Aerospace and Electronic Systems - June 2019 - 48

Prospects of Distributed Wireless Sensor Networks for Urban Environmental Monitoring
achieving ultrahigh capacitance resolution, which is significantly below 10-18 F. Such a miniaturized device (see
Figure 4) can be produced at negligible cost and can be
located in personal and wearable air quality monitors as
well as on board of a UAV. Power dissipation is below
100 mW, being negligible with respect to the pump dissipation. The pump is required to aspire the air volume required
for statistical significance (at a flow rate of about 1 l/s). The
capacitive approach is characterized by a minimum detectable size of about 1 mm, analogous to what achievable with
optical techniques based in CMOS cameras [30]. In order to
detect nanoparticles, silicon resonators appear to be one of
the most promising directions of development [29].

ENERGY HARVESTING

Figure 4.
Dynamic air quality monitoring network leverage a single-chip
dust monitor able to perform granulometric spectroscopy of PM
(down to 1 mm diameter) thanks to high-resolution microcapacitance measurement [27].

Another important aspect is data analysis. It is evident
that the most relevant value of a pervasive environmental
network is not only in mapping in real time the status of
all parameters in all nodes (alerting in case of faults), but
in processing the measured data in a proper way to feed a
model of the dynamic system under investigation. In the
case of water and air quality, complex chemo-physical
models have been developed, whose main goal is prediction. Machine learning (such as clustering and principal
component analysis to elicit correlations) and "big data"
techniques need to be adopted for the elaboration of dense
networks [24], [25].
The second example refers to an application for air pollution monitoring, where a similar matching between the
sensitive volume and a small particle to be detected has
been adopted within the MINUTE Project [26]. It led to the
realization of microcapacitive sensors enabling the detection (i.e., counting the concentration) and the granulometry
(i.e., the determination of the size spectrum) of PM on a single chip [27]. In fact, several efforts have been devoted to
the realization of dense networks of air quality monitors
based on commercial low-cost devices [28]. Among different technologies, the deepest miniaturization can be
achieved by solid-state detectors [29]. The detection
approach is based on the measurement of the capacitance
variation produced by the deposition of a single PM particle
in a microcapacitor. Assuming an average dielectric constant of the dust, it is possible to extract the size class of single dust particles. The sensor is implemented in a single
CMOS chip [27], hosting both the electrodes with a sensitive area of 1 mm2 and the low-noise front-end electronics,
48

The minimization of energy consumption and the possibility to harvest energy from the environment are fundamentals aspects for both static and mobile WSNs [31], [32]. For
the former, it is a key point to avoid battery replacement,
while, from the standpoint of the latter, the reduction of
energy consumption will allow to extend the duration of the
missions. The largest pressure is thus on battery technologies, harvesting solutions and low-power design of the electronics, and of the radio protocols. Developments have been
advancing from both sides: more efficient energy harvesting
approaches to be combined with the decreasing power dissipation of lower power circuits.
Energy can be harvested from several sources: solar
power, electromagnetic radiation, mechanical vibrations
(induced by sounds, wind, people walking, waves, passing
vehicles, etc.), flow of mass (as in the case of water) or
heat, and temperature gradients. In the operation of a wireless sensing node, the two major sources of power dissipation are processing and data transmission. The former is
addressed in two ways: either the traditional digital processing is replaced by analog processing (as in [33]), or
processing is completely shifted to the receiving units.
The major remaining power-hungry operation is wireless
data transmission. Thus, power-aware solutions are under
development [4]. The typical strategies to reduce power
dissipated for communication are: i) distance reduction
(possible if several nodes are pervasively distributed
across the area of interest), ii) reduction of the transmission frequency, iii) adoption of low-power protocols, such
as Sigfox [34], LoRa [35], BLE, NB-IoT, requiring the
installation of additional radio infrastructures with respect
to the current LTE one, iv) adoption of suitable grouping
topologies (mesh, clustering, see Figure 5 as an example).

DATA FUSION
The combination of heterogeneous sensing sources is
clearly a major challenge. An example is given by the

IEEE A&E SYSTEMS MAGAZINE

JUNE 2019

Aerospace and Electronic Systems - June 2019

Table of Contents for the Digital Edition of Aerospace and Electronic Systems - June 2019