1,721,033 research outputs found
Investigating the Resilience Source of Classification Systems for Approximate Computing Techniques
Full text linkDuring the last decade, classification systems (CSs) received significant research attention, with new learning algorithms achieving high accuracy in various applications. However, their resource-intensive nature, in terms of hardware and computation time, poses new design challenges. CSs exhibit inherent error resilience, due to redundancy of training sets, and self-healing properties, making them suitable for Approximate Computing (AxC). AxC enables efficient computation by using reduced precision or approximate values, leading to energy, time, and silicon area savings. Exploiting AxC involves estimating the introduced error for each approximate variant found during a Design-Space Exploration (DSE). This estimation has to be both rapid and meaningful, considering a substantial number of test samples, which are utterly conflicting demands. In this paper, we investigate on sources of error resiliency of CSs, and we propose a technique to haste the DSE that reduces the computational time for error estimation by systematically reducing the test set. In particular, we cherry-pick samples that are likely to be more sensitive to approximation and perform accuracy-loss estimation just by exploiting such a sample subset. In order to demonstrate its efficacy, we integrate our technique into two different approaches for generating approximate CSs, showing an average speed-up up to ≈18
METODO E SISTEMA PER GENERARE UN'IMMAGINE DI USCITA RAPPRESENTATIVA DELLO STATO DI UN IMPIANTO FERROVIARIO
Full text linkUn metodo per generare un'immagine di uscita rappresentativa di uno stato di un impianto
ferroviario comprende una fase di preparazione di una struttura di dati grafici. La
preparazione include le fasi di: rendere disponibile un'immagine di riferimento (3) per
l'impianto, includente simboli (30); una scansione di un'immagine di riferimento (3) per
identificare i simboli (30); una generazione della struttura di dati grafici includente una
pluralità di registri di dati grafici (4), in funzione dei simboli (30). Il metodo comprende una
fase di: predisposizione di istruzioni di gestione, di caricamento delle istruzioni di gestione
della struttura di dati grafici in un terminale di calcolo (2); ricezione, al terminale di calcolo
(2) di una successione temporale di insiemi di dati di ingresso (100), ciascun insieme
essendo rappresentativo di uno stato aggiornato dell'impianto; generazione, da parte di
un'unità d'elaborazione (200) del terminale di calcolo (2), dell'immagine di uscita, sulla
base della pluralità di registri di dati grafici (4) e sull'insieme di dati di ingresso (100)
corrispondente allo stato aggiornato dell'impianto, secondo le istruzioni di gestione. [FIG. 1
Advancing synthesis of decision tree-based multiple classifier systems: an approximate computing case study
No full textSRAM-PUF Authentication Schemes Empowered with Blockchain on Resource-Constrained Microcontrollers
No full textThe pervasive presence of Internet of Things (IoT) devices across diverse critical industrial contexts underscores the crucial role of authentication mechanisms to protect applications from misuses and violations. Those authentication schemes not relying on passwords are built by using cryptography protocols so as to achieve high security levels. However, their efficacy relies on the confidentiality of cryptographic keys and has limitations when implemented on resource-constrained devices. This paper introduces a novel approach to devise an authentication scheme without relying on the storage of keys in device memory and without the presence of a centralised entity. We develop a lightweight mutual authentication scheme utilizing Static Random Access Memory (SRAM)-Physical Unclonable Function (PUF), which leverage on the inherent randomness of SRAM obtained during its manufacturing process to create keys. The proposed solution is further fortified by blockchain technology that is used to provide a decentralised management authority. By incorporating decentralization, scalability, freshness, and non-repudiation, this research contributes to the advancement of secure authentication protocols for IoT-enabled critical industrial applications
Enforcing Mutual Authentication and Confidentiality in Wireless Sensor Networks Using Physically Unclonable Functions: A Case Study
No full text
- …
