• Development of new sensors including nucleic acid-based and tissue based sensors
• Pathogen countermeasures
• Medical treatment(s) and evaluation of long term effects
• Vaccine and drug development
• Protection and decontamination systems
• Domestic preparedness/incident response evaluation
• Collection and information systems development

• Development of new sensors including nucleic acid-based and tissue based sensors
• Remote sensing, instrumentation, microfabrication, microsensors, optical materials, taggants
• Materials and coatings, chemistry and chemical engineering Platform technologies including ROVs and UAVs
• Specialized robotic systems for collection, analysis and countermeasures
• Testing facilities and instrumentation suitable for testing weapons effects against CB warheads and facilities
• Situational awareness information systems development including intelligent decision tools, C4I, and sensor fusion
• Molecular and cellular biology including target detection based on protein and nucleic acid detection, identification of gene mechanisms and tissue-based detection technologies
• Assay and drug development including vaccine therapy development
• Development of sample collectors including aerosols

Key advantages of the UCP technology include:

• High sensitivity (single-phosphor particle)
• Many colors for multiplexing (10 unique colors currently)
• Robust, no photobleaching
• Diode laser excitation (compact sensors)

Upconverting Phosphor Technology

SRI International's Upconverting Phosphor Technology (UPT) has the following superior advantages for BW agent detection and identification:

• High sensitivity due to zero optical background because no other materials in nature upconvert
• Many unique colors excited by the same laser for multiplexing
• Robust--no photobleaching
• Diode laser readout providing for compact, long lifetime sensors.

Samples can be archived at room temperature for subsequent analysis. Two instrument platforms are at the prototype stage of development.
 
SRI, under DARPA support, developed a battery operated, handheld sensor that allows highly sensitive, rapid detection of multiple pathogens (bacteria, viruses, and toxins) simultaneously. The system uses UPT to color-code different pathogens in a test strip. Prototype units are being tested.
 
Also, a compact UPT-based flow cytometer capable of simultaneous detection and identification of multiple antigens is being developed under co-sponsorship of DARPA and the Defense Threat Reduction Agency. Increased sensitivities with significantly improved assay reaction times have been demonstrated. Prototype units are undergoing testing.

Standoff Detection Systems (Lidar)
 
SRI International developed and proved the feasibility of using infrared Differential Absorption Lidar (DIAL) with CO₂ lasers to remotely detect CW agent vapors at ranges up to tens of kilometers. Both airborne and ground-based standoff detection systems for CW agents have been developed and tested for the U.S. and French Governments. Work on standoff detection using lidar systems is continuing with the development of newer laser transmitters and more robust detection algorithms.
 
The detection and identification of biological warfare agents presents a more challenging problem. Under U.S. Army sponsorship, SRI investigated the fluorescence spectra of biological agents to measure critical spectral parameters for BW detection. Under a DARPA SBIR Phase II contract, EOO Inc. (269 N. Mathilda Ave., Sunnyvale, CA 94086, 408-738-5390), with SRI International as a subcontractor, tested a breadboard hybrid biological weapons (BW) agent light-detection-and-ranging (lidar) system at the Summer 98 Standoff Detection Joint Field Trials (JFT). These tests, sponsored by the Joint Program Office for Biological Defense, occurred at the U.S. Army's Dugway Proving Ground (DPG).
 
This hybrid system is the smallest, first ever standoff detector that combines:

1. Aerosol cloud mapping based on infrared (IR) elastic backscatter detection;
2. Ultraviolet (UV) induced fluorescence measurements to determine whether the cloud is biological or not; and
3. Determination of the wind field (speed and direction) using edge-filter Doppler information.

These capabilities are essential for early warning of a BW attack and assessment of downwind hazard conditions. A common IR laser source is used for all three measurements with nonlinear shifting to obtain the UV. Where expedient, commercial-off-the-shelf components were used in the breadboard that was mounted in a van to show concept feasibility. Follow-on improvements at EOO with SRI support brought the system performance up to the design criteria. Parametric studies provided performance vs. form/fit trade-offs that show the design pathway for mounting a hybrid system on a UAV.

Right: Hybrid Lidar Image

Signal Processing/Algorithm Development
With the frequency agile laser (FAL) advantage of tuning to more than 60 wavelengths in less than a second, the opportunity exists to improve upon traditional two-wave-length DIAL for CW detection. To address shortcomings in traditional pattern recognition algorithms based on linear or quadratic discriminants, SRI has developed compelling multivariate statistical inference techniques, particularly estimation and hypothesis testing, that can be used to construct optimal detection algorithms based on the likelihood ratio test methodology. Maximum likelihood estimates of vapor concentration pathlength (CL) and its uncertainty are derived as byproducts of the detection test. This approach has been successfully applied to multicomponent CL detection and estimation using lidar with FAL sources.
 
Specifically, methods to optimally detect and estimate vapor concentration from FAL lidar data addressed signals with a fluctuating component caused by the shot-to-shot variations in the transmitted energy for situations where a local measurement of this energy is made. Previous methods, in which the received lidar signal is ratioed to the energy monitor data, are not only suboptimal, but can degrade lidar performance below that obtained without normalization. This optimal approach, exercised on simulated and actual FAL data, is a linear correction to the received signal that is proportional to the monitor data. The estimated correlation between the transmitted and received signals serves as the optimal proportionality factor for each wavelength.
 
This work, which was performed in fixed-size data, has been extended to address the time series aspects of data collection. The vapor CL is modeled as a simple random walk process in time, thereby leading to a replacement of the maximum likelihood estimates with extended Kalman filter estimates. Not only is the CL estimation variance reduced, but also the Kalman filter approach is better suited to real-time implementation because it does not require the iterative solution of nonlinear likelihood equations encountered in the earlier approach. These techniques are being applied in the Department of Defense's Joint Service Chemical Warning and Identification Lidar Detector (JSCWILD) program.
 
In a complementary SRI thrust, it is important to note that traditional pattern recognition algorithms based on linear or quadratic discriminants force the data into linear vector space frameworks that are amenable to convenient linear processing. This strategy, however, is only valid for weak absorption conditions. Strongly absorbing components can lead to higher false alarms, both negative (missed detection) and positive (interference detection). To address this type of shortcoming, signal models can be constructed for the data that recognize the essentially nonlinear mapping between the path-integrated concentration (CL) and the observed spectral signature. Using such models, the aforementioned estimation and hypothesis testing techniques can be used to construct optimal detection algorithms. As in the case of multicomponent CL detection and estimation for FAL sources, maximum likelihood estimates of vapor CL and its uncertainty are obtained. These techniques may be useful in passive infrared remote sensing applications.

For Further Information on Chem/Bio Defense Applications: Contact Us
Dr. David E. Cooper, Director
Sensor Systems Laboratory
Physical Sciences Division
650-859-3742
david.cooper@sri.com