top - download
⟦e3a8e431f⟧ Wang Wps File
Length: 9435 (0x24db)
Types: Wang Wps File
Notes: .of meetg. 30-31.10.80
Names: »0294A «
Derivation
└─⟦303071de8⟧ Bits:30006071 8" Wang WCS floppy, CR 0027A
└─ ⟦this⟧ »0294A «
WangText
…02…GJ/801106…02……02…#
Minutes of Meeting 30. - 31.10.80
CAMPS meeting no. 2…02……02…CAMPS
Participants: SHAPE: - Mr. Marsh
Mr. Earl
ARC : - Mr. Philip Becker
CR : - GJ, BHB, OE
S̲u̲b̲j̲e̲c̲t̲:̲ ̲C̲A̲M̲P̲S̲ ̲E̲M̲I̲ ̲R̲A̲C̲K̲S̲
-------------------------------------------------------------------
1 A̲R̲C̲ ̲R̲E̲P̲O̲R̲T̲ ̲O̲F̲ ̲S̲E̲P̲T̲E̲M̲B̲E̲R̲ ̲1̲9̲8̲0̲
The report was handed over to Mr. Marsh and Mr. Earl
who then read the report. The following issues were
brought up:
a) SHAPE: The racks should be welded into a complete
rack so the top and buttom is not assembled
after shipping.
ARC: This is proposed by the various vendors.
b) SHAPE: Do we need to be able to open the entire
back of the rack?
CR action CR: The
design
has
to
be
studied
before
the
answers
can
be
given.
c) SHAPE: Will shielding effectiveness be specified
in the technical specifications to the
rack suppliers?
ARC: This cannot be recommended. The specifications
will be a mechanical specification.
d) ARC pointed out that the cage solution is less
risky than the rack solution.
e) SHAPE: Finger stock gaskets tend to wear out.
ARC: The gaskets will be designed as good as
possible. A 3 - 5 years life time is recommendable.
Spares shall be available for replacements.
f) ARC clarified that the door will have a latch in
each side. Then the door can be pulled out 3 -
4" before it is turned on the hinges. SHAPE agreed
with the general approach to the door gasketing.
g) SHAPE: Have you considered QA inspections at the
rack vendor?
ARC: Either CR or ARC will inspect the racks
at the vendor prior to shipment to CR for
integration.
h) SHAPE: SHAPE is confident that if the methods
proposed are put into effect, then ACE
COMSEC can avoid any tests prior to the
test at CR.
i) SHAPE: It is recommendable that any unneccessary
holders/blinds into the racks should be
avoided. The lifting eyes should not penetrate
the racks.
ARC: This will be satisfied.
2 R̲A̲C̲K̲ ̲W̲E̲I̲G̲H̲T̲
CR action The weight of one rack is estimated to approximately
350 kg. Therefore some additional lifting arrangements
are needed.
3 H̲A̲R̲D̲W̲A̲R̲E̲ ̲C̲O̲N̲F̲I̲G̲U̲R̲A̲T̲I̲O̲N̲
3.1 O̲p̲t̲o̲f̲i̲b̲e̲r̲ ̲C̲o̲n̲f̲i̲g̲u̲r̲a̲t̲i̲o̲n̲
SHAPE stated that the LTUX/OPTO configuration, where
only one connection between the opto link and the terminal,
is a good one. It gives maximum security. Patching
facilities are not recommendable. The LTUX design is
a good approach.
All signal lines entering or leaving the CAMPS racks
are red.
3.2 P̲o̲w̲e̲r̲ ̲S̲t̲a̲b̲i̲l̲i̲z̲e̲r̲s̲
The power stabilizers cannot be fitted into the existing
racks. Therefore they will have to be placed in an
additional rack within the CAMPS room. This would require
additional filters in the ABC bay, if the main filter
is placed outside the ABC bay. The stabilizer rack
does not necessarily have to be a tempest rack.
SHAPE SHAPE needs to tell CR if all sites need power
action stabilizers.
3.3 C̲o̲l̲l̲o̲c̲a̲t̲e̲d̲ ̲T̲a̲r̲e̲
When a tare is remote, the modem and the Black Patch
Jack Field (BPJF) cannot be installed in the crypto
rack. It shall be placed in a separate 19" rack. Advanced
copies of drawings showing the layout is attached in
appendix A. The drawings will be formally forwarded
by SHAPE with the comments to the Installation Plan.
The cable coming out of the IDF is a 9 wire cable.
This does not agree with the info CR
CR action currently has. This must be investigated.
3.4 R̲e̲m̲o̲t̲e̲ ̲T̲e̲r̲m̲i̲n̲a̲l̲s̲
The modem and BPJF shall be placed in a separate rack.
At the remote location the crypto and RPJF could be
placed in a separate rack from the statistical Mux
and opto Mux/Demux. This is site dependent. All connections
to the crypto is SHAPE's responsibilities.
3.5 O̲p̲t̲o̲ ̲F̲i̲b̲e̲r̲s̲
At the moment no cable length is above 1000 m. Most
lengths are 50 - 200 m, exceptions are approx. 500
m.
Repeaters have to be installed if the cable length
exceeds 800 - 1000 m. Splicing may be needed since
it is difficult to install a long cable.
The fiber link itself will be a dual monofiber.
3.6 L̲i̲n̲e̲ ̲D̲r̲i̲v̲e̲r̲s̲
ARC pointed out that the 1 mA requirement asked for
by SHAPE placed significant limitations on the driving
distance. ARC also pointed out that the tempest requirements
could be solved by different methods (a combination
of wave shaping, filters, and other). Therefore ARC
would like to recommend that line drivers like Fairchild
9636A or similar should be used. The line driver characteristic
should be wave shaping adjustment and possible linear
wave shaping instead of exponential.
It was suggested that CR could ask for a change on
the 1 mA requirement of the signal in order to get
a longer driving distance. Then SHAPE will evaluate
if the tempest requirements are met.
3.7 I̲D̲F̲
It was recommended by ARC that the IDF should be build
in such a way that the low, medium, and high speed
line connectors are separated in 3 separated areas
within one box.
For each line there are 2 connectors. One for the computer
connection and one for the terminal connection. The
signal filters are placed within the connectors on
the computer side.
The demarcation line of tempest responsibility between
CR and SHAPE is the IDF.
The approach sketched by ARC was acceptable to SHAPE.
3.8 C̲o̲n̲n̲e̲c̲t̲i̲o̲n̲ ̲B̲e̲t̲w̲e̲e̲n̲ ̲T̲w̲o̲ ̲R̲a̲c̲k̲ ̲B̲a̲y̲s̲
The connection between the two rack assemblies must
be done in such a way that it is a complete part of
the rack. If the rack approach is approved this would
mean an independent duct between the two rack assemblies
used exclusively by the interrack cabling. ACE COMSEC
will forward details on ducting available. This ducting
SHAPE will form part of the Civil Work Package, but it is
ACE
action COMSEC's recommendation that the ducting should be
purchased centrally for standardization and security
purposes.
3.9 P̲o̲w̲e̲r̲ ̲S̲u̲p̲p̲l̲i̲e̲s̲
SHAPE confirmed that the power input lines will be
filtered as described in SRS, section 3.4.5.1 p).
For the rack-solution filtering on the power-lines
for the CAMPS room can be achieved in 2 ways to meet
the TEMPEST requirements:
a) SHAPE supplies separate filters for the rack-assemblies
and separate filters for terminals and lighting.
b) SHAPE supplies a common filter for all equipment
in the room. In this case sufficient filtering
to prevent the power lines from radiating shall
be included in the racks.
4 T̲E̲S̲T̲I̲N̲G̲
Three solutions to testing were proposed.
a) For the cage solution, the needed test certificate
will be issued by the cage manufacture.
In the rack solution you need some tangible assurance
that the racks are highly effective. First, a review
of the completeness and suitability of the rack
specifications, second test data showing the shielding
effectiveness of a sample rack (3 bay), third a
suitable QA plan.
Then ACE COMSEC will measure and judge that the
CR factory location chosen for TEMPEST test is
suitable regarding ambient noise. If it is found
not suitable, then solution b or c will be employed.
CR and ACE COMSEC, as part of the overall test
plan, will define installation, method of exercising,
exact configuration, and logistics for the factory
TEMPEST test.
ACE COMSEC will perform TEMPEST tests within the
framework of the test plan. If emanations are found
and judged unacceptable, ACE COMSEC will request
a second opinion from a NATO COMSEC organization.
If the judgements stands, CR will be required to
take corrective action.
ACE COMSEC approval will be given as a result of
the successful completion of factory TEMPEST test.
ACE COMSEC feels that no formal certificate is
required for the mainframe equipment since that
constitutes the major part of the system.
b) For the event that the environmental ambient is
unacceptable in approach a) as defined by ACE COMSEC,
CR is obliged to either recommend a different test
site acceptable to ACE COMSEC or to take steps
to reduce the environmental ambient of the factory
site. If the test site is moved, CR will install
the CAMPS in the approved location.
c) In the event that certification based on laboratory
test is required, CR will install the equipment
in an approved TEMPEST laboratory.
The testing laboratory (like ARC) will prepare
and submit a TEMPEST test plan for approval along
with facility certification data.
The testing laboratory will perform laboratory
tests in accordance with AMSG 720, and prepare
a report with complete data documenting the testing.
The document will be submitted to DANISH COMSEC
for certification, or if DANISH COMSEC is unwilling,
direct to ACE COMSEC for certificate.
The test results and certificate will be supplied
to ACE COMSEC to fullfil contract requirements.
It is proposed that the laboratory TEMPEST testing
will replace the factory TEMPEST testing. ACE COMSEC
will be invited to witness tests or do their own
TEMPEST tests during the laboratory tests.
A P P E N D I X…01……01…A
