t
SYSTEMS CONTROL, INC.1801 pagemill road
Palo Alto.California 94304
TELEX 348433Telephone (41s)
494-1 16511 September 1975
Proposal for Research and Development
MPL 76-040
TECHNICAL SUPPORTFOR THE
AI/ASW APPLICATION PROGRAM
*Prepared for: Approved by:
Defense Advanced Research Projects AgencyTactical Technology Office
R. P. WishnerVice PresidentDefense Directorate
SUMMARY
Systems Control, Inc. (SCI) proposes to serve as the Transfer Agentin support of the DARPA Program that is directed towards application ofartificial intelligence to the Navy antisubmarine warfare mission (AI/ASWApplication). The primary role of the Transfer Agent will be to provideassistance to the AI Research Group in the achievement of a significantfeasibility demonstration. The first year effort to be undertaken by theTransfer Agent will include the following areas:
First, the generation of simulated data sets to support developmentof the AI System and the assessment of the conversion processrequirements to translate real data to usable AI development format.Second, the formulation and development of feasibility demonstrationStructure and plan including the definition of applicable measuresof effectiveness.
Third, provide assistance to the AI Research Group and designatedDARPA agents in the achievement of an efficient integrated develop-ment team.
The Transfer Agent's role as it is defined is an essential one tothe achievement of defined program goals - a feasibility demonstration.SCI is the logical choice to be the Transfer Agent because day-to-dayinteractions must be maintained between the AI Research Group and theTransfer Agent to accomplish the interactive data generation, AI develop-ment, and AI System capability assessment tasks that will lead to asuccessful demonstration.
As an option, SCI also proposes to undertake, in addition to theTransfer Agent's tasks, the survey, review, and documentation of availabledata sets that are suitable for support of AI System development.
-
1.0 INTRODUCTION
SCI proposes to provide technical support to the DARPA Program that
is directed towards application of artificial intelligence (AI) to the
Navy antisubmarine warfare (ASW) mission. The proposed SCI effort is
based upon program guidelines set forth by DARPA in their briefing to
potential contractors on 21 August 1975.
SCI recognizes, and fully supports, DARPA's desire that the program
be organized around an integrated team modeled after other successful AI
application efforts such as the DENDRAL and HEARSAY Projects. An AI
Research Group will provide the overall system structure and will be
supported in the development of a knowledge-based system by experts in
areas including current Navy undersea surveillance systems and operations,
projected undersea surveillance missions (systems and operations), advanced
signal and information processing techniques, underwater acoustics relating
to the long range surveillance environment, and availability and constraintsof data bases required for AI application developments.
SCI also recognizes and supports DARPA's desire that the program should
focus on a key ASW problem, the solution of which is central to the success
of the development of the Integrated Undersea Surveillance System (lUSS) as
projected for the 1980s . Thus, the program's initial development emphasis
will be on alerted regional multi-array target tracking and sorting.
Furthermore, a feasibility assessment can only be properly made on real
V
acoustic data rather than simulated data. Thus, the development data
base must evolve from "clean" simulated TARF-like data sets to more
realistic "noisy" data sets and, then, to data sets generated from real
acoustic data. Finally, the program must provide a technical feasibility
demonstration that serves to identify the utility of the AI technology to
the IUSS concept.
The AI/ASW Application Program will capitalize on the results already
achieved by the on-going Heuristic Adaptive Surveillance Process (HASP)
Program in the transfer of AI technology to the undersea surveillanceproblem.
The SCI proposed technical support effort is based upon the DARPA
guidelines and is presented in the following sections. These sections
are summarized below.
Section 2 provides a refined description of the overall DARPA Program
structure as recommended by SCI. Section 3 states proposed SCI technical
support role and describes the specific objectives and deliverables of
each task, together with key project milestones. Section 4 presents the
proposed SCI project organization and key personnel.
2.0 RECOMMENDED PROGRAM STRUCTURE
A view of the overall AI/ASW Application Program structure as recom-
mended by SCI is shown in Figure 1. The diagram depicts the technical
feasibility demonstration as the major program milestone of AI applied to
accomplishing a selected undersea surveillance function: the alerted
regional multi-array target tracking and sorting. Three major task groups
are to be members of the DARPA Program team. These task groups are:
(a) The AI Research Group will be the lead group of the overall
program. This group will produce an AI system that accomplishes
the selected undersea surveillance function.
(b) The Expert Support Groups will provide technical knowledge in
response to the requirements specified by the AI Research Group.
Initially, the identified Expert Support Groups comprise the
signal and information processing group, the undersea surveillancesystem group, and the data base group. Other expert groups will
be identified as future AI developments dictate.
(c) The Transfer Agent will provide assistance to the AI Research
Group in the achievement of a significant feasibility demonstra-
tion.
2:>- oh- "-"»—I I—-J <"—" c_CQ f—■1—« COco 2:<C Ouj 2:U. LU
Q
<:i-<
LU
LU©3 -a:CO H~2: O<C id-JQ.o_ j—
co
1 1
! !
II
UJon
I H- !Xl UJ CO LU I
1 a<o. iI X l-H |
- y*. lv co rD 1/ 1 o_ o* 1f 1 a: d v 1
1 uj o c_r 11 xo. 1I (— CD CO II O <C I
I I
LUQ.ZDI—C_>_3C£H-CO
I .]_:<ca:cdoor
Q_.LUU_ h-co 2:2: UJ<: CDC_ <I—
II
o-UJor
<UJDt— CO o<C <C DCO DD CD
OLUQ2:LU
oc_>UJa: i—
Q
iii
OLUI—<c2:CD COh-i f—co 2:UJ UJQ CD
11
11I "ZZ.I oI I—lI h--1 <=c1 2:« "—"1 Q
1 or1 o1 o1 c_>
"LU__:IDCDi—.U.
I—_TLUSIa.o_iUJ>UJa
2:UJh-CO>-CO
LU Dl.O _z>
<: 2: olv <c ck:co _i o_£ __JLUmZa uj lv2: >" v~z_» q_: co
ZD >-CO CO
CLo_: LU
iii... -- „«._
V V
reow <C r_><UJOco or.
uj c_o_: I
I
T "UJor.
CDo rrU- "—--2: co a.."—1 co 1_""->.UJ oCD CJ> CC*-* O CDco cc:
CL
L I
The tasks of the Expert Support Groups and the Transfer Agent are
summarized in Table 1. A general indication of the first and second year
efforts is provided in the table.
The Transfer Agent, as indicated in Table 1, will be responsible for
four tasks during the first year. These tasks are:
Generate simulated TARF-l ike data for initial AI system
development,
Define and select measures-of-effectiveness for the AI feasibility
demonstration,
Develop a demonstration plan and structure,
Assist the AI Research Group and DARPA-designated agents in
coordination and integration of expert support efforts.
The Transfer Agent's second year tasks will include:
Generate TARF-l ike data from real world data sets,
Prepare and implement the feasibility demonstration,
TABLE 1. TASK OUTLINE, TIMING, AND TASK GROUP RESPONSIBILITIES
TIMING EXPERT SUPPORT GROUPSTRANSFERAGENTTASKS
FIRSTYEAR
SIGNAL/Informationprocessing
FOLLOW-ONYEARS
SURVEILLANCESYSTEM
DATABASE
1. Generate simulated TARF-like data for initialAI system development
2. Survey and review available data sets andprepare data base compendium
3. Generate TARF-like data from data setsdefined in (2)
4. Define and select measures-of-effectivenessfor AI feasibility demonstration
5. Develop feasibility demonstration planand structure
6. Prepare and implement demonstration
7. Establish quantitative values of systemmeasures-of-effectiveness to supportdemonstration evaluation (quantify AIutility in lUSS)
8. Evaluate feasibility demonstration
9. Develop advanced signal processingknowledge base
10. Assess relative utility of AI feedbackcontrol of array processing
11. Plan and Implement special data collectionin support of (9 and 10)
12. Develop advanced TARF processing, TARF-TARFlinking, and integrate AI controlled arrayprocessing
13. Assess impact of differences in operatorexpertise (heuristic power fromstation-to-station)
14. Establish knowledge base for broadtactical use of surveillance resources
15. Assist AI Research Group and ARPAdesignated agents in coordination andintegration of expert support efforts
X X X
Continue assisting the AI Research Group and DARPA-designated
agents in coordination and integration of expert support efforts.
While the primary role of the Transfer Agent is to assist the AI
Research Group in achieving a significant feasibility demonstration, the
Transfer Agent is also the logical choice to undertake the generation of
development and test data sets. The TARF-like data sets, whether simulated
or developed from real data, are a driving factor in the evolutionary
development of AI System capability. The incremental increase in complexity
of the input data sets and the AI System's ability to handle the increased
complexity has been the primary measure of the evolving AI System capability
However, the measures-of-performance must be related to lUSS performance
and formally demonstrated. Since the formulation of the feasibility
demonstration will require a comprehensive understanding of the state of'
the AI System capability, it is imperative that the Transfer Agent be
continually apprised of the evolution of AI capability. An effective
mechanism to achieve the required understanding will be to assign the task
of generating the TARF-like data to the Transfer Agent. Day-to-day AI
development interactions can then be maintained between the AI Research
Group and the Transfer Agent to accomplish the iterative data generation,
AI development, and capability assessment tasks that will lead to a
successful demonstration.
In order to successfully implement the integrated team approach, SCI
recommends that some formal mechanism be established, rather than depending
on ad hoc coordination, to transfer AI development information requirements
and information generated as a result of the requirements. Such a formal
mechanism will ensure that all team members are working from the same data
base and will ensure the efficient and timely update of the data base.
The Transfer Agent, who must comprehend all developments of the Expert
Support Group, is the logical choice to assist the AI Research Group and
DARPA in developing and implementing an integration and coordination
mechanism.
The tasks of the Data Base Expert Support Group will be to survey and
review the available data sets and prepare a data base compendium. This
will assist the Transfer Agent in the generation of TARF-like data from
the data sets. In addition, they will develop the plan for and will
implement the special array data collection as required and specified by
the Signal and Information Processing Expert Support Group.
3.0 SCI TASK ORGANIZATION AND INFORMATION INPUT REQUIREMENTS
SCI proposes two roles in support of the DARPA AI/ASW Application
Program during the first year. First, SCI proposes as a baseline role
to become the Transfer Agent and assume the associated tasks as indicated
in Table 1. Second, as an option, SCI proposes to undertake the Transfer
Agent role plus the Data Base Expert Support role (as defined in Table 1).
The following provides a brief description of the proposed tasks, infor-
mation input requirements, schedules, and major milestones.
3.1 SCI Baseline Support Role
SCI estimates that approximately 24 man-months of effort will be
required over the initial 12 month period to perform the Transfer Agent's
tasks. The specific tasks to be conducted during this period will include
1. Generation of simulated data to support continued AI System
development.
2. Development of demonstration plan and structure.
3. Identification of data conversion requirements.
4. Assistance in coordination and integration of expert support
efforts.
A milestone chart indicating estimated activity periods for eachtask is presented in Figure 2. Key elements of each task are presentedin the following discussion.
3.1.1 Generate Simulated Development Data
The objective of this task is to provide the AI Research Group withTARF-like data sets representing artificial scenarios to be specified by
that group. In particular, this task should relieve the tedious and
time-consuming hand-generation of data sets through the provision of
an automatic data set generator. In addition, these data sets should
become increasingly realistic by the automatic incorporation of noisy
data, false contacts, and ambiguous target parameters. This task will
consist of two parallel subtasks.
The first subtask will design, code, and document an automaticTARF-like. data set generator ("AUTOTARF"). The input to this program
will be the target-mix geometry, the target source levels, and thetemporal and spatial anomalies (i.e., fades) as a function of time.The program will follow a specified data set generation protocol (to beprovided by the AI Research Group). The output will be a data set filein the format currently used by the AI Research Group (and available ashard-copy for distribution and examination).
"I 'I i
1 ! 1- r 4 LI. ! j
! I1... t . . ' j" '....j....... ..... I
..!_.: j
..{...., . - - i --"[--1
i-I
; — -j
-+-■-)-- »"--.-
-4 t 4t _
j '
.. |_...
I-I
I
-j--
-4- " . 4■4- ■ f--i j- --
, j ... 1 p
...4 ....L.i Li
Si■-H- i-li I Lx
j--r -- -j j—"* j pp -j
i h i ! ; rxt j-Ipp ! |"t I n -!!Ii X X i i]-; --) -x ; t 4- ■ i
...J ...p. ..-p. ...Ij ; Ii" 'j :'"!{--'p-- ;
i"1i '"1
\"
i -
I—I—4--, ,—....! : ; i
p.— h . 1 4x
...4... .4 ......j-... p- i-
"}--m1 X XI- !-4 --j -"|--H'-(—.).--i 1 :j-
... ....j... , +i j
i .4—-I-—.; 4—-'. ..;... .J--...!-.-.-
---4xl~i HI 1 i
I_LJo;- - --H-"
1 : o----: t!> ■-
4-_i.__.:ti:p;:
j j4|445:
j o. I ; :_rr?......I f n~
-o
44i x xMX Ml4 i 4
_._.;.._: ; L- 4L ..L.l■ - i'll!: X! ! \
:■' i I !!_■_■_! i i ! i
"■":---■ pp pp p~ j
... ..
f "
...)..
-
..— j..... :. ._..;I I !I ! j
ill:j j !:x::q! ! I
4.44!;]I
—4 ;.~Hi : <— i—r-r>
!_4—>-1.—L.4 a
■4-
~j4
-! -I---'
!i iI r-i-4 —i!-4",4 i1. 1\ L-;li-l—4.(__,
X
...j...! 4 1 i : ' : j_. ■
~"\'_'\ 4444\~Tj4~^^-^-: 1 ; 4'y £ i r '
--4-4 —4—f--r <C-Q viJ .-> -4 4 4 i Y-.PiC^ 4 X
qpXx-M^^xL-4-. _;-.,-_ 4-_-p2 VV --/"o \-i—i---—4--4-&-&- p -a-iA:4
M X ! £_ fiv_j , . _g.Ay,.. - I 1HJxz£4-----zn
-I -4- iIN -i_;_ -1.-§-r-—^| L—L-M_._~|": £51 ~r—T-j--... *. _j __^_j — __________p_
— i-
_.! ..._-(._.
I | ; : I■; : . ■■
._..._!_. 2....
__..._.__.__ hM «<" 2""r !?"§-p-r-1-5_ Mi
- - »%3=» <i
.._._..; 1
4-4--1 I! Il-r—f "-
.... ,
I
i !" !'
■" "
■ ! 4
i
I !_ __! I
— j - ;■+
"
■r-l H-r-ri -rI ;-i_
h- -
r-:^-x—-
■ .-_.y_...
_\ H- -1
-M:3—:—■ <r! ; J-rr- 1
X ; tt\ X vi—7'__L_4„^.
i : «ilh—J.—4— 2,__...._......0
iii" - 4-- ~t-!
1..j --- _ .. ._..__.
..... ....__..__
...;.. p.. _.._..__.-v--i I""i=
; 1 :■
...... vj.---;.....0
4 !":x:_
>a _ 1
I i
LM- r^—xx- j L, :. S' ■ "XI =■"
..._.^.o—r-
.-£--s5"
/\yr~— -t g 4- ~r-X2 VP|X
0
el- - --<! -a" X
_o COp u> .h v
(S I-
::xMdpzxJM-$-M--; _ x-K2--i- -30
A-^y -;-C5 f.:■'!!/■ SX: ■ : i
4 Q :
'44--.4-2- -X. r-
■4--CT"-'--.!.ot_.X.
'■ p.p-A-X--
- _t—-i—uj—J !_^4i 'l "2
-H--x-^f-xpf^-X---^»-o.»-§XJX^2.._...- __.__;_ _ -p g Ul
-■——-^4 t-H-- g"-__ M X »
..._^_r_nr|.._r_r ._1
_r_2
- -4 —. —J--6-*" --!—i H—;--ti
ntfe TMIH4
.__ ...___ :_>.! til illiwiu. _..__.. : ... V_P
.___. : _.-_ 7.4. ": -4 !— fl A)
X < u%'X:p:e
x_ _.^>U-
■■■=> S Ul
XIP-p■ffl
: ■ !>i" 4j "" J
i\^> HrX ; X4——H; i
I1 1 ;;.....i_..r-..H
—V5■■-<— 00--_<t...l
-9-_..\u...
4—1—
...X.. ._.: ___._.__l__i___.___LJ
—!—:—i—pp—ii—r \—
; i [ I ■ j ! j
,_p_ ;! _^_j_|_|_j_ ,
!I;j i :' I
; x i : : .] ; ".
4~TZT~^TT~M—ITX!
M ;-4-hi—l—
—txr-r~i-4
4-M-443u.
2 ;_
...h-H--! '
.. ..: .. i-^ l__... .>T_^_J_...__^__X_
l «._;___.ALi>4ox'if4£|_|[-
-fff-}~H—-r—
>AP——V
I
-">,___y._-^_< I— 1— 14
HHv4iri>:i
I—J—,—j—„.
!
—-v^. X'- X- X.iX:Sji..:X../j : e\—Y-~. s^xi -%. -r _->.. -_7 _n
- . X Xl
s 9,4 \U« 0 *< z ,9 o.
--■7^-r\-~--Xj X..<.-!.._-; x ._-\
<.5..0.PMJ .._ _
__ i i I
: j
! j
4xl
—_).__
_.... Lx44~4xSty-zY-S-iyA--1 A_^ j__pj_|_ j ] I -
- ! ■
±4[xEr.-LCfJ: :4_4_J._-.u<, D..._l__p._
l4£X/:
...I 4
! i _p
T"vs~ T"
4-H -■ -I—t -"--—i ! !
IiI:: i i ' ! i
SSSSYYSYSs:\1 : / _ \ i '
-7-4 xYd 4 3x/! |41 "1 '
I v -r^vi
*$■■).V- ..a..
-j-
-1 i !! i ( 4
—, ■ " '"4 '
- -r-- 4—j_
--'—1 I—p1—p
!
i'.
1I ; ■__ -f- i 'X-h-f- -II ._L_ ; />I !...i_x.p._.._. i_..X-i—I :I i !.1... i ...I. 14—j i
-* — i -
i r I-
f
I '""'j'\\ \~r-M......i I :
i4..<: 414---
-■ ■!--, .....a i -. -}---h
M ■ A i Ii i te"-Mx-i u-"1 !xy| %■ AA
! ! [4-1-4- ~l■4- j-# <C -4--,X* Q XX■ v* v_*
US v* X !
! 1 "I [ " T !""( 1 !
I
! : I
I I
-t i
...
v I !X^T4~Lri~t~.. _t 1 1 r=z itt "~
-itI t-Xc-tXH--'-if £f 7 x[«. -rr o ■■■ -!~IV $t ; -
r «t"S.'-"'t : ~X<T5? h
M^SY\e%S->
j+r-x--,—f -4-- 1-.-.. , —I._ ..__V ./; w-^! ! "■© : 3 <...... _|^ ....__ ..
1 2,' _ 5 O--|uj 5 ■ ? c" ■
r* -K-H"-f 44§ iXI1 Ci> '^
■ £---■» —■ ; > > x .|
I 1_ 1
■ - 4 -4-— ■
I "111
_ _ I4§4
—. : , :——_
. ; 4-U.4... 4! I I
"--7---' I.| ... (—: i ]
k.- . '!'"j I L.I ; L if -...j.. -J [4 ■ ' .4... I 1 J X -4- ■ i '■"''
""»—
V
The second subtask will design, code, and document an automaticTARF-like data set transformer that will map the current TARF-like data
sets and the future AUTOTARF data sets into noisy data sets. The
principle feature of this program will be the introduction of ambiguous
target parameters (bearing center frequency, bandwidth, and stability)
into the data sets according to empirical statistics of those parameters.
The resulting data sets will provide a close approximation to the quality
of data to be encountered with real acoustic data.
3.1.2 Development of Demonstration Plan and Structure
The objective of this task is to develop and specify the plan for
the demonstration to be performed in support of the contention that the
field of AI has applicability to the undersea surveillance problem of
(multi-array) regional target tracking and sorting.
This task will begin with a description of the projected AI ProgramV
capabilities, the corresponding of all available (real) data sets, the
specified measures of effectiveness, and a summary of suitable advanced
processing techniques. SCI will then develop the demonstration methodo-
logy and structure; in particular, SCI will specify the input data
requirements, the output data statistics, and the demonstration perfor-
mance measures (incorporating the established measures-of-effectiveness).
SCI will also propose a plan-of-operations for the demonstration including
demonstration facilities, personnel, components, and schedule. The
ARPA Research Center will be examined as a primary candidate facility
to implement the feasibility demonstration.
3.1.3 Identify Data Conversion Requirements
The objective of this task is to identify the processing require-
ments and the estimated effort and schedule for converting the real
acoustic data sets into the TARF-like format required by the AI Research
Group.' The basis for this task will be the data base compendium generated
by the Data Base Expert Support Group.
3.1.4 Assist in Coordination and Integration
SCI will provide assistance to the AI Research Group and DARPA
designated agent in providing a mechanism for the timely transfer of
information and requirements between the program team members. SCI will
be responsible for the maintenance of a project library (to contain all
relevant technical memoranda and data sets) for ready access by all
project participants. An updated index to this library will be distri-
buted periodically.
3.2 SCI Support Role Option
It is estimated that approximately 30 man-months of effort will be
required over a 12 month period to undertake the Transfer Agent's tasks,
-
as described in Section 3.1 above, plus the survey, review, and prepara-
tion of a data base compendium. The estimated schedule for completionof the data base compendium is estimated to be six months after project
initiation as it is shown in Figure 2. This schedule can be kept onlyif DARPA can immediately provide SCl's access to all available data sets
V
4.0 SCI PROJECT MANAGEMENT AND KEY PERSONNEL
The proposed SCI technical support effort will be conducted by the
members of the staff within SCl's Defense Directorate Organization,
directed by Dr. R. P. Wishner.
The overall responsibility for carrying out a project and for
meeting project objectives is placed with the Project Supervisor.
Specifically, the Project Supervisor has the direct responsibility and
authority for assuring that the project is adequately staffed, that all
efforts are directed towards fulfilling project objectives, and that all
milestones are achieved. He also monitors technical progress on eachtask of the contract. The SCI Project Supervisor will be Mr. A. Bien,
Manager, Naval Analysis Programs.
Tne SCI Project Leader is directly responsible for the detailed
technical quality of all deliverables, the technical direction of all
project personnel, the detailed monitoring of technical progress on each
task, and the evaluation and control of project funds. The Project
Leader is assisted in his control of project funds by SCl's Project Labor
and Cost Reporting System, which assures an up-to-date reporting of all
project charges. Separate accounts are established and monitored for
each assigned task.
These reports will be used as the basis for a periodic reporting
of the fiscal status, as required by the sponsor. Regular status review
of individual tasks with the SCI Project Supervisor will be required as
another assurance of satisfactory performance. The SCI Project Leader
will be Dr. W. J. Bangs, 11.
Other key staff members expected to be major contributors to the
proposed technical support effort include Dr. J. M. Clark and Mr. M. W.
Zumwalt. Brief summary biographies of the proposed project personnel
are presented below.
Mr. A. Bien, Project Supervisor, is the manager of the Naval Analysis
Programs. Mr. Bien has had 10 years experience prior to joining SCI at
the Naval Warfare Research Center, Stanford Research Institute, where he
developed and managed the undersea warfare program. Mainstream effort of
the program was directed towards establishing requirements, and formu-
lating and evaluating concepts to counter the Soviet cruise missile sub-
marine threat. Mr. Bien conducted and led a number of studies that
required the formulation and use of measures-of-effectiveness (MOE's)
and development of methodologies required to assess systems performance
within the framework of the defined MOE's. Since joining SCI, Mr. Bien
has continued developments in the ASW areas with emphasis on undersea
surveillance system integration to perform the tactical surveillance
mission.
Dr. W. J. Bangs, 11, Project Leader, was formerly AssistantProfessor of Mathematics at the College of William and Mary. There,he developed and taught graduate-level courses in artificial intelligence,
simulation, automata theory and formal languages, and problem solving.
Since joining SCI in 1973, Dr. Bangs has had a variety of technicaland managerial responsibilities. He has been responsible for the design
and development of real-time spectral analysis algorithms for the ARPA
Acoustic Research Center (ARC). He has played a central role in the
requirements analyses and preliminary designs of the ARC signal processing
capabilities (including RTSP, MSDR, ARCLT, and MATE).
From July 1974 to May 1975, Dr. Bangs was the Project Leader for
the System Validation Exercises Project on the ARC Program. He had the
responsibility for defining the RTSA system test and the New Facility
Demonstration. From June 1975 to the present, Dr. Bangs has been Project
Leader for a Signal Processing System Test Bed Design Contrabt with NRL.
He has been responsible for the creation of a data management model of
signal processing processes and a preliminary design of a software
structure for the creation and execution of signal processing. In each
of his roles as a Project Leader, Dr. Bangs has successfully demonstrated
technical leadership, personnel management, and fiscal responsibility.
..
Dr. J. M. Clark has had many years experience prior to joiningSCI in the areas of computer systems and signal processing. Duringhis six years with TRW Systems, he held many positions associated withdevelopment of the undersea surveillance system. He was manager of theAcoustic Signal Processing Projects Office, the Project Manager ofSystem Engineering Support for the ARPA Research Center, and the managerof a signal processing section. He conducted numerous studies involvingundersea surveillance signal processing. While at SCI, Dr. Clark
continued his involvement in the ARPA Research Center as well as othersignal processing activities.
Mr. M. W. Zumwalt has been extensively involved, for over a decade,in the operational and cost-effectiveness analysis of naval systems. Inparticular, a large portion of Mr. Zumwalt's analytical efforts have beendirected towards the investigation of the effectiveness of the variousNavy surveillance systems; in both the tactical and strategic contexts.For example, Mr. Zumwalt has participated in studies designed to deter-mine the effectiveness of the SOSUS System when employed either as adefense against the ballistic missile launching submarine or as atactical adjunct to Naval Task Forces at sea. Most recently, Mr. Zumwaltorganized and conducted a series of studies of the capabilities, bothtactical and in the surveillance role, of several of the Navy's currentand proposed towed array systems. All of these studies have necessitatedthe formulation of relevant resources of system's effectiveness and thestructuring of appropriate analytical models in order to properly quantifythe effectiveness exhibited or expected from each of the various systems.