This paper describes the Advanced Packet Vault, a cryptographically secured archiver of network packet data that reliably captures all packets on a 100 Mbps Ethernet network, encrypts them, and writes them to long-term magnetic tape storage for later analysis and evidentiary purposes. Based on a previous prototype, the APV provides an enhanced cryptographic organization that allows site-specific selection of the encryption format and that permits selected traffic to be made available without compromising the security of other traffic. The APV operates reliably under a continuous 100 Mbps load. We conclude with a discussion of future work necessary to scale the APV beyond 100 Mbps.

Covert communications through digital networks are a small subset of covert channel communications through all other networks such as human contacts, postal, telephone, ham radio and other networks. The repertoire of options available to anyone for covert communications through digital networks is limited only by imagination; covert communications are therefore inherently uncontrollable.

Information dominance, the superior ability to perceive, correctly interpret, and rapidly respond, is a central strategic goal for the United States at the dawn of the information age. To achieve information dominance, it is necessary to control the global information lattice through comprehensive strategic mechanisms supported by National Policy. To maintain control requires technologies to measure, analyze, prevent, predict and respond to information warfare in a manner that is commensurate with the evolving scale of threat. Unfortunately, little of the research conducted over the last 25 years has focused on response. As nations around the world grow increasingly sophisticated technologically, and become interconnected in the global information lattice, there is now a need to consider a methodical scientific exploration of response technologies. This exploration is intended to exercise and improve defensive capabilities, expand and improve offensive capabilities, and examine the relationship between defense and active response.

Digital evidence is information of probative value that is either stored or transmitted in a digital form. Digital evidence can exist as words (text), sound (audio), or images (video or still pictures). Law enforcement and forensic scientists are faced with collecting and analyzing these new forms of evidence that previously existed on paper or on magnetic tapes. They must apply the law and science to the processes they use. Extrapolating the old processes into the new formats has been proceeding since the 1980's. Regardless of the output format, all digital evidence has a certain commonality. One would assume that the rules of evidence and the scientific approach would also have some common characteristics. Obviously, there is also a divergence due to the differences in outputs. It is time to approach the issues regarding digital evidence in a more deliberate, organized, and scientific manner. The program outlined by the NCFS would explore these various formats, their features common to traditional types of forensic evidence, and their divergent features and explore the scientific basis for handling of digital evidence. Our web site, www.ncfs.org, describes our programs.

Early warning of active worm propagation over the Internet is of vital importance to first responders. Knowing an active worms characteristics very early in its propagation can significantly reduce the damage it may cause. In this paper we propose an early warning system that uses ICMP Destination Unreachable (ICMP-T3) messages to identify the random scanning behavior of worms. Participating routers across the Internet send Blind Carbon Copies of all their locally generated ICMP-T3 messages to a central collection point. There all the incoming messages are compared for similarities. Incoming messages are abstracted and patterns identified. Using the methods discussed in this paper we identify 'blooms' of activity that are a clear signature of worm propagation. Preliminary test results have shown that actively spreading worms can be identified in the first few minutes after they are launched. By using the characteristics gathered in those early stages, action can be taken and widespread damage might be avoided.

This paper discusses a framework for conduct of information security trend analyses. While several organizations are performing such analyses, there is wide disparity between the level of the analyses, the applicability of results, and the assumptions involved in properly interpreting the results. The framework offers a common ground in which these issues may be resolved. An example analysis process is presented in the paper. The paper includes a discussion of cautionary factors in the application of this framework.

Sexually explicit Internet chat rooms are increasingly used by pedophiles to reach potential victims. Logging and linking suspects to chat room conversations and e-mails exchanged with undercover detectives are crucial to prosecuting travelers, i.e., pedophiles who travel across state lines to engage in sexual acts with minors. This paper describes two tools, a chat room monitor and a remote fingerprinter, for acquiring and preserving evidence. The chat room monitor logs online communications as well as screen images and keystrokes of the undercover detective. stored to allow the chronological reconstruction and replay of the investigation. The remote fingerprinter uses sophisticated scanning techniques to capture and preserve a unique fingerprint of the suspect's computer over the Internet. Once the suspect's computer is seized, it is scanned again; matching this new fingerprint with the remotely acquired fingerprint establishes that the suspect's computer was used to communicate with the detective.

The importance of glass as evidence of association between a crime event and a suspect has been recognized for some time. Glass is a fragile material that is often found at the scenes of crimes such as burglaries, hit-and-run accidents and violent crime offenses. The physical and chemical properties of glass can be used to differentiate between possible sources and as evidence of association between two fragments of glass thought to originate from the same source. Refractive index (RI) comparisons have been used for this purpose but due to the improved control over glass manufacturing processes, RI values often cannot differentiate glasses, even if the glass originates from different sources. Elemental analysis methods such as NAA, XRF, ICP-AES, and Inductively Coupled Plasma Mass Spectrometry (ICP-MS) have also been used for the comparison of trace elemental compositions and these techniques have been shown to provide an improvement in the discrimination of glass fragments over RI comparisons alone. The multi-element capability and the sensitivity of ICP-MS combined with the simplified sample introduction of laser ablation prior to ion detection provides for an excellent and relatively non-destructive technique for elemental analysis of glass fragments. The methodology for solution analysis (digestion procedure) and solid sample analysis (laser ablation) of glass is reported and the analytical results are compared. An isotope dilution method is also reported as a high precision technique for elemental analysis of glass fragments. The optimum sampling parameters for laser ablation, for semi-quantitative analysis and element ratio comparisons are also presented. Finally, the results of a case involving the breaking of 15 vehicle windows in an airport parking lot and the association of a suspect to the breakings by the glass fragments found on his person are also presented.

In law enforcement there is much textual data which needs to be searched in order to detect new threats. A new methodology which can be applied to this need is the automatic searching of the contents of documents from known sources to construct a lexicon of words used by that source. When analyzing future documents, the occurrence of words which have not been lexiconized are indicative of the introduction of a new topic into the source's lexicon which should be examined in its context by an analyst. A system analogous to this has been built and used to detect Fads and Categories on web sites. Fad refers to the first appearance of a word not in the lexicon; Category refers to the repeated appearance of a Fad word and the exceeding of some frequency or spatial occurrence metric indicating a permanence to the Category.

Bayesian Network Fragments (BNFrags) provide a practical, computational methodology to encode a distributed library of computer-usable knowledge patterns for automated reasoning about aspects of homeland defense against terrorism. Multi-Entity Bayesian Networks provide a means of encoding repeated patterns and relationships in the form of BNFrags having variables that range over entities of a given type. New evidence either is matched to existing entities or triggers new entities to be hypothesized. BNFrag instances are created by replacing the variables by the names of entities in the situation. These BNFrags are combined to form situation-specific Bayesian networks (SSNs). We propose the use of MEBNs as the inferential cornerstone of a cumulative national, distributed knowledge base (KB) for homeland defense. In this paper we illustrate the use of MEBNs for these purposes with an example concerning a multi-city coordinated biowarfare attack. We show how current trends in the use of on-line reporting by health care and related facilities has the potential to enable opportunistic detection of and response to low probability, high consequence events for which it would otherwise be a practical impossibility to set up specifically directed monitoring capabilities.

A software application has been developed to aid law enforcement and government intelligence gathering organizations in the translation and analysis of foreign language documents with potential intelligence content. The Automated Document Analysis System (ADAS) provides the capability to search (data or text mine) documents in English and the most commonly encountered foreign languages, including Arabic. Hardcopy documents are scanned by a high-speed scanner and are optical character recognized (OCR). Documents obtained in an electronic format bypass the OCR and are copied directly to a working directory. For translation and analysis, the script and the language of the documents are first determined. If the document is not in English, the document is machine translated to English. The documents are searched for keywords and key features in either the native language or translated English. The user can quickly review the document to determine if it has any intelligence content and whether detailed, verbatim human translation is required. The documents and document content are cataloged for potential future analysis. The system allows non-linguists to evaluate foreign language documents and allows for the quick analysis of a large quantity of documents. All document processing can be performed manually or automatically on a single document or a batch of documents.

Both concealed weapons detection and through the wall surveillance are significant problems for both law enforcement and military personnel. While on the surface it would appear that these two problems are unrelated technologically, they do, in fact, share some common ground. A concealed weapon acts as resonant object, exhibiting electromagnetic resonance peaks at frequencies characteristic of the weapon's major dimensions. For handguns the frequency range of interest lies approximately between 450 MHz and 2 GHz. As it turns out, this is also a region over which many common building materials are largely transparent. As part of grant 97-IJ-CX-K013 from the National Institute of Justice, AKELA, Inc. has developed a stepped-frequency, CW radar that covers this frequency range. The radar is digitally synthesized and controlled and has a range resolution of approximately 4'. Digital waveform control gives the radar the ability to avoid interference with other electronic devices, to tailor data collection for signal processing requirements, and to change its sweep time in response to operational requirements. AKELA has developed a brassboard concealed weapons detector that uses this radar. A through the wall imaging system that uses the radar is currently in development under AFRL Contract F30602-00-C-0205.

A prototype through-the-wall surveillance system is being developed for use by law enforcement personnel that utilizes high power ultrasonic transducers to detect and locate stationary or moving people inside either metallic or non-metallic walled enclosures. Details of the prototype design will be presented together with recent data and analyses. The sensor is being designed as a field-portable unit with a real-time display of the location, in range and azimuth, of each detected individual. Applications of this technology include detecting and locating unconscious, sleeping, hiding, tightly bound or otherwise stationary people, as well as moving people, inside an enclosed area. The sensor should also prove useful in border patrol applications for inspecting vehicles and shipping containers at points of entry.

A portable Through-the-Wall Surveillance System is being developed for law enforcement, counter-terrorism, and military use. The Motion and Ranging Sensor is a radar that operates in a frequency band that allows for surveillance penetration of most non-metallic walls. Changes in the sensed radar returns are analyzed to detect the human motion that would typically be present during a hostage or barricaded suspect scenario. The system consists of a Sensor Unit, a handheld Remote Display Unit, and an optional laptop computer Command Display Console. All units are battery powered and a wireless link provides command and data communication between units. The Sensor Unit is deployed close to the wall or door through which the surveillance is to occur. After deploying the sensor the operator may move freely as required by the scenario. Up to five Sensor Units may be deployed at a single location. A software upgrade to the Command Display Console is also being developed. This software upgrade will combine the motion detected by multiple Sensor Units and determine and track the location of detected motion in two dimensions.

A handheld, battery-operated prototype of a concealed weapon detector (CWD) has been built and tested. Designed to detect both metallic and non-metallic weapons, the sensor utilizes focused ultrasound (40 kHz frequency) to remotely detect concealed objects from beyond arm's length out to a range of about 25 feet (8 meters). Applications include weapon detection in prison settings, by officers in the field for stand-off frisking of suspects, and as supplemental security at courthouse entrances and other monitored portals. The detector emits an adjustable, audible alarm (with provision for an earphone jack) as well as a visible light-bar indicator when an object has been detected. An aiming light, with momentary switch, allows the user to accurately determine the location of the concealed object. A presentation of the detector's capabilities and limitations will be presented along with probability of detection (PD) data obtained using the latest prototype version.

As a result of its relatively short wavelength coupled with relatively high penetration of many materials, millimeter- wave imaging provides a powerful tool for the detection of concealed articles. By using a passive approach such as that implemented here, it is possible to image (detect) concealed weapons and articles or look through certain types of walls, all without generating any form of radiation that might raise health concerns. In this paper we will show images from our current first generation unit and discuss the technology and performance of the second generation state- of-the-art unit currently being built.

Different types of shape parameters based on circularity, Fourier descriptors, and invariant moments are studied for the automatic detection of weapons in Millimeter-wave data. First, performance of the shape descriptors is evaluated on simulated objects. The best performing shape descriptors are then tested on the automatic recognition of weapons in real data.

A concealed weapons detection technology was developed through the support of the National Institute of Justice (NIJ) to provide a non intrusive means for rapid detection, location, and archiving of data (including visual) of potential suspects and weapon threats. This technology, developed by the Idaho National Engineering and Environmental Laboratory (INEEL), has been applied in a portal style weapons detection system using passive magnetic sensors as its basis. This paper will report on enhancements to the weapon detection system to enable weapon classification and to discriminate threats from non-threats. Advanced signal processing algorithms were used to analyze the magnetic spectrum generated when a person passes through a portal. These algorithms analyzed multiple variables including variance in the magnetic signature from random weapon placement and/or orientation. They perform pattern recognition and calculate the probability that the collected magnetic signature correlates to a known database of weapon versus non-weapon responses. Neural networks were used to further discriminate weapon type and identify controlled electronic items such as cell phones and pagers. False alarms were further reduced by analyzing the magnetic detector response by using a Joint Time Frequency Analysis digital signal processing technique. The frequency components and power spectrum for a given sensor response were derived. This unique fingerprint provided additional information to aid in signal analysis. This technology has the potential to produce major improvements in weapon detection and classification.

Successful demonstration of a video rate 94 GHz camera for concealed weapons detection has lead to interest in commercialization of this system. Besides the basic physics of object detection, there are many details a practical system needs to consider for the transition from experiment to product. We describe improvements in the RF modules, back end electronics, and user interface, along with additional video images captured with the improved system. RF module improvements include better low noise amplifiers and RF switches, along with a two-temperature calibration method resulting in a noticeable improvement in image quality. Analog electronics now have a more rapid offset correction, better stability, and better dynamic range. The user interface will now permit image fusion and other user- friendly features. Indoor experiments with active illumination are reported.

This paper discusses the TWS data collected with a state-of- the-art 100 GHz radar imager developed for law enforcement use by Millivision, PPC. The system collects a cube of data consisting of 16 azimuth elements by 16 elevation elements by 256 range elements. The cube represents 11 degrees by 11 degrees by 25 m of coverage. The relatively narrow field-of- view (fov) was extended by physically moving the antenna in 11 degree segments and collecting data which is stitched together into larger images, e.g. a 3X3 stitched image represents 33 degrees by 33 degrees by 26 m of coverage. Unfortunately, this stitching process required up to 5 minutes to collect a single (3X3) stitched image. Thus, motion had to be simulated. The paper will discuss the phenomenology of the MMW radar return from various objects including walls, wall-corners, desks and other furniture, and persons simulating walking. Successive frames from a simulated move of a man and woman walking will be presented, and the actual movie shown at the presentation. Comments will be offered as to the practicality of active MMW imaging for TWS application.

This paper describes new class of high-resolution electro-optical imaging search and surveillance system (visual and/or IR). This new system uniquely provides continuous real-time situation-awareness and response, with immediate detection of approaching/ emerging threats, and rapid assessment of a situation, permitting immediate response/ reaction to counter the threat. As such it is uniquely suited for a very wide variety of diverse surveillance applications, especially in the areas of security and defense. The system provides an uninterrupted real-time, effectively live, imaging-display coverage of the entire 360 degree(s) panorama, all evaluated in real time at the full resolution of the imaging sensor. The image resolution for the full 360 degree(s) panorama collected by the sensor, and displayed by the display module, is unprecedentedly high, typically at least 20 megapixels (e.g., 2,000 pixels vertically by 20,000 pixels in circumference). The associated scene-scan cycle time is typically a second or less. This level of performance assures earliest detection of distant targets and highest-quality evaluation of the targets. Two alternative means of surveillance evaluation are available: the Direct-View (Observer) Surveillance Evaluation, and the Semi-Automated Surveillance Evaluation, with visual verification by the observer personnel. The resulting surveillance capability is unprecedented, simultaneously providing a) assured target detection anywhere within the 360 degree(s) panorama, b) classification, c) localization, d) tracking, and e) target status and activity monitoring, all without the need for any external tip-off or cueing inputs to the system. The greatest value of this unique situation-awareness capability is primarily in situations requiring high-confidence protection of high-value vehicles or installations located inside a relatively open and unobstructed region wherein hostile threats could quickly appear at any time and from any direction. An important practical aspect of the system is that implementation is straightforward and quite cost-effective, fully leveraging current technology in a unique combination of commercially available components.

Airborne electro-optic surveillance from a moving platform currently requires regular interaction from a trained operator. Even simple tasks such as fixating on a static point on the ground can demand constant adjustment of the camera orientation to compensate for platform motion. In order to free up operator time for other tasks such as navigation and communication with ground assets, an automatic gaze control system is needed. This paper describes such a system, based purely on tracking points within the video image. A number of scene points are automatically selected and their inter-frame motion tracked. The scene motion is then estimated using a model of a planar projective transform. For reliable and accurate camera pointing, the modeling of the scene motion must be robust to common problems such as scene point obscuration, objects moving independently within the scene and image noise. This paper details a COTS based system for automatic camera fixation and describes ways of preventing objects moving in the scene or poor motion estimates from corrupting the scene motion model.

Low cost, low power uncooled thermal imaging sensors have completely changed the way the world views security and law enforcement. A technology has emerged that has furthered lowered the cost and power of thermal imaging sensors. These sensors utilize low cost, ultra low power, high volume, vacuum packaged amorphous silicon technologies. Manufacturing and packaging discoveries have allowed infrared sensitive silicon arrays to be produced with the same methods that have driven the rapidly advancing digital wireless telecommunications industries. It is currently available and in production for the Fire Fighting and Security markets. This technology will allow insertion into Law Enforcement and Security areas that have previously been unable to use IR technology due to cost barriers. These amorphous microbolometers contain no choppers or thermoelectric coolers, require no manual calibration and use readily available commercial off-the-shelf components.

During the early 1990's, uncooled microbolometer thermal imaging technology began a journey from Government and corporate laboratories to practical application in addressing military, Government, and commercial customer needs. Today, that transition could arguably be considered complete, punctuated by BAE SYSTEMS' delivery of the 10,000th microbolometer camera on 12 February 2002. While microbolometer developmental research continues to advance the state-of-the-art at an ever increasing pace, uncooled infrared cameras are widely deployed serving society in meaningful ways; from preventative maintenance and process inspection to law enforcement and rescue operations. Following last years terrorist attacks in New York and Virginia, President Bush appointed Governor Ridge to lead federal coordination efforts for defense of the homeland. While uncooled microbolometer sensors served in Homeland Security long before September 2001, it is certain that new applications will be identified for surveillance, security, law enforcement and protection needs. In this paper we will describe advances in military uncooled infrared sensor technology and how these sensors can serve in the role of Homeland Defense. Developments in uncooled sensors that will be described include the rugged performance validation of a thermal weapon sight and head-mounted imager. We will look at those areas of Homeland Defense that are most likely to benefit from the application of uncooled microbolometer thermal imaging sensor technology. These include: a) search & rescue camera systems, b) handheld surveillance systems and c) hands-free camera systems.

We present a plan for the integration of geometric constraints in the source, sensor and analysis levels of sensor networks. The goal of geometric analysis is to reduce the dimensionality and complexity of distributed sensor data analysis so as to achieve real-time recognition and response to significant events. Application scenarios include biometric tracking of individuals, counting and analysis of individuals in groups of humans and distributed sentient environments. We are particularly interested in using this approach to provide networks of low cost point detectors, such as infrared motion detectors, with complex imaging capabilities. By extending the capabilities of simple sensors, we expect to reduce the cost of perimeter and site security applications.

The BioCapture^TM is a handheld commercially available product for the collection and concentration of airborne particulates. Designed for use by first responders in an emergency, the BioCapture^TM allows for forensic analysis of any environment, providing a means for determining the characteristic and extent of human health threat. In the present study, the basic technologies that comprise the BioCapture^TM are discussed. Also included are presentations of data that compare the performance of the BioCapture^TM with traditional aerosol collectors.

MSAD is being developed as a low-cost point detection chemical and biological sensor system designed around an information fusion inference engine that also allows additional sensors to be included in the detection process. The MSAD concept is based on probable cause detection of hazardous chemical vapors and aerosols of either chemical or biological composition using a small portable unit containing an embedded computer system and several integrated sensors with complementary capabilities. The configuration currently envisioned includes a Surface-Enhanced Raman Spectroscopy (SERS) sensor of chemical vapors and a detector of respirable aerosols based on Fraunhofer diffraction. Additional sensors employing Ion Mobility Spectrometry (IMS), Surface Acoustic Wave (SAW) detection, Flame Photometric Detection (FPD), and other principles are candidates for integration into the device; also, available commercial detectors implementing IMS, SAW, and FPD will be made accessible to the unit through RS232 ports. Both feature and decision level information fusion is supported using a Continuous Inference Network (CINET) of fuzzy logic. Each class of agents has a unique CINET with information inputs from a number of available sensors. Missing or low confidence sensor information is gracefully blended out of the output confidence for the particular agent. This approach constitutes a plug and play arrangement between the sensors and the information pattern recognition algorithms. We are currently doing simulant testing and developing out CINETs for actual agent testing at Edgewood Chemical and Biological Center (ECBC) later this year.

Plants are ubiquitous in the environment and have the unique ability to respond to their environment physiologically and through altered gene expression profiles (they cannot walk away). In addition, plant genetic transformation techniques and genomic information in plants are becoming increasingly advanced. We have been performing research to express the jellyfish green fluorescent protein (GFP) in plants. GFP emits green light when excited by blue or UV light. In addition, my group and collaborators have developed methods to detect GFP in plants by contact instruments and at a standoff. There are several law enforcement applications for this technology. One involves using tagging and perhaps modifying drug plants genetically. In one instance, we could tag them for destruction. In another, we could adulterate them directly. Another application is one that falls into the chemical terrorism and bioterrorism countermeasures category. We are developing plants to sense toxins and whole organisms covertly. Plants are well adapted to monitor large geographic areas; biosurveillance. Some examples of research being performed focus on plants with plant pathogen inducible promoters fused to GFP for disease sensing, and algae biosensors for chemicals.

ENSCO has been at the forefront in the development of a novel technology for human presence detection in vehicles, including automobiles, vans, trucks and semi-tractor trailers. This technology, by its very nature, is quite practical and less costly than other methods because it is based on a fundamental physiological characteristic common to all humans. Based on the ballistocardiogram force it measures the vehicle vibration produced by the individual's heart. The force produced by the heartbeats of humans will move vehicles, even large semi-tractor trailers, enough to be easily detectable by small seismic sensors. MicroSearch finds stowaways in a matter of seconds. No need to unload the vehicle, no need for manual searches.

The objective of the Department of Defense Physical Security Equipment (DoD PSE) RDT&E program is to provide end users within the four Services with the most efficient and productive physical security (PS) at the most reasonable cost to ensure the effective protection of DoD resources. These resources include personnel, nuclear weapons, classified information, materiel, and readiness assets. As a result of the1996 Khobar Towers terrorist bombing incident, the DoD PSE program began to receive additional funding in 1997 for Force Protection Commercial-Off-The-Shelf (FP COTS) equipment evaluation and testing. The FP COTS testing applies to all available technologies, which are considered effective for DoD use. Successive terrorist incidents occurring since Khobar Towers have resulted in increasing focus on the demonstration and validation of equipment necessary to combat the ubiquitously asymmetric terrorist threat.

This paper discusses the USSOCOM Special Operations Technology Objectives with regard to Unattended Ground Sensors. It covers capabilities, objectives and challenges. Special operations are conducted by highly trained, organized and equipped military and paramilitary forces to achieve military, political, economic or informational objectives by unconventional military means in hostile, denied, or politically sensitive areas. Political-military considerations frequently shape special operations, requiring clandestine, covert, or low visibility techniques and oversight at the national level. Special Reconnaissance, a core task, for USSOCOM are those surveillance activities conducted by SOF to obtain or verify by visual or other collection methods, information concerning the capabilities, intentions, and activities of enemy forces or to secure data concerning the meteorological, hydrographic, or geographic characteristics of a particular area. It includes target acquisition, area assessment and post-strike reconnaissance. This is a general overview; it does not discuss specific mission requirements or scenarios.

Technological advances in a number of fields have allowed SenTech to develop a highly capable Unattended Ground Sensor (UGS) able to perform a number of critical missions such as ground and air vehicle surveillance, personnel detection and tracking and sniper localization. These sensors have also been combined with electro-optic sensors to provide target images and improved tracking accuracy. Processing is done in a highly integrated processing module developed under DARPA's IUGS program. Acoustic sensors have been engineered to achieve a three-pound unit with a 15 day field life and long range VHF communications. These sensors will be delivered in early 2002 for testing during field exercises. Extensive testing of the algorithms and software has been conducted over the last few years at a variety of government-sponsored tests and demonstrations. A Gateway unit has been developed which can manage the operation of an eight-sensor field and perform two-dimensional sensor fusion.

Recent events have increased interest in the use of sensors by law enforcement and homeland defense related organizations. Autonomous sensors such as those under development for the Unattended Ground Sensor (UGS) program are suitable for some of these applications. The operational lifetime of a UGS depends on the power consumption of the package and the space allocated for batteries. We survey and assess options for powering these devices ina long-term scenario. These alternatives are in various stages of development, and range from conventional batteries and solar cells that are ready for deployment and are now commercially available; to technologies developed for other applications (e.g., power for deep-space probes, man portable power for soldiers, or for sensors in oil drilling bore holes) that would need to be adapted to UGS's; to new and often speculative concepts that are in the laboratory or are still on the drawing board. Ideally, unattended ground sensors do not require servicing, re- energizing or refueling; and are capable of autonomous operation for weeks or even years. Further, UGS's may need to be used covertly, which restricts schemes that would provide a detectable signature. Reliability, ruggedness, cost, weight, size, camouflaging, use of toxic materials and other safety or disposal aspects, restrictions on their deployment (e.g., whether UGS's can be dropped form the air or whether they need to be uprighted or favorably oriented), storage and inventorying considerations, temperature ranges of operation, and complexity of associated electronics are also important issues. In this paper, we will limit the discussion to systems where operating power does not exceed 5 watts since larger systems are commercially available. Some subjectivity in comparisons is perhaps inevitable, but despite the disparate physics upon which these devices are based, a few common criteria can be invoked for discussing their suitability for energy storage and powering UGS's. Metrics can be developed to assess and compare options, but since most of the options are in very different stages of development, one is sometimes forced to use performance specifications that are predicted, rather than demonstrated. Thus, in some cases the comparisons are tentative or speculative.

Under National Ground Intelligence Center (NGIC) sponsorship, BBN Technologies (BBN) has developed a physical model and a software utility to estimate the range and orientation of a target from video images and known geometric information about the target. This capability can be used in conjunction with law enforcement sensor systems to enhance situation assessment and facilitate the associated decisions process. The target range and orientation estimation system (Video Range Finder) allows the user to associate identifiable features of the video image with physical components of the target through an interactive process. The three-dimensional polyhedron defined by the location of these components in the coordinate frame of the target, is projected along the viewing axis onto a two-dimensional polygonal image at the focal plane of the video camera. The azimuth, elevation, and viewing-axis rotation angles are obtained by solving numerically the associated non-linear projection equations. The calculated angles are subsequently used in the estimation of the target range. The paper discusses the solution method and demonstrates the application of the system using both simulated and real objects.

Accurate personnel and vehicle tracking has been achieved using networks of small, unobtrusive, low-cost wireless sensors. The wireless MSTAR sensors developed in this work are based on previous pioneering MEMS sensing and TinyOS communications software work completed at UC Berkeley. The works has been funded under the DARPA SensIT, SensorWebs, and on-going DARPA NEST programs. These MSTAR sensors deliver around the clock all-weather surveillance and perimeter protection for field environments, including buildings, camp and tent locations, streets, mountainous regions, and other geographies. These capabilities satisfy many on-going intelligence and warfighter safety requirements. The MSTAR sensors are quickly deployed by hand emplacement or air-drop from a UAV or other airborne platform. The combination of multimode sensing on each wireless MSTAR sensor and multiple MSTAR sensors in the environment yields low false detections within the network perimeter. A low-power spread spectrum wireless link is used for communication across the MSTAR sensor network. Satellite exfiltration of data provides real-time access to the data on a worldwide basis. Future work includes additional field trials and the incorporation of acoustic capture, video capture, and biosensors into the MSTAR wireless sensor platform.

Catastrophe-theory-based Autonomous Terrain-Feature UAV Relative (CATFUR) navigation is geolocation without the Global Positioning System (GPS). As fully autonomous navigation based only on recognition of terrain features, it can be integrated with GPS or other state-of-the-art navigation systems, or can be independent. CATFUR navigation is based on integration/comparison and sensor fusion of DEM (digital elevation map) 3-D data, processed by commercial off-the-shelf geographic information system (COTS GIS) environments into a vectorial graph. CATFUR obtains data from the vertical takeoff unmanned air vehicle (VTUAV) COTS inertial and visual sensors, and from components of an azimuth-elevation local positioning system (LPS). Real-time data processing could perform on highly parallel 2 in. x 3 in. Application specific hardware. Typical point or line catastrophic singularities on surfaces are edges, ridges, wrinkles, and surface cracks. Such singularities have a fixed location on the surface. In contrast, catastrophes have the unexpected property of not being fixed to a surface. Catastrophes can be the basis of GPS-independent relative navigation, based only on the existence of a folded terrain, even without landmarks. Since mountains do not move, we can use mountain guidance, much as star guidance has been used for centuries to navigate the oceans.

Today's society places an extremely high price on the value of human life and injury. Whenever possible, police and paramilitary actions are always directed towards saving as many lives as possible, whether it is the officer, perpetrator, or innocent civilians. Recently, the advent of robotic systems has enable law enforcement agencies to perform many of the most dangerous aspects of their jobs from relative safety. This is especially true to bomb disposal units but it is also gaining acceptance in other areas. An area where small, remotely operated machines may prove effective is in local aerial surveillance. Currently, the only aerial surveillance assets generally available to law enforcement agencies are costly helicopters. Unfortunately, most of the recently developed unmanned air vehicles (UAVs) are directed towards military applications and have limited civilian use. Systems Planning and Analysis, Inc. (SPA) has conceived and performed a preliminary analysis of a low-cost, inflatable, lighter- than-air surveillance system that may be used in a number of military and law enforcement surveillance situations. The preliminary analysis includes the concept definition, a detailed trade study to determine the optimal configuration of the surveillance system, high-pressure inflation tests, and a control analysis. This paper will provide the details in these areas of the design and provide an insight into the feasibility of such a system.

The Law Enforcement and Military technology development communities have a growing common interest in the technologies associated with gunshot detection and localization. These common interests include urban warfare, community-oriented policing and sniper location. Technologies of interest include those associated with muzzle blast and bullet shockwave detection and the inter-netting of these acoustic sensors with electro-optic sensors. To date, no one sensor technology has proven totally effective for a complete solution. PSI has a muzzle blast detection and localization product which is wireless, highly mobile and reconfigurable, with a user-friendly laptop processor and display unit, which is currently being demonstrated in two different implementations: 1) A one-year, and on-going urban gunshot detection system installed in Austin, Texas, that began July 2001; and 2) A counter sniper system demonstration conducted at both the Aberdeen Proving Grounds and at an Israeli Defense Force firing range in the second half of the year in 2001. The former topic is under the auspices of a National Institute of Justice Cooperative Agreement with PSI and the Austin Police Department, and the latter topic was managed by the Army Research Laboratory and co-funded by DARPA/ATO and PSI. This paper will discuss successful aspects of the demonstrations to date, operational conclusions, and the development directions indicated for the future.

The Enhanced Consequence Management, Planning and Support System (ENCOMPASS) was developed under DARPA's Advanced Consequence Management program to assist decision-makers operating in crisis situations such as terrorist attacks using conventional and unconventional weapons and natural disasters. ENCOMPASS provides the tools for first responders, incident commanders, and officials at all levels to share vital information and consequently, plan and execute a coordinated response to incidents of varying complexity and size. ENCOMPASS offers custom configuration of components with capabilities ranging from map-based situation assessment, situation-based response checklists, casualty tracking, and epidemiological surveillance. Developing and deploying such a comprehensive system posed significant challenges for DARPA program management, due to an inherently complex domain, a broad spectrum of customer sites and skill sets, an often inhospitable runtime environment, demanding development-to-deployment transition requirements, and a technically diverse and geographically distributed development team. This paper introduces ENCOMPASS and explores these challenges, followed by an outline of selected ENCOMPASS deployments, demonstrating how ENCOMPASS can enhance consequence management in a variety real world contexts.

The tremendous increase in satellite imagery resolution and accuracy has made it reasonable to exploit such imagery in GIS-based command and control, fusing data streams from UAV/UGV and satellite sensors for C4ISR systems. To fuse imagery from satellite and tactical sensors in real time, the authors apply probabilistic analysis, polyquadrically interpolating evidential reasoning about non-numeric tactical factors and superimposing them on Digital Terrain Models and satellite imagery. Doing this in real time for C4ISR requires highly parallel computation, implemented on Video/Imagery Super-Parallel and Supercomputer-Path (VISP) hardware. Another approach is integrated view analysis for purposes of command decision making missions, making use of GIS environments, Dynamic Time Multiplexed Holographic (DTMH) screens, and 3-D projection for precise visualization of accurate GIS-based imagery. This technology incorporates non-numeric real-time information into the command and control processes with simultaneous integrated 3-D views. Theory and experiment both indicate that the methods and algorithms described here are feasible.

Acoustic sensors have been used to monitor firefighter and soldier physiology to assess health and performance. The Army Research Laboratory has developed a unique body-contacting acoustic sensor that can monitor the health and performance of firefighters and soldiers while they are doing their mission. A gel-coupled sensor has acoustic impedance properties similar to the skin that facilitate the transmission of body sounds into the sensor pad, yet significantly repel ambient airborne noises due to an impedance mismatch. This technology can monitor heartbeats, breaths, blood pressure, motion, voice, and other indicators that can provide vital feedback to the medics and unit commanders. Diverse physiological parameters can be continuously monitored with acoustic sensors and transmitted for remote surveillance of personnel status. Body-worn acoustic sensors located at the neck, breathing mask, and wrist do an excellent job at detecting heartbeats and activity. However, they have difficulty extracting physiology during rigorous exercise or movements due to the motion artifacts sensed. Rigorous activity often indicates that the person is healthy by virtue of being active, and injury often causes the subject to become less active or incapacitated making the detection of physiology easier. One important measure of performance, heart rate variability, is the measure of beat-to-beat timing fluctuations derived from the interval between two adjacent beats. The Lomb periodogram is optimized for non-uniformly sampled data, and can be applied to non-stationary acoustic heart rate features (such as 1st and 2nd heart sounds) to derive heart rate variability and help eliminate errors created by motion artifacts. Simple peak-detection above or below a certain threshold or waveform derivative parameters can produce the timing and amplitude features necessary for the Lomb periodogram and cross-correlation techniques. High-amplitude motion artifacts may contribute to a different frequency or baseline noise due to the timing differences between the noise artifacts and heartbeat features. Data from a firefighter experiment is presented.

Optical tagging methods have the advantage that they cannot be detected by a suspicious criminal or terrorist using a radio frequency (RF) sensitive device to scan his vehicle for the presence of an RF emitting tag. We will describe two optical tagging methods in which the presence of the tagging marks can be visually discovered only by very close observation. On the other hand, the tags can be readily recognized by a surveillance team through the use of infrared imagers, either in the longwave infrared (LWIR) or in the near infrared (NIR). The first approach uses a clear coating that has a higher thermal emissivity than the glass window to which it is applied. This coating can be viewed with a thermal imager that operates in the LWIR, with the tags appearing as bright marks on a dark background. The second method uses an NIR laser illuminator and also quarter-wave thick layers applied to the license plate of a vehicle. When viewed with a polarization-sensitive imager that operates in the NIR, these quarter-wave tags appear as bright marks on a dark background. We will show sample images of both of these optical tags, as viewed in the LWIR and NIR regions, respectively.

In this paper, a new approach to Wireless Local Area Network (WLAN) encryption based on Local Positioning System (LPS) is proposed.

We propose the new secure key management scheme for SDH-based WDM AON. The problem that is to have no overhead data for security service on SDH format may be overcome easily by using the proposed key management scheme. The major advantages of the proposed architecture are that the end-to-end security service can be offered and the key management of a cryptosystem can be implemented easily. The proposed secure SDH-based WDM AON architecture have the property of tapping resistance : the key management channel can monitor any tapping attack using the cutoff wavelength of the fiber ; the encryption function of the optical cryptosystems can prevent eavesdroppers from understanding information by tapping. In addition, the proposed architecture can ensure that synchronization of key stream sequence between the optical cryptosystems is maintained perfectly and security management for APS can be performed easily. The proposed architecture can be applicable to the secure node system.

The DARPA funded sensors developed for the UGS program have provided a good starting point on the physical infrastructure protection capability. Now we need an ability to merge the sensor data from multiple sources into a knowledge base that allows the monitoring personnel to easily view the entire situational awareness picture. This will entail the development of a command and control suite for use in the other agency operations. Presently there is a developmental gap between the sensor raw data and the full up military system that feeds global command and control networks. This gap is the small software-based command and control system with identification algorithms that can be tailored to the small operational scenario such as embassy protection or border patrol monitoring and be capable of residing on a laptop computer. From a systems engineering viewpoint, this command and control suite could be the key ingredient in the transition of the sensor technologies into the other agency applications. The transition of the sensor technology to numerous other applications can be hastened by the quick development of a C2 suite for use in the present high terrorist threat environment.

The Defense Advanced Research Projects Agency (DARPA) envisioned a system to assist decision-makers during crisis situations ranging from terrorist attacks to large-scale natural disasters. This system would provide the tools for responders, incident commanders, and officials at all levels to share vital information during the planning and execution of a coordinated response. The system would offer custom configuration of components with capabilities including map-based situational awareness, situation-based response checklists, casualty tracking, and epidemiological surveillance. On-scene commanders would use this system to document the progress of a response, direct and coordinate responder activities, and manage the response as a whole. Off-scene responders (hospitals, command centers, and local, state and federal agencies) would have the ability to visually assess the state of assets and casualties to better anticipate the need for personnel and supplies. DARPA's Enhanced Consequence Management, Planning and Support System (ENCOMPASS), successfully demonstrated all of these capabilities. ENCOMPASS was successfully transitioned to a commercial program: the Lightweight Epidemiology Advanced Detection and Emergency Response System, otherwise known as LEADERS.