Digital Fixed-Point Low Powered Area Efficient Function Estimation for Implantable Devices

This article introduces a new multiplier-less 32-bit fixed point architecture for estimating complex non-linear functions based on adapted shift only series expansions. This novel hardware structure has been proposed for use as a dedicated core unit in implantable medical devices. Its implementation...

Descripción completa

Detalles Bibliográficos
Autores: Romaine, James Brian, Ashley, Thomas Ian, Pereira Martín, Mario
Tipo de recurso: artículo
Fecha de publicación:2022
País:España
Institución:Universidad Loyola Andalucía
Repositorio:Brújula
OAI Identifier:oai:repositorio.uloyola.es:20.500.12412/5074
Acceso en línea:https://hdl.handle.net/20.500.12412/5074
Access Level:acceso abierto
Palabra clave:Hardware
Power demand
Field programmable gate arrays
Estimation
Epilepsy
Implants
Table lookup
Descripción
Sumario:This article introduces a new multiplier-less 32-bit fixed point architecture for estimating complex non-linear functions based on adapted shift only series expansions. This novel hardware structure has been proposed for use as a dedicated core unit in implantable medical devices. Its implementation in FPGA produces a mean squared error of 0.23% over the functions sin(x),cos(x),eix and tan−1(x) when compared to unrestricted CPU implementations. These results are achieved with the use of only 133 sliced registers and 399 Look-up-tables (LUTs). Furthermore, the hardware performs extremely well in our hardware-in-the-loop real use case application for the detection of epilepsy by correctly detecting true positive seizures. When implemented into 130 nm technology via GOOGLE Sky130 PDK and Openlane EDA tools, the ASIC occupies a space of 0.0625 mm2 which represents a 47% reduction when compared to competitors. In addition, its power consumption is reduced to 6.46 mW at 100 MHz fo and just 0.4 μW at 1KHz fo .