Shapelet-based methods have emerged as accurate and interpretable tools for time-series classification in machine learning applications. Although these methods find wide applicability, nowadays their use is restricted due to the heavy computational burden associated with numerous computations of z-score normalization and Euclidean distance. The current article addresses this issue with the proposal of a parameterizable parallel hardware accelerator to perform the aforementioned computations. Results show that our dedicated shapelet distance engine can significantly reduce the run time when compared with a software implementation.

Highly-abstracted hardware operation

Entity Interface

Control Finite-State Machine (FSM) diagram

Block diagram

Hardware vs software timing

Total power and area proportions for different parameterizations

Power and area proportions among the accelerator's components for one configuration of parallelism and maximum length

The described shapelet distance hardware accelerator is licensed under the CERN Open Hardware License S v2, as described below.

Copyright GMicro UFSM 2020. This source describes Open Hardware and is licensed under the CERN-OHLS v2 You may redistribute and modify this documentation and make products using it under the terms of the CERN-OHL-S v2 (https:/cern.ch/cern-ohl). This documentation is distributed WITHOUT ANY EXPRESS OR IMPLIED WARRANTY, INCLUDING OF MERCHANTABILITY, SATISFACTORY QUALITY AND FITNESS FOR A PARTICULAR PURPOSE. Please see the CERN-OHL-S v2 for applicable conditions. Source location: https://github.com/vctrop/shapelet_distance_hardware_accelerator As per CERN-OHL-S v2 section 4, should You produce hardware based on these sources, You must maintain the Source Location visible on any product you make using this documentation.