A Nose for Security

New sensor technology designed to monitor toxic and explosive terrorist threats

• By Hugh Greville
• Nov 01, 2006

FOR the past 15 years, the global terrorist threat has manifested itself through many different targets and methods. These are as diverse as crude truck bombs in front of buildings, chemical attacks in subways, threat of aircraft terrorism, attacks on public transport and mass-gathering attacks. The constant development of new and insidious methods for penetrating detection barriers was highlighted by the most recent discovery in London of a plot to bring onto airplanes liquids that would later be combined in-flight to form explosives.

The diversity of terror methods and targets presents challenges for currently-deployed detection technologies. It also is a significant opportunity for developers of new technologies to confront the challenges in innovative ways.

The diversity of terror methods and targets presents challenges for currently-deployed detection technologies. It also is a significant opportunity for developers of new technologies to confront the challenges in innovative ways. However, for technology developers, the challenge will not be met if the developers do not fully understand the environments in which the equipment must work, the range of tasks that must be fulfilled and, most importantly, the need for a single, low-cost detection platform capable of detecting a variety of disparate threats.

Providing low-cost technologies is not just a case of better economics for the user, it also allows the movement toward a completely new paradigm of detection capabilities in the form of multiple, distributed sensor networks in key places for continuous monitoring of broader areas. This impact can be seen in a number of as yet undefended arenas.

One core platform technology in advanced stages of development is the electronic, or artificial, nose for the detection of odors or volatile chemicals. A truly biomimetic device, it can detect many chemicals without the need for pre-concentration or pretreatment of the sample. These devices are inexpensive and adaptable to almost any threat. There are many environments that benefit from the deployment of low-cost sensor technologies.

Understanding the Environment
At airports, major events and large buildings, there are typically entry checkpoints. People passing through a checkpoint undergo screening procedures. All activities are controlled in an effort to screen baggage and people for recognized threats and to observe people in critical ways. Although these procedures are necessary, the process is slow, cumbersome and costly. Screening is often the only time that authorities have any contact with the threats.

Detection technologies used in controlled environments are typically static, with the most commonly used devices for explosives being particle analyzers, such as ion mobility spectroscopy (IMS) and X-ray equipment, which detect suspicious-looking articles, but not the explosive materials themselves. Both provide reasonable security against recognized threats, but the technologies are expensive, may require that surfaces be wiped for sampling and have been less reliable than hoped for in some situations. These technologies also have certain explicit limitations such as the inability to identify potentially harmful bottled liquids.

New technologies are being tested, but they tend to be specific to a particular task. The use of Raman spectroscopy for on-site analysis of liquids in containers is an example. As with IMS technologies, Raman requires direct contact of the suspect object and is fairly costly.

The checkpoint alone, although critical, does not provide comprehensive security, as interrogation occurs as a single point of contact and relies on equipment that does not permit continued or redundant monitoring once the examination portal is passed. Indeed, the checkpoint itself may represent a target, as it is a singular place where people are guaranteed to congregate.

For a comprehensive security solution, technologies must be able to detect threats beyond the static checkpoint at multiple sites and be able to cope with open, uncontrolled environments. Open environments can range from airport or railroad station concourses, where people are continually moving in many different directions, to open-field HazMat spill situations. These environments are often dirty, windy, wet, as well as uncontrolled with regard to temperature and are therefore not suitable for inspection by current, static checkpoint detection equipment and procedures. New technologies should be able to handle a range of conditions and have the ability to constantly monitor throughout the dynamic space of the open environment.

Currently, the primary security/detection tool is human observation, primarily carried out by using human/canine patrols, cameras and human observers. Although not sensitive enough to detect hidden chemicals and objects, human surveillance takes advantage of trained personnel to interpret associations made across various observations. This way, threats are identified by security personnel seeing, hearing or smelling something out of the ordinary. Some new sensor technologies are designed to reflect human sensing paradigms and most detection devices are extensions of the senses used by humans to observe their surroundings. For example, X-ray machines are used to enhance visual inspection, and CCTV cameras are installed in city centers, sporting arenas and other areas. Similarly, specialized microphones can be used to pinpoint the direction of inappropriate noises such as gunfire.

Detection devices that mimic or amplify other human senses are less developed. Apart from the use of canines, there is no technology that provides enhancements of the sense of smell, even though many explosives, chemicals and people give off signature odors that can be detected by dogs, often in the presence of masking agents and other interference.

One system that aspires to mimic and enhance the sense of smell is a biomimetic device called an electronic or E-nose. These machines generally employ low-cost sensor technologies that can be deployed in a distributed fashion at multiple locations in ways that should help to solve security problems in both closed and open environments.

The E-Nose
For detecting, discriminating and identifying volatile compounds in the air, one of the most highly developed detection devices is the olfactory system of animals. The way olfactory systems encode odors is via broadly responsive receptors, rather than specific ones, that then produce widespread patterns that are interpreted by parallel processing elements in the olfactory areas of the brain. Analogous with playing notes on a piano, odors are represented by chords rather than single notes. The distributed process is thought to give rise directly to the broad discrimination ability and fault tolerance characteristic of the sensitive olfactory systems of animals like dogs.

E-noses mimic the biological aspects of the olfactory system by using arrays of broadly responsive sensors and pattern recognition methods for processing sensor output. E-noses also may incorporate other biological mechanisms, such as sniffing and exploitation of the aerodynamics in the nasal cavity, in order to maximize odor detection efficiency. Such odor detection systems offer a number of advantages over traditional analytical devices: they directly sample the air, so no sample preparation is necessary, and odor detection time is limited only by the time required for the sensors to respond and calculate result (in present devices as little as 2 to 3 seconds or less). In addition, due to the nature of detection by the sensor array, relatively small arrays of broadly responsive sensors are theoretically capable of discriminating among large numbers of different compounds with the flexibility of detecting and discriminating compounds of future interest. These advantages have an impact on the performance and cost of the technology. In turn, lowering the cost of detection -- by at least 75 percent over current technologies -- will have a significant impact on the ability to move detection away from the single, controlled checkpoint scenario to the deployment of multiple detection locations in uncontrolled environments.

Sniffing Out Trouble
One important application of low-cost sensor technology is the detection and identification of vapor-phase hazardous materials. If toxic chemicals are present, HazMat specialists must suit-up in full breathing apparatus and HazMat gear before identifying and quantifying the gas phase threat. This is a severe logistical problem for the HazMat team, as they not only have limited working time with the breathing apparatus, but the suits are cumbersome to work in and limit their ability to perform complex tasks. The use of multiple, low-cost chemical detectors that can identify, quantify and discriminate chemicals can alert HazMat teams as to when protective gear is required, allowing them to set perimeters of the area and to permit the mapping of the spatial distribution of the threat in real time. This mitigates the logistical problem and identifies the boundaries at which safety and security measures need to be implemented.

Other detection applications include a ballpark, office building or other sensitive arena. Remote detection units that are sensitive to explosions can be deployed to monitor these controlled areas. Individual and/or networked odor detection devices are desirable in other, less-controlled areas, such as in an airport or airplane lavatory, where a terrorist might go to mix explosive constituents. Since there is availability and easy access to materials and methods for constructing cheap, homemade bombs, high concentrations of precursors or products are likely to be present during the assembly process. This indicates that odor detection is a viable approach to identifying these sites and activities. This approach may provide defenses against terrorists using small-scale production of explosives made in concealed and seemingly innocuous venues, as was the case in the alleged London terror plot.

Development of low-cost platforms for E-nose devices can move the sense of smell into the detection arena now only occupied by the many devices that mimic sight and hearing. There is potential to fully integrate distributed networks with hybrid sensor systems for audio, visual and odor detection, transmitting information back to a central location. Multi-modal monitoring of remote locations is useful in enclosed spaces such as rail cars, waiting areas, ventilation systems and meeting centers. Indeed, as such distributed technologies become more commonly used, high-traffic areas may be custom engineered to take advantage of these integrated capabilities in the same way that casinos are designed around video monitoring technologies.

The ability to deploy inexpensive sensors that enable continuous, remote and distributed monitoring of both open and restricted environments is a desirable goal for the detection of threats that have an odor signature. In the security field, this ability is restricted by the detection capabilities of currently devices that are limited both by performance and high cost. As technologies evolve, they must be flexible enough to continuously improve, prompting more deployments at a greater amount of venues. Recent, state-of-the-art electronic noses are among the most attractive technologies for achieving these goals.