Alex Veith, Patrick R. Carney, Aiqing Wu, Brenda L. Rojas ¡ 9 authors
Scientific progress benefits from the sharing of "research assets" such as data, reagents, models, and experimental samples. To improve asset shareability, we evaluated the availability, quality, and characterization of recombinant DNA molecules, recombinant mouse models, and tissue samples described by our laboratory in ten publications spanning over thirty years. Employing state-of-the-art molecular technologies, we identified existing samples, updated their localization, generated modern sequences and maps of recombinant models, and ported the associated metadata to an internal blockchain-dependent resource using a standardized description for each asset class. We also created non-fungible tokens representing research assets on the public blockchain network Solana. In addition to providing an audit of previously reported shareable assets and improving the value of recombinant models, this re-analysis also provides evidence for the utility of extant tissue samples that may be difficult and expensive to regenerate. The results demonstrate how retrospective analysis can improve and expand upon the spectrum of shareable research assets through updates on molecular characterization and physical location, as well as improving the availability of biological samples of potential high experimental value. Moreover, the development of a decentralized ledger harboring this revised metadata provides a path to the description and tokenization of scientific assets and provides a strategy to extend the life of scientific assets even after laboratories or sources close.
The increasing penetration of renewable sources introduces new challenges for power systemsâ stability, especially for isolated systems characterized by low inertia and powered through a single diesel power plant, such as it happens in small islands. For this reason, research projects, such as the BLORIN project, have focused on the provision of energy services involving electric vehicles owners residential users to mitigate possible issues on the power system due to unpredictable generation from renewable sources. The residential users were part of a blockchain-based platform, which also the Distributors/Aggregators were accessing. This paper describes the integrated framework that was set up to verify the feasibility and effectiveness of some of the methodologies developed in the BLORIN project for fast frequency response in isolated systems characterized by low rotational inertia. The validation of the proposed methodologies for fast frequency response using Vehicle-to-Grid or Demand Response programs was indeed carried out by emulating the dynamic behavior of different power resources in a Power Hardware-in-the-Loop environment using the equipment installed at the LabZERO laboratory of Politecnico di Bari, Italy. The laboratory, hosting a physical microgrid as well as Power Hardware-in-the-Loop facilities, was integrated within the BLORIN blockchain platform. The tests were conducted by assuming renewable generation development scenarios (mainly photovoltaic) and simulating the system under the worst-case scenarios caused by reduced rotational inertia. The experiments allowed to fully simulate usersâ interaction with the energy system and blockchain network reproducing realistic conditions of tracking and remuneration of usersâ services. The results obtained show the effectiveness of the BLORIN platform for the provision, tracking and remuneration of grid services by electric vehicles and end users, and the benefits that are achieved in terms of reducing the number of diesel generating units that need to be powered on just to provide operational reserve due to the penetration of renewable sources, resulting in fuel savings and reduced emissions.
The Android platform is undoubtedly the most popular platform for smartphones, with thousands of new applications becoming available daily and billions of app installations each year. Ethereum is the most popular smart contract platform, with thousands of applications on the blockchain serving as trading platforms and providing other functionalities. Due to these platformsâ popularity, security issues in their applications may reach a catastrophic scale with ease. Several prominent automated techniques help to reveal security problems in applications at the early stages of expansion. One such technique is static analysis. This thesis focuses on the design of static analysis techniques for Android apps and smart contracts distributed in the form of low-level code (bytecode).After installation, an Android app may get access to a set of sensitive information sources (e.g., location data). Unfortunately, exposure of such information to third parties has led in the past to several cases of privacy breach, and continues to be a serious threat. In this thesis, we tackle information flow propagation in the bytecode of Android applications by sound Horn-clause based abstraction techniques. This work will be the first to use Horn-clause based techniques in the context of security analysis. Moreover, we prove that our approach is sound, that is, our approach provides guarantees for its results. As a consequence, it can be used to show the absence of explicit data leaks in an app. Furthermore, Horn-clause based abstraction techniques are not limited to information propagation tasks, that is, our techniques can be used to show any kind of program property expressed as a reachability property. In addition, our Horn-clause based techniques scale to large codebases, benefit from the advancements in Satisfiability Modulo Theory solving, and allow for favorable performance with respect to the state-of-the-art. We instantiate the principles that were obtained while developing the analysis techniques for Android applications in the context of Ethereum smart contracts distributed in the form of Ethereum Virtual Machine (EVM) bytecode. Smart contracts are programs mainly used to perform financial operations (e.g., auctions) on cryptocurrency blockchains (e.g., Ethereum). Recent attacks demonstrate that certain vulnerabilities in smart contracts might cause severe money loss and an overall decrease of trust in the technology. Therefore, security analysis of EVM bytecode is in the focus of the research community. This thesis presents two results which establish the foundations for sound security analysis of EVM bytecode. First, the semantics of EVM bytecode is mechanized for the first time and tested against the official Ethereum test suite. This result facilitates both the design of analysis techniques and establishing their correctness properties. Second, this thesis provides the first sound Control Flow Graph reconstruction solution for EVM bytecode, that is, our analysis guarantees that reachable parts of the code are never pruned. This guarantee is required by a number of security properties for smart contracts. We also develop a tool implementing our analysis and successfully evaluate it on a big collection of real-world contracts.