The new horizon: transferring defense technology to law enforcement.
The Government Technology Transfer Program(1) has made another promising approach available to law enforcement. This initiative enables Department of Defense and commercial organizations to work together to assist law enforcement through the application of defense-related technology.
As part of this initiative, the National Institute of Justice (NIJ) operates the National Law Enforcement and Corrections Technology Center (NLECTC),(2) as well as four regional technology centers across the country. These regional centers use existing facilities and resources to provide specialty support to NIJ's Office of Science and Technology and to the law enforcement and corrections field. Each center has a specific technological focus.
Rome Laboratory hosts the Northeast Regional Center. For more than 40 years, Rome Laboratory has developed the technologies that have provided the vital eyes, ears, and voices for the American military. This article describes some of the defense technologies being converted for law enforcement uses by this regional center.
PARALLEL OPERATIONAL STRATEGIES
Law enforcement and defense missions share similar concerns and strategies. A key concept in the defense community is command, control, communications, and intelligence, known collectively as C3I. C3I includes a broad range of techniques and technologies that increase the effectiveness of a deployed force. It enables troops to perform operations more rapidly and safely and allows actions to be contained within a desired area or to a specific group of combatants.
Command and control, the first two components of C3I, address resource allocation and general mission planning - aspects shared by most law enforcement operations. As forces execute the plan, commanders monitor progress and issue corrective orders to deal with the changing scenario.
The intelligence aspect of C3I refers to covertly acquiring, cataloging, and using relevant information about the enemy or its environment. In a military scenario, intelligence could include maps, pictures, or the results of interviews. For law enforcement, it also could encompass street maps, train station locations, pictures of known suspects, fingerprint files, or any other information that might provide a clue or help to determine an optimum course of action.
Closely related to intelligence is surveillance, which the military most often uses to identify both hostile and friendly forces. A radar or multispectral device used to detect an airborne threat would be one type of surveillance sensor. Law enforcement applications could include video cameras for street surveillance and multifrequency sensors for contraband detection.
The final element of C3I is communications, the infrastructure that ties everything together. Anything related to the exchange of information falls into this category, such as computer links, printed text, voice transmissions, photographs, and other imagery, to name a few.
The parallels between the military C3I concept and a similar law enforcement C3I concept easily can be recognized. Law enforcement applications include, for example, riot control, mission planning, timely decisionmaking, covert surveillance, and illegal drug interdiction. As more and more law enforcement agencies with adjacent or overlapping jurisdictions join forces to combat crime, C3I technologies will become particularly useful for coordinating activities and making the most effective use of resources.
COMMAND AND CONTROL
A good plan can make all the difference in whether an operation succeeds or fails. Similarly, having the pertinent facts about a situation and its participants affects the decisionmaking process. Law enforcement commanders can take advantage of this to ensure that they have access to the information they need to control their operations effectively.
Planning Complex Operations
Many law enforcement operations, such as installing listening devices pursuant to a court order or responding to a widespread civil disturbance, require coordination among commanders at multiple locations or even in other governmental agencies. A distributed collaborative planning (DCP) process can make strategic deployment and crisis management tasks easier.
In the DCP process, "distributed" means that it links commanders at multiple locations and enables them to share data, software decision models, and other information on a real-time basis. "Collaborative" indicates that planners communicate with each other via digital video teleconferences and shared computer "desktops" and databases, passing textual, verbal, and pictorial information to one another instantly.
Having a DCP capability allows police commanders to coordinate activities and responsibilities among agencies and response teams and to distribute imagery, including surveillance and suspect photographs, and other information as the situation unfolds. For example, headquarters personnel, en route response cars, helicopters, and other field units responding to a civil disturbance could share up-to-date, as well as archived, information drawn from diverse locations, both prior to and during operations. Each unit in the operation could provide real-time situation reports and work through problems as they developed.
Sharing Information About Offenders
The inability to access critical information about offenders quickly and accurately represents a significant hindrance to law enforcement today. Traditionally, law enforcement agencies have developed information systems peculiar to their unique needs, making multimedia information sharing among agencies nearly impossible. Joint automated booking stations (JABS), originally a DEA-Rome Laboratory pilot project in the Miami area, help overcome this obstacle by enabling the five Federal law enforcement agencies in the region(3) to share information more effectively.
JABS combines multimedia information systems, image- and text-oriented databases, image exploitation (enhancing images for identification, detection, and dissemination), and multisource fusion (combining information from many sources). Using computer workstations installed in each agency, agents can share unified text, photograph, and fingerprint information through a centralized database. Each workstation consists of an IBM-compatible computer, a digital video camera, a live-scan fingerprint system, and both black-and-white and color printers. A system administrator manages the centralized database and provides round-the-clock, on-call assistance to the agencies should any problem arise.
The shared data encompass prisoner case information, biographical statistics, voice prints, and images, such as facial photographs, fingerprints, and pictures of evidence. Eventually, advanced signal and image exploitation capabilities will enhance the system's ability to identify subjects using speaker identification, facial recognition, and fingerprint matching. This electronic booking process will replace the former paper method of booking arrests, although a printout of arrest information can be made. System designers project that JABS will reduce the time it takes to process prisoners by 75 percent, significantly cut the number of fingerprint cards rejected by the FBI, improve the quality of prisoner photographs, and make it easier to access information.
Intelligence on suspects, victims, and crime trends constitutes a critical law enforcement resource. A number of technological capabilities can make it easier to obtain and analyze intelligence information.
Many aspects of intelligence gathering revolve around monitoring conversations or coordinating complex operations using voice links among operatives. The needs for high sound quality and the capability to identify and understand speakers have led to the development of several speech-related capabilities.
Enhancing Voice Transmissions
Noise and other types of interference often make it difficult to understand what people say during phone, radio, or other voice transmissions. Speech enhancement technology, currently used in military operations to clean up noisy radio communications, reduces noise and interference and enables users to recover conversations that would otherwise be unintelligible. In airborne operations, the equipment is on a 6" x 9" printed circuit card; it also comes in a 19" rack-mountable box or as a software package for operation on a personal computer with a co-processor.
Speech enhancement technology offers several benefits. It works in real time with only a 200 millisecond processing delay. It reduces interference caused by a variety of equipment, atmospheric conditions, and other sources, including receivers, wire and radio links, tape recorders, automobile ignitions, and power-line hums. It has been used to recover conversations lost due to low-level recordings, malfunctioning equipment, environmental noise, and ground loop connections. Voice transmissions can be recovered using this enhancement process regardless of the language or dialect being spoken or the person talking.
Automatic speaker identification technology determines the identity of the speaker in a live or taped conversation. Speakers can be identified with as little as 4 seconds of their speech used to characterize the voice for comparison. Identification does not depend on the language or dialect the person uses or which words are spoken. Identification decisions can be made using as little as one word (approximately one-third of a second).
Currently only available in a laboratory setting, the military uses this technology to identify speakers on a military communication network where the communications have been recorded. A field version will be available in 1996. Law enforcement agencies could use automated speaker identification technology in a number of ways, such as tracking individuals using wire or cellular phones, recording and identifying suspects in wire-tapping and other monitoring operations, and using voiceprints for police sorting and booking operations.
Translating Spoken Conversations
Machine voice translation equipment takes in spoken voice in one language and translates it to another language. It provides the results in printed text or in audible spoken language form. As with automated speaker identification, the system does not depend on the speaker.
Three components operate the translation system - a commercial word recognizer, a personal computer that acts as a translator and system manager, and a voice synthesizer. Currently, it translates only Spanish and English in limited applications, but researchers are developing several other language translations.
Police departments in localities that have a large Spanish-speaking population would have an immediate interest in this technology. It can help officers collect information at crime scenes, reduce the time needed to acquire critical time-sensitive information, and make interrogating suspects less difficult and costly.
Another key aspect of the intelligence component of C3I is knowing where to find the enemy. A variety of sensors can locate and track suspects. Some sensors also can locate concealed weapons, peer through walls, and see at night.
Passive sensor technology will provide law enforcement with a covert means for identifying and tracking suspects by air and by sea. It especially applies to the drug interdiction arena where covert detection and tracking of suspected drug running aircraft is essential. It merges two complementary technologies - electronic support measurement (ESM) and bistatic radar.
ESM enables operators to receive and analyze any signal transmitted in the radio frequency (RF) spectrum, such as communications or radar signals. Analyzing such signals reveals the angle of arrival, frequency, pulse width, and any characteristics unique to a transmitter. This information provides a profile of the targeted emitter, which can be used later to re-identify the target.
Bistatic radar uses existing sources of illumination to detect and track targets passively, instead of the more conventional monostatic radar systems that actively send a signal and wait to receive a return echo. Bistatic radars track targets using signals from television stations or from Federal Aviation Administration (FAA) en route surveillance radars to provide the ambient illumination. The FAA radar energy, for example, reflects off the target in many directions, including the direction of the bistatic radar receiver. The receiver intercepts this energy and determines the location and characteristics of the target. This passive radar system cannot be detected by the target because it does not send signals, it only receives them.
Conventional radar systems work on the principle of line-of-sight detection and surveillance, which limits their range of effectiveness. Over-the-Horizon (OTH) radar, however, exploits the refractive properties of the ionosphere at low level frequencies (between 3 and 30 megahertz) to provide coverage far beyond line-of-sight distances.
The lower frequencies associated with OTH radar can bounce off the ionosphere, whereas the very high frequencies (above one-half gigahertz) at which conventional radar operates cannot. Similar to the way pool players bank billiard balls against a railing to get around their opponent's balls, the lower frequencies associated with OTH bounce off the ionosphere to reach around line-of-sight obstructions. In addition, the lower frequencies result in significantly wider beam widths, which in turn allow a much wider area to be monitored.
OTH radar performs three functions better than conventional radar-detection of targets at the source, continuous tracking of targets from take-off to landing, and routine surveillance of airfields suspected of being used by drug traffickers. At present, OTH systems are used in California for Mexican border surveillance.
Concealed Weapons Detection
RF sensors also can provide law enforcement agencies with two significant capabilities - concealed weapon detection and wall-penetrating surveillance. Law enforcement officers could use RF sensors to detect hidden weapons in crowded areas, such as airports or street parties, and to conduct surveillance of a building's interior and surrounding environment during hostage or cornered-fugitive situations.
These sensor technologies can be divided into two categories - passive and active. Passive sensors do not illuminate the targets; instead, they detect thermal energy generated from within the target and therefore can be used to find concealed weapons. For example, the human body emits electromagnetic energy, which can penetrate most types of clothing. Weapons concealed from view by clothing become visible to the sensor because they block some of the energy coming from the body.
Active sensors, on the other hand, illuminate the target with radio frequency energy for through-the-wall surveillance. Operators select a frequency that can penetrate the wall. The RF energy reflects off the people and objects in the observed area. The radar receiver then interprets the reflections to depict what is hidden from view.
Infrared Night Vision
Existing techniques, such as night-vision goggles and low light-level television, depend on some form of light source, such as the moon, stars, or distant city lights, and are subject to saturation and "blooming," which can make them ineffective. The new infrared sensors, however, are completely passive and inherently antiblooming because they do not rely on a light source. Instead, they sense the heat radiated by the subject and produce its image on a standard television monitor. They can reveal clandestine operations without alerting subjects that they are being observed.
The military uses such sensors on a number of aircraft and weapons navigation systems. Law enforcement agencies could use infrared sensing for passive border surveillance and drug interdiction on land or water.(4) Marine vessels can use infrared sensors to find survivors in water during both day and night searches. Limited viewing of concealed articles, including weapons, also might be possible.
Investigators must closely examine the data collected during an investigation. In complex, on-going cases involving multiple suspects and broad geographical regions, a picture of the collected information can be worth the proverbial thousand words. Sometimes, however, a single piece of evidence can provide the link that solves the case. The following computerized techniques can help.
Displaying Data Visually
Rummaging through piles of reports, interview transcripts, interrogation results, and surveillance information can make it difficult to see patterns and cause-and-effect relationships in cases. Using the Timeline Analysis System (TAS), investigators can add a visual dimension to the collected information that can help bring those patterns into focus.
The system consists of a set of software tools originally developed to help intelligence analysts understand a foreign country's military and political behavior and to project possible intentions. TAS software runs on a personal computer and represents each observed event as a meaningful icon on timelines and maps. For example, in a drug-running case, investigators could record the origins, destinations, and frequency of known drug flights, movements of suspects, and other information. The system would then graphically display each event on a timeline and/or a map, showing patterns of behavior.
Scientists designed the Timeline Analysis System to be flexible, so it can be tailored easily to support various types of cases. Useful during investigations, it also can serve as an effective tool for presenting cases to prosecutors and jurors.
To assist the FBI with forensic identification of firearms, researchers designed a system to enhance the existing Drugfire(5) system by automating the matching process. Currently, firearms experts must manually compare the characteristics of spent shell casings - a task that grows more and more daunting as the size of the database increases. The FBI's local database in Washington, DC, for example, contains more than 2,000 shell casings.
Computerized Automatic Target Recognition (ATR) speeds up the process by narrowing the number of potential matches for experts to examine. ATR uses a parallel, neural network-based system, which learns to recognize patterns rather than requiring operators to program the patterns. By eliminating obvious mismatches, the system can reduce by as much as 98 percent the number of images that must be examined manually. When tested on the LAPD database of more than 6,000 spent shell casings, the Automatic Target Recognition system linked five homicides.
The sensitive nature of law enforcement operations often requires agencies to restrict access to certain areas. Two promising technologies currently under development include an infrared facial recognition system and an optical correlator with a phase-only filter. Both can be used to control access to secure areas. One major application of the optical correlator is in detecting counterfeited valuables.
All people have unique facial signatures determined by their underlying vascular structure. State-of-the-art infrared cameras, in conjunction with computerized image processing software, can be used to recognize facial signatures. Researchers envision using this infrared facial recognition system to establish automated control of access to secure areas.
In a police department, for example, cameras would record the patterns of heat radiated from the facial area of employees authorized to enter the evidence room. The patterns, or thermograms, then would be stored in a computer database connected to the evidence room's locked doorway. As someone approached the doorway, an infrared camera would capture the person's facial image. The image processing software then would compare it to the database of previously stored thermograms, and in just a few seconds, determine whether it matched the thermogram of an employee authorized to enter the area.
This technology distinguishes itself from other biometrics approaches because it is passive, not intrusive, light-independent, and invulnerable to disguises. When completed, it could be employed for any military, law enforcement, or civilian use where personnel need to be identified.
To deal with the growing problems of counterfeited currency and other valuable items, researchers have developed a new pattern recognition device known as an optical correlator. It uses a laser to compare a stored reference image to an unknown image to determine their similarities.
Image processing usually involves transforming an image into a frequency spectrum representation. The components of this representation can be thought of as a set of ripples on the surface of a pond. The ripples have a magnitude (or height) and a phase (or relative time delay) associated with them. An optical correlator has been developed based on a phase-only filter, which disregards the magnitude and only uses the phase information. This filtering technique is more effective than other image processing techniques and requires far less information storage.
Developed originally for military use, the optical correlator based on the phase-only filter has been used by the Army Missile Command to track targets. For law enforcement, Rome Laboratory has built a prototype that performs real-time analysis of fingerprints. This could be used to control access to a secure area or a computer file.
As crime spreads beyond traditional boundaries, criminal justice agencies across jurisdictions must join forces to enforce the law. In large part, this requires enhanced communications capabilities, in terms of both speed and of compatibility. The technologies described below will help law enforcement stay one step ahead of crime.
The advent of the National Information Infrastructure, a seamless web of broadband communications networks, computers, and databases, will provide law enforcement agencies with vast amounts of multimedia information. These networks will allow federal, state, and local agencies to share text, voice, image, and video data in a timely manner through one network.
In a military environment, these networks allow for rapid exchange of critical information, such as intelligence resources and weapon quantities, drawn from sources in diverse locations. In the law enforcement arena, high-speed networks will enable agencies to access FBI databases to perform rapid fingerprint identification, conduct live teleconferences with other agencies in situations that require shared planning and coordination, and provide immediate, widespread dissemination of pictures of wanted or missing persons.
Compatible Communications Systems
Components of the C3I system use both wide- and narrow-band services, which flow across landlines, satellites, fiber optic cables, and terrestrial radio links. The divergent characteristics of each of these media have required users to obtain many types of often incompatible equipment. Rome Laboratory researchers are working to mitigate this problem with both short- and long-term solutions.
A quick fix for the problem of communications systems that cannot interact is a rapidly deployable set of radio switching and computer equipment that can serve as an interface among systems. Under computer control, this central communications center enables the exchange of information among virtually all forms of transmission media, including facsimile, data, and voice. It can be configured quickly for a variety of communication capabilities. The U.S. Coast Guard currently uses a central communications capability in its drug interdiction and other law enforcement operations, as well as during responses to natural disasters.
A longer range approach involves development of a new concept for radio systems. Sponsored by the Advanced Research Projects Agency, the "Speakeasy" program seeks to standardize radio equipment to establish common and flexible systems for radio communications among the various military services.
Speakeasy is a modular system in which many of the modules have multiple uses and can serve a variety of radio types. It takes advantage of the newest microcircuit technology in several ways. First, it employs an open systems architecture, meaning that interface specifications at all layers and connection points are published in open salutations and U.S. standards documents. With widely published specifications, more than one vendor can design and improve components that will be mutually compatible.
Second, the Speakeasy system will be programmable, enabling the same radio to be configured for different operations. It also will be multiband, that is, capable of operating in a variety of frequencies, and will operate simultaneously in more than one mode.
For law enforcement, this system will provide the capability for radios designed for different purposes to operate together. It also will provide the foundation for a greatly improved, easily upgradable radio system for all types of law enforcement applications.
Over the past 40 years, researchers at Rome Laboratory have developed a vast array of technological tools for the military to employ in our national defense. Within the shared framework of command, control, communications, and intelligence, many of those technologies apply to the domestic law enforcement mission as well. As one of NLECTC's regional law enforcement technology centers, Rome Laboratory will continue to make substantial contributions to the war on crime by developing technologies that meet the needs of law enforcement.
1 Federal Technology Transfer Act of 1986, P.L. 99-502.
2 For more information on the regional technology centers, write to NLECTC, Box 1160, Rockville, MD 20849, or call 1-800-248-2742.
3 The five agencies currently participating in the project are the Bureau of Prisons, the Drug Enforcement Administration, the Federal Bureau of Investigation, the Immigration and Naturalization Service, and the U.S. Marshals Service.
4 Because of its current high cost, this important technology has not been applied widely to law enforcement missions. However, Rome Laboratory has developed a new, more affordable infrared sensor technology. The equipment consists of a palm-sized video camera that uses standard rechargeable batteries.
5 Drugfire is a system that matches spent rounds to the weapons that fired them in order to identify firearms used in crimes.
Members of the Rome Laboratory Law Enforcement Technology Team - John Ritz, Donald Spector, Joe Camera, Fred Demma, and Warren Debany - collaborated on this article, with the assistance of Rome Laboratory researchers Wayne Bonser, Hunter Chilton, Ed Cupples, Dave Ferris, Paul Gilgallen, Joseph Horner, Robert Kaminski, John Mucks, Paul Pellegrini, Antonette Pettinato, Fred Rahrig, Lee Uvanni, Bill Wolf, and Frank Zawislan.
|Printer friendly Cite/link Email Feedback|
|Publication:||The FBI Law Enforcement Bulletin|
|Date:||Apr 1, 1996|
|Previous Article:||Above and beyond the call of duty: preventing off-duty officer deaths.|
|Next Article:||Prosecuting cases without victim cooperation.|