-JDG_11015-4_SPIE_OPTICAL.pdf

Modular open systems architecture for optical networks in spacecraft r1

James Lyke, Christian Peters, Derek Buckley, and Zachary Bergstedt (Air Force

Research Laboratory)

SPIE 2019, 16 Apr, Baltimore, MD

This paper describes an architecture intended to support all-optical signaling and routing to and through spacecraft. The approach is based on a dense wavelength division multiplexing (DWDM) router, transceivers, and internal optoelectronic translators that support the interface of high-performance sensors and processing arrays. Key to this approach is the novelty of an optical middle box (OMB) scheme.

Each OMB contains a number of optical ports, electrical ports, and reconfigurable optical transmitter and receiver elements. Laser communications transceivers interface to the optical router and support dynamic reconfiguration. A basic interoperability protocol (BIP) defines a type of "optical dial tone" through which metadata is relayed to provide communications management cues to a control processor that manages the entire payload enclave (router and several OMB-based devices, with a spacecraft host interface). This paper describes details of the optical router, OMB, reconfigurable transceivers and supporting protocols.

Outline

•Motivation

•Cable replacement and flexible relay mission

•Spacecraft and payload

•Elements of optical network

Combat Cloud Information for Warfighting

•United States Air Force Combat Cloud Operating

Concept defines the Combat Cloud as “an overarching meshed network for data distribution and information sharing within a battlespace, where each authorized user, platform, or node transparently contributes and receives essential information and is able to utilize it across the full range of military operations”.

“Combat Cloud must exhibit critical attributes such as the ability to be self-forming, self-healing, gracefully degradable, and redundant. Under this construct, the ability to collect data and integrate it in an open, adaptive information system will significantly enhance C2 and operational agility for the US military and other US governmental agencies across the range of military operations.”

From The Combat Cloud Across the Range of Military Operations: Interagency Coordination By Shaun Williams and Jacob Hess

Broad opportunities for optical network

•Highest theoretic performance (energy/bit)

•Potential for long distance (e.g. interplanetary, 2x geostationary orbit)

•Bandwidth multiplication through dense wavelength division multiplexing (DWDM) – e.g. ~200 wavelengths in ITU grid (20Tbps @ 10gbps line rate)

DWDM in space

Cable Replacement consideration - 1

• Single cable (routing traffic from two ground points)

• Smaller constellation

• Disadvantage (site occlusion due to weather)

Cable Replacement consideration -2

• More dispersed satellite constellation and more ground sites

• Traffic model more random ground to ground site bulk data transport

• More agility, site diversity improves resiliency to weather, with one other concept…..

Cable Replacement consideration -3

• Supernodes – several sites on ground act as a larger virtual node

• Sites are ground connected with high-rate backhaul

• Ex. a->d is equivalent to b->d, c-f

Spacecraft concept

• Spacecraft = bus and payload

• We want a payload optimized for optical networks

• Orbit geometries – LEO preferred for time-of-flight reasons

Electrical Power Subsystem_

Command & Data Handling (C&DH) Computer

Health & Status

Sensors

Guidance, Navigation, & Control (GNC

Thermal Control

Timing, Tracking, & Telemetry (TT&C)

Processor

Payload Configuration for flexible data relay

Payload Global Network Access Terminal (GNAT)

Router

H o st

In te rf ac e

Laser transceiver

Laser transceiver

Laser transceiver

Laser transceiver

Laser tran sceiver

Laser tran sceiver

Inter-satellite crosslinks

Uplink/downlink transceivers

Vehicular Bus

Notional Concept for Constellation /Network of Optical “Space GNATs”

Processor

Data server in space? Extended payload concepts

Payload Global Network Access Terminal (GNAT)

Router

H st

In te rf ac e

Laser transceiver

Laser transceiver

Laser transceiver

Laser transceiver

Laser tran sceiver

Laser tran sceiver

Inter-satellite crosslinks

Uplink/downlink transceivers

Processor

Processor

Processor

Processor

Processor

Storage

Storage

Storage

Storage

Storage

Sensor

Optical Middle Box (OMB)

Optical Middle Box (OMB)

Optical plane

Timing Layer

Control/configuration plane

Power Distribution

Optical router

Optical plane

Timing Layer

Control/configuration plane

Power Distribution

Processor

Router – Vertical “sandwich” model

•Multi-port internal nexus for uniting all information management devices (transceivers, optional data servers and sensors)

•Features (data, control, timing, power distribution)

Router – key points

•Establish an all-optical pathway to preserve optical bundles in as close to native form as possible

•Support precision timing, data transport in/out of spacecraft through signal conversion: optical <-> electronic (OE,EO)

•Have ability to mix devices of “lesser” ability (e.g.

lower performance optical modems, radio-frequency transceivers)s

Control/configuration plane

Power Distribution

Egress

Ingress

EO/OE

conversion

Reconfigurable

Optical

Transceiver

Example Device and Connection to Router Reconfigurable hi-capacity laser transceiver

Control/configuration plane

Power Distribution

EO/OE

conversion

Traditional

Lasercom

Transceiver

Single channel laser transceiver

Control/configuration plane

Power Distribution

EO/OE

conversion, other inter-domain operations

Egress

Ingress

Processor

Processor

Processor

Processor

Processor

Data Center (processor farm)

Key Router Fabric Concepts for Optical Relay

•Optical bundles to (egress) and from (ingress) each major port connection (for devices supporting optical layer)

•Partition bundle into k subbands (e.g., k=4 up to

50 wavelengths per sub-band)

•Employ continuous optical paths using programmable switch fabric in optical domain

(electronically controlled through control place) a b c d e a b c d e

Ingress

Egress

Switch fabric for cross-connect

W

SS

W

SS

W

SS

W

SS

W

SS

W

SS

W

SS

W

SS

PC PC PC PC PC PC PC PC

P0.in

P0.out Ingress

Egress

P1.in

P2.in

P3.in

P4.in

P5.in

P6.in

P7.in

P1.out P2.out P3.out P4.out P5.out P6.out P7.out λA λB λD λC λA λB λD λC λA λB λD λC λA λB λD λC λA λB λD λC λA λB λD λC λA λB λD λC λA λB λD λC

(a) (b) (c) λA λB λC λD λA λB λC λD λA λB λC λD λA λB λC λD λA λB λC λD λA λB λC λD

Switch population strategies

•“Linear” vectorized: bundle i goes to bundle i only

•Ad hoc: allows some bundle sub-band group

•Fully connected: Brute force allows all possibilities at high switch cost

Introducing optical middle box (OMB) concept

•OMB is a device that manages groups of OE and EO conversions

•Flexibility analogous to the reconfigurable optical add drop multiplexers (ROADMs) used in long-haul telecom industry

•Roles:

– Sensor conversion EO

– Data Centers in space (fluid manipulation of many groups of simultaneous EO,OE transfers)

– “Wavelength conversion” (requires OE-EO back-to-back for each wavelength)

Wavelength selective switch array Passive optical coupler R ec ei ve r

O E co n ve rs io n

R

XO

E) optical elec.

R ec ei ve r O

E co n ve rs io n

R

XO

E) optical elec.

R ec ei ve r O

E co n ve rs io n

R

XO

E) optical elec.

R ec ei ve r O

E co n ve rs io n

R

XO

E) optical elec.

Optical switch λi,1, λi,2 λi,3, λe,1 λm

Tr an sc ei ve r

EO

co n ve rs io n

TX

EO

optical elec.

Tr an sc ei ve ve rs io n

TX

EO

optical elec.

Tr an sc ei ve ve rs io n

TX

EO

optical elec.

λe,2 λe,nλi,m-1, Internal (Optical Middle Box) Finite State Machine / Processor

C o n fi gu ra ti

M an ag em en t

P ro ce ss o r

Control Processor Connection

External Digital Interface (to optional processors and sensors)

OMB elements •Wavelength selective switch (WSS) for ingress

•Passive optical coupler for egress

•Shunt switch to pass through optical bundle

“remnant”

•Banks of reconfigurable receiver opto-electronic

(RXOE) modules (each “picks off” one wavelength)

•Banks of reconfigurable transmitter electro-optical

(TXEO) modules (each “synthesizes” one wavelength)

•High-performance datapath manipulation through finite state machine (e.g. field programmable gate array)

Reconfigurable Lasercom transceiver •Critical element of optical network infrastructure

•Performs capture and launch of optical bundles

•Sends / receives bundles to/from router without

OE/EO conversion (potential need for erbium-doped fiber amplifier (EDFA))

•Extracts one designated wavelength (λ0) as the

“service channel” which is promptly EO-converted to extract bundle meta-data for control plane

•Service channel (λ0) is likewise synthesized for output bundle to send control plane information to next relay point λ0

Optical splitter

Optical combiner

Optical path assembly

OEEO

Pointing, acquisition, and tracking subsystem (PATS)

Timing Extraction/Injection

Modem λ0

Optical Egress

Optical Ingress

Configuration Management Engine

Timing Layer Interface

Control Plane Interface

O p ti ca l L ay er In te rf ac e

Power interface

Power mgt PATS/optical assembly control

Ethernet backbone

EDFA

EDFA

(optional)

Control plane extraction

Reconfigurable Lasercom transceiver Challenges

• Pointing and tracking subsystem (PATS) – “stuffing a fiber” from thousands of kilometers

• Atmospheric issues for terrestrial facing links

• Compact EDFA and service channel processing

– Lasercomm system within a lasercomm system

– Opportunities of integrated / silicon photonics

Where are we?

• Under support of DoD’s Defense Optical Channel Program (DOC-P) we are doing some key exploratory work over next two years

– Modeling and simulation of constellation

– Analysis work to arrive a spacecraft basing approach (bus and allocations for payload)

– Optical network component and subsystem development in laboratory

– Possible radiation-testing of some critical components

Conclusions

• Describe an ambitious concept for DWDM in space

• Ambitious mission and ambitious technology portfolio required to address mission

• Not “rocket science” in that most of the key technologies are mature/emerging for terrestrial networks

• Translating a viable approach for space-based implementation remains critical challengt