Printer Friendly

Perimeter protection - define, detect, and defend.


THE VILLAGE OF ST. BASILE-LE-Grand in Quebec was an innocuous settlement within commuting distance of Montreal until August 23, 1988. On that date, one heedless act of arson in an unguarded warehouse storing waste transformer oil laden with PCBs caused the evacuation of 3,500 people from the surrounding area for over two weeks. The incident triggered speculation that the area might be reduced to an uninhabited wasteland for as long as 10 years.

Rather than a dramatic act of organized terrorism directed against a nuclear establishment, this obscure event in a remote village illustrates a modern dilemma. Today's society randomly produces a myriad of sites ranging from the residences of public officials to nuclear generators and chemical dumps, and all are potential targets.

Historically, security forces used formidable physical structures and inexpensive personnel to secure perimeters. Today's high cost of employees and proliferation of high security sites renders such traditional means of securing perimeters impractical. Appropriate technological solutions are sought that can dramatically reduce staffing requirements without sacrificing performance. Before inspecting some promising solutions, let's look at the basics.

Surveillance and detection. Proper protection of an asset requires a defined surveillance zone and some means of detecting intrusions through, under, or over that zone. Observing the US military standard for effective human surveillance requires eight guard towers and approximately a 40-person staff on a perimeter of one mile--a distance that only encompasses a comparatively small facility.

Also, humans are excellent detectors when confident and alert, but their efficiency rapidly falls off under conditions of fatigue, boredom, or stress. Frequent rotation of guards helps correct this problem but at the expense of using even more personnel.

Assessing alarms. Not every intrusion is hostile and not every alarm represents a threat. A human's greatest talent--the ability to assess the nature of a threat quickly and to respond realistically--is seriously degraded by nonvalid alarms.

Nonvalid alarms fall into two categories--nuisance alarms and false alarms. Nuisance alarms include nonvalid threats from discernable causes, such as a marauding raccoon climbing a sensor-equipped fence. False alarms include situations where no valid threat or cause of the alarm can be discerned. A high rate of nonvalid alarms has two negative aspects. First, it increases a security guard's work load because good security demands assessment of every alarm; second, it diminishes that person's confidence and responsiveness. Aesop's fable about the boy who cried wolf still holds true. High nuisance alarm rates (NARs) and false alarm rates (FARs) condition humans to discount all alarms--including valid ones.

Delay. Ideally a perimeter should have a barrier that permits time to assess the nature of the threat and marshal an adequate defense if necessary. This barrier should be sufficiently formidable and remote from the asset under protection that potential intruders (or escapees) cannot achieve their objective by surprise and main force--as happened in the infamous truck bombing of the US Marine base in Beirut.

Adequate means of response. A total perimeter security system obviously must provide for an appropriate and timely response to an intrusion assessed as threatening.

THE SEARCH FOR SOME TECHNOlogical means of reducing the exorbitant amount of personnel required to protect society's vital resources has concentrated on the first two elements of classic perimeter defense--detection and assessment of threats. Today's best perimeter security research is directed toward two goals. The first goal is to develop devices that assess alarms automatically and only alert security guards when a valid intrusion is likely. The second goal is to develop means of presenting alarm information to guards in ways that assist the assessment process.

The objective is to reduce staff required for perimeter protection by making better use of fewer people. A dramatic improvement in detection and assessment capabilities could almost dispense with any required ratio of personnel to perimeter length and permit the establishment of strategically located security control centers.

Before proceeding to examples of workable high-tech solutions, let's clarify the situation by examining an ancient fallacy that has recently reappeared. It's called "the quest for a perfect sensor."

Historically, much imagination has been spent on devising sensors to overcome human sensing deficiencies; geese, dogs, and trip wires strung to bells are among the many avenues explored. Not surprisingly, therefore, most security-oriented companies have focused on developing better sensors. This focus has produced a whole spectrum of sensors that can detect parameters such as heat, motion, pressure, and sounds.

Most of these sensor-focused approaches have failed because they do not account for the power of nature. Outdoor security is rife with causes of nuisance alarms. High winds trigger audio sensors on fences, animals walking through free zones trip motion detectors, and rapidly changing cloud and thermal conditions fool heat detectors. The list is endless. High alarm rates erode confidence in a system. Retaining a guard's confidence in a system by reducing false alarms is more important than a system's technological brilliance.

One example of an inappropriate application of brilliant technology was the urine sensor, which was developed for the famous MacNamara Fence in Vietnam to detect the enemy moving through dense foliage. Once the working principle was understood, the Vietcong easily thwarted the sensors by hanging buckets of urine on trees in desolate jungle areas, which subsequently were bombed out of existence. Note that the sensors worked perfectly, but the system failed.

The point is that detectors that function perfectly in the laboratory are often swamped by stimuli originating outdoors or by counterresponses in the real world. The result is an overload of detection with little assessment. For proper assessment to occur, the work load placed on guards has to be kept to reasonable levels. Two goals mentioned earlier are capable of dramatically reducing the work load and hence, staffing requirements--developing detectors that include an automatic screening (partial assessing) of alarms and developing means of presenting alarm information in ways that greatly assist the assessment process.

IN PERIMETER SECURITY, THE KEY TO automatically screening alarms is to apply an advanced technology to detect a human characteristic that seldom occurs elsewhere in nature. One perimeter intrusion detection system uses guided radar technology. Guided radar literally, and accurately, detects specific radar cross section characteristics of humans. Keying in on these characteristics renders this technology immune to most environmental effects such as blowing debris, high winds, birds, small animals, rain, temperature extremes, dense foliage, or changes in light intensity.

Guided radar technology creates a detection zone between a pair of ported coaxial cables that have been constructed to leak electromagnetic energy at a low VHF frequency. Pairs of such cables are buried around the protected perimeter and, when energized, the electromagnetic coupling between the cables (one functions as a transmitter and the other as a receiver) produces a detection zone or radar fence. Interaction with local soil patterns and other mediums forms a sensitivity profile unique to each perimeter cable pair. A digital copy of this profile is stored in a computer, which monitors changes.

When this detection zone is interrupted by an intruder, the computer notes the change in the profile and raises an alarm if the magnitude of change exceeds a predetermined threshold. Additionally, the computer determines the position of the potential intruder by analyzing the return signal and visually presents the location to the guard on a display.

A low VHF frequency is optimal for detecting humans as radar targets. At this frequency, small animals and most other nonthreat targets have a much smaller radar cross section, and therefore, response to these objects falls below the human detection threshold.

A 10-lb. animal, for instance, appears to be significantly smaller than an average human at this operating frequency. The automatic screening of detections not only reduces the work load (and hence, personnel) needed for long perimeter surveillance but also increases guards' confidence in the validity of alarms, so they take a careful look at all reported detections.

Two cautions are in order. The addition of automatic prescreening of detections and an enhanced information display cannot remedy deficiencies in an inadequate security system. Also, no single detection system gives perfect security.

Guided radar works well because it has fundamentally good security attributes. It is covert so an intruder is unaware of being detected and surprise is on the side of the guard forces. In addition, it creates a zone that is physically too high and wide to jump. Attempts at bridging the detection zone increase an intruder's visibility and chances of detection by other means.

One type of detector alone is rarely sufficient for the highest level of security outdoors. Weather changes and unpredictable stimuli all go with the territory. Detectors are most helpful when they are part of a system that processes information from multiple detectors, correlates the results to the actual situation, and presents information about the threat to the guard in an enhanced format. Digital processing offers the key to such a system.

HUMANS HAVE AN AFFINITY FOR visual identification and assessment. Consequently, closed-circuit television (CCTV) is extremely popular among security people. However, CCTV is not innately compatible with an outdoor environment. Changing shadows during the day and poor lighting at night adversely affect video performance. Reflections, wind-induced camera motion, electromagnetic noise, birds, small animals, insects, rain, fog, snow, and dust constitute only some of the difficulties for video surveillance outdoors. However, improvements in video camera performance and the use of special technologies, such as infrared bandwidths, overcome most obstacles.

Existing CCTV installations address two important requirements. First, they reduce the personnel formerly required for visual surveillance of large facilities. Second, they offer guards the instant ability to assess situations before responding.

The greatest weakness of CCTV is that guards are subjected to the mind-numbing monotony of scrutinizing banks of video monitors for hours at a time. In fact, the human inability to sustain the task makes unaided video surveillance almost useless. This is where digital signal processing comes in.

A patented application of digital processing to CCTV, known as DAVID (digital automatic video intrusion detector), offers a better human-machine interface. This interface effectively equips the CCTV system to monitor video images from the surveillance zone automatically. It detects movements that require human attention and furnishes graphic enhancements to help a guard assess the nature of the intrusion. The concept of automatic video change detection is not new, and over the years a variety of techniques have been developed to look for changes in the video scene. As mentioned earlier, the outdoors has an immense capacity to produce video changes. In the past, the missing component has been a cost-effective intelligent processor that can recognize the difference between environmentally caused video changes and changes that indicate a potential intrusion.

Modem digital signal processing has this capability to analyze the size, shape, and movement patterns of the video change. Digital processing can scan multiple cameras and point to the exact location in the picture of a potential intrusion. The processor's analytic capability combined with the graphics support results in a very low alarm rate and a work load that a single operator can handle with ease and confidence.

The technology is compatible with virtually all video standards and camera types. In long-term field tests using DAVID-processed CCTV in all kinds of weather, this technology achieved a detection rate higher than 99 percent and a false alarm rate of less than one occurrence every 25 days. The highly successful filtering of nonvalid threats reduced the strain on guards while justifiably increasing their confidence in the system. The same field tests proved that adding computer graphics to CCTV substantially improved a guard's ability to track and assess a detected threat situation.

Perimeter protection is an ancient as society itself, and the basics have not changed over the years. Today, the creative addition of digital processing to perimeter security systems can reduce the number of personnel required to protect the essential assets of modern society. Digital processing also enables carefully selected technologies to enhance the unparalleled human ability to assess and respond to a potential threat.

Technology is transforming both our society and the ways in which we protect its vital assets. Yet good perimeter protection usually does not result from a flash of technological genius. Rather, it is the reward for many years of heavy R&D investment in painstakingly plugging the loopholes nature unerringly finds in every technology.

John C. Alexander is manager of security systems market development for Computing Devices Company in Ottawa, Canada, a Control Data Company that specializes in advanced digital signal processing applications.
COPYRIGHT 1989 American Society for Industrial Security
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Alexander, John C.
Publication:Security Management
Date:Aug 1, 1989
Previous Article:The private perspective.
Next Article:Targeting professionalism.

Related Articles
The first line of defense.
Intrusion detection: inside and out.
CCTV watches the world go by.
Sensing your way to security.
Getting a sense for danger.
Prevent viruses on enterprise WLANs: security gateways provide protection from within the network perimeter.
The security consultant.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters