Succeeding in 21st Century Network Competitions

John Stillion and Bryan Clark Center for Strategic and Budgetary Assessments

1 • Battle Network (BN) definition:

– A combination of distributed target acquisition sensors (finders and damage assessors), command and control (deciders), (shooters), and the electronic communications linking them together. • Essential BN attributes:

– Enable shooters to engage targets they cannot “see” far more effectively than would otherwise be possible

– Enable finders to achieve much higher levels of effectiveness as a group than they possess organically

– Enable deciders to coordinate and prioritize tactical engagements at a much higher level of efficiency to achieve the desired operational effects

– Enable those assessing the results of these operations (damage assessors) to determine their relative success with far greater accuracy than would otherwise be possible • BNs first emerged about 100 years ago but were relatively rare until recently due in part to the high cost of transmitting and processing information

– This limited the number of BNs and the instances of BN competition • Declining cost and increasing power of information transmission and processing systems will likely spur BN proliferation, and with it BN competition

2 • Network attributes depend heavily on operational metrics

• Tempo of operations influences decision to exploit or disrupt opposing network

• “Virtual Attrition” is often more cost-effective than platform destruction

• Competitions accelerate and culminate, then jump to new mode

• In some cases one side or the other is “saved by the bell” when a conflict ends just before a competition jumps to a new mode

3 • Submarines vs. ASW – Examine competition with focus on BMC2, multi- domain elements, success of networked vs. autonomous attackers

• IADS vs. Air Attack – Explore how “cutting-edge” technologies advantage first one side, then the other in Battle Network Competitions

4 • Submarines and commercial/escort ships are relatively slow – Dependent on cueing to get in position – Difficult to evade attack – Results in “slow-motion” operations (better to exploit comms)

• Submarines have limited situational awareness – EM sensor range <10 nm; sonar ranges <30 nm normally – Difficult to determine if incoming will be successful

• Submarines lack self defense – Some decoys, but little else to defeat torpedoes and depth charges

Fundamentals of ASW-sub competition remain largely unchanged.

5 1.00 250000

0.90 Sub speed disadvantage (u/v) Mean Transoceanic Merchant Ship Size 0.80 200000

0.70

0.60 150000

0.50

0.40 100000

0.30 MeanMerchant Size Ship (tons) 0.20 50000

Maximumstealthy sub speed /Merchant ship speed 0.10

0.00 0

1946 1964 1990 2008 1940 1942 1944 1948 1950 1952 1954 1956 1958 1960 1962 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1992 1994 1996 1998 2000 2002 2004 2006 2010 2012

Subs maintain speed disadvantage & larger ships less vulnerable.

6 Sep ’39– U-boat bases Sep ’39– Mar ’40 Sep ’39 & after Mar ’41 Jun ’44 Apr ’41–Jun ’44 Jun ’44– Apr ’45 U-boat bases Jul ’40–Dec ’41 Jan ’42– Jul ’40–Jun ’44 Jan ’44 Transatlantic convoys 1939–1945

Jan ’42–Sep ’42 Apr ’41–May ’43

Coastal convoys mid-1942–1945

U-boat patrol areas

Jul ’43–Jun ’44

Long range & slow speed of adversaries shaped competition. 900000 120 Bombing of French U-boats to European U-boat bases coast and southern Atlantic 800000 Supply submarines start ops off Azores 100 700000 Overall shipping losses=shipping 600000 80 construction U boats present or lost permonth Start max submergen More escorts 500000 ce through for convoys Bay of U-boats deploy to Evasive routing Biscay coast of France 60 based on shore HFDF “Wolf Pack” tactics start 400000 U-boats evacuate begins German decryption of French bases U-boats shift to convoy orders H-K groups sink surfaced night attack S-band 300000 Convoy escorts Naval Enigma code broken supply submarines 40 due to threat Subs shift to fully Shipping losses (tons) losses Shipping proficient with U.S. coast U.S. coast fielded radar S-band GSR U-boats begin using convoys 200000 H-K carrier groups deployed X-band GSR deployed Norwegian bases Enigma HF/DF, radar on L-band GSR Invasion of code X-band radar 20 U-boats begin escort deployed “Saved by Norway & upgraded Allies crack new Enigma deployed 100000 using French bases low countries Escort CV Hedgehog code the Bell” & L-band radar a/c Continuous air escort New convoy codes Snorkel deployed 0 0 Sep-39 Jan-40 May-40 Sep-40 Jan-41 May-41 Sep-41 Jan-42 May-42 Sep-42 Jan-43 May-43 Sep-43 Jan-44 May-44 Sep-44 Jan-45 May-45

Shipping losses (tons) U-boats at sea in Atlantic U-boats sunk per month Key metric for allies: overall shipping losses. Overall shipping losses (X 1000 tons) 1200 120 Losses on U.S. East Coast and Carribean (X 1000 tons) Losses per month per U-boat (X 100 tons) Coastal vessels stay in inland U-boats at sea in Atlantic waterways, protected by minefields 1000 100 U-boats sunk off East Coast and in On U.S. entry into , Caribbean 6 aircraft and ~12 By 1 April, 170 planes and 94 escort ships available escorts of various types U boats present or lost permonth (but not equipped) on East Coast 800 80 During May radar-equipped aircraft begin deploying to East Coast U-boats shift back to central and eastern Atlantic Coastal convoys start on East coast Enigma code broken 600 60 Radar equipped aircraft and Shipping losses Shipping escorts start deploying

Coastal convoys start in Caribbean 400 40

Enigma code Effective weapons upgraded (hedgehog/) deployed 200 20

0 0 Jan-40 Apr-40 Jul-40 Oct-40 Jan-41 Apr-41 Jul-41 Oct-41 Jan-42 Apr-42 Jul-42 Oct-42 Jan-43 Apr-43 Jul-43 Oct-43 Jan-44 1942 spike around U.S.; stopped by convoys w/out sinking subs. 120 Bombing of French U- U-boats to European coast boat bases and southern Atlantic Supply submarines start ops off Azores More escorts 100 for convoys Evasive convoy “Wolf Pack” routing based on tactics start

shore HFDF begins U U boats present or lost permonth German 80 U-boats shift to decryption of Start max submergence surfaced night attack convoy orders U-boats deploy through Bay of Biscay due to sonar threat to coast of France Convoy escorts U-boats evacuate 60

boat per monthper boat proficient with

- French bases radar H-K groups sink S-band radar supply submarines Naval Enigma Hedgehog fully code broken Subs shift to 40 U.S. coast fielded S-band GSR U.S. coast U-boats begin using convoys deployed X-band GSR deployed Norwegian bases Enigma H-K carrier groups HF/DF, radar L-band GSR code X-band radar

Shipping losses per U losses Shipping U-boats begin on escort deployed Invasion of upgraded deployed Norway & low using French bases Escort CV Hedgehog 20 countries Leigh Light & L-band Allies crack new Snorkel deployed radar a/c Enigma code

Continuous air escort New convoy codes 0 Sep-39 Feb-40 Jul-40 Dec-40 May-41 Oct-41 Mar-42 Aug-42 Jan-43 Jun-43 Nov-43 Apr-44 Sep-44 Feb-45

Shipping losses per U-boat per month U-boats at sea U-boats sunk per month Key metric for Axis: U-boat productivity. North Atlantic convoy results

Convoys % Convoys Convoys Ships Ships Sunk % Ships Sunk Sighted Sighted

HX (9 kts) 23 923 8 12 35 1.3 SC (7 kts) 24 991 14 45 58 4.6 ON (9.5 kts) 24 897 11 29 36 3.2 ONS (7 kts) 23 836 11 31 48 3.7

Japanese convoy results 1400 25.0

1200 Number of convoys 20.0 1000 Percentage of M/V sunk

800 15.0

600 10.0

400 Numberofconvoys

5.0 Percentage ofships lost 200

0 0.0 1 2 3 4 5 6 7 8 9 10 11 12 13 Convoy size But convoys alone reduced submarine effectiveness. 11 Location of attacks Independent Coordinated U.S. attacks, Atlantic and Mediterranean; Jan 1943–Feb 1944 Number of incidents 176 18 Number assessed as sunk or probably sunk 9 3 Percent successful 5 17 U.S. attacks, Atlantic and Mediterranean; March 1944–May 1945 Number of incidents 41 38 Number assessed as sunk or probably sunk 5 21 Percent successful 12 55

Probability of regaining contact

Single Ship Coordinated

Jan 1943–July 1943 0.54 0.8

Aug 1943–Feb 1944 0.68 0.9

Engagement platform sensors not able to gain & maintain contact. Lethal # of charges Weapon Weapons effectiveness per radius (ft) / barrage

1st half ’43 2nd half ’43 1st half ’44 2nd half ’44 1st qtr ’45

Depth 21 9 5.4 4 6.4 5.1 7

Mousetrap 0 24 7.5 15.4 28.1 23

Squid 0 16 33.3 62

Depth Charge (1939) Hedgehog (1941) Mousetrap (1942) (1943) Smaller, aimed contact weapons more effective. 13 Sep ’39–Mar ’41 Jun ’44–Apr ’45

Brest

Lorient Jul ’40–Dec ’41 St. Nazaire

La Pallice

Bordeaux

Apr ’41–May ’43 “Virtual attrition” in Bay of Biscay much more important than kills. 50

U-boats return to 120

max submergence 45

U -

40 boat losses / 100

35

Bombing of French U- U-boats ordered to surface Percentage of deployed U 80 boat bases in day, submerge at night 30

duration (hours) duration U-boats start max New detector slows boats, submergence Covert detection increases Time to introduce RWR kills, but only modestly increases presence in Bay takes half as long as 25 60 increase presence in Bay and reduces Atlantic previous time Radar detector protects presence U-boats but reduces 20 productivity U-boats fight back with FLAK 40 - 15 boats

Enigma code broken Naxos S-band GSR boat presence / transit presence boat

- deployed, not effectively X-band radar in Bay U Hedgehog deployed deployed 10 20 Enigma code upgraded S-band radar H-K groups start Naxos S-band GSR L-band radar L-band GSR deployed 5 more effective deployed in Bay Leigh Light deployed deployed in Bay S-band radar deployment starts 0 0 Jan-42 Mar-42 May-42 Jul-42 Sep-42 Nov-42 Jan-43 Mar-43 May-43 Jul-43 Sep-43 Nov-43 Jan-44

Uu-boat losses per month Transit Duration (hours) U-boats at sea in Atlantic Percentage of deployed subs in Bay Metric was kills; should be “virtual attrition” of U-boat presence. Lifetime of Advancements in II ASW 35

30 Convoy order decryption

25 L-band ASW radar 20

Metox L-band GSR S-band radar Months 15

Upgraded Enigma code 10 Original Enigma code S-band ASW radar

5 X-band ASW radar L-band RWR S-band RWR Naxos S-band GSR Tunis X-band GSR 0 Sep-39 Mar-40 Oct-40 May-41 Nov-41 Jun-42 Dec-42 Jul-43 Jan-44 Aug-44 Mar-45

Counters take less time until competition “jumps” to a new band. 16 Diesel submarine detections by U.S. during 1967 Arab- Sensor Israeli War Visual 9 250 40 Radar

35 Broadband Sonar 200 161 Narrowband sonar 150 74 Other

100

50

0 54 214 15 36 Surface ship Aircraft Submarine Intelligence

Diesel subs in 1967 Arab-Israeli War tracked as in WWII. 17 Nuclear submarine detections by U.S. during 1967 Sensor Arab-Israeli War 12 Visual 6 6 Radar 10 Broadband Sonar 11 8 13 Narrowband sonar

Other 6

4

2

0 8 11 11 6 Surface ship Aircraft Submarine Intelligence

Nuclear subs tracked during 1967 war by passive sonar. 18 1970 1978

1978

SOSUS array 1958 coverage 1970 1970 1978

1978

SOSUS and submarine arrays enabled near-continuous track. 1978

1958 1978

1978 1970

SOSUS array coverage SOSUS array decreasing with quieter coverage threat subs

Passive sonar competition nears “culmination” in 1980s.

EM competition Passive sonar competition Active sonar competition?

e e e

Snorkeling Nuclear and German, Diesel subs Soviet, noisy Soviet quiet Rest-of-World (ROW), non- non-nuclear U-boat (including Soviet) Nuclear subs nuclear subs nuclear submarines (some AIP) subs ~1975 1944 1980s Major U.S. acoustic U.S./UK forces Significant USSR 1960–70s advantage over ROW subs noisy, win Battle of quieting effort More Soviet Soviet SS/G/BN continuing Atlantic following Walker nuclear subs 2003 passive regime spy revelations 1958 deployed Fielding of VDS, LFA, Task Force Alfa multi-statics raises U.S. ASW performance “Saved by the Bell”

1940 1950 1960 1970 1980 1990 2000 2010 “Saved by 1980s the Bell” As Soviet SSNs became harder to 1962 track, primary SSKs detected in effort became Cuban Missile holding SSBN at risk Crisis, but only with in bastions enormous effort ~1995 1942 ~1955 Limited ASW Disastrous U- Large Soviet SSK capability boat campaign force; U.S. reliant against quieter off U.S. East on snorkel for Soviet sub force Coast detection

U.S. was lucky in last two shifts; how to prepare for next one? 21 • Competition will shift away from detection of noise from sub – Culminated in 1990s, but U.S. got a reprieve – Quieting techniques known and achievable by key competitors

• Competition will shift toward: – Active sonar: Low-frequency (100–300 hz) ranges in 10s–100s of miles – Non-acoustic phenomena: Wakes, chemicals, radiation, etc. – Background noise

• Shift enabled by improved processing and modeling – Increasingly can fit onto mobile platforms, co-locating sensor and shooter

• Effective competitors will exploit fundamentals, “virtual attrition” – Active sonar impacts sub operation and behavior, even if inaccurate – Smaller, “mission-kill” weapons can be longer range, effective vs. subs

22 • Submarines vs. ASW – Examine competition with focus on BMC2, multi- domain elements, success of networked vs. autonomous attackers

• IADS vs. Air Attack – Explore how “cutting-edge” technologies advantage first one side, then the other in Battle Network Competitions

23 • Speed

• Access

• Fragility

• Combination makes air • Relatively low payload forces extremely sensitive to attrition Combat Sorties per Month US AAF vs. Germany March 1944 26,411 Rolling Thunder (1965–86) 9,468 • Relatively low combat persistence IAF 1973 18,131 Desert Storm 51,840 Allied Force 10,231 24 Analog computing

1000 ft radius USAAF “Target Area”

1200 ft CEP still sufficient to attack major plants

Early losses drove both the and RAF to adopt night bombing. 25 Electronic Competition Phase Stream Program Start GEE Testing GEE Jammed 100% 400,000

90% H2S: 3 GHz ground H2S Program Start H2S Testing mapping radar Naxos-Z detector fielded 350,000 Program Start Oboe Test Oboe Mk I 200 MHz Karl Jammer Oboe Mk II 3 GHz 80% Serate Mandrel Counter Freya, etc. Counter VHF Communication Jammer 300,000 Khammhuber 100 Group Jammers and Night Fighters 70% Line IOC Airborne Cigar, Corona Windo Test Policy Delay w Conops Degrade

Battle of 250,000 60% /Zame Sau Average Chance of Surviving 30 Missions 51%

50% Freya: 120–130 MHZ pulsed system. ~ 75 mile range "Mandrel" jamming EW modified to operate 107–158 MHZ to reduce jamming effectiveness 200,000

Kleine Heidelberg: Bi-Static radar leveraging Signal. Range ~180 miles 40% Windo 150,000 Wurzburg Test Wurzburg w Dopler and Resonance anti-Window Mods

30% Lichtenstein: 490 MHz pulsed system. ~3 mile range "Jammed" by Window SN-2 85 MHZ Jammed 100,000 Naxos-Z H2S detector 20% “” Running Commentary Jammed Initial Day Introduced 50,000 10% Operations Khammhuber 100 Group vs. Germany Line FOC Dedicated Comm. Jamming Aircraft

0% 0 1939 1940 1941 1942 1943 1944 1945 26 BC Probability of Completing Tour Fraction of Bombs w/I 1,000 ft BC Cum. Sorties Bomber Command operations through early 1942 relied on crews independently finding targets with aircraft widely Typical RAFseparated Bomber inCommand space and Bombing time Accuracy Pre-1942 • Navigation and bombing highly inaccurate • German defenses develop slowly but grow increasingly effective

Typical RAF Bomber Command Bombing Accuracy Pre-1942

RAF bombing pre-1942 largely ineffective. 27 March 1942 British field GEE navigation aid after 21-month development period • Accurate to about 5 nm at maximum range of 250 sm • Enabled concentration of attackers in time and space to overwhelm linear Himmlebett defenses • Germans identified its use and purpose on May 29 (82 days) • “Henrich” jammers developed and fielded by August 4, 1942 (67 days) • By December 1, 1942 GEE useless over Germany (267 day operational life)

Maximum GEE Range Maximum GEE Range

28 H2S required 15 months from program start to first combat use on January 30, 1943 • Provided navigation and bombing accuracy independent of range from UK • About twice as accurate as GEE • Germans reconstructed set from aircraft downed on February 2, 1943 (second combat use—3 days!) • Countermeasures team formed February 22, 1943 (24 days) • Naxos airborne homing device fitted to German night fighters from September 11, 1943 (222 Days)

Maximum GEE Range

H2S S-Band Navigation Radar H2S Scope Presentation of Zuider Zee 29 July 1943 RAF introduces “Window” in “Battle of Hamburg” German Low-Tech Response Restores Defense • 746 attacking bombers dispensed one 2-pound packet of Window Effectiveness Rapidly every minute • Average time of fall was 15 minutes • Produced about 11,000 “bomber-like” returns • German radars “clustered” near 500 MHz • Foiled GCI, gunlaying and AI radars (all used similar frequencies) • Reduced effectiveness of defenses by about 75 percent

July 24, 1943 Raid on Hamburg “Zahme Sau” (Tame Boar) tactics: Ground controllers provided general running commentary on the course, speed, height, and location of the bomber stream rather than single bombers.

“Wilde Sau” tactics: Single seat day fighters converged on target city and engaged bombers visually in the target area under a ‘freelance” system

Window defeats Himmelbett system, but Germans rapidly adopt new tactics. 30 ?

One of four 2.5 kW/650lb. Jostle VHF communications jammers in bomb bay of 100 Group B-17 100 Group B-17 Fortress III ECM Aircraft Fielding 250+ dedicated ECM aircraft defeats new German tactics by denying communications. 31 Brief bombing halt Intense attacks on Hanoi Bombing halted above 20o North

100%40 400,000 SA-2 Battalions in

SAM Hunter - Killer North Vietnam 90% ECM Pod Formation 35 350,000

AGM-45 Shrike ARM 80% 35AGM5 -78 Standard ARM 30 Dev. F-100F Wild Weasel (APR-25/26 RHWR) 300,000 70% Dev. F-105F Wild Weasel 30 25 250,000 60% 25

EB-66B Standoff Jamming vs. Fire Can and Fan Song EB-66B Shift to Jamming EW/GCI 50%20 QRC-160-1 Development QRC-160-1 ECM Pod In- Testing QRC-160-1 Repeater Jammer - Denied Range to Fan Song and Fire Can 200,000 SA-2 Command Uplink First Detected by Firebee Drone - QRC-160-8 Dev. Starts QRC-160-8 Jams SA-2 Missile Beacon

40% 18 15 Flak Trap 150,000 15 Use MiGs to Break-Up Pod Formations 30% SA-2 Manual Track Tactics SA-2 Track on Jam Tactics 10 QRC-160-8 Recovered - New SA-2 Missile Beacon Dev. Starts 100,000

20% MiG-21 Introduced

5 Fan Song EMCON "Dummy Load" Modification 50,000 10% 5 Flat Face L/L Acq. Radar Added to SA-2 Regiment

EW/GCI and SAM radars integrated 0%0 0 19651965 SepSep NovNov 19661966 MarMar MayMay JulJul SepSep NovNov 19671967 MarMar MayMay JulJul SepSep NovNov 19681968 MarMar MayMay JulJul SepSep Nov

Probability of Completing Tour Strike Sortie FractionSA-2 Bat. Total Strike Sorties Flown vs. N. Vietnam "Virtual Attrition" Mid-1965 Early 1968

SA-2 drives U.S. to introduce extensive dedicated ECM support. 33 SA-2 Site Target SA-2 Site

“Iron Hand” “Iron Hand” SEAD SEAD

Strike Force

EB-66 Standoff Jammer “MiG CAP” “MiG CAP” EC-121 AEW Fighter Escorts Fighter Escorts By 1968 support aircraft often outnumber strike aircraft resulting in “virtual attrition.” 34 WWII Night Bombing Competition • Time from introduction of an innovative system to fielding of a countermeasure generally shorter than development time of the innovation • Cycle time decreases as conflict duration increases Rolling Thunder Competition • Systems working in new ways (H2S, SA-2) take longer to counter • Anticipated measures (EB- 66 Jamming) take less time

35 • U.S. experience in Vietnam and IAF experience in October War showed: – Even with massive support packages and advanced ECM attrition is still a concern – Introduction of new systems (e.g. SA-2) can dramatically change combat outcomes in short period of time – Big ‘last mover’ advantage in ECM/ECCM competition • U.S. begins serious search for a “new approach” to aircraft survivability – Eventually invests in stealth to change the nature of the competition from active vs. active to active vs. passive in the EM realm • What is next? – Increasing reliance on active ECM as counter-LO EM sensors proliferate – Increasing centrality of IR sensors and weapon seekers as LO aircraft proliferate on both sides • Calls into question utility of supercruise and afterburners Have Blue Combat Aircraft (DARPA Initiative 1974) Design Transformed ?

36 • Aircraft get shot down over enemy territory providing adversary physical access to innovative systems—this often aids in fielding countermeasures – Corollary: It is better to wait until innovative systems can be fielded “en mass” to achieve significant results than to introduce them piecemeal and risk early compromise

• IADS goal is to minimize damage to important assets, not (necessarily) to shoot down aircraft

• Great sensitivity to “actual attrition” makes air attackers very susceptible to “virtual attrition” – Rising support : strike sortie ratios likely indication that a “jump” is required

37 • Network attributes depend heavily on operational metrics – Avoiding subs or night fighters is “good enough”—don’t have to kill them all

• Tempo of operations influences decision to exploit or disrupt opposing network – Short duration of air operations makes disruption more attractive – Slower pace of makes exploitation more attractive • “Virtual Attrition” is often more important and cost-effective than platform destruction – Forcing opponent to operate less effectively or efficiently

• Competitions accelerate and culminate, then jump to new mode – Competitions “jump” when one side fields systems that defeat the opposing network (usually wide area and/or localization sensors) at the physics level – Signs a competition may be nearing culmination include: • Increasing support requirements to enable operations (Rolling Thunder) • Useful life of innovations measured in weeks • In peacetime, inability to significantly improve KPPs at affordable cost and/or in a reasonable time

• In some cases one side or the other is “saved by the bell” when a conflict ends just before a competition jumps to a new mode – Identifying and preparing for new basis of competition important for future success 38 39