Instrumentation & Measurement Magazine 23-3 - 36

demining teams often encounter between 100 and 1,000 false
alarms per mine [2]. Since the standard operating procedures
most often require that each piece of metal is removed from the
ground (no metal = no mine policy), this means that hundreds
of innocuous items must be prodded and excavated, which increases a deminer's fatigue and may eventually lead to loss of
concentration, missed mines or casualties.

Ground Penetrating Radar (GPR)
Motivated by the need to reduce the false alarm rates in humanitarian demining, researchers have investigated other
sensing modalities to improve clutter rejection. A general idea
behind most of these efforts was to identify a viable sensing
modality that could ultimately lead to development of a portable, real-time landmine confirmation sensor. Such a sensor
would not be used as a standalone device, since the probability of detection close to that of an MD is extremely difficult
to achieve, but rather as a tool to confirm or reject the alarms
coming from a metal detector. The idea of applying GPR to
landmine detection came from the fact that it can be observed
as a sensing modality complementary to MD. Whereas MDs
detect metallic content of buried landmines, GPR could be
used for detection of their plastic casings (and potentially explosive charges). Simultaneous detection of metal parts and
the surrounding plastics could then be used as a more reliable
indication of a buried landmine.
GPR operates by emitting electromagnetic waves (between
few hundred MHz and several GHz) into the ground and analyzing the return signals generated by reflections of the waves
at the boundaries of materials with different indexes of refraction. Weak return signals are picked up by receiving antenna(s)
and processed to create either an image of an underground object or an audible detection signal. In general, GPRs require
very sensitive receivers and sophisticated signal processing to
extract the target signal from background interference and signals corresponding to clutter objects such as rocks, plant roots,
or pockets of water [2], [3].
Although dual-sensor detectors, combining MDs and
GPRs into a single device, have been developed and used in
the field for some time, GPRs are still considered a rather expensive technology, since the price of a typical handheld GPR
device is an order of magnitude higher than that of a stateof-the-art MD. Consequently, dual-sensors detectors are
nowadays predominantly used by the military, so the engineering challenge is to devise affordable dual-sensor detectors
for the humanitarian demining market. A potential solution
might be to take advantage of relatively inexpensive electronic
components used in consumer products in a rapidly developing wireless communications market [3].

Lessons Learned
Other sensing technologies such as acoustic/seismic sensing,
infrared cameras, nuclear quadrupole resonance for bulk explosive detection, different chemical/biological methods for
explosive vapor detection, etc. have also been investigated in a
context of close-in landmine detection [2]. Although valuable
36	

in a scientific sense, these research efforts did not result in detection tools that could be reliably used in practical demining
scenarios. Such experimental devices were either not sensitive
enough, too bulky, too power-hungry, too expensive, difficultto-use, or just incapable of operating in adverse conditions of
a minefield.
Motivated by the low take-up of new technologies, there
has been a notable decline in funding of research activities for
humanitarian demining, starting from the mid-2000s to present days. Consequently, the demining industry has turned to a
concept of incremental improvements to established detection
technologies and methods, such as MDs, instead of relying on
development of completely new technologies [2], [8].

Improving the Abilities of Conventional
Metal Detectors
As already pointed out, MDs are relatively simple and inexpensive devices, and there is generally a good understanding of
their benefits and drawbacks, both in the scientific and demining community. Compared to first models designed back in the
times of World War II, modern state-of-the-art MDs are more
sensitive, low-power, more convenient to use, as well as capable
of compensating the electromagnetic effects of mineralized soils
(albeit with some loss of sensitivity). In general, the MD industry
is still considered rather conservative, so up until very recently
it was not so uncommon to find that fully analog models of MDs
are still being produced. On the other hand, the industry has
accumulated a lot of practical/empirical knowledge through
field exploitation of these devices over the past decades. Such
know-how is priceless for applications such as humanitarian
demining. Some of the well-kept secrets of the industry relate
to simple and proven electronic circuit designs, mechanical/
thermal designs of coil assemblies that deliver greater stability during sweeping motion, coil shielding to reduce capacitive
effects without disturbing the basic sensitivity of a device, etc.
Building new I&M concepts on top of existing knowledge seems
therefore like a promising approach aimed to provide the incremental improvements the demining industry is looking for.

Signal Level
Observing conventional MDs at the level of raw signals, we
can identify three specific areas where improvements can be
made both in a context of basic metal detection, as well as metal
characterization or discrimination:
◗◗ Excitation signal chain,
◗◗ Receiver signal chain,
◗◗ Search head geometry.
When it comes to the type of excitation, there are currently two kinds of MDs: the ones using continuous wave
(CW) signals of sinusoidal waveform and those employing
pulsed current excitation [2]. CW detectors used in demining operations typically operate at a single frequency, which
is convenient since relatively simple circuits can be used. On
the other hand, discriminating different types of metals (e.g.,
aluminum detonator caps from steel coins) is not possible unless multiple excitation frequencies are applied [3], [8]. Moving

IEEE Instrumentation & Measurement Magazine	

May 2020
Instrumentation & Measurement Magazine 23-3

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 23-3
