Ten Ideas for Designing Next Generation Communications Network

07_Ten_Ideas_FA.FH11 24/4/08 3:03 pm Page 1

Composite

C M Y CM MY CY CMY K

Ten Ideas for Designing

Next Generation Communications

Network


Composite


ABSTRACT

The realisation of a robust, integrated and intelligent next

generation communications network will form a solid

foundation for the Singapore Armed Forces’ Integrated

Knowledge-based Command and Control. Many challenges

remain in our way – a harsh environment, immature science

and the limits imposed by the laws of physics. Innovative

strategies and cutting-edge technologies are needed to

overcome them.

This paper discusses the collective wisdom that emerged from

the Communication Architecture and Strategy Forum conducted

by the Future Systems Directorate in January 2007. Participants

of the forum included representatives from various groups

within the defence eco-system as well as external partners.

The top ten ideas that were voted by the participants are

presented here. Undoubtedly, some of these ideas are difficult

to realise but if we can do so, they will benefit us for a

long time.

Lee Kok Thong


Composite


80


Next Generation Communications Network

INTRODUCTION

The study of network-centric warfare (NCW)has its origins in the 1990s. Network CentricWarfare: Developing and LeveragingInformation Superiority, by Stein, Alberts andGarstka, is generally considered by many to bethe seminal work in this field. IntegratedKnowledge-based Command and Control(IKC2) is the Singapore Armed Forces’ (SAF)initiative in taking NCW to the next level – theknowledge edge (Lee, Jacqueline, 2003). IKC2is one of the key drivers of the Third GenerationSAF fighting force. It seeks to organiseknowledge and augment battlefield awarenessthrough networking of military systems.

Force is increasingly being realised throughinformation sharing over the communicationsnetwork. This philosophy places a premium oninformation superiority on the battlefield andis based on the ability to achieve an Internet-like secure capability in operational areas,providing ubiquitous network access to enable“anytime, anywhere” communications.

CHALLENGES INCOMMUNICATIONS

Building a networked force requires carefulconsideration of the types of assets that haveto be connected. The huge repertoire of entitiesincludes ground troops, armoured and non-armoured vehicles, naval platforms as well asairborne units. In addition, there are commandand control (C2) and intelligence, surveillance,and reconnaissance (ISR) assets that may befixed or mobile. Moreover, the tacticalenvironment is extremely harsh, in sharpcontrast to commercial environments.

From marching soldiers to supersonic tacticalaircraft, the huge extent of the mobility gapintroduces even more challenges (Tong, Swami,2003) in protocol designs. Undulating terrainand extreme weather conditions furtherobscure wireless communication. The result

can be intermittent connectivity with wideranging communication delays - frommilliseconds to days. The wide operationalscenario also presents significant RadioFrequency (RF) propagation and signalcharacteristics, resulting in poor channel (airor water medium) capacity.

In addition, information assurance presentsmore stringent considerations. The risk ofcapture and fear of compromised unattendedsensors further limit the usage of sensors. Thesecurity elements introduced to thecommunications system will constituteadditional costs to the system.

The introduction of new electronic systems(and consequently more portable powerneeded) will further add on to the load thatsoldiers are already carrying. For Special Forcesand long-deployed systems, the tactical domainwill require even more high energy-efficientsolutions. Unlike equipment in the commercialworld, military communications equipmenthave to survive rough handling, heat, dust,sand, salt water, and high humidity.Figure 1 illustrates some of the difficultiesthat the tactical space imposes oncommunications technology.

NETWORKING FORUMAND SYSTEMS ENGINEERING APPROACH

In a bid to solicit new paradigms towards thedesign of next generation wireless tacticalcommunications, the Future SystemsDirectorate (FSD) organised a technical andstrategy forum in early 2007. A wide spectrumof participants - ranging from military users,partners from the local defence ecosystem(DSO National Laboratories, DSTA and STEngineering) to research academics (I2R) andexternal participants (Cisco Systems and GeneralDynamics Canada) – was invited. This was aninclusive effort to bring all key stakeholderstogether to discuss and debate aboutnew concepts.


Composite


81

Many ideas were mooted and there was agenerous exchange of views during the forum.Many of these ideas raised remain hard torealise. They can be described as “FSD-hard”1

kind of problems. However, the payoffs arebig even if we are able to unlock just a fewamong them. Some of them might emerge asbreakthrough ideas that will go a long wayfor us.

The participants voted for the 10 mostpromising ideas at the end of the forum. Thefollowing is a discussion of these top 10takeaways.

TOP 10 IDEAS

Idea#1: Adapt Cross-LayeringAgainst Stove-pipe OpenSystems Interconnection Layers

The fundamental protocol building block forthe Internet is the Transmission ControlProtocol/Internet Protocol (TCP/IP). Originallyconceived for the wired Local Area Networks,TCP/IP was re-used when protocols for thewireless network were needed. Experimentshave shown that the transmission of video

over current wireless media (such as WIFI) isriddled with unsatisfactory experiences suchas video “freeze” and jerky images.

To start with, the TCP/IP protocol designers didnot take into account that the wirelesstransmission medium is impaired by negativeeffects such as fading and co-channelinterference. Moreover, the bandwidth of theradio-frequency transmission is limited andpower is usually too scarce to satisfy thedemands of video. The advancement ofcomputing and programming tools has madeapplications even richer. Today’s wirelessnetwork will need to carry different types oftraffic - voice, video and data, each having itsown characteristics.

Furthermore, we have to re-examine thefundamental way protocols were designed inthe Open Systems Interconnection (OSI) Model.The OSI layer consists of seven layers, as seenin Figure 2. Data going to and from thenetwork is passed from layer to layer. This way,each layer is written as a modular softwarecomponent. From the software design pointof view, it is efficient but as a system, theindependent design of different layers mayyield a grossly sub-optimal network.

Figure 1. Challenges faced in a wireless tactical environment (Burbank et al, 2006)


Composite


Inevitably, as a new design paradigm, cross-layering architecture will raise other issues:which and how many layers should beintegrated? How do we manage cross-layeroverheads and parameters exchange andconsequently the delay to end-to-endapplications?

In the course of FSD experiments, an innovativeand successful start-up company VSEE wasdiscovered. VSEE exploits cross-layeringtechniques to enhance video conferencingexperiences over the Internet. VSEE reserves adynamic and small portion of the networkbandwidth to do video conferencing; it doesso by constantly monitoring the amount ofinstance capacity available at the physicalchannel and adaptively adjusts the video streamusing its proprietary compression technique.VSEE was able to maintain a good video qualitydespite the low-bandwidth of no more thana few 100Kbps .

Idea#2: Tailored Solution toSpecific Ops Context

The concept of mobile ad hoc networks(MANET) is not alien to military users today.A MANET network consists of a collection ofmobile nodes communicating in a multi-hop

82

Figure 3. Cross layering optimisation

To break out of these limitations, the new goalfor communication designers is to analyse across-layer (Lin, Shroff, Srikant, 2006) that willyield jointly optimal physical, medium accessand routing layers for wireless networks.Cross-layering will allow for sharing and betterunderstanding of different layer performanceto make system-level decisions. Figure 3 depictshow cross-layering architecture can occur inthe TCP/IP architecture. Currently, cross-layeringarchitecture is a hotly researched topic whichhas already begun to show some promisingresults (Nahm, Helmy, 2005).

Figure 2. OSI and TCP/IP


Composite


83



new nodes to identify themselves, hencereducing the controlled traffic albeit withslower convergence. Many research effortshave been poured into deciding the “winner”.Nonetheless, a single optimal solution doesnot exist for all military contexts.

To build a practical, operational MANETsolution is to tailor specific solutions accordingto the relevant operational context.

Idea#3: Put Commercially-off-the-Shelf Strategy at the Core

Metcalfe's law states that the power (value)of a telecommunications network isproportional to the square of the number ofusers of the system (n2). The value grows non-linearly with the number of nodes in thesystem. Likewise for IKC2, the power of thecommunications network grows with theproliferation of tactical radios all the way tothe “last mile” i.e. the lowest fighting echleon.Traditional Very High Frequency (VHF) / UltraHigh Frequency (UHF) / High Frequency (HF)military radios have limited capacity.Consequently, the commander’s situationawareness is largely impaired. The developmentand advancement of tactical radios have beenslow, hampered by the lack of demand due to

Figure 4. Mobile ad hoc network

way without any fixed infrastructure such asaccess points or base stations. This is particularlyuseful in military applications or disasterrecovery operations.

Commercial interest in such networks has beengrowing steadily. However, conflicting needsin designing MANET parameters such asrouting, security, Quality of Service (QoS) andpower consumption often mean thatcompromises have to be made. As a result, anydirect adoption of commercial MANET solutionsis often met with unsatisfactory outcomesespecially in the military operationalenvironment. Experiments have furtherindicated that commercial MANET solutionsare still a long way from maturing intodeployable solutions that can scale and workwell under more stringent operational context.Customisation of Commercially-off-the-Shelf(COTS) MANET products to the specificoperational context in which they will bedeployed is therefore needed.

As an example, consider the choice of routingprotocols which are critical to MANET scalability.Two broad types exist: proactive or reactiveones. Proactive routing protocols activelybroadcast to achieve a converged network atthe expense of more controlled traffic whilereactive routing protocols passively wait for


Composite


84

the small market. Militarytactical radios remaine x p e n s i v e , m a k i n gproliferation to the lowestechelon difficult. SeeF i g u r e 5 f o r c o s tdevelopment trends. Incontrast, cost reductionfor commercial radio isoften fuelled by civiliann e e d s , w h i c h f a routweigh that of themilitary.

COTS Wireless LAN such as WIFI is several orderscheaper yet provides several orders morebandwidth than any of the digital tactical dataradios available. Even before WIFI was adapted,IEEE had already ratified a newer and betterstandard, WiMAX, in 2006, which offered upto seven times more bandwidth than WIFI perclient and other advanced features such assecurity and QoS. There are many morebreakthroughs coming from the commercialsector as the amount of research funds andeffort is tremendous. COTS communicationsolutions put more pressure on military vendors.We should find means to exploit theseexploding trends. They will assist us to realisetrue affordability and massive proliferation.“Put COTS at the core” represents a giganticstep in the changing of mindset towardsmilitary acquisition strategy.

Idea#4: Identify CorrectRequirements through IterativeProcess and Ops-Tech Integration

Besides playing its traditional roles, today’sSAF is also called upon to perform other taskssuch as humanitarian and relief efforts. Thisexacerbates the difficulty of military plannersin predicting the needs of future operations.The difficulty is often due to the unpredictablenature of conflict and the inadequacy ofinformation at one’s disposal.

The BOWMAN project is a case in point.BOWMAN is the UK Armed Forces’ multi-billionbattlefield digitalisation programme. The

programme, which commenced in the late1980s, was due for full system delivery in 2007.In the 2006-7 report, the UK Public AccountsCommittee (Page, 2007) reported thatBOWMAN suffered from several lapses andserious delays in delivering the system. One ofthe reasons cited was the lack of proper initialproject understanding between theprogramme’s technical side and the Ops users.

The SAF believes that technology is a force-multiplier. Investing correctly in coretechnologies will be critical to the SAF. DSTAis poised to identify and manage technology.This helps the SAF to better respond to the“gaps” caused sometimes by poorly definedor unexpected requirements. A tight and equalOps-Tech integration partnership, coupled withiterative rounds of open examination, istherefore key to identifying the rightrequirements for future communication needs.

Idea#5: Use Systems EngineeringSolution to Solve ComplexProblems

Building an IKC2 communications network tosupport the Third Generation SAF is a complextask. Interoperability with legacy systems andstriking a balance between the Services’divergent needs are just some teething issuesthat need to be overcome. For example, navyships move significantly slower than airplanesand are more widely spread out than armytroops. Employing a systems engineeringapproach is therefore akin to identifying the

Figure 5. Platform cost over time


Composite


85

Figure 6. Defence-In-Depthmulti-layer defence concept



right ingredients that will make up the futurecommunications solution.

Consider the problem of jamming to tacticalradios at the command post; one non-systemsapproach would be to consider more poweror sophisticated anti-jamming algorithms. Theoutcome is a more expensive radio withoutany real capability improvement. An expandedapproach is to include possibilities derived fromsystems analysis such as non-related domainsor even non-technical solutions. In this case, anon-RF solution such as using Free Space Optics(FSO) can be cheaply implemented for inter-command post connectivity. FSO is frequency-jam free and more secure due to its limitednarrow beam.

Another useful concept is defence-in-depth. Itis the practice of layering defence to providea robust, comprehensive, systems-levelprotection against attacks and intrusionattempts. Each defence strategy may not becomprehensive and is probably effective againstonly specific attacks (and therefore cheaper toimplement). Collectively, they are verypowerful. The defence-in-depth strategy slowsdown the attack while allowing the system torecover and perform counter measures. People,technologies and operations are three mainareas that the defence-in-depth strategy coulduse to provide a robust protection architecturefor any IKC2 network.

Idea#6: Overcome Trade-offs byWorking in Alternative Spaceand Adaptive Radio Software

“An ideal wireless communications systemwould provide high data rates at very longrange and yet use minimum power andbandwidth. Such a system would overcomeproblems and perform well despite thechallenges of channel weakness such as signalfading and interference. Adding on to thesefeatures, the cost of fabricating such acommunications system with transmitters andreceivers would be kept at a minimum andaffordable for mass proliferation.”

Figure 7. Trade-offs – “triangular” relationships


Composite


86

This is a description of a system that has yetto be built because multiple and mutuallyexclusive objectives such as low cost, small size,high reliability, and large capacity make itdifficult to do so. The systems architect wouldneed to study the trade-offs among differentcombinations of modulation, coding, multipleaccess, and antenna techniques to determinethe best design. Figure 7 illustrates some ofthese fundamental “triangular” trade-offrelationships - “spectral efficiency, cost andpower” or “Range, mobility and bandwidth”2.

How can we achieve breakthroughs despitethese constraints? The basic assumption on theusage of military radio is that most researchershave tried to solve the power-range trade-offon the ground. Radio propagation on theground does not travel far as it encountersobstacles like foliage and urban static features.The signal’s penetration is low and its quickattenuation with distance reduces thecommunication range. With the advancementof microcomputers and unmanned air vehicles(UAVs), the cost of providing an airbornecommunications service over a broad area canbe drastically reduced. It is akin to a mobileGSM base station mounted on a UAV. In fact,such a strategy completely transforms the“triangular” relationship.

The fabrication of traditional radios is fixedonce the radio engineer has selected a fixedsolution (triplet) on the “triangle”. Fixing theradio solution limits the context in which theradio can be used. Software-defined radios(SDR) and cognitive radio (CR) concepts seekto break out of this restrictive approach. TheSDR concept reduces radio cost by re-using thesame radio chassis for different waveformsadapted for different scenarios. The waveformsare digitally inserted into the radio and canbe re-loaded at any time. Cognitive radio isanother paradigm for wireless communicationin which wireless nodes adaptively changetheir transmission or reception parameters tocommunicate efficiently without interferingwith licensed or unlicensed users. CR adaptsand increases communication capacityaccording to the feedback received.

Idea#7: Use Spiral DevelopmentAgainst Fast Obsolescence

Newer variants of radio are coming out at afaster rate due to COTS advancement, notablyin electronics. Many military programmes canbecome hostages to the speed of advancementin commercial standards and demands.Moreover, the existing quantity of the radiosin the military inventory means that the militarycannot afford to replace them overnight. Sucha strategy presents a high risk in that a wrongbet would result in an instant write-off ofhundreds of millions of dollars. Moreover,massive re-hauling of one’s radio capability isoften a long and tedious process.

The global war on terror as well as OperationIraqi Freedom has prompted the USDepartment of Defense to expedite the deliveryof the latest weapons systems and protectiveequipment to the troops deployed at thefrontline3. Rather than developing, testing,and then fine-tuning systems before sendingthem to the field, the priority is to get newtechnologies to the troops as quickly as possible,while continuing to improve on them in thelaboratories. Such is the essence of the spiraldevelopment approach.

Spiral development offers mitigation againsttechnological obsolescence. It does so bycontinually allowing the insertion of technologyinto long-term projects. Coupled with thestrategy of COTS exploitation, this presents apowerful acquisition approach.

Idea#8: Standardise Only at theLowest Denomination

The navy, air force and army conduct theiroperations in vastly different tacticalenvironments. To fully equip the army, it wouldrequire a large number of communicationdevices whereas to fully equip the navy andthe air force, it would require significantly lesseffort. Fighter aircraft and UAVs are high-speed platforms whereas ships and convoysmove much slower on the ground. The infantry


Composite


87Figure 8. Delay Tolerant Networking, (Source: ATO, DARPA)



has to navigate through vegetation andundulating terrain whereas aircraft have tograpple with cloud and airspace coordination.Underwater communications is altogetheranother domain of expertise as acoustic wavestrump electromagnetic ones. In short, eachService has its own specific requirementsused in the design of their optimalcommunications radio.

However, the realisation of a networked forcemeans that some form of interconnectivity andinteroperability among the services is needed.This will likely require some form ofstandardisation. Yet standardisation has thepitfall of satisfying the common needs butfailing to deliver optimal solutions specific toeach individual Service. Where should we drawthe line?

As the Third Generation SAF grows, it will getmore connected and complicated in theprocess, while inter-Service traffic profile willgain more importance and the conventional“80-20” wisdom will apply less. A more suitableresponse would be to adopt the policy of“standardisation only as a last resort”. Morespecifically we ought to standardise designs atthe lowest common denominator applicable.

The Internet was designed on the principle ofstandardisation at the lowest commondenominator. The usage of TCP/IP providedbaseline stability. It is anticipated that open,v o l u n t a r y s t a n d a r d s a n d f l e x i b l ecommunication protocols, such as IP, will playa key role in fulfilling the goal of designingmultimedia-rich information exchangeapplications and services based onmechanisms and frameworks that are hiddenfrom the users.

Idea#9: Delay-TolerantNetworking Against ExtremeNetwork Conditions

The assumption made by Internet designersfor TCP/IP is that the end-to-end round triptime of the data is short; no more than a fewseconds. This is valid as devices within theInternet are generally connected via hardphysical wires, which makes transmission qualityquite reliable. Otherwise, the circuit will bedropped. Hence it would be catastrophic forNASA to design a communications protocolusing TCP/IP for communication between Earthbase stations and probing devices for Mars.


Composite


88

Harsh operational environments characterisedby non line of sight, high terrain and highmobility makes no guarantee of end-to-endconnectivity. This results in unreliable networksand high link error rates. Moreover,intermittent links are sometimes required tosave power. It is also caused by scheduledtransfers of unattended devices, so as to limitcongestion. A military communicationsnetwork consists of a number of independentnetworks, each supporting specialisedcommunication requirements. These networksdo not necessarily use IP and they are oftenmutually incompatible in that each is good atpassing data within the network but is notable to exchange data among one another.

Delay-Tolerant Networking (DTN) (Warthman,2003) or Opportunistic Routing is the study ofnew innovative techniques to helpcommunications engineers design betternetwork appliances to fight these challenges.DTN is an overlay on top of the heterogeneous“ reg iona l” ne tworks . I t suppor t sinteroperability of these “regional” networksby accommodating long delays betweenregional networks and within each networkvia the translation of regional networkcommunication characteristics. DTNs overcomethe problems associated with intermittentconnectivity, long delays, and high error ratesby using store-and-forward message switching,caching of in-transit data packets, messageferrying and connection state hibernation forsubsequent reactivation. Studies conducted byIntel and Berkeley (Demmer, 2004) haveindicated a significant improvement when DTNtechniques are employed. In some cases, DTNproduces throughput that is close to thetheoretical maximum throughput achievable.

Idea#10: Manage MobileBandwidth as a Scarce Resource:Policy-based Quality of Service

Bandwidth is a scarce resource that has to bemanaged carefully as more applications suchas video streaming, which make intensive useof bandwidth, become commonplace.

QoS refers to “control mechanisms that canprovide different priority to different users ordata flows, or guarantee a certain level ofperformance to a data flow in accordance withrequests from the application programme.”QoS guarantees are important especiallyif network capacity is limited, as seen inFigure 9.

The current generation of Internet is still “best-effort”. QoS will be particularly important inthe military context. Data in military networkscomes in various forms, ranging from chats,emails to command and control data andintelligence reports. Building separate networksfor each of these applications runs against thespirit of a networked force. Data differentiationor prioritisation will make our network moreresponsive and intelligent, especially insituations where data traverse non-homogeneous environments.

There are however considerable challenges inimplementing practical QoS due to the highand unpredictable mobility of troop forces andaerial platforms. Enabling QoS in tacticalcommunication environments requires someform of “network stability”. QoS degradesquickly as the mobility of end nodes increases.In turn, re-transmission of dropped data leadsto a worsening of the traffic congestion. Oneidea for ensuring QoS in ad hoc mobileenvironments is to build a core set of wirelessnodes that are relatively static - controlledmobility nodes - in order to provide a baselinesupport backbone. Implementing QoS policies

Figure 9. Where QoS matters


Composite


89



within this backbone is then achievable andallows the system to maintain a high standardof service.

Policy-based QoS refers to the middlewareservices that decision-makers such ascommanders can use to interface directly withtheir network resources, such that they canassign and control these resources whileunderstanding the implications of their actions.Policy-based QoS can be made on the fly andquickly disseminated to all system nodes. Thisensures that the underlying network resourcesdo not operate in a black box.

Routers located at the edge of the networkare the first recipients of data from the enduser devices. They simply accept and relay thedata to its designation. Even when the networkis congested, edge routers continue to injectdata indiscriminately into the system. This canresult in sudden network outage. Networkadmission conditions and traffic shaping tools(Bidgoli, 2007) are powerful techniques thatcan help to manage congestion. Data that donot fulfil criteria ought to be prevented fromentering. Policy-based schemes couldimplement the supportability of certaindata/video formats. The result will be anintelligent network capable of recovery fromcongestion.

CONCLUSION

We have presented the top 10 strategiesfor designing next generation tacticalcommunications networks, based on thecollective wisdom of diverse stakeholders.

We recognise that many hard challenges stillneed to be overcome before the ideas can berealised. Some of the suggested technologiesare still in their infancy while others are moreready to be adopted. These are very toughissues, and once we are able to overcome someof them, they will emerge as potentialdisruptive breakthroughs for us.

These strategies are by no means exhaustive.However, together they represent a formidableplan to move ahead as we seek to design andbuild a robust, intelligent and dynamic IKC2network for the Third Generation SAF.

07_Ten_Ideas_pg90.FH11 6/5/08 4:13 pm Page 1

Composite


90

END NOTES

1. Note: Borrowing a phase which currentDirector of US DARPA, Dr Tony Tether likes tosay, “DARPA-hard” questions, which meansproblems that are so hard to overcome thatonly DARPA will contemplate.

2. Henry Lee et al. Connectivity for NetworkCentric Warfare is still a Far-Fetch Dream?,Cutting Edge Communication ArchitectureForum, 2007. Employing a highly spectralefficient modulation technique such as 64-QAM means that for every bandwidth(measured in Hz) used, we are able to derivemore data (in terms of bits). However thisoften means that sacrifices have to be madeon cost efficiency (more expensive and lessstable electronics) as well as power efficiency(a higher signal-to-noise figure is thus neededand this translates to high receiver sensitivity,often harder to achieve.)

Note: In 2005 and 2006, the Aberdeen TestCenter's scientists, technicians and engineerstested about 30 rapid fielding initiatives aweek, with more than 1,400 tests conductedin 2006 alone. There has been an 87 percentincrease in range activity here since fiscal 2001.Source: http://www.technologynewsdaily.com/node/6336

REFERENCES

Bidgoli. (2007), The Handbook of ComputerNetworks, Network Traffic Management. Wiley.

Burbank, Chimento, et al. (2006). KeyChallenges of Military Tactical Networking andthe Elusive Promise of MANET Technology. TheJohns Hopkins University Applied PhysicsLaboratory, IEEE Communication Magazine.

Forrest Warthman. (2003). Delay TolerantNetworking: A Tutorial. Retrieved on Sep 2007from http: / /www.ipns ig .org/reports /DTN_Tutorial11.pdf

Lee, Jacqueline et al. (2003). RealisingIntegrated Knowledge-Based Command andControl, Transforming the SAF. PointerMonograph, Singapore Armed Forces.

Lin, Shroff and Srikant. (2006). A Tutorial onCross-Layer Optimization in Wireless Networks.IEEE Journal on Selected Areas inCommunications on ``Non-Linear Optimizationof Communication Systems'', vol. 24, Issue 8,[pp].

Michael Demmer, Eric Brewer, Kevin Fall,Sushant Jain, Melissa Ho, Rabin Patra. (2004).Implementing Delay Tolerant Networking. IntelResearch Berkeley.

Nahm, Helmy et al. (2005). TCP over Multihop802.11 Networks: Issues and PerformanceEnhancement, MobiHoc.

Page, Lewis. (2007). MPs: UK defence projectwas crap. Retrieved on 1 Oct 2007 fromhttp://www.theregister.co.uk/2007/03/08/bowman_cpac_pasting/.

Ripley, Tim (2007), Bowman misses the target.R e t r i e v e d o n 1 5 S e p 2 0 0 7 f r o mhttp: / /www.global-defence.com/2001/CSpart1.html.

Tong, Swami et al. (2003). Future Challengesof Signal Processing and Communications inWireless Networks. NSF/ONR/ARL Workshop.

07_Ten_Ideas_pg91.FH11 6/5/08 4:15 pm Page 1

Composite


Lee Kok Thong is Principal Engineer (Systems Architecting) and was AssistantDirector (Future Systems Directorate). In that capacity, Kok Thong helped theFuture Systems Architect formulate alternative concepts that were not in themainstream, conducted experiments for the validation of new concepts throughexploiting technology, performed the role of institutional red-teaming andoversaw the collaboration in experiments with external partners. He graduatedwith a Master of Science in Telecommunications from King’s College, Londonand Diplome D’ingenieur (French), Fondation EPF in 1998 under the OverseasGovernment PSC Scholarship. He obtained a Master of Science in DefenceTechnology and Systems from National University of Singapore and Masterof Science in Computer Science from Naval Postgraduate School in 2004. Hewon the Defence Technology Prize (Team) in 2002, as well as the DSTA TeamExcellent Award in 2002, 2003 and 2006.

BIOGRAPHY

91



Documents

Ten Ideas for Designing Next Generation Communications Network