The United States Army has embarked on a revolutionary transformation path towards a future, general purpose Objective Force that is strategically responsive and dominant across the entire spectrum of military operations. Key to the success of the Objective Force will be the ability to generate unprecedented battlespace understanding and situational awareness. A layered surveillance approach, one that relies on a full range of sensor assets from all echelons, is required to paint a battlespace piocture of sufficient fidelity. The Army's science and technology program continues to research and develop promising unattended ground sensor technologies that support Army transformation objectives.

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.

This paper presents issues and algorithms for the problem of source tracking with a network of aeroacoustic sensors. We study fusion of data from sensors that are widely separated, and we give particular attention to the important issues of limited communication bandwidth between sensor nodes, effects of source motion, coherence loss between signals measured at different sensors, signal bandwidth, and noise. We compare the tracking performance of various schemes, including joint (coherent) processing of all sensor data, as well as data-reduction schemes that employ distributed computation and reduced communication bandwidth with a fusion center. The key result of our analysis is a quantification of the potential gain in source tracking accuracy that is achievable with greater communication bandwidth and joint processing of sensor data. We show that the potential gain in accuracy depends critically on the scenario, as determined by the source motion parameters, signal coherence between sensors, bandwidth of the source signals, and noise level. For scenarios that admit increased accuracy with joint processing, we present a bandwidth-efficient algorithm that involves beamforming at small-aperture sensor arrays combined with time-delay estimation between widely-spaced sensor arrays.

Measurements of different types of aircraft are performed and used to obtain information on target characteristics and develop an algorithm to perform classification between jet aircraft, propeller aircraft and helicopters. To obtain a larger detection range, reduce background noise and to reduce classification errors in a multi-target environment, a real time adaptive beamformer algorithm is developed for a three microphone array. The output of the beamformer is submitted to a tracking algorithm. Acoustic signals from identified tracks are submitted to the classification algorithms. The algorithm is tested on data recorded during various field trials. The objective of the research, which is part of a research program for the Dutch Army, is to detect the passage of an aircraft with one or more mechanical wave sensors, either acoustic or seismic. After detection of a target, classification of the type of aircraft is requested (for example: helicopter-jet-propeller-rpv). If possible type identification is also requested. Earlier work showed promising results for detection and classification of helicopter targets. The projects resulted in an algorithm that can detect and classify helicopters, but it was developed to reject other targets. The chosen approach is to combine new aircraft detection and beamforming algorithms with the existing algorithms.

Acoustic sensors are becoming more and more applicable as a military battlefield technology. Those sensors allow a detection and direciton estimation with low false alarm rate and high probability of detection. The recent technological progress related to these fields of reserach, together with an evolution of sophisticated algorithms, allow the successful integration of those sensoe in battlefield technologies. In this paper the performances of an acoustic sensor for a detection of avionic vessels is investigated and analyzed.

An interactive platform has been developed for simulating the detection and direction-finding performance of battlefield acoustic ground sensors. The simulations use the Acoustic Battlefield Aid (ABFA) as a computational engine to determine the signal propagation and resulting frequency-domain signal characteristics at the receiving sensor array. There are three components to the propagation predictions: the transmission loss (signal attenuation from target to sensor), signal saturation (degree of signal randomization), and signal coherence across the beamforming array. The transmission loss is predicted with a parabolic solution to the wave equation that accounts for sound refraction and ground interactions; signal saturation and coherence are predicted from the theory for line-of-sight wave propagation through turbulence. Based on these calculations, random frequency-domain signal samples are generated. The signal samples are then mixed with noise and fed to the selected detection or beamforming algorithm. After averaging over a number of trials, results are overlaid on a terrain map to show the sensor coverage. Currently available algorithms include the Neyman-Pearson criterion and Bayes risk minimization for detection, and the conventional, MVDR, and MUSIC beamformers. Users can readily add their own algorithms through a 'plug-in' interface. The interface requires only a text file listing the algorithm parameters and defaults, and a Matlab routine or Windows dynamic link library that implements the algorithm.

There are lots of uncertainties with ground vehicle classification when such a classification is based only on acoustic emissions. To handle such uncertainties, we apply type-2 fuzzy system theoreies to the designs of FL rule-based vehicle classifiers, using amplitudes of the 2nd through 12th harmonics as the features. Statistical analysis of these features for both tracked and wheeled vehicles demonstrates that their standard deviations vary as much as their means; hence, the membership functions for the antecedents of the type-2 FL rule-based classifier were chosen to be Gaussian primary memberships with uncertain means of standard deviations. We constructed three classifiers for tracked/wheeled vehicle classification, namely Bayesian, type-1 and type-2 FL rule-based, and used the leave-one-out scheme to evaluate these classifiers. Our experiments demonstrated that the average false alarm rates of the type-1 and type-2 FL rule-based classifiers are much smaller than that of the Bayesian classifier; and the average false alarm rate of teh type-2 FL rule-based classifier is smaller than that of the type-1 FL rule-based classifier.

This paper presents the results of an installation of four acoustic/seismic ground sensors built using COTS computers and networking gear and operating on a continuous basis at Yuma Proving Grounds, Arizona. A description of the design can be found as well, which is essentially a Windows 2000 PC with 24-bit data acquisition, GPS timing, and environmental sensors for wind and temperature. A 4-element square acoustic array 1.8m on a side can be used to detect the time and angle of arrival of the muzzle blast and the impact explosion. A 3-component geophone allows the seismic wave direction to be estimated. The 8th channel of the 24-bit data acquisition system has a 1-pulse-per-second time signal from the GPS. This allows acoustic/seismic 'snapshots' to be coherently related from multiple disconnected ground sensor nodes. COTS 2.4 GHz frequency hopping radios (802.11 standard) are used with either omni or yagi antennas depending on the location on the range. Localization of the artillery or impact can be done by using the time and angle of arrival of the waves at 2 or more ground sensor locations. However, this straightforward analysis can be significantly complicated by weather and wind noise and is also the subject of another research contract. This work will present a general description of the COTS ground sensor installation, show example data autonomously collected including agent-based atmospheric data, and share some of the lessons learned from operating a Windows 2000 based system continuously outdoors.

In the late 1990s, ARL:UT developed a real-time system capable of detecting and classifying a variety of vehicles using a single 3-axis seismic sensor. At that time, the detection and classification software was designed to classify vehicles for a variety of features, including the number of cylinders in the engine, number of axles, engine type, traction mechanism, and relative weight. Using a single 3-axis seismic sensor, buried approximately 3 feet deep, the demonstration system showed that it was possible to detect and classify a single time-critical target, moving with constant bearing and speed. Since then, modifications have been made to the detection and classification software in support of a transition of the algorithms into unattended MASINT sensors (UMS). This paper describes the next evolution of the detection and classification system, which focuses on generating highly accurate traction mechanism and relative weight classifications. The current build of the software has proven capable of detecting vehicles and generating high confidence classifications for a wide range of vehicles and test scenarios, and performs well for both buried and hand-emplaced sensors. Included in the paper is a description of the upgrades made to the classification algorithms, event switching, and graphical user interface (GUI), an outline of the processing and analysis of seismic data, and a summary of the software performance for a wider range of vehicles, test scenarios, and sensor configurations.

The development of a novel, biologically inspired acoustic sensor is presented. The primary goal of this effort is to construct a miniature device that is capable of detecting the orientation of an incident sound source with an accuracy of 2°. The design approach follows from our investigation of the mechanics of directional hearing in the parasitoid fly, Ormia ochracea. This animal has been shown to be able to detect changes in the line of bearing of an incident sound that are as small as 2°. The tympanal structures of the ears of this animal suggest a novel approach to designing very small directionally sensitive microphones. A microphone diaphragm design is presented that has been fabricated using silicon microfabrication technology. Measurements of the static deflection due to intrinsic stress and of the response to sound are shown to be in excellent agreement with predictions. Predicted results indicate that this microphone concept could lead to a practical differential microphone with self-noise as low as 20 dBA.

The U.S. Army is transforming from a massive, heavy force that fought by finding the enemy and defeating them with overwhelming force to a strategically responsive force that wins battles through superior maneuverability and, more importantly, through superior situational awareness. This transformation is driving requirements for improved sensor capabilities across the spectrum of intelligence capabilities. REMBASS-II, as the type-standard unattended ground sensor system of the U.S., Army has evolved to provide a baseline capability that satisfies the immediate needs of the warfighter and provides a flexible platform to incorporate new unattended sensor technologies as they evolve. This paper reviews the development, capabilities, and flexibility of the U.S. Army's REMBASS-II system.

We consider the problem of locating and orienting a network of unattended sensors that has been deployed in a scene with unknown sensor locations and orientation angles, when no 'anchor' nodes are present. Many localization problems assume that some nodes have known locations and propagate location information about other nodes using triangulation procedures. In our formulation, we do not require such anchor nodes, but instead assume prior probability density function for the nominal locations of a subset of the nodes. These nominal locations typically have high uncertainty, on the order of tens of meters. The self-calibration solution is obtained in two steps. Relative sensor locations are estimated using noisy time-of-arrival and direction-of-arrival measurements of calibration source signals in the scene, and absolute location calibration is obtained by incorporating prior nominal location knowledge. We consider a Bayes approach to the calibration problem and compute accuracy bounds on the calibration procedure. A maximum a posteriori estimation algorithm is shown to achieve the accuracy bound. Experiments using both synthetic data and field measurements are presented.

This paper reports progress in the development of a miniature, monolithic AlGaAs/GaAs photovoltaic solar cell array, used in combination with scintillating fibers, and its application to recharging batteries of unattended ground sensors. Improvements in the design and processing of these photovoltaic arrays resulted in significant increases in both voltage and current. Moreover, new polycarbonate scintillating fibers that are more durable and flexible than polystyrene ones were tested. Output electric power density as high as 73 mW/cm² for total area and 105 mW/cm² for active area was measured outdoors for a 6-cell array integrated with a 1 foot long fiber bundle.

Remote, automated, day/night detection, classification, and tracking of personnel and vehicles is vital to a wide variety of security, law enforcement, and military applications. A practical system must consider cost, power, size, and operational limitations. Video tracking offers the ultimate in definitive identification and tracking capabilities, but also maximizes cost and power consumption. 360 degree views from a single infrared camera maximize the ability to monitor and track targets in a wide area. Other technologies using acoustic, seismic, and magnetic sensing can be used to extend the range or resolution of the system as well as preserving power by performing initial detection of targets to trigger video sensors. A variety of detection, fusion, and tracking algorithms can be applied to monitor targets around these sensors and across a wide area using multiple sensors.

The proliferation of small infrared cameras in high-volume commercial applications (e.g. firefighting, law-enforcement, and automotive) presents a tremendous opportunity for truly low-cost military micro-sensors. Indigo Systems Corporation's UL3 OmegaTM camera is a commercial off-the-shelf (COTS) thermal imager that offers ultra-small size, light weight, and low power. It employs a 164×120 microbolometer focal plane array (FPA) and is currently entering full-scale production. Furthermore, a 324×240 upgrade is in development. While aimed primarily at the commercial market, small size and low-power consumption make UL3 well-suited for other applications, including miniature unmanned aerial vehicles (UAVs) weapon-sights, and unattended ground sensors (UGS). This paper focuses on the key features of the UL3 family of miniature IR cameras and their utility in soldier systems.

Use of infrared cameras for unattended ground sensors and related applications provide challenges not normally required in hand held systems. This paper discusses the current state of amorphous silicon (a-Si) microbolometer technology as it relates to unattended ground sensor applications. Raytheon has used a-Si detector technology to develop a 160×120 a-Si based infrared camera. This camera, which was originally targeted towards unattended applications, is now in production at Raytheon for hand-held commercial applications. This paper discusses the typical requirements that face unattended ground sensors and how Raytheon's current 160×120 a-Si camera is positioned to meet these requirements.

320×240 and 640×480 small pixel uncooled microbolometer focal plane arrays have been developed that reduce overall sensor size, weight, power consumption, and cost. At the same time, these sensors still provide the high quality image resolution needed for target recognition and identification. These newly developed small uncooled thermal imaging sensors are being demonstrated in several attended and unattended sensor applications that include Unattended Ground Sensors, Micro Air Vehicles, and Infrared Helmet Sights. This paper describes recent developments at BAE SYSTEMS in uncooled microbolometer sensor technology for unattended sensor applications and presents the latest performance and image data for our 2nd generation systems.

We recently implemented a heterogeneous network of infrared motion detectors and an infrared camera for the detection, localization, tracking, and identification of human targets. The network integrates dense deployments of low cost motion sensors for target tracking with sparse deployments of image sensors for target registration. Such networks can be used in tactical applications for local and distributed perimeter and site security. Rapid deployments for crisis management may be of particular interest. This paper focuses particularly on the need for applications that deal with relatively dense and complex source fields such as crowds move through sensor spaces.

In this paper, we present the Video Flashlight System and 3D Visualization Display for providing total battlefield situation awareness by integrating a blanket of ground and aerial video cameras, and UGS data within a 3D model of the site. The system enables visualization of an integrated view of a scene, combining video and sensor data from multiple cameras and UGS. Users can move seamlessly in space -- monitor a site from an aerial view and fly-down to examine suspicious activity up close. The system detects moving objects from all cameras and provides an integrated view of all motion in the monitored zones. Users can click on a moving object to get a zoomed-in view or updated data from the sensor. The aerial and ground videos are geo-registered to a world coordinate system. GPS-located UGS data is correctly positioned on the 3D display. Finally, the system can be used to track multiple objects from camera to camera and make measurements such as velocity, etc., and to fuse data from other emplaced sensors into the displayed scene.

Spin Dependent Tunneling (SDT) magnetic field sensors are under development as high performance magnetometers for Unattended Ground Sensing applications. These sensors have been successfully incorporated into a 3-axis magnetometer circuit for prototype level demonstrations with noise floors on the order of 1 nT / root Hz. at 1 Hz. Future versions of these magnetometers will require lower power and lower noise floor than the existing prototype. This paper examines near term developments in the SDT magnetometer power requirements, considering the transducer + amplifier combination as the basic unit. An emphasis is placed on discussing the trade-offs between low power and high sensitivity. The impact of both SDT transducer impedance and amplifier power and noise are considered. Projections show that a 1 pT / root Hz at 1 Hz noise floor is achievable in a transducer + amplifier unit consuming 1 to 10 mW. SDT sensors are being made in an effort to reach these goals. Special emphasis in this paper is on efforts to make an intrinsically low power transducer. Further discussion will explain some fundamentals of SDT sensor operation and how they impact ultimate expected magnetometer performance.

Magnetic sensors are able to detect, track and characterize targets without line of sight, vegetation or weather constraints. This paper discusses performance models, experimental data and the capabilities of a specific magnetic sensor design. These factors are combined to infer the operational parameters of an effective future magnetic sensor array.

Arrays of vector magnetometers employing matched field processing have demonstrated the ability to accurately detect, track, and characterize the magnetic signature of vehicles traveling within range of the sensor field, regardless of weather conditions. This processing works on all types of vehicles, including passenger cars, light trucks, tractor-trailers, tanks, armored personnel carriers, etc. Consistency among the magnetic dipole estimates from similar vehicles has led to investigations of the ability to classify using magnetic information obtained from this process. Preliminary results suggest that separation of the permanent and induced portions of a vehicle's magnetic moment can provide the basis for an accurate, all-weather vehicle classifier.

Currently, the detection of Chemical and Biological agents on the battlefiedl is limited to the response from point and standoff C/B agent detectors. Speculation has been raised, however, that non-C/B sensors could assist in the detection process. Data from unattended ground sensors, infrared imagers, and radar may provide useful information to support a deteciton and may actually provide some warning capability. Ideally, these devices would provide this informiaotn while providign the core Force Protection functions assigned to them. Over the last years, the Soldier and Biological Chemical Command has conducted ouitdoor testing to collect seismic, acoustic, infrared, and radar signatures from conventional and non-conventional events to determine if tehse events can be separated. Preliminary results using several high resolution-laboratory grade seismic, acoustic and IR sensors would indicate that these sensors provide useful discrimination information.

Plants are ubiquitous in the environment and have the 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 tactical uses for this technology. Some obvious applications are using plants as sentinels for detecting biological and chemical warfare agents or their derivatives from a remote platform, as well as detecting explosives. Another tactical application is covert monitoring using individual plants. Different methods to detect GFP in transgenic plants will be discussed.

Tomorrow's battlefields will be dominated by the force that is able to obtain the greatest amount of information. Although battlefield situation information can come from a number of source, one particularly important segment of intelligence gathering is aerial surveillance. For the small, lightly armed unit in the field, the location and force structure of local enemy troops is critically important. Systems Planning and Analysis has conceived and performed a preliminary analysis of a low-cost, lighter-than-air surveillance system. The preliminary analysis includes the concept definition, a detailed trade study of determine the optimal configuration of the surveillance system, high-pressure inflation test, and a flight stability analysis. This paper will provide details in these areas of the design and provide insight into the feasibility of such a system.

In this paper we will discuss DARPA's interest in short-wavelength ultraviolet (UV) semiconductor emitters, and their potential application to biological agent detection, as well as other military applications. Specifically, there is considerable interest in miniaturizing 'detect-to-warn' bio-sensors. Currently, DoD is developing a sub-system known as the Biological Aerosol Warning System (BAWS) for the Joint Biological Point Detection System (JBPDS). The BAWS works on the principle of laser induced fluroescence (LIF), in which an ultraviolet laser beam interacts with a flowing particle stream such that fluorescence is generated when particles containing biological agents are present. A future goal is the reduction in size, weight, power consumption, and cost of this system. Semiconductor UV optical sources afford the opportunity to achieve these new miniaturized system goals.

In this paper, a new WLAN integrating local positioning with Unattended Ground Sensors is described.

The US Army's Future Combat Systems (FCS) and Objective Force will rely heavily on the use of unattended sensor networks to detect, locate and identify enemy targets in order to survive with less armor protection on the future battlefield. Successful implementation of these critical communication networks will require the collection of the sensor data, processing and collating it with available intelligence, then transporting it in a format conducive to make quick and accurate command decisions based on the latest tactical situational awareness. The networked communications must support both static deployed and mobile ground and air robotic sensors with secure, stealthy, and jam resistant links for sensor fusion and command and control. It is envisioned for broadest application that sensor networks can be deployed in a two-tiered architecture. The architecture includes a lower sensor sub-layer consisting of mixes of acoustic, magnetic and seismic detectors and an upper sub-layer consisting of infrared or visual imagers. The upper sub-layer can be cued by the lower sub-layer and provides a gateway link to the higher echelon tactical maneuver layer networks such as the Tactical Internet. The sensor deployments, networking constraints and reach back distances to Command and Control (C2) nodes will be mission scenario specific, however, the architecture will also apply to tactical unattended sensor, munition and robotic applications. Technologies from the Army Research Laboratory, Defense Advance Research Projects Agency (DARPA), and commercial will be leveraged for this effort.

The technical objectives of this effort are to develop low cost sensor packages optimized for three types of unmanned platforms: UGVs, SUAVs and UGS. Additional goals are to develop robust communications to network these sensor systems throughout complex terrain, develop command and control software tools to incldue mission planning, monitoring, dynamic re-planning, sensor planning and management functions; and to demonstrate a system-of-systems capability when fusing information from these various unmanned sensor systems. These capabilities provide the battlefield commander organic unmanned sensor network assets to compelte his Battlespace Situational Awareness picture for targeting, direct and indirect-fire weapons, and threat avoidance. The networked sensors will provide remote monitoring of areas of interest out to approximately 10km not covered by higher echelon surveillance assets and without placing soldiers in harm's way, will increase unit areas of coverage and will provide near real time BSA and targeting data for early warning to speed decision making and reaction time.

The objective of the Joint Robotics Program (JRP) is to research, develop, acquire and field unmanned ground vehicle systems for the United States Armed Forces. The program is structured to field first generation Unmanned Ground Vehicle (UGV) systems, mature promising technologies and then upgrade capabilities by means of an evolutionary strategy. The JRP has a history of adapting to new mission area requirements. An emerging mission area is the employment of UGV systems in conjunction with Unattended Ground Sensors. The author will provide an overview of Service perspectives on sensors and intelligence, recent UGV/UGS efforts, and offer comments on interoperability.

With its focus on the Future Combat Systems (FCS), the U.S. Army has embarked on an important campaign to field a lighter force, capable of being deployed in a fraction of the time currently required. The survivability of this force, however, will be based more heavily on the use of integrated command and control capabilities with unsurpassed situational understanding for all levels of commanders. Arrays of small, low cost, low power sensors will play a key role in detecting, locating, tracking and identifying targets, particularly in areas where the terrain or other circumstances prevent traditional high performance sensors from providing critical information. There has been, and continues to be, a number of DoD programs aimed at fielding Unattended Ground Sensors (UGS). While much of the underpinning technology exists to field such systems, many technology and operational barriers still remain. This paper will attempt to discuss the challenges to overcoming these barriers.

The US Air Force's TASS (Tactical Automated Security System) program has been in existence since 1996. The TASS program meets the growing need to supplement security personnel with modern technology, when these forces are deployed around the world. TASS combines five equipment elements into an integrated security solution, providing both a detection and an assessment capability. TASS does this in a way which maximizes the mobility and user friendliness objectives of the system. In this paper, we will take a closer look at TASS. We will examine the concepts that drive the TASS development process. We will provide an overview of the TASS technical elements, and provide a roadmap for further development of those elements. Finally, we will provide recommendations to security providers who aim to have their products included in the TASS baseline of equipment.

The Submarine . . . The Road Ahead document establishes the desired capabilities of the submarine in extending its reach and becoming a fully netted node within network-centric operations. As a network node, the submarine can receive, process, and share mission-critical distributed sensor data. Additionally, submarines can act as platforms for the deployment of submarine tactical unmanned aerial vehicles (STUAVs). STUAVs extend the capability to communicate critical data and deploy off-board sensor systems. Acting as a 'system administrator' of unattended ground sensors (UGS), the submarine extends its support to special operating forces (SOF) and clandestine intelligence surveillance and reconnaissance (ISR) in access-denied areas. The submarine's stealthy long on-station presence provides for a non-provocative observation of adversary actions on land using unattended ground sensor systems which detect, monitor, and communicate targets of interest. The intent of this paper is to review the needed capability to maximize UGS capability while minimizing volume. UGS system deployment from a STUAV must be packaged in such a manner that the UGS system 1) will be a feasible payload of the STUAV, 2) will be deployed as part of a network grid, 3) will be embedded in the ground, 4) will support the needed duration, and 5) will be camouflaged for low probability of detection (LPD) and low probability of interception (LPI). Once deployed, UGS capabilities to detect, monitor, and relay information will provide another cued asset for intelligence, surveillance, reconnaissance and targeting (ISRT).

Present employment tactics for acoustic sensors in Unattended Ground Sensor (UGS) networks bear no resemblance to similar broadband acoustic sensor tactics for anti-submarine warfare. The tactical thought processes for the employment of both networks do not explain the differences even taking into account the variance in speed of sound in water and other environmental factors. The use of sonobouy experience appears to be a valid source of information on which to base tactical theory for acoustic sensor placement in the land warfare scenario. The development of tactical scenarios for land sensor operations requires knowledge of a set of factors parallel to the at sea scenario. Reverberation characteristics of local terrain, foliage attenuation of sound, weather background noise and other environmental issues must be considered in the placement of sensors to maximize effectiveness. Land-based sensors are being deployed in a single formation for both search and for track of targets. Examination of the operational experience of at sea deployment of acoustic sensors can greatly accelerate the deployment success of land-based unattended ground sensors in tactical situations.

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. In addition, using the geopgraphy dispersion and networked algorithms, it is possible to estimate the target's speed, direction, and loosely classify the target. 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.

We demonstrate an operationally ready system for multi-sensor command and control, display, fusion, and analysis. The demonstration will show integration of seismic, acoustic, passive infrared, magnetic, electromagnetic, electro-optical, and special purpose sensors in an operational scenario against time-critical targets and underground facilities. Unlike other proposed ground stations, DIGS is capable of performing automated target identification, location, and characterization based on all the accumulated evidence from the entire sensor field using a hybrid expert system classifier and a configurable characterization mechanism. These automated capabilities are made possible by operating primarily under an event model rather than a sensor detection model of site activity. DIGS provides all necessary displays for operations including event displays, detailed information, sensor status, sensor control, and a GIS-based map display. In addition, DIGS provides analysis tools to calculate array bearings and Time Difference of Arrival (TDOA) locations, and perform signal analysis of time series or spectral data. DIGS provides an additional layer of fusion to analyze the behavior of targets over time. DIGS is capable of accepting numerous data types including feature vectors, identifications, simple alarms, spectral data, pictures, video, biological and chemical agent information, frequency information, and raw time series data.

Under National Ground Intelligence Center (NGIC) sponsorship, BBN Technologies (BBN) has developed a physical model and a software system to estimate the range and orientation of a target from video images and known information about the target and video camera. This capability may be exploited in an unattended ground sensors field to enhance situation assessment for subsequent response decisions. The target range and orientation estimation system (Video Range Finder) is based on an interactive procedure enabling the user to associate identifiable features of the video image with physical components of the target. These points define a three-dimensional polyhedron in the coordinate frame of the target, which is projected onto the two-dimensional polygon image at the focal plane of the video camera. The associated projection equations are non-linear with respect to the azimuth, elevation, and viewing-axis rotation angles of the video camera, therefore, a solution for these angles is obtained numerically. The calculated angles are subsequently used in the estimation of the target range. The paper discusses the physical model, describes the algorithm for estimating target range and orientation, and demonstrates its application using both simulated and real objects.

In the development of distributed security sensor networks a large variety of prototype systems have been implemented and tested. However these systems tend to be developer specific and require substantial overhead in demonstrating more than one application. To bridge the gap between embedded, networked systems and desktop simulation environments, systems are necessary which are easily deployable and allow extended operation of distributed sensor networks, while allowing the flexibility to quickly test and evaluate a variety of operational algorithms. To enable fast optimization by leveraging the widest development community, open standards for such a portable development system are desired. An open development system allows individual developers and small groups to focus on and optimize specific aspects of a distributed sensor network within realistic deployment constraints, prior to complete integration and deployment of a system within a specific application. By providing an embedded sensor and processing platform with integrated wired and wireless networking, a modular software suite separating access and control of individual processes, and open APIs, algorithm development and software optimization can be greatly accelerated and more robustly tested. To meet the unique needs of distributed sensor network applications, additional separation must be provided between the access to various subsystems, for example real-time embedded control versus tasks with less stringent timing requirements. An open platform that separates these requirements allows developers to accelerate testing and development of applications by focusing on individual components of the distributed sensor system, such as target tracking or low power networking. The WINS NG 2.0 developer's platform, provided by Sensoria Corporation for the DARPA/ITO Sensor Information Technology (SensIT) program, provides one example of such a system. This systems bridge the gap between dedicated desktop development environments and embedded application-specif