Instrumentation & Measurement Magazine 26-1 - 19

Table 2 - Key technologies and their effects on the evolution of I&M systems
Technologies, features
New materials and related
Sensor technology
processing technologies,
micro-electromechanical
systems (MEMS), energy
harvesting
Electronics manufacturing
technology
Flexible and 3D printed
electronics, high-density
packaging technologies
Data processing technology
High-performance
embedded computing
platforms, blockchain
technology
Networking technology
IoT technologies (LoRaWAN,
NB-IoT, SigFox), 5G
networks
system transfer function H(t) (Fig. 2). In a given approach,
measuring the parameter of interest (e.g., in-situ stress of rock
formation, or some specific machine health indicator) requires
deconvolving the measured signal and its transfer function,
and then further characterizing the obtained source signal, or
calibrating it against a known reference source.
Conventional Approach
In traditional I&M systems, two distinct groups of AE signal
processing methods can be identified. The first group involves
emission counting schemes, where the frequency of acoustic
event occurrence is used to assess the average emission energy.
Although easy to implement and effective in certain scenarios,
such methods struggle when additional information, such
as event location or its characteristic acoustic signature, is required,
for instance to differentiate between the measured and
interfering emission sources [8].
Reduced size, energy
autonomy, robustness
Reduced size, tighter
integration of sensors and
electronics
Complex model-based and
data-driven approaches
to I&M feasible without
additional computing
resources
Scalable sensor networks,
wireless control loops
Fully-passive / " zero
energy " sensing
Smart textiles / surface
coatings, robotic skin
AI-enabled instrumentation,
metrological traceability
using blockchain
technology
Machine-machine interfaces,
human-robot interaction
(Industry 5.0)
In the latter cases, the second group of methods are used
which rely on AE signal spectro-temporal decomposition and
interpretation of obtained spectrograms, [8], [9]. This normally
requires a high-bandwidth signal chain that is able to
capture important transient features of AE signals, fast ADC
and a powerful digital signal processing (DSP) engine for
spectrogram computations. These requirements lead to high
energy consumption, which becomes problematic when devices
need to operate for prolonged periods of time, often in a
hard-to-reach places, such as within rotating parts of machinery.
Off-loading the DSP part of the system to a remote, less
power-constrained location can be of limited help, as this usually
leads to increased energy cost of wireless transmissions.
Contemporary Approach
Fig. 2. A representation of the detection signal VAE(t) as a convolution of
the source signal VS(t) (corresponding to strain process in the material) and
the transfer function H(t), which depends on material properties and detector
characteristics. (Adapted from [6], used with permission under Creative
Commons Attribution 4.0 International License.)
February 2023
Ultra-low Power Wake-up Signal Chains: A sporadic nature of
AE led to the idea of deploying a specific, always-on ultralow
power circuitry that could wake up the main DSP engine
from sleep only when an acoustic event of interest has been detected.
The concept is essentially the same as the one employed
in wireless sensor networks where the main power-hungry
RF receiver is awakened by an auxiliary ultra-low power receiver
circuit. In its simplest variant, AE wake-up signal chain
can be made of a single high-speed comparator indicating the
start of an individual emission. If selective wake-up event detection
is required, a possible solution might be to perform a
spectro-temporal decomposition using an analog filter bank,
typically covering a few characteristic frequency bands of interest
[8] (Fig. 3). In the wake-up architecture shown, a critical
portion of power consumption is due to the analog filter bank,
IEEE Instrumentation & Measurement Magazine
19
Key effects on I&M systems Emerging I&M applications
Instrumentation & Measurement Magazine 26-1

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 26-1
