Design of Energy Efficient and Dependable Health Monitoring Systems under Unreliable Nanometer Technologies

Abstract

In this paper we investigate the impact of potential hardware misbehavior induced by reliability issues and scaled voltages in wireless body sensor network (WBSN) nodes. Our study reveals the inherent resilience of popular algorithms in cardiac monitoring applications and argues that by exploiting the unique characteristics of such algorithms the energy efficiency and reliability of such systems can be significantly improved. This is achieved by developing a cross-layer design paradigm that utilizes low cost techniques at the hardware and software layers and by optimizing the synergy between them in order to provide intelligent trade-offs between energy, performance and quality. The main idea of the proposed approach is the selective application of costly robust techniques only to the most critical tasks identified at the application layer that are detrimental for obtaining sufficient output quality. Our results show that by ensuring the correct operation of only 37% of the total computations in an electrocardiogram (ECG) monitoring WBSN node we can achieve up to 70% power savings with only 9% degradation in ECG output quality.

Heterogeneous Error-Resilient Scheme for Spectral Analysis in Ultra-Low Power Wearable Electrocardiogram Devices

“…As an energy-saving strategy, the approximate computing paradigm relaxes reliability constraints when errors have a negligible impact from an application perspective. Algorithms in the embedded health monitoring domain operate on noisy acquisitions, while often presenting statistical or qualitative outputs [18]. In such scenarios, in this paper we extend the observations from our previous work [18], and advocate for not needing to provide 100% exactness in all cases, which is also extremely expensive from an energy efficiency viewpoint.…”

“…Algorithms in the embedded health monitoring domain operate on noisy acquisitions, while often presenting statistical or qualitative outputs [18]. In such scenarios, in this paper we extend the observations from our previous work [18], and advocate for not needing to provide 100% exactness in all cases, which is also extremely expensive from an energy efficiency viewpoint. In this paper, we propose to investigat of the approximate paradigm in bio-signal ana extraction applications and utilize their stat for limiting the overhead of classical EC contributions can be briefly described as:…”