is the congestion control mechanism that is termed as the “TCP-friendly Rate Control (TFRC) and has been accepted as an international standard by the Internet Engineering Task Force (IETF). In the area of multicast transmission the TCP-friendly Multicast Congestion Control (TFMCC) has also become acceptable as an experimental standard from IETF. Nevertheless, laboratory studies and experiments have shown that both TFRC and TFMCC are not the most suitable adaptation mechanisms for multimedia transmission. The main problems have to do with its friendliness towards the Transmission Connection Protocol (TCP) and the sudden fluctuations in the transmission. rate. These sharp variations of the transmission rates are an attribute non desirable by multimedia applications and particularly by real-time applications. In the area of wireless networks the problems with the transmission of multimedia data are not directly linked to the congestion of the network (this mainly occurs in wired networks) as the packet losses are a direct result of the free space propagation. The approach so far has aimed at the individual optimization of the various protocols of the OSI model so as to reduce the transmission problems and minimize packet losses and the delays from the sender to the end user. In recent years, however, a different approach which has prevailed to be termed internationally as “cross-layer optimization-adaptation” has earned more and more space. Under this approach we could be able to succeed the optimization of the service, regarding the quality of multimedia applications, by means of some adaptation mechanisms which will involve more than one of the OSI layers to current network conditions. The methodology, the challenges, the restrictions and the applications of cross layer adaptation constitute an open research area which is currently in progress. The aim of this dissertation is firs the study of the existing congestion control mechanisms which mainly concern best-effort wired networks, such as the Internet. In this direction we evaluate the existing congestion/flow control mechanisms and record the main problems related to the quality of service. The performance evaluation is based on criteria relating both to TCP-friendliness and the quality of service of multimedia applications. This performance evaluation leads us to the design of new protocols which promise greater TCP-friendliness and better quality of service. An important element that distinguishes these protocols of the other approaches is the “smooth” behavior by which we minimize the high oscillations of the transmission rate, which are not desirable by multimedia applications, while maintaining a high response to sudden changes of network conditions. A second important element of this dissertation is the additions we have made to the libraries of the ns-2 simulator which are already exploited by other researchers. For this purpose the new protocols are fully defined and incorporated into the ns-2 libraries so as to be available to the research community as part of the simulator, for further studies and evaluation. At the same time we expand existing research tools in order to enable the analysis and evaluation of existing and future mechanisms based on quality criteria specific to multimedia applications, along with network-centric criteria. Regarding the wireless networks we study the cross layer adaptation and how it is possible to achieve the increase in the quality of service by implementing such a design. We study the various ways and design methodologies of a cross layer adaptation and propose a new framework with which it is possible to increase the quality of service in hybrid networks consisting of both by wired and wireless users.

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2018-03-20T16:23:19Z No. of bitstreams: 1 HelberWagnerDaSilva_TESE.pdf: 5778578 bytes, checksum: 90e873919b0c427d1742037fea6faf9d (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2018-03-21T17:00:22Z (GMT) No. of bitstreams: 1 HelberWagnerDaSilva_TESE.pdf: 5778578 bytes, checksum: 90e873919b0c427d1742037fea6faf9d (MD5) / Made available in DSpace on 2018-03-21T17:00:22Z (GMT). No. of bitstreams: 1 HelberWagnerDaSilva_TESE.pdf: 5778578 bytes, checksum: 90e873919b0c427d1742037fea6faf9d (MD5) Previous issue date: 2017-11-29 / As aplica??es de miss?o cr?tica (AMC) representam um dos casos de uso mais promissores da Internet das Coisas (do ingl?s, Internet of Things ? IoT) ao prometer impactos em ?reas vitais, como videovigil?ncia inteligente em ambientes com alta densidade humana, tr?fego de ve?culos aut?nomos com seguran?a, cirurgia remota com precis?o, dentre muitas outras. As AMC devem explorar o conte?do disponibilizado por plataformas IoT em cen?rios de rede IoT softwarizada (IoTS), que nesta tese representa um sistema IoT executando sobre uma infraestrutura de rede cuja arquitetura inclui substrato de rede definida por software para permitir a flexibilidade nas opera??es de controle. Entretanto, as AMC possuem fortes requisitos de Qualidade de Servi?o (do ingl?s, Quality of Service - QoS), tais como lat?ncia, jitter e perdas, al?m da alta demanda por recursos de rede (e.g., processamento em n?s, caminhos e largura de banda nos enlaces) que precisam ser assegurados pela IoTS para garantir efici?ncia e precis?o. A variabilidade e a dinamicidade de requisitos dos servi?os nesse cen?rio s?o muito elevadas, que podem ir desde a coleta de dados escalares (e.g., sensores ambientais, etc.) at? o processamento digital de multim?dia (e.g., v?deo e ?udio) em tempo real. Nesse cen?rio de IoTS com AMC sens?vel a QoS, torna-se necess?rio um plano de controle que tenha capacidade de prover um servi?o de transporte mais refinado com qualidade garantida, de forma otimizada e aut?noma. Esta tese vai al?m do estado da arte ao definir um arcabou?o hol?stico para o controle de sess?es multiusu?rios (agregando m?ltiplas AMC compartilhando conte?do de uma plataforma IoT) altamente sens?veis a qualidade em uma IoTS, com m?todos refinados para orquestra??o auto-organizada, controle e gest?o de recursos orientados a QoS. O arcabou?o, chamado CLASSICO (Cross-LAyer SDN SessIon COntrol), permite acoplar a variabilidade e a dinamicidade da IoTS, alocando dinamicamente recursos para satisfazer requisitos das AMC em termos de largura de banda elevada e lat?ncias muito baixas durante todo o tempo de dura??o da sess?o, alavancando substrato das Redes Definidas por Software (do ingl?s, Software Defined Networks - SDN) para flexibilidade e modularidade. Para alcan?ar esses objetivos, CLASSICO define um plano de controle cross-layer integrado ? IoTS que considera os par?metros do conte?do (ao n?vel de aplica??o) requerido pelas AMC para construir e manter sess?es multiusu?rios orientadas a QoS, e induz uma IoTS multiusu?rio otimizada atrav?s do transporte baseado em grupos (ao n?vel de rede), buscando ao mesmo tempo aumentar a escalabilidade do sistema IoT. Para valida??o, CLASSICO foi prototipado e avaliado em um testbed real em um caso de uso de v?deo. Os resultados da avalia??o revelam os ganhos de CLASSICO em termos de QoS e Qualidade de Experi?ncia (do ingl?s, Quality of Experience - QoE), em compara??o a uma solu??o baseada em SDN Multicast. / Mission Critical Applications (AMC) represent one of the most promising use cases in the Internet of Things (IoT) as they promise to impact vital areas, such as smart surveillance in environments with high human density, autonomous vehicle traffic with security, remote surgery with precision, among many others. AMC are expected to exploit the content made available by IoT platforms in softwarized IoT networks (IoTS) scenarios, which represents an IoT system running over a network infrastructure which includes software-defined network substrate to allow flexibility in control operations. However, AMCs have strong Quality of Service (QoS) requirements, such as latency, jitter and losses, as well as the high demand for network resources (e.g., node processing, paths and link bandwidth) that need to be guaranteed by IoTS to ensure efficiency and accuracy. The variability and dynamicity of service requirements in this scenario are very high, ranging from scalar data collection (e.g., environmental sensors, etc.) to digital multimedia (e.g., video and audio) processing in real time. In this IoTS scenario with QoS-sensitive AMC, it is necessary to have a control plan with the capacity to provide a fine-grained transport service with guaranteed quality, in an optimized and autonomous way. This thesis goes beyond the state of the art by defining a holistic framework for controlling multiuser sessions (aggregating multiple AMCs sharing content from an IoT platform) highly quality-sensitive in an IoTS, with fine-grained methods for quality-oriented self-organized orchestration, resource control and management. The framework, called CLASSICO (Cross-LAyer SDN SessIon COntrol), allows to couple IoTS variability and dynamicity, dynamically allocating resources to meet AMC requirements regarding high bandwidth and very low latencies throughout the session duration, leveraging the substrate of the Software Defined Networks (SDN) for flexibility and modularity. To achieve these objectives, CLASSICO defines a cross-layer control plan integrated with IoTS that considers the content parameters (at the application level) required by the AMC to build and maintain QoS-oriented multiuser sessions, and induces an optimized multiuser IoTS through group-based transport (at the network level), while increasing the scalability of the IoT system. For validation purposes, CLASSICO has been prototyped and evaluated on a real testbed in a video use case. The results of the evaluation reveal the gains of CLASSICO regarding QoS and Quality of Experience (QoE), in comparison to an SDN Multicast-based solution.