Networks take center stage in ship systems

Navy experts rethink long-term plans for shipboard networking by embracing fast Ethernet and gigabit Ethernet over ATM/Sonet and FDDI to move naval surface and subsurface forces toward the new era of network-centric warfare and the digital battlefield.

Mar 1st, 2002
Th 89418

By Edward J. Walsh

Navy experts rethink long-term plans for shipboard networking by embracing fast Ethernet and gigabit Ethernet over ATM/Sonet and FDDI to move naval surface and subsurface forces toward the new era of network-centric warfare and the digital battlefield.


The U.S. Navy's Mk 41 Vertical Launch System, waiting for shipboard installation, is a networked group of missile launchers consisting of one eight-cell module that holds eight missiles — one in each cell.
Click here to enlarge image

U.S. Navy warships are set for a substantial influx of commercially developed data networking technology on new vessels as well as on shipboard systems upgrades. U.S. Navy officers who manage battle-management and command-and-control systems aboard surface warships say they expect to quicken the pace of using commercial network architectures and interface technologies this year to control tactical and non-tactical systems. They are working towards building distributed "total-ship" engineering architectures for future ships.

Navy shipbuilders and systems engineers have focused on the use of networks that use commercial routers, switches, and interface software to integrate combat, command-and-control, damage-control, navigation, and other systems. Shipboard systems integrators have aimed advanced distributed networks at high-profile programs like the now-terminated DD-21 land-attack destroyer, the San Antonio-class (LPD-17) amphibious transport dock ship, and the last of the Nimitz class aircraft carriers, the so-called CVN-77.

The Navy also is installing new commercial off-the-shelf (COTS) network technology aboard active-fleet ships to link ship combat systems to weapons, sensors, and non-real-time command, control, communications, computers, and intelligence (C4I) systems. Ship systems designers say commercial networks can multiply response time and eliminate costly and obsolescent point-to-point Navy-unique hardware and software.

AN/UYQ-70 centerpiece
Supporting many shipboard tactical and non-tactical networks are variants of the Navy's new UYQ-70 display-processor family from an industry team led by Lockheed Martin Naval Electronics & Sensor Systems-Eagan in Eagan, Minn. Also on the Q-70 team are DRS Technologies of Parsippany, N.Y., and Raytheon Naval & Maritime Integrated Systems in Portsmouth, R.I.

Navy leaders fielded the original Unix-based Q-70, which incorporated Hewlett Packard 743 processors, to replace Navy-standard OJ-194 and OJ-451 displays built for years for ship use by Hughes (now Raytheon). Lockheed Martin subsequently upgraded the system to the VME-based HP 744 processor, and plan to move it to a commercially developed commercial VME processor architecture.

The Aegis combat system aboard the U.S. Navy's Ticonderoga-class cruisers and Arleigh Burke-class destroyers is the biggest Q-70 customer. The first Q-70 application supported Aegis combat system baseline 6 phase 1, which is aboard two Aegis cruisers and is set for six Burke-class destroyers.

The Q-70 has been extended to the Aegis weapons control system (WCS) and command & decision (C&D) system. Currently, the real-time mission-critical SPY-1 Aegis phased array radar remains tied to a dedicated special-purpose processor.

Hewlett Packard officials plan to abandon their VME product line. As a result, experts in the Navy's Detection, Navigation, and Processing program office (PMS-440) in the Navy's program executive office for expeditionary warfare, which manages the UYQ-70 program, plan to upgrade the Q-70 system to a commercial architecture. Lockheed Martin officials recently made their final lifetime buy of HP VME boards.

The new Q-70 architecture uses VME single-board computers from Themis Computer of Fremont, Calif., which are based on the Sun UltraSPARC microprocessor. The boards run on the Solaris operating system.

The consolidation of tactical and non-tactical networks remains a central goal for ship programs. To demonstrate the performance of the Q-70 for command and control applications, the San Diego-based Space and Naval Warfare Systems Command (SPAWAR) in late 2000 installed two Q-70 "mission essential variant" (MEV) server racks aboard the command ship USS Coronado (AGF 11). Both MEVs are linked to a shipwide ultra-thin client network and flat-panel displays for monitoring operational and non-real-time ship systems data. One of the MEV racks supports a Sun E4500 processor and acts as a back-end network-control server. The second MEV ties in to two Unix and two Windows NT processors.

The demonstration, while aiming primarily at improved shipboard C4I dataflow aboard Navy vessels, also permitted a closer integration of processing for non-real time "network-centric warfare" operations, and real-time mission-critical operations, such as weapons engagement, Navy officials say. The Coronado retains the MEVs in a permanent configuration.

In a separate initiative that will lead to a second Coronado demonstration, officials of PMS-440 are collaborating with SPAWAR on using Q-70 technology for a long-term IT-21 block upgrade initiative that aims at using a common processing technology baseline to upgrade all SPAWAR-managed systems.

Among the SPAWAR programs to be affected is the automated digital network system (ADNS). This aims at providing "seamless" management of ship internal and external communications through intelligent management of network elements, such as routers and switches. NX Networks Inc. of Herndon, Va., originally provided the ADNS router infrastructure; SPAWAR information technology experts say they plan to continue the router engineering in-house.

Q-70 network management
Program managers say the ADNS will distribute shipboard communications among all the available wireless radio frequency links based on message type, urgency, bandwidth requirements, and system availability. This will eliminate the need for communications operators to perform those tasks. The ADNS should be as easy to use as the commercial phone system, such that operators will not know or care which system they are using, Navy officials say.


The networked array of missile launchers aboard U.S. Navy cruisers and destroyers ties into the latest shipboard computer systems.
Click here to enlarge image

The ADNS is being installed on surface combatants — particularly amphibious assault ships and aircraft carriers — through low-rate initial production and will be fielded eventually to every ship in the Navy, SPAWAR officials say. The system went through a successful operational evaluation in May 2000, and program managers say they expect approval for full-rate production early this year.

An increasing number of combat systems are adopting the UYQ-70 for mission-critical network management functions. Officials of the Lockheed Martin Marine Systems division in Baltimore have started delivering a baseline 6 Mk 41 configuration to international customers that introduces a Q-70-based dual-redundant new launch control unit (LCU), says Dan Heller, technical director for the Mk 41 vertical launch system (VLS) at Lockheed Martin.

Mk 41 VLS is a canister launching system designed to launch missiles quickly against hostile threats. Each missile launcher consists of one eight-cell module that holds eight missiles — one in each cell. The Mk 41 is fielded to all but five Ticonderoga-class cruisers and Burke-class Aegis destroyers (eight 8-cell modules forward and aft on the cruisers, and four 8-cell modules forward, eight 8-cell modules aft on the destroyers). The LCU acts as the control interface between the ship's weapons-control systems and the launch sequencers that are mounted to each launcher module.

For the baseline 6 configuration the new LCU is based on the HP 744 microprocessor and runs software written in the C++ programming language. This architecture replaces the obsolescent Navy-unique UYK-44 shipboard computer that runs proprietary code written in the old CMS-2 programming language. The guided missile destroyer USS Pinckney (DDG-91) will be the first U.S. Navy ship to get the baseline 6 VLS.

Fast Ethernet replaces NTDS
The new baseline also links the LCU to the launch sequencers with a fiber optic Fast Ethernet local area network, which replaces the old Navy Tactical Data System (NTDS) interface. Each of the eight sequencers links to one of the eight cells via a 145-pin cable, which provides the missile-cell support functions necessary for all Mk 41-compatible missiles.

The design, Heller says, enables Navy personnel to load any VLS-compatible missile into any cell without any modifications: the LCU interrogates each cell to identify the missile loadout and automatically configures it for launch.

One exception involves the new Evolved Sea Sparrow (ESSM) area air-defense missile from by Raytheon Missile Systems in Tucson, Ariz.; the ESSM launches from a Mk 25 "quadpack" canister that fits in one VLS cell. The Mk 25 requires a 1-megabit-per-second MIL-STD 1553 digital interface hookup and customized sub-routine commands to produce a new "multi-pack" setting that could be used for future loadouts of more than one weapon per cell.

A subsequent baseline 7 architecture, which will incorporate Intel Pentium-based PC-104 single-board computers for the launch sequencers, will be fielded starting with DDG-96, set for delivery in 2003.

In a similar missile-systems upgrade, officials of the NATO Sea Sparrow program office, who manage the ESSM, are pushing through an extensive re-architecture of the below-deck control electronics for the currently fielded RIM-7 NATO Sea Sparrow missile system, also built by Raytheon. The Mk 91 NSSM "re-arch" is aimed at integrating the system into the Navy's Mk 2 ship-self defense system (SSDS).

The re-arch will consolidate the firing-officer and radar-set display consoles in a single UYQ-70 console. This electronics re-design is now installed aboard the aircraft carriers USS Enterprise, USS Nimitz, and yet-incomplete USS Ronald Reagan, as well as the big deck amphibious assault ship USS Wasp (LHD-1). It also will modify the NSSM signal data processor and eliminate the computer signal data converter and system evaluation and trainer by transferring their functions to SSDS microprocessors.

The re-arch also eliminates the analog point-to-point NSSM control architecture and dramatically increases processing throughput. Navy leaders say they plan to field the re-arch NSSM to four ships per year — two new-builds and two backfits.

Introduction of AmpNet
The key new network innovation of the effort is the introduction of a high-speed advanced multi-processor network, called AmpNet. Belobox Networks Inc. of Irvine, Calif., a subcontractor to Raytheon for the re-arch NSSM program, developed the AmpNet with support from Raytheon and Johns Hopkins University's Applied Physics Lab in Baltimore.

Larry Wilbur, chief technology officer for Belobox, says the AmpNet is a fiber-optic fast Ethernet network that replaces the old NSSM Fiber Distributed Data Interface — better known as FDDI. The AmpNet, he says, runs TCP/IP protocols and supports middleware in VX Works, Unix, HPX, Linux, and Windows. Belobox engineers last fall ported the network into Windows 2000 and are configuring it for the SGI IRIX operating system.

The AmpNet is laid out in a star configuration that replaces the FDDI counter-rotating ring network for the VME-based NSSM. The AmpNet provides interfaces to 14 NSSM nodes: two missile launchers, four trackers, and eight Q-70 consoles. The chassis of each node contains one VME printed circuit board with links to two or more fiber channels. Wilbur says his engineers can build the AmpNet with a dual-, triple, or quadruple-redundant fiber infrastructure.


The newest version of the Vertical Launch System links to launch sequencers over a fiber-optic Fast Ethernet local area network, which replaces the old Navy Tactical Data System (NTDS) interface.
Click here to enlarge image

The AmpNet uses a micro-packet format consisting of very short data messages, as short as a single word. The approach, Wilbur says, "makes the network work like a computer," to provide the deterministic time-critical weapon-sensor-processing linkage required for air-defense engagements. The network performs fault control using a "roster" methodology to monitor status. Within two or three milliseconds of detecting a fault the AmpNet shifts data transfer functions to the alternate fiber ring.

Belobox experts are under contract to use their AmpNet technology to develop telemetry for the Naval Air Warfare Center at Patuxent River Naval Air Station, Md. AmpNet networks are part of the Eurofighter flight-test center in Italy, and several large commercial steel mills in Germany, company officials say.

The AmpNet links to the full ship self-defense (SSDS) fiber-optic network through one of a LAN access units (LAUs) that serve as system interfaces between the fiber cable plant and the SSDS weapon and sensor suite.

The SSDS, managed by the Navy's program executive office for theater surface combatants, is developed as a COTS-based combat system for aircraft carriers and big-deck amphibious-assault ships. An early version, SSDS Mk 1, is fielded to Whidbey Island-class (LSD-41) dock landing ships. Raytheon N&MIS is the prime contractor and systems integrator for the SSDS hardware and software.

Navy program officials say that Mk 1 is "ship-centric," in that it supports only data exchange from ship sensors to weapons. The Mk 1 fiber-optic network links the rolling airframe missile, Phalanx terminal self-defense gatling gun, Mk 36 decoy launcher, an identification friend/foe (IFF) system, as well as the SPS-49 air-defense radar, SPS-67 surface-search radar, and SLQ-32 electronic warfare system.

Program officials say that much of the Mk 1 SSDS software is being "plagiarized" through code reuse to develop the second-generation Mk 2 SSDS, eliminating the need for developing an entirely new program.

Because of software-development problems and funding constraints, the Mk 2 SSDS program has been forced to field the system into three separate variants in order to line up with ship programs and meet ship-delivery schedules. A one-of-a-kind Mod 0 variant has been installed aboard the aircraft carrier USS Nimitz. Mod 1 will be fielded to aircraft carriers and Wasp-class amphibious assault ships, and Mod 2 to the new San Antonio-class (LPD 17) amphibious transport dock ships.

All three SSDS Mk 2 variants will incorporate command-and-control and detect-to-engage capabilities, providing combat-systems connectivity "beyond the lifelines" of individual SSDS ships. The system has been developed to operate with the cooperative engagement system (CEC) to participate in battle group and potentially also in joint-service integrated air-defense networks.

A baseline goal of the SSDS Mk 2 is to replace Navy-proprietary combat systems interfaces with COTS technology, and in the process eliminate the costly and obsolescent Navy-proprietary advanced combat direction system (ACDS), developed in the mid-1980s by Hughes (now Raytheon N&MIS).

While Raytheon engineers continue development and testing for the full-up COTS architecture, they produced Mod 0 to meet the Nimitz's overhaul timetable at Newport News Shipbuilding in Newport News, Va. The overhaul was completed last summer. The Mod 0 retains the ACDS to share combat systems functions with CEC and SSDS. The ACDS will perform target tracking, air control, and identification tasks. CEC and SSDS will manage weapon and sensor control.

Raytheon is scheduled to start factory qualification testing next November for the Mod 1 SSDS, which is slated for the carriers and the Wasp-class amphibious assault ships.

The Mod 1 architecture eliminates the ACDS and, in the Navy's current plan, incorporates CEC baseline 2.1 software to enable the Mod 1-equipped ships to participate in CEC air-defense networks. The Mod 1 will add a new set of six commercial-grade cabinets, called a multi-LAU cabinet (MLC), which will house Mod 1 LAUs. The LAUs accommodates new circuit cards that are necessary to support upgrades to SSDS weapons and sensors.

The Mk 2 Mod 1 configuration serves as a foundation for the Mk 2 Mod 2 SSDS, which also transitions the system from Ada used in the Mk 1 to C++ object-oriented software modules, which will run on UYQ-70 consoles. Mod 2 also introduces Motorola PowerPC processors.

Raytheon is developing interfaces that will link the SSDS variants with the non-tactical shipboard network architectures aboard the carriers and the Wasp and San Antonio classes. The carrier program office is backfitting aboard the newer flattops an interior communications and navigation network (ICAN) during overhauls at Northrop Grumman Newport News and managed by the shipyard.

The ICAN, developed jointly by the Naval Sea Systems Command's carrier program office and SPAWAR, links shipboard communications systems on a core network that also supports machinery and navigation control and interfaces with tactical systems. The Navy and Newport News have introduced for ICAN the use of blown-in fiber-optic cable provided by General Cable of Highland Heights, Ky., which acquired blown-fiber developer BICC plc, a U.K. company, in May 1999.

With the blown-in technique, pneumatic devices pump the fiber cable through cableways to cable-interface boxes. The company installed some 47 miles of blown fiber aboard the aircraft carrier USS Harry S. Truman (CVN-75) in March 1998 and has completed blown-fiber installs for five more carriers. General Cable also blew-in fiber cable for several big-deck amphibs, and the technique is being adopted for other surface combatants.

The San Antonio-class shipwide area network (SWAN), an ATM-Sonet mesh network running on a dual fiber plant, links ship damage control, engineering control, navigation and steering, and C4I. The SWAN will provide a critical combat-system interface the SSDS Mk 2 Mod 2.

Navy program officials say that the original vision for the San Antonio class was to use the SWAN to "seamlessly" integrate tactical and non-tactical systems. They retreated from that goal after the SWAN design work started, saying that because both systems are new developments it was considered prudent to leave combat systems on a dedicated network. A third network will support the battle force tactical trainer (BFTT), used for collaborative battle group tactics training.

Fast Ethernet
Navy priorities for ship networks, meanwhile, are moving beyond the ATM Sonet technology planned for the SWAN to gigabit Ethernet (GbE). In late 2000 the Navy's Space, Command and Control, and Information Warfare directorate in the Office of the Chief of Naval Operations (N6) in the Pentagon endorsed gigabit Ethernet, supported by a mesh-network topology, as the Navy's standard shipboard network architecture. The decision reversed a 1997 plan that endorsed ATM Sonet as a replacement for standard Ethernet and FDDI networks. ATM networks have been installed on aircraft carriers, amphibious ships, Ticonderoga-class Aegis cruisers, and Los Angeles-class attack submarines.

Boeing Marine Systems in Anaheim, Calif., developed the FDDI copper-based data multiplex system called DMS, installed aboard the first 27 Arleigh Burke-class Aegis destroyers, as well as an improved fiber-optic version (FODMS) for DDG-79 and forward. Two years ago Boeing Marine officials announced they would seek to switch the system to ATM. The Navy provided no money for that effort.

OPNAV officials said that ATM initially provided dramatic improvements in speed and quality of service over Ethernet, but proved complex and difficult to maintain.

OPNAV N6 officials announced that 10-megabit-per-second standard Ethernet would be replaced by new 10/100-megabit-per-second fast Ethernet as the standard design for the IT-21 network, now called the integrated shipboard network system (ISNS).


Vertical launch Systems of the future will carry a wide variety of missiles, which will identify themselves automatically over shipboard networks to battle-management computers.
Click here to enlarge image

The decision, Navy officials say, was based on comparative studies of ATM, fast Ethernet, gigabit Ethernet, and other network architectures in designs developed by MIT, Stanford University, and the University of Illinois/Urbanna. Cisco Systems, Telecordia, Lucent, Sun Microsystems, AT&T, MCI WorldCom, Hewlett Packard, and other companies helped with the analysis, as did the Army's Signal Command and SPAWAR Systems Center supported the analysis.

The timing of the decision to move to fast Ethernet may benefit the high-profile cruiser conversion program for Ticonderoga-class cruisers, which currently includes plans to install the ISNS as an element of the ship's "core conversion" that will go aboard the entire class. The conversion program, now slipped from 2004 to 2006, also is expected to incorporate the tactical Tomahawk weapons control system, land-attack missile fire-control system, and a Naval Fire-Control System (NFCS). All three systems are integrated in a top-level system called TLN. The core conversion also is scheduled to add a COTS-based computing plant

The scope of the conversion program may be affected by the termination in December 2001 of the Navy's area theater ballistic missile program, planned for Aegis baseline 6 phase 3, and considered a major software development challenge. Edward Aldridge, the Under Secretary of Defense for Acquisition, Technology, and Logistics has directed the newly formed Missile Defense Agency to look at new approaches to developing a Navy terminal missile defense capability that, like the Area TBMD, would be hosted by Aegis.

The Aegis program office also is working to switch the Aegis combat system hardware and software architecture award from the current Navy-proprietary CMS-2 code that runs on obsolescent proprietary UYK-43 computers to an architecture of distributed processing elements, based on UYQ-70 technology.

A Navy integrated information networks integrated product team (NIIN IPT) continues to work on new shipboard network concepts. NAVSEA and SPAWAR officials established the NIIN IPT in 1996 to developed command standards and guidelines for ship networks. New working groups from the amphibious shipping, aircraft carrier, and surface combatant program office are working with the group on network design approaches.

In 1999 the NIIN IPT and the office of the chief information officer of the Navy produced an Integrated Technical Standard Guidance (ITSG) document in 1999 that provides functional requirements for ship networks.

The NIIN effort has no program budget. NIIN IPT officials point out that because networks are considered "infrastructure" that provide a broad capability for multiple ship programs rather than systems with an OPNAV sponsor and a programmed budget, the effort has been forced to seek money piecemeal from the shipbuilding program offices. Top Navy leaders lost interest in the program in the late 1990s.

In January 1999, the chief engineers for NAVSEA, SPAWAR, NAVAIR, and the Marine Corps Systems Command, as well as the Assistant Secretary of the Navy for Research, Development, and Acquisition signed a memorandum of agreement to re-start the program. Since then, the expeditionary warfare, mine warfare, and surface combatant program offices have set up working groups to begin meeting with the IPT.

More in Computers