Design of a Transoceanic Cable Protection System
Isaac Geisler, Dane Underwood, Kumar Karra, Felipe Cardenas
System Engineering & Operations Research, George Mason University
CONTEXT
Underwater cables’ main purpose is to
transmit valuable financial data and
government communications.
Transoceanic Bandwidth Projected Growth 20072020
faults, increase the rate of detection, and improve the mean notification time of damaged cables.
500
400
300
100
2007
2008
2009
2010
2011
Transatlantic
Pan-East Asian
North America - South America
2012
2013
2014
2015
2016
2017
2018
2019
Transpacific
South Asia & Middle East Intercontinental
Austrailia & New Zealand Intercontinental
Time
7
6
5
4
3
Current
Desired
2
1
0
2015
2020
2025
2030
Year
Underwater Surveillance and Threat Detections
PLATFORM DESIGN ALTERNATIVES SONAR DESIGN ALTERNATIVES
ACTIVE:
Autonomous Underwater Vehicles
Raytheon AQ/ANS-20A Minehunting
Sonar
Kongsberg REMUS 6000 AUV
Kongsberg HUGIN AUV
Klein System AUV 5000 V2
Remote Operated Vehicles (ROV)
ASI Falcon ROV
Oceaneering NEXXUS ROV
Oceaneering Millenium Plus ROV
PASSIVE:
Node Network
2020
70% of faults occur in water of less
than 200m deep.
Sources of 2,162 Faults, 1959-2006
Synthetic Aperture Sonar (SAS)
300m signal range; 3cm resolution;
6,000m depth rating
Compressed High Intensity Radar Pulse
(CHIRP)
High-resolution images; 300m signal
range
Side-scan and Multibeam
Lower resolution but greater coverage
area
Hydrophone
16km listening range; Uses localization
techniques; 3,500m depth rating
Each fault costs millions of dollars in
lost bandwidth revenue and repair
costs.
1200
1000
Current
Desired
Approximately 150 faults reported per
year. Over 20% are unknown causes.
When the cause is known, identifying a
responsible party is difficult.
200
0
Cable Damages
Activated Capacity in Tbps
$11.8 billion investment in new cables
from 2008-2014
31 New cable systems worth $4.8
billion will come online by 2017.
Marine Traffic Monitoring and
Warning
Automatic ID System (AIS)
Capabilities
Required on all ships over 299 tons
Tracks location, speed, ID
GPS updates every 10-180 seconds
Expected Mean Notification Time
Expected Cable Damages vs. Time
700
600
ALTERNATIVES & SIMULATION
NEED STATEMENT: There is a need to increase surveillance of cables in order to decrease the number of
Notification Time (Days)
800
NEED STATEMENT & OPSCON
962
Increasing fears of intentional cable
sabotage and espionage.
800
600
No central database or logging of
threats/faults exists.
460
400
315
155
200
149
Current cable protection technologies
(cable armor and burial) and
international organizations and treaties
(ICPC, ACMA, and NASCA) are not
impactful.
116
0
Fishing
Anchoring Component Natural
Failure
Causes
Other
Unknown
RESULTS
Mean per 10 years
Fishing
56.65
Anchoring
4.00
Mean per 10 years
Mean per Fault
Threats
66.65
-
Faults
4.11
-
Detection probabilities:
Based on platform, sonar, other parameters
Interaction of UISS Agent and threat type, location and depth
Still being implemented
Delay, Travel, Repair and Downtime Calculations
downtime = notifyDelay + travelTime + repairTime
Threat Type
Component
0.47
Totals (AS-IS)
Downtime (hrs)
1236.63
301
Prevention
Natural
4.68
Repair Cost
$9,971,023.31
$2,426,960.00
Espionage
0.66
CONCLUSIONS & FUTURE WORK
In Progress
Complete implementation of agents
Calculate detection probabilities of alternatives
Account for various movement patterns of AUVs/ROVs
Determine costs for alternatives
Model additional cable systems
Utility
Identification
Downtime
Lifespan
Sabotage
0.20
Lost Bandwidth Cost
$10,107,639.98
$2,460,212.67
Stakeholder
Prevention
Identification
Downtime
Lifespan
Private
0.40
0.29
0.23
0.18
Government
0.40
0.23
0.26
0.11
Cost-Utility Analysis
1
0.9
0.8
0.7
0.6
Utility
Lost Bandwidth and Repair Cost Calculations
capacity = 2.4 Tbps
10 Gbps rental rate = bandUnitCost = $25,000 (est)
shipCost = $12,000 per hour (est)
bandwidthCost = downtime * bandUnitCost * capacity
repairCost = (travelTime + repairTime) * shipCost
FA-1 Cable
COST-UTILITY ANALYSIS
0.5
0.4
0.3
0.2
0.1
0
0
5
10
15
20
25
Cost (in millions)
30
35
40
45
50
To be Implemented
Add dimensions to cable model to account for vertical and lateral
movement
Add movement of appropriate threats