Luca Rizzon

Energy Neutral Wireless Sensing for Server Farms Monitoring

Energy harvesting techniques are consolidating as effective solutions to power electronic devices with embedded wireless capability. We present an energy harvesting system capable to sustain sensing and wireless communication using thermoelectric generators as energy scavengers and servers CPU as heat source. We target data center safety monitoring, where human presence should be avoided, and the maintenance must be reduced the most. We selected ARM-based CPUs to tune and to demonstrate the proposed solution since market forecasts envision this architecture as the core of future data centers. Our main goal is to achieve a completely sustainable monitoring system powered with heat dissipation of microprocessor. To this end we present the performance characterization of different thermal-electric harvesters. We discuss the relationship between the temperature and the CPU load percentage and clock frequency. We introduce a model to simulate the power characteristic of the harvester and a prototype has been realized to demonstrate the feasibility of the proposed approach. The resulting system achieves a minimum 5 min sampling frequency of environmental parameters such as temperature, humidity, light, supply voltage, and carbon-monoxide/volatile organic compounds gases using a MOX sensor mounted on a commercial wireless node with a power budget in the microwatt range.

Wireless sensor networks for environmental monitoring powered by microprocessors heat dissipation

We present an energy harvesting solution for a wireless sensor network for indoor environmental monitoring in data centers. The energy that supplies the nodes is harvested from the heat generated by the server microprocessors using Thermo Electric Generators (TEG), which convert a temperature gradient into electrical energy. We present a performance comparison between two commercial TEGs under different server processor load profiles. We focus our attention on server boards based on ARM CPUs (Arndale with ARM Cortex A15 and Pandaboard with ARM Cortex A9), supplying nodes equipped with gas sensors. From our results and simulations, we are able to demonstrate the possibility of powering a perpetual environmental monitoring WSN with a 0.0027% duty-cycle with the energy scavenged from computationally intensive embedded platform.

Self-Powered WSN for Distributed Data Center Monitoring.

Monitoring environmental parameters in data centers is gathering nowadays increasing attention from industry, due to the need of high energy efficiency of cloud services. We present the design and the characterization of an energy neutral embedded wireless system, prototyped to monitor perpetually environmental parameters in servers and racks. It is powered by an energy harvesting module based on Thermoelectric Generators, which converts the heat dissipation from the servers. Starting from the empirical characterization of the energy harvester, we present a power conditioning circuit optimized for the specific application. The whole system has been enhanced with several sensors. An ultra-low-power micro-controller stacked over the energy harvesting provides an efficient power management. Performance have been assessed and compared with the analytical model for validation.

Embedded soundscape rendering for the visually impaired

The objective of this work is to improve the quality of life for the visually impaired by enhancing the ability of self navigating. Our system provides a 3D audio representation of the environment by synthesizing virtual sound sources corresponding to obstacles or as a guide for a safe path. The key characteristics of our system are low computational complexity and a simple user customization method. Low complexity makes our system suitable for a resource constrained embedded platform, such as a portable device, while assuring the real time reproduction of the auditive stimuli. In the paper, we discuss the basic perception model, its implementation, and experimental results that show the effectiveness of the approach.

Spatial Sound Rendering for Assisted Living on an Embedded Platform

3-D sound can be used to synthesize audio stimuli able to describe spatial information. This can be used as a sensorial substitution of sight for the visually impaired to help them in the task of autonomous orientation and mobility. However, commonly used techniques are computational demanding, therefore not optimal for being implemented in embedded systems. Moreover, the sound localization is specific to each individual and complex to measure or customize. We chose to develop a bottom-up physical model to synthesize a simplified transfer function and playback audio signals over headphones. The model permits the computational requirements to be reduced at the cost of lower accuracy of representation. Still the proposed system can meet the goal of describing spatial information to the listener. Moreover, it can be a promising solution for on-the-go individualization. In this paper we describe the algorithm, the implementation on an embedded platform and present the comparison between HRIR-based synthesis and the proposed simplified physical approach.

Sensory stimulation for human guidance in robot walkers: A comparison between haptic and acoustic solutions

We compare two different solutions to guide an older adult along a safe path using a robotic walking assistant (the c-Walker). The two solutions are based on tactile or acoustic stimuli, respectively, and suggest a direction of motion that the user is supposed to take on her own will. We describe the technological basis for the hardware components, and show specialised path following algorithms for each of the two solutions. The paper reports an extensive user validation activity, with a quantitative and qualitative analysis.

A Smart Walking Assistant for Safe Navigation in Complex Indoor Environments

Large and crowded public places can easily disorientate elderly people. The EU FP7 project Devices for Assisted Living (DALi) aims at developing a robotic wheeled walker able to assist people with moderate cognitive problems to navigate in complex indoor environments where other people, obstacles and multiple points of interest may confuse or intimidate the users. The walking assistant, called c-Walker, is designed to monitor the space around the user, to detect possible hazards and to plan the best route towards a given point of interest. In this chapter, an overview of the system and some of its most important functions are described.

Cyber/physical co-design in practice: Case studies in metroII

To analyze embedded systems, engineers use tools that can simulate the performance of software components executed on hardware architectures. When the embedded system functionality is strongly correlated to physical quantities, as in the case of Cyber-Physical System (CPS), we need to model physical processes to determine the overall behavior of the system. Unfortunately, embedded systems simulators are not generally suitable to evaluate physical processes, and in the same way physical model simulators hardly capture the functionality of computing systems. In this work, we present a methodology to concurrently explore these aspects using the METROII design framework. In this work, we provide guidelines for the implementation of these models in the design environment, and discuss the results gathered with the simulator for two case studies.

Self-powered Active Cooling System for High Performance Processors

Thermal stability in datacenter’s computing units is fundamental to ensure reliability, and durability of the equipment, besides, environmental concern and new regulations require a reduction of the power used. For these reasons, a novel energy neutral hybrid cooling system is proposed. We describe the design, and the prototype’s performance evaluated both in passive and active cooling modes. During normal operating conditions, the thermo-electric energy harvesting system transforms wasted heat into electric energy, and stores it in super-capacitors while the system is providing passive cooling. Active cooling can be activated when a boost in performance requires CPU overclocking, using free energy from the passive step. After the choice of the most suitable harvesting system we designed and tested the prototype on an ARM based CPU, the future core of low-power server architectures. The proposed governor switches to active cooling mode based on customizable thermal management policies. Experimental results demonstrate good passive cooling performance, and several minutes active cooling exploiting the recovered heat.

Efficient Neural Computation on Network Processors for IoT Protocol Classification

The Internet of Things (IoT) brings forth pressing requirements on the service providers in terms of service differentiation, which plays an important role in pricing policies as well as network load balancing. In this paper, we consider differentiation of application level protocols for IoT from general application protocols through flow classification. We implement a neural network classifier that can run at wire speed reaching 100 Gbps on a network processor. In particular, we study approximations which allow us to efficiently compute the neural network output, while complying with the network processor limitations, which does not provide multiplication or other complex mathematical operations. The results show that the implementation is efficient and that the classification error is negligible.