A Platform for Adaptive Audio Sensing in IoT

Abstract

The Internet of Things (IoT) is the technology which connects Things that surround us to the Internet. The interest in this technology has been increasing along with its fields of interest in recent years. This project focuses on checking the feasibility of a sensing audio system in the IoT environment. We think that an audio sensing system composed of lots of small and low-cost devices has potential in several scenarios such as dBA noise measurement and room occupation. We implement a prototype of a sensor node able to detect the human audible spectrum, which we use as testbed. We selected hardware and software technologies in the way that the sensor node can work in an IoT environment: low-energy consumption, small footprint and wireless connection are the characteristics that the sensor node must have.

The key feature of our sensing system is the use of the audio spectrum analysis. In fact it allows us to work with audio chunks of tens up to hundreds of milliseconds to be transmitted instead of streaming the audio continuously, reducing the information to send and the active time of the node. The resulting sensing pipeline consists of Detection, Sampling, Fast Fourier Transform (FFT) and Transmission stages, continuously iterating after an idle state period. The firmware for the Detection, Sampling and FFT stages was developed. Random Access Memory (RAM) and Read Only Memory (ROM) occupation, time occupation and power consumption measurement were performed on different test cases and sample lengths. The FFT stage can also be performed in a more powerful node and the pros and cons were evaluated.

The measurement tests provided evidence on efficient RAM usage on the testbed while simultaneously working with very small audio chunks. However, this deteriorated when we performed the FFT on board because it requested extra space on the RAM. From the gathered data from the time occupation and power measurement tests, we developed an energy model of our sensor node that helps to configure the energy profile of the sensing system. Performing the FFT is a hallmark of our sensing system, its performance on board is justified only if the data are needed locally or for privacy-related reasons. We also suggest some proposals to improve the FFT stage.