Dependability for future wireless Machine-Type Communication networks

In this thesis, a cohesive dependability-theoretic framework is established for the design and evaluation of future wireless networks supporting ultra-reliable, low-latency and secure communication crucial for Industrial Internet of Things (IIoT). Several IIoT applications operate under stringent execution deadlines during which even minor communication failures are unacceptable. The conventional understanding of reliability for Machine-Type Communication (MTC) often overlooks the inherently time-varying aspects of wireless fading channels. Moreover, the prior literature mainly neglects the temporal dynamics of the fading channel and its crucial impact on mission-critical applications. Therefore, we present a new theoretical framework for statistical Quality-of-Service (QoS) for MTC networks based on dependability theory. For this purpose, we derive novel closed-form expressions (CFE) for second order statistics such as Level crossing rate (LCR), and Average Fade Duration (AFD). These statistics help to formulate mission reliability and Mean Time-to-First Failure (MTTFF) as core dependability attributes to characterize the probability of failure-free transmission over the specified mission duration. Furthermore, we expand the traditional Effective Capacity (EC) to amalgamate mission reliability and MTTFF in the Finite Blocklength (FBL) regime, exhibiting the mission Effective Capacity (mEC) which not only relies on the average Signal-to-Noise Ratio (SNR) but also identifies fading-induced failure events in the wireless fading channel. The proposed framework is applied to two key technologies: i) Fluid Antenna System (FAS), which offers real-time spatial reconfiguration to improve reliability, throughput, and energy efficiency; ii) Multi-connectivity (MC)-enabled industrial wireless systems. To capture energy efficiency under bursty traffic and latency constraints, we also develop a mission Effective Energy Efficiency (mEEE) metric for FAS and formulate its maximization as a non-convex fractional optimization problem. Moreover, we also analyse security as a crucial component of dependability. We specifically examine physical layer security for the Multiple-Input Multiple-Output Multi-Antenna Eavesdropper scenario that involves a friendly multi-antenna jammer. A novel and exact CFE for the secrecy outage probability is derived for the proposed scenario, conditioned upon a message actually being transmitted. As a by-product, we provide a simpler CFE of the secrecy outage probability when all nodes are equipped with a single antenna. Furthermore, we also examine secure effective capacity for an ON-OFF transmission under jamming while analysing the traffic originated by a massive number of IoT devices through Markovian sources.