14 may 08...rigid body (rb) movements (tape measure) form basis of the algorithm zbody encountering...

SIGNAL PROCESSING MEANS FOR DETECTING AND SIGNAL PROCESSING MEANS FOR DETECTING AND DISCRIMINATING BETWEEN STRUCTURAL DISCRIMINATING BETWEEN STRUCTURAL

CONFIGURATIONS AND GEOLOGICAL GRADIENTS CONFIGURATIONS AND GEOLOGICAL GRADIENTS ENCOUNTERED BY KINETIC ENERGY PENETRATING ENCOUNTERED BY KINETIC ENERGY PENETRATING

PROJECTILESPROJECTILES

14 May 08

Expanding the Realm of Possibility

Program OverviewDevelop a Signal Processing Method to:

Detect and Discriminate betweenNatural & Structural Layers Encountered During Terra-Flight

Distinguish Gradients Encountered During Flight

Verify the Algorithm (method) Against Multiple Real World EventPatent the Algorithm


Program GoalsSet-up a Matlab Simulink Algorithm SimulatorVerify the Algorithm (method) against Multiple Real World Events (digital signals)Transfer Algorithm to ‘g’ Hardened DSP Controller Chip Develop Operator Interface for DSP ChipVerify Chip Operation with Real World Analog Signal Inputs


The Acceleration MeasurementThe Acceleration Measurement

Rigid Body (RB) Movements (Tape Measure) form basis of the Algorithm

Body encountering a resisting target with a force of opposite sense as velocity and proportional to V2

Rise time equal to crater depth (Nose Length)RB Waveform a non-structured trapezoid (= low frequency <100 Hz)


Time

Dec

eler

atio

n

ReboundSpringEnergy

Release

Lock-upProportional to VViscous/StokesDrag

Nose Entry

Penetration Phase

Characteristics of RB Acceleration Records

End of EventIgnored inAlgorithm


Superimposed On the Superimposed On the Measurement & Masking the Measurement & Masking the Rigid Body are:Rigid Body are:

KHz signal movements representing (i.e., measured by micrometers)

Vibrational modes of the penetrator structureMounting of the transducerInternal Component Freight Training

Some MHz signals representingSteel on Steel contactKinematic Mass Property Changes 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 0.055

-2

-1.5

-1

-0.5

0

0.5

1

1.5

x 104

time(seconds)

g's


UnUn--Masking the Rigid Body MovementMasking the Rigid Body Movement

It is known (One Way) that an integral will detect layers

But slope detection merely returns the signal

What Venue supports detection of targets masked by High Frequency Content signals?

AnswerRadar TargetsLFM Acquisition (Purposely Masks LF Target) and detected by Pulse Compression (An Integration)

4000 5000 6000 7000 8000 9000 10000 11000 12000 13000-2

-1.5

-1

-0.5

0

0.5

1

1.5

x 104

G's


Pulse CompressionPulse Compression ItIt’’s Analog to Decelerations Analog to Deceleration

Energy to a target has been purposely modulated with multiple frequencies. Multiple Freq Sources in Projectile.The target is hit with the energy and returns a high frequency content (HFC) signal. Penetrator strikes target & returns HFC SignalEach ‘f’ contains target information, both cases – Goal = ExtractThe return pulse is compressed by slowing down the high frequencies and speeding up the low frequencies using analog devices or

A matched filter used but this is an autocorrelationDue to the high frequency of the return radar signal analog pulse compression is easier then digitizing and running an autocorrelation.Due to the relatively low frequency of a penetration event KHz versus GHz, digitization and autocorrelation is more appropriate.Results are the same:

Target resolved in range (ringing removed and info stored in low freq range resolved pulse)


Pulse Compression = Auto CorrelationPulse Compression = Auto Correlation

Pseudo


0.75ft (5ksi) + 13ft air + 0.75ft (5ksi) + 13ft air + 0.75ft (5ksi) + 13ft air + 4 ft (10ksi) + 1.7ft air + 4ft (10ksi)

0 2000 4000 6000 8000 10000 12000 14000 16000-1

-0.5

0

0.5

1x 10

4

Counts = Time*200000

g's

Multi-Layer Target

0 200 400 600 800 1000 1200 1400 1600 1800 2000-0.5

0

0.5

1

1.5

2

2.5

3x 107

Counts = Time*25000

g2

Autocorrelation

+ Nice Range Resolution


Where is the Uniqueness?Where is the Uniqueness?In a sphere, cylinder, plane, concrete?In a radar system each target is identified by a unique number (its RCS)In a penetrating system each target is identified by a unique unit less value called peak ‘g’This identifier is independent of striking or draining velocity and only dependent on geometry and target resistanceTherefore penetrator targets (for the same penetrator geometry) that are not the same are different

0

10000

20000

30000

40000

50000

1500

2000

2500

30003500

2030

4050

60

Peak

g's

velo

city

ft/s

ec

Penetration Inches

Peak 'g' Convergence

0 10000 20000 30000 40000 50000

0

1000

2000

3000

4000

5000

6000

7000

500

600700

800900100011001200

1015

2025

30354045

Peak

g's

Velo

city

Penetration Depth


0 1000 2000 3000 4000 5000 6000 7000


2

)42.32( 1

1gtVstNL −=

2)(2.32 2

2tgNLDF =−2

]2[2.32 21 ttgVs −=

G’s

1t 2t

Vs=Striking Velocity NL=Nose Length

DF=Final Depth of Penetration


If Correct Data Should ConvergeIf Correct Data Should Converge to Single Valueto Single Value

0

100

200

300

400

0

500

1000

1500

2000

050

100150

200

Peak

g's

Velocit

y ft/s

ec

Depth of Penetration ft


0 100 200 300 400

0 500 1000 1500 2000-50

0

50

100

150

200

250

Velocity (feet/ sec)

Dep

th o

f Pen

etra

tion

(fee

t)

GENPEN 9.25 CRH 350#L/D = 10 Antelope Lake, Nevada

GENPENFit

Velocity Stepped thru Natural Cemented Sand Target


0

10000

20000

30000

40000

50000

1500

2000

2500

30003500

2030

4050

60

Peak

g's

velo

city

ft/s

ec

Penetration Inches


0 10000 20000 30000 40000 50000

1500 2000 2500 3000 35000

10

20

30

40

50

60

70

Velocity (ft/ sec)

Pene

trat

ion

Dep

th (i

nche

s)

3CRH, 1Kg, L/D=10,Limestone

Concrete DataFit

Velocity Stepped thru Limestone Target


0

1000

2000

3000

4000

5000

6000

7000

500

600700

800900100011001200

1015

2025

30354045

Peak

g's

Velo

city

Penetration Depth


0 1000 2000 3000 4000 5000 6000 7000

0 200 400 600 800 1000 1200-10

0

10

20

30

40

50

Velocity (ft/ sec)

Pene

trat

ion

Dep

th (i

nche

s)

3CRH, 13Kg, L/D=7, 3.5Ksi Concrete

Concrete DataFit

Velocity Stepped thru Concrete Target


7x10^6


Program AccomplishmentsA Robust Matlab Simulink Algorithm Simulator was set-upThe Algorithm was Tested against Multiple Real World Events (digital signals)The Algorithm was transferred to a ‘g’ Hardened Freescale DSP Controller Chip with Operator InterfaceChip Operation was verified with Real World Analog Signal Inputs to the ChipThe Algorithm is Patent Pending

14 may 08...rigid body (rb) movements (tape measure) form basis of the algorithm zbody encountering...

Documents