Control Algorithms and Architectures for Resource Management in Multi-Layered Systems: Application to SatCom, Security and Manufacturing Domains

Last two decades have seen an exponential increase in the availability of innovative and cost effective technological solutions in a wide variety of fields, ranging from electronics to chemistry, from mechanics to computer science. These new potentialities have led to a growth in systems’ complexity, thus making their control and optimization a challenging task, eager for new methodologies, approaches, paradigms. With respect to this context, the doctorate research presented in this thesis aims at providing the scientific and industrial communities with enriched solutions to cope with a specific class of control applications, resource management, in a well-defined class of complex systems, the multi-layered ones. Multi-layered systems are commonly obtained by incremental design or hierarchical approaches: several layers, each one assigned to a specific task, are put together to jointly create an enriched, vertical or end-to-end, behaviour. In such conditions, resources are necessarily distributed between layers and information sharing is limited, thus resulting in an underutilization of system’s potentialities and poor performances: efficient resource management solutions are needed. Resource management is in fact the set of mechanism, procedures and algorithms that allow to control the allocation, distribution or utilization of systems’ capabilities (being them hardware or software functionalities, physical or logical resources): for this reason, they play a key role for the overall system’s performances; in recent years, control and optimization theories have proven to be the best candidates to address this class of problems. This doctorate work provides original results in the following selected multi-layered domains: - Satellite Communication, with the design of innovative cross-layer algorithms - Security, with the formalization of the “composable security” by means of control theory - Manufacturing, with the identification of a closed-loop “cognitive architecture” for the Factories of the Future. To achieve these results, original research has been performed in the areas of: i) modelling techniques; ii) control algorithms; iii) optimization algorithms. Some of these results have been jointly developed in the scope of European funded research, in particular in the MONET Project (Satellite Communication) and in the p/nSHIELD Projects (Security in Embedded Systems), where they have represented an original contribution with promising industrial exploitation perspectives.