PUBLIC DISTRIBUTION
Pulsed Microwave Data Link-
Smart Wireless Switched BuildingBlocks for IVHM Sensor Interoperability
Pulsed Microwave Data LinkPulsed Microwave Data Link--
Smart Wireless Switched BuildingSmart Wireless Switched BuildingBlocks for IVHM Sensor InteroperabilityBlocks for IVHM Sensor Interoperability
Sandeep GulatiJohn Mai
Tom GeorgeGreg Cardell
ViaLogy2400 Lincoln Ave.
Altadena, CA 91001
28 March 2007
Sandeep GulatiSandeep GulatiJohn MaiJohn Mai
Tom GeorgeTom GeorgeGreg Greg CardellCardell
ViaLogyViaLogy2400 Lincoln Ave.
Altadena, CA 91001
28 March 2007`̀
PUBLIC DISTRIBUTION
ViaLogyViaLogy
Altadena, California operations
Original Discovery – QRI Core Technology
12 Patents Issued to ViaLogy
First Patents Issued with ‘No Prior Art Status’
Core Competencies:IP-based interoperable sensor networks & policy engines
Sensor fusion
Weak signal detection
PUBLIC DISTRIBUTION
OverviewOverview
Demonstration Scenario2.45 GHZ RFID system scanning tags behind Kevlar and other composite materials.
ViaLogy’s Active Signal Processing Solution (the “A------” solution)
Qualitative Casual Physics and Quantum Resonance Interferometry
ViaLogy’s Network Interoperability Solution. (the “B------” solution)
Sensor Policy Manager (SPM) and IPICS
Summary: Building Blocks Available Today.
Demonstration Scenario2.45 GHZ RFID system scanning tags behind Kevlar and other composite materials.
ViaLogy’s Active Signal Processing Solution (the “A------” solution)
Qualitative Casual Physics and Quantum Resonance Interferometry
ViaLogy’s Network Interoperability Solution. (the “B------” solution)
Sensor Policy Manager (SPM) and IPICS
Summary: Building Blocks Available Today.
PUBLIC DISTRIBUTION
Experimental ProcedureExperimental Procedure
45o
30o
15o
0o
0.3m 0.6m 0.9m
A
1. Antenna mounted on tripod
2. Various types of tags are positioned at different distances and angles from the center of the antenna
3. DAQ is turned on for 30 seconds, sampling at 2 Hz.
4. Record the total number of tag reads after 30 seconds. Calculate the Read Percentage.
5. Repeat with tag positioned behind different composite materials.
PUBLIC DISTRIBUTION
Sample ResultsSample Results
45o
30o
15o
0o
0.3m 0.6m 0.9m
Plastic Tuned Tag
100% 48% 3%
100% 3%0%
65%
3%
0%
92%
62%
0%
A
45o
45o
30o
15o
0o
0.3m 0.6m 0.9m
Plastic Tuned Tag + ¼” of Kevlar
90% 90% 0%
72%
18%
0%
35%
25%
0%
A
PUBLIC DISTRIBUTION
Contradictory Results?Contradictory Results?
45o
30o
15o
0o
0.1m 0.3m
Glass Tuned Tag
68% 0%
0%
0%
7%
0%
A
45o
45o
30o
15o
0o
0.1m 0.3m
Glass Tuned Tag + ¼” Kevlar
95% 60%
37%
20%
43%
20%
A
PUBLIC DISTRIBUTION
Active Signal ProcessingActive Signal Processing
Is it logical to use low value sensors to protect high value assets?
Yes, if we can implement a mathematical solution that can deal with highly non-linear outputs.
What we really need is a Situational Calculus Solver
• automatically reasoning about the kinds of changes that occur and their effects [Frame Problem, McCarthy & Hayes]
• interpreting what changes and what does not
• reasoning under uncertainty
Is it logical to use low value sensors to protect high value assets?
Yes, if we can implement a mathematical solution that can deal with highly non-linear outputs.
What we really need is a Situational Calculus Solver
• automatically reasoning about the kinds of changes that occur and their effects [Frame Problem, McCarthy & Hayes]
• interpreting what changes and what does not
• reasoning under uncertainty
PUBLIC DISTRIBUTION
Qualitative Causal Physics [J De Kleer, Xerox]
Understanding the paths of causal interaction formulated as a constraint network is fundamental to robust monitoring and control.
Adjacency: basis for all interaction - physical, electromagnetic, thermal, electrical, ..., adjacency
difficulty: multiplicity of interactions possible, some of which are unusual or subtle
no-function-in-structure principle laws of the subsystem may not presume the functioning of the whole system
locality principlelaws for a subsystem cannot specifically refer to any other subsystem
Qualitative Causal Physics [J De Kleer, Xerox]
Understanding the paths of causal interaction formulated as a constraint network is fundamental to robust monitoring and control.
Adjacency: basis for all interaction - physical, electromagnetic, thermal, electrical, ..., adjacency
difficulty: multiplicity of interactions possible, some of which are unusual or subtle
no-function-in-structure principle laws of the subsystem may not presume the functioning of the whole system
locality principlelaws for a subsystem cannot specifically refer to any other subsystem
Theoretical Basis of ApproachTheoretical Basis of ApproachTheoretical Basis of Approach
locus of causal interactions must be determined a priori
PUBLIC DISTRIBUTION
So how should we model this?So how should we model this?
si sj sk
sp sn
sl
smsqO1
O2
a typical subsystem Γi:: AACS
si...or...os...smΓAACS
si...op...oq...smΓPPS
...
Θα(or) Θβ(os)⊗ Θα(or) Θβ(os)⊗
lag Δ lag ΨOperability DesignΘ, ⊗, Δ
si: sensor Ok: critical observable
ΨΨ
Ψ,Ψ,
Spectral Coherence Density estimatesCausality estimates
Track Temporal Moment Oscillation
Power Source16
Power ControlShunt
Regulator
1773 ProtocalChip
FPGA
DATABUSI/F
PMADPMAD
PMADPMAD
PMADPMAD
PMADPMAD
Pyro Electronics
Pyro
DriveCircuit
Hybrid
Power
Converter
DATABUS I/F
1773 Protocal
Chip
Power ControlShunt
Regulator
1773 ProtocalChip
FPGA
PMAD
Hybrid
Converter
DATABUSI/F
PMADPow
er Switch
Control ASIC
PWR MOSFETS 1
thru 16
FPGA
Hybrid
Power
Converter
DATAB
USI/F
PMADPMAD
Power
Switch
Control ASIC
PWR MOSFETS 1
thru 16
FPGA
Hybrid
Power
Converter
DATAB
USI/F
Pyro Electronics
Laser DiodeCircuit
Hybrid
Power
Converter
DATABUS I/F
1773 Protocal
Chip
Hybrid
Converter
PUBLIC DISTRIBUTION
PMDL Computation SummaryPMDL Computation Summary
CORE HYPOTHESISSpacecraft/system treated as arbitrarily large, complex multidimensional signal where the information actually resides within a low dimensional spaceinitially the total diagnostic space covers all sensors, command and executing flight softwarerigorously developed algorithms to automatically determine the “information space” of the system over all operating modes
formally showed that “reduced and reconstructed information space” can actually and quantitatively capture all diagnostic and degradation information (guarantee complete state coverage with arbitrary fidelity)
PUBLIC DISTRIBUTION
Step 1: Develop a “Quantum Expressor Function” (QEF) i.e. QSR model characterizing the system noise, ultimate limit of detection and precision.
Step 1: Develop a “Quantum Expressor Function” (QEF) i.e. QSR model characterizing the system noise, ultimate limit of detection and precision.
Quantum Resonance Interferometry
Step 2: “Destroy” the QEF model by injecting signal.
Three QRI Modes:a. Generate a QSR for signal of interest
and match transition rate to QEFb. Feed the output of signal QSR to input
of QEF and adjust transition ratec. Construct an idealized QEF based on
theoretical principles, inject the real data and track the deviation
PUBLIC DISTRIBUTION
PMDL COMPUTATIONAL ELEMENTSPMDL COMPUTATIONAL ELEMENTS
ON-BOARDFUSION
OPERATOR
Sensors: 1,2,.....NExecuting Software
Commands
OK!PROBLEM!
EMERGENCY!
ON-GROUNDBACK PROJECTION
OPERATOR
BBEE
AA MM
problem diagnosis and basis on which “problem
decision was made”
PROBLEM!PROBLEM!EMERGENCY!EMERGENCY!
HYPERCOMPACTOPERATING
MAP
• Impact Reporting• Time-to-criticality• Degradation Assessment• Residual-life• CBM (inspection) Tasking
TestbedTestbed
ON SPACECRAFT/AIRCRAFTON SPACECRAFT/AIRCRAFT ON GROUNDON GROUND
PUBLIC DISTRIBUTION
Existing Example: Quantum Resonance Interferometryand Gene Expression
QuantumInterferometric
Interaction(few milliseconds)
Spatial Random Field (SRF)Device, Dye ResponseStatistics
Quantum ExpressorFunction
Complex Mathematical
Coding
Digitized Microarray Output from Lab Sample
ExpressedBixels
Empty &Non Specific
“Training” Data Sets
QuantumExpressorFunction
Gene Expression Pattern
QRIProcessing
(few ms)
Microarray Dataset
PUBLIC DISTRIBUTION
Wireless adds another degree of freedomWireless adds another degree of freedom
Why do I need “artificial intelligence?” –you “paid” for the wireless innovation, why not take full advantage of it.
Wireless networks allow us to integrate “expert opinion” (real people) into the decision loop.
ViaLogy’s Sensor Policy Management (SPM) and Sensor Fusion
RFID tags is just another sensor input
Why do I need “artificial intelligence?” –you “paid” for the wireless innovation, why not take full advantage of it.
Wireless networks allow us to integrate “expert opinion” (real people) into the decision loop.
ViaLogy’s Sensor Policy Management (SPM) and Sensor Fusion
RFID tags is just another sensor input
PUBLIC DISTRIBUTION
SPM/IPICS ArchitectureSPM/IPICS Architecture
Sensor Policy Manager (SPM) And IPICS
Communications Interoperability
GPSAdaptor
Sensor Channel
Configuration & Management
VIPERSoftware
Sensor Policy Manager
SP
MIn
tegr
atio
nA
PI
IPICS MasterPolicy Engine Generic
Adaptor
Radar
IPICS Services
Stack
WebBrowser
NotificationDevices
Sensor Channel
EO/IRRFID Adaptor
Imaging Adaptor(EO/IR, Video)
Video
WirelessNetwork
PUBLIC DISTRIBUTION
Why is this valuable?Why is this valuable?Data combined with knowledge creates valuable information.
IP Interoperability & Collaboration System (IPICS)/ Sensor Policy Manager (SPM) for communicating
the right information
in the right format,
to the right person,
at the right time
using diverse sensor/communication networks.
A logicial extension of the Skype, Wikipedia, IM, youTube, and email-based technology work culture of today…
Data combined with knowledge creates valuable information.
IP Interoperability & Collaboration System (IPICS)/ Sensor Policy Manager (SPM) for communicating
the right information
in the right format,
to the right person,
at the right time
using diverse sensor/communication networks.
A logicial extension of the Skype, Wikipedia, IM, youTube, and email-based technology work culture of today…
PUBLIC DISTRIBUTION
What does SPM/IPICS Enable?What does SPM/IPICS Enable?
PSTN CellPC VoIP
Comprehensive Voice Interoperability
Radio Interoperability
UHF VHF
NXTL
Contextually Integrates Multiple Modalities of Communication for Interoperability and Collaboration
GIS and Data SensorVideo
Policy-Based Communications Interoperability
Automation, Inference Engine, Workflow
Integration
PUBLIC DISTRIBUTION
AdvantagesAdvantages
IPICS/SPMSensors are cheap. People add value.
Human in the loop confirms decisions- experts establish alert criteria.
But all data is available for review- experts can subscribe to data channels
Decisions and data can be reviewed anywhere at anytime
Flexible use of experts only at critical events.
IPICS/SPMSensors are cheap. People add value.
Human in the loop confirms decisions- experts establish alert criteria.
But all data is available for review- experts can subscribe to data channels
Decisions and data can be reviewed anywhere at anytime
Flexible use of experts only at critical events.
PUBLIC DISTRIBUTION
SummarySummary
Available TodayQRI – Quantum Resonance InterferometryIPICSSPM
Video adaptors (2.1 to 5.1 megapixel)GPS adaptorChemical adaptor (O2, CO2, methane)Bio adaptor (PCR)
Custom SPM sensor adaptors within 72 hrs.
Coming SoonMobile Access Router (PC104 form factor) with video streaming.
Available TodayQRI – Quantum Resonance InterferometryIPICSSPM
Video adaptors (2.1 to 5.1 megapixel)GPS adaptorChemical adaptor (O2, CO2, methane)Bio adaptor (PCR)
Custom SPM sensor adaptors within 72 hrs.
Coming SoonMobile Access Router (PC104 form factor) with video streaming.