HEAL-WEAR: An Ultra-Low Power Heterogeneous System for Bio-Signal Analysis

Abstract

Abstract-Personalized healthcare devices enable low-cost, unobtrusive and long-term acquisition of clinically-relevant biosignals. These appliances, termed Wireless Body Sensor Nodes (WBSNs), are fostering a revolution in health monitoring for patients affected by chronic ailments. Nowadays, WBSNs often embed complex digital processing routines, which must be performed within an extremely tight energy budget. Addressing this challenge, in this paper we introduce a novel computing architecture devoted to the ultra-low power analysis of biosignals. Its heterogeneous structure comprises multiple processors interfaced with a shared acceleration resource, implemented as a Coarse-Grained Reconfigurable Array (CGRA). The CGRA mesh effectively supports the execution of the intensive loops that characterize bio-signal analysis applications, while requiring a low reconfiguration overhead. Moreover, both the processors and the reconfigurable fabric feature Single-Instruction / MultipleData (SIMD) execution modes, which increase efficiency when multiple data streams are concurrently processed. The run-time behavior on the system is orchestrated by a light-weight hardware mechanism, which concurrently synchronizes processors for SIMD execution and regulates access to the reconfigurable accelerator. By jointly leveraging run-time reconfiguration and SIMD execution, the illustrated heterogeneous system achieves, when executing complex bio-signal analysis applications, speedups of up to 11.3x on the considered kernels and up to 37.2% overall energy savings, with respect to an ultra-low power multicore platform which does not feature CGRA acceleration.

i-DPs CGRA: An Interleaved-Datapaths Reconfigurable Accelerator for Embedded Bio-Signal Processing

“…A similar conclusion is drawn in [4] and [5], which propose an embedded platform for bio-signal processing in personal health monitors, an increasingly relevant domain with ultra-low power constraints [8]. Similarly to us, [5] describes a CGRA with multi-DP RCs operating in SIMD (single instruction-multiple data) mode. Their proposed strategy, however, is only beneficial when the same acceleration is requested by different processors, themselves executing in SIMD mode.…”

“…Similarly to [5], we characterized the system components (including processors, memories and the i-DP CGRA) at the post-synthesis level, targeting a 65 nm UMC technology. The obtained energy parameters were then employed to annotate a cycle-accurate virtual platform (specified in SystemC), allowing fast whole-system simulations of entire applications.…”

“…First, 8-lead Compressed Sensing (8L-CS) [5] derives a number of random features as linear combinations of input samples [6]. In the targeted implementation, we adopted signal windows of 1024 samples, and a compression ratio of 50 %.…”

“…As opposed to [5], our approach results in high efficiencies regardless of the structure and operating states of other system components (e.g., processors) at run-time. Our contribution is two-fold:…”

Heterogeneous and Inexact: Maximizing Power Efficiency of Edge Computing Sensors for Health Monitoring Applications

“…However, bio-DSP often involves complex functions and hence introduces high power requirements. To address this challenge, previous works have shown that task-level parallelism, typically present in bio-DSP applications, can be leveraged by multicore architectures with Single-Instruction Multiple-Data (SIMD) capabilities [1], while the execution of computational hotspots (kernels) can be effectively supported by Coarse-Grained Reconfigurable Array (CGRA) accelerators [2] [3] [4]. Thanks to the faster and more efficient execution achieved with the aforementioned techniques, smart bio-DSP IoT systems can remain longer in deep-sleep mode, consequently reducing their energy consumption.…”