Wireless Body Area Network (WBAN) facilitates efficient and cost-effective e-health care and well-being applications. The WBAN has unique challenges and features compared to other Wireless Sensor Networks (WSN). In addition to battery power consumption, the vulnerability and the unpredicted channel behavior of the Medium Access Control (MAC) layer make channel access a serious problem. MAC protocols based on Time Division Multiple Access (TDMA) can improve the reliability and efficiency of WBAN. However, conventional static TDMA techniques adopted by IEEE 802.15.4 and IEEE 802.15.6 do not sufficiently consider the channel status or the buffer requirements of the nodes within heterogeneous contexts. Although there are some solutions that have been proposed to alleviate the effect of the deep fade in WBAN channel by adopting dynamic slot allocation, these solutions still suffer from some reliability and energy efficiency issues and they do not avoid channel deep fading. This thesis presents novel and generic TDMA based techniques to improve WBAN reliability and energy efficiency. The proposed techniques synchronise nodes adaptively whilst tackling their channel and buffer status in normal and emergency contexts. Extensive simulation experiments using various traffic rates and time slot lengths demonstrate that the proposed techniques improve the reliability and the energy efficiency compared to the de-facto standards of WBAN, i.e. the IEEE 802.15.4 and the IEEE 802.15.6. In normal situations, the proposed techniques reduce packet loss up to 61% and 68% compared to the IEEE 802.15.4 and IEEE 802.15.6 respectively. They also reduce energy consumption up to 7.3%. In emergencies, however, the proposed techniques reduce packets loss up to 63.4% and 90% with respect to their counterparts in IEEE 802.15.4 and 802.15.6. The achieved results confirm the significant enhancements made by the developed scheduling techniques to promote the reliability and energy efficiency of WBAN, opening up promising doors towards new horizons and applications.

Dans le cadre de cette thèse, on se proposait de développer de nouveaux mécanismes de radiolocalisation, permettant de positionner les nœuds de réseaux corporels sans-fil (WBAN) mobiles, en exploitant de manière opportuniste des liens radio coopératifs bas débit à l'échelle d'un même corps (i.e. coopération intra-WBAN), entre réseaux distincts (i.e. coopération inter-WBAN), et/ou vis-à-vis de l'infrastructure environnante. Ces nouvelles fonctions coopératives présentent un intérêt pour des applications telles que la navigation de groupe ou la capture de mouvement à large échelle. Ce sujet d'étude, par essence multidisciplinaire, a permis d'aborder des questions de recherche variées, humine-biomécanique et de ayant trait à la modélisation physique (e.g. modélisation spatio-temporelle des métriques de radiolocalisation en situation de mobilité, modélisation de la mobilité groupe...), au développement d'algorithmes adaptés aux observables disponibles (e.g. algorithmes de positionnement coopératifs et distribués, sélection et ordonnancement des liens/mesures entre les nœuds...), aux mécanismes d'accès et de mise en réseau (i.e. en support aux mesures coopératives et au positionnement itératif). Les bénéfices et les limites de certaines de ces fonctions ont été en partie éprouvés expérimentalement, au moyen de plateformes radio réelles. Les différents développements réalisés tenaient compte, autant que possible, des contraintes liées aux standards de communication WBAN émergeants (e.g. Impulse Radio - Ultra Wideband (IR-UWB) IEEE 802.15.6), par exemple en termes de bande fréquentielle ou de taux d'erreur. / The PhD investigations aim at exploring new WBAN cooperative localization mechanisms, which could benefit jointly from on-body links, body-to-body links between distinct mobile users or off-body links with respect to the infrastructure. Following a multidisciplinary approach, we have thus addressed theoretical questions related to physical modeling or to algorithmic and cross-layer design. A few more practical aspects have also been dealt with. More specifically, based on WBAN channel measurements, single-link ranging error models are first discussed for more realistic performance assessment. Then a Constrained Distributed Weighted Multi-Dimensional Scaling (CDWMDS) positioning algorithm is put forward for relative MoCap purposes, coping with on-body nodes' asynchronism to reduce system latency and exploiting the presence of constant-length radio links for better accuracy. Subsequently we consider extending this algorithm for larger-scale asbolute MoCap applications within a 2-step localization approach that incorporates additional off-body links in a heterogeneous WBAN framework. Then, both individual and collective kinds of navigation are addressed. In both MoCap and navigation scenarios, low-complexity solutions exploiting on-body deployment diversity enable to combat error propagation and strong range biases due to body shadowing, relying on on-body nodes' dispersion or graph neighborhood to approximate the corrupted distances. Finally, experiments based on real IR-UWB radio platforms validate in part the previous proposals, while showing their practical limitations.

Increasing number of ageing population and people who need continuous health monitoring and rising the costs of health care have triggered the concept of the novel wireless technology-driven human body monitoring. Human body monitoring can be performed using a network of small and intelligent wireless medical sensors which may be attached to the body surface or implanted into the tissues. It enables carers to predict, diagnose, and react to adverse events earlier than ever. The concept of Wireless Body Area Network (WBAN) was introduced to fully exploit the benefits of wireless technologies in telemedicine and m-health. The main focus of this research is the design and performance evaluation of strategies and architectures that would allow seamless and efficient interconnection of patient’s body area network and the stationary (e.g., hospital room or ward) wireless networks. I first introduce the architecture of a healthcare system which bridges WBANs and Wireless Local Area Networks (WLANs). I adopt IEEE 802.15.6 standard for the patient’s body network because it is specifically designed for WBANs. Since IEEE 802.15.6 has strict Quality of Service (QoS) and priorities to transfer the medical data to the medical server a QoS-enabled WLAN for the next hop is needed to preserve the end-to-end QoS. IEEE 802.11e standard is selected for the WLAN in the hospital room or ward because it provides prioritization for the stations in the network. I investigate in detail the requirements posed by different healthcare parameters and to analyze the performance of various alternative interconnection strategies, using the rigorous mathematical apparatus of Queuing Theory and Probabilistic Analysis; these results are independently validated through discrete event simulation models. This thesis has three main parts; performance evaluation and MAC parameters settings of IEEE 802.11e Enhanced Distributed Channel Access (EDCA), performance evaluation and tuning the MAC parameters of IEEE 802.15.6, and designing a seamless and efficient interconnection strategy which bridges IEEE 802.11e EDCA and IEEE 802.15.6 standards for a healthcare system.

Increasing number of ageing population and people who need continuous health monitoring and rising the costs of health care have triggered the concept of the novel wireless technology-driven human body monitoring. Human body monitoring can be performed using a network of small and intelligent wireless medical sensors which may be attached to the body surface or implanted into the tissues. It enables carers to predict, diagnose, and react to adverse events earlier than ever. The concept of Wireless Body Area Network (WBAN) was introduced to fully exploit the benefits of wireless technologies in telemedicine and m-health. The main focus of this research is the design and performance evaluation of strategies and architectures that would allow seamless and efficient interconnection of patient’s body area network and the stationary (e.g., hospital room or ward) wireless networks. I first introduce the architecture of a healthcare system which bridges WBANs and Wireless Local Area Networks (WLANs). I adopt IEEE 802.15.6 standard for the patient’s body network because it is specifically designed for WBANs. Since IEEE 802.15.6 has strict Quality of Service (QoS) and priorities to transfer the medical data to the medical server a QoS-enabled WLAN for the next hop is needed to preserve the end-to-end QoS. IEEE 802.11e standard is selected for the WLAN in the hospital room or ward because it provides prioritization for the stations in the network. I investigate in detail the requirements posed by different healthcare parameters and to analyze the performance of various alternative interconnection strategies, using the rigorous mathematical apparatus of Queuing Theory and Probabilistic Analysis; these results are independently validated through discrete event simulation models. This thesis has three main parts; performance evaluation and MAC parameters settings of IEEE 802.11e Enhanced Distributed Channel Access (EDCA), performance evaluation and tuning the MAC parameters of IEEE 802.15.6, and designing a seamless and efficient interconnection strategy which bridges IEEE 802.11e EDCA and IEEE 802.15.6 standards for a healthcare system.

The trend towards personalized ubiquitous computing has led to the advent of a new generation of wireless technologies, namely wireless body area networks (WBANs), which connect the wearable devices into the Internet-of-Things. This thesis considers the problems of relay selection and power control in fading WBANs with energy-efficiency and security considerations. The main body of the thesis is formed by two papers. Ideas from probability theory are used, in the first paper, to construct a performance measure signifying the energy efficiency of transmission, while in the second paper, information-theoretic principles are leveraged to characterize the transmission secrecy at the wireless physical layer (PHY). The hypothesis is that exploiting spatial diversity through multi-hop relaying is an effective strategy in a WBAN to combat fading and enhance communication throughput. In order to analytically explore the problems of optimal relay selection and power control, proper tools from game theory are employed. In particular, non-cooperative game-theoretic frameworks are developed to model and analyze the strategic interactions among sensor nodes in a WBAN when seeking to optimize their transmissions in the uplink. Quality-of-service requirements are also incorporated into the game frameworks, in terms of upper bounds on the end-to-end delay and jitter incurred by multi-hop transmission, by borrowing relevant tools from queuing theory. The proposed game frameworks are proved to admit Nash equilibria, and distributed algorithms are devised that converge to stable Nash solutions. The frameworks are then evaluated using numerical simulations in conditions approximating actual deployment of WBANs. Performance behavior trade-offs are investigated in an IEEE 802.15.6-based ultra wideband WBAN considering various scenarios. The frameworks show remarkable promise in improving the energy efficiency and PHY secrecy of transmission, at the expense of an admissible increase in the end-to-end latency.