AD-A202668
`
`
`
`
`
`EXHIBIT,
`
`
`{pg
`V
`GHARTMENT OF THE AIR FORCE
`AIR FORCE INSTITUTE OF TECHNOLOGY
`
`AIR UNWERSIXY
`
`ELECTE
`“.5: ‘1 9 JAN 3989
`
`
`
`Wright-Patterson Ah Force Base, Ohio
`
`
`
`. Page 1 Of155
`
`PETITIONER'S EXHIBIT 1017
`
`
`
`
`
`
`
`EXHIBIT,
`
`
`{pg
`V
`GHARTMENT OF THE AIR FORCE
`AIR FORCE INSTITUTE OF TECHNOLOGY
`
`AIR UNWERSIXY
`
`ELECTE
`“.5: ‘1 9 JAN 3989
`
`
`
`Wright-Patterson Ah Force Base, Ohio
`
`
`
`. Page 1 Of155
`
`PETITIONER'S EXHIBIT 1017
`
`
`
`AFIT/GE/ENG/B8D—ll
`
`A COMPUTER SIMULATION ANALYSIS OF CONVENTIONAL
`AS” TRUNKED LAND MOBILE RADIO SYSTEMS AT
`WRIGHT YATTERSON AIR FORCE BASE
`
`THESIS
`
`Thomas C Farrell
`Captain, USAF
`
`AFIT/GE/ENG/88D- 1 1
`
`
`
`Approved for public release; distribution unlimited
`
`Page 2 0f155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`U' “.A‘. akin». an uth,
`
`‘ :..»‘.‘-..L .15 "g.'b_"»'.~1a.
`
`. ‘.
`
`'
`
`AFIT/GE/ENG/SflDoll
`
`A COMPUTER smuwnon ANALYSIS OF Ccmnmmu AND TRUNKED
`
`LAND MOBILE RADIO’SYSTEMS AT QRIGHT PATTERSON AIR FORCE BASE
`
`THESIS
`
`Presented to the Faculty of the School of Engineering
`
`of the Air Force Instituté of Technology
`
`Air University
`
`In Partial Fulfillment of the
`
`Raquirements fur the Degree of
`
`flasher of Science in.Electrical Engineering
`
`Thomas C Farrell, 3.8.
`
`Captain, USAF
`
`November 1988
`
`.w«.~w___,____*_,~__,~__*
`Aoaession For
`
`
`GRA&I
`NTIS
`DTIC TAB
`
`Unannounced
`Jus$1ficatio
`
`
`Distribution]
`
`
`Availability Cadas
`“AQaZYHéfid/ofi
`Special
`
`
`
`
`
`Approved far public release; distribution unlimited
`
`Page 3 of 155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`
`
`Emfiam
`
`3y interest in land mobile radio (LMR) began in Europe when, as an
`
`additional duty,
`
`I became our unit’s Site Security DICK
`
`Subsequent
`
`exercises and real vorid events demonstrated tha need for reliable
`
`intra-base communications, and how easily the communicaticn systems
`
`(public telephone, field phone, and radio) could become saturated with
`
`calls in an emergency.
`
`Hybrid trunked LflR should go a long way to sélving these prcblems.
`
`Although this thesis explores the effects of some increases in loading
`
`on fleets of a cranked system, more research on LMR loads during
`
`exercises would be profitable, of particular interest Would be the
`
`probability distributipns and statistics (described in Chapter V} of
`
`various LMR nets currently in use at Air Force bases during exercises,
`
`In conducting this research I have been helped by many people.
`
`In
`
`particular,
`
`I would like tn express gratitude to my sponsor,
`
`Mr Gardner, who provided much of the background information about LMR
`
`systems and answered many questions, and to my committee, Maj Prescott,
`
`Ha j Norman, and CPT Shaw.
`
`(DPT Shaw deserves special thanks for the
`
`time he spent and advice he gave, both on the queueing aspects of this
`
`thesis, and on good engineering practices in general.
`
`I would also
`
`like ta thank my parents who,
`
`through example, dgmonstrated the
`
`benefits of academic discipline and self motivation‘ Finally,
`
`I would
`
`like to thank the technical people I have knownj and learned from, who
`
`are serving in the United States armed forces around the world.
`
`Thomas C Farrell
`
`ii
`
`Page 4 0f155
`
`PETITIONER'S EXHIBIT 1017
`
`
`
`.mj
`
`
`
`..1V,
`
`' 3*3-fl‘mk ’ A» w»- 1.1.
`
`Wflm
`
`Preface .
`
`List of Figures .
`
`List of Tables
`
`Abstract
`
`.
`
`.
`
`1
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`1
`
`I.
`
`Introduction
`
`.
`.
`Background .
`Problem and Scope
`Approach
`.
`.
`.
`Assumptinns
`Equiyment
`
`.
`
`.
`
`,
`
`II.
`
`Litararure Review .
`
`.
`
`.
`.
`.
`.
`.
`Trunking Schemes
`1
`Air Force Requirements .
`Description of the Hybrid Trunkad System .
`Load Analyses
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`,
`.
`,
`
`.
`
`.
`
`.
`
`.
`
`.
`
`Page
`
`ii
`
`vii
`
`ix
`
`vmwwr—oH
`
`w
`
`10
`12
`
`16
`
`16
`17
`19
`21
`
`25
`
`25
`25
`30
`33
`
`39
`
`39
`39
`40
`41
`43
`
`III.
`
`Conventional Model_...
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`Introductinn ,
`
`.
`
`.
`
`.
`.
`
`.
`.
`
`Description of the Computer Model
`Discussion of the Madel
`.
`.
`Mathematicai Verificat1on of the Made1
`
`.
`
`.
`
`1V.
`
`Trunked Model
`
`Introduction .
`
`.
`
`.
`
`Description of the Computer Mudel
`Discussion of the node}
`,
`.
`Mathemgtical Vetification of the Model
`
`Analysis of Data Collected Via fionitoring
`
`.
`Objectiveg .
`Procedure Used to Collect Data .
`
`Monitering:
`Monitoring;
`Results
`
`Phase I
`Phase II
`
`.
`
`iii
`
`Page 5 of 155
`.A.~__.m..~-..4w
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`" “’“W "i
`
`M '"
`
`'Qamfitn‘fimv‘ngJy $.3‘m..f§'>. 13:“5!» ". 8 .‘u ‘ a
`
`<
`
`VI.
`
`Narmal Configuration Runs
`
`,
`
`.
`.
`.
`’
`t
`.
`.
`.
`Overview .
`Comparison of Conventional
`and Trunked Systems
`,
`.
`Interpretation of Results
`Sub-fleets »
`‘
`.
`.
`,
`Priority .
`.
`a
`Sensitivity w
`Optimum RU !
`
`VII‘
`
`Contingency Hodel Runs ,
`
`a
`.
`w
`.
`.
`,
`.
`.
`.
`Overviaw ‘
`Increase in Load; No Increasa
`in Sub-fleets
`p
`.
`.
`.
`,
`.
`;
`Creaticn of Naw Sub~fleets .
`Failure of Parts of the
`
`a
`
`.
`.
`
`Trunked System ,
`
`.
`
`‘
`
`VIII» Conclusions and Recommendations
`
`‘
`Summary
`.
`.
`.
`.
`.
`.
`.
`.
`.
`,
`Concluaions
`Recommandations For Further Work .
`
`a
`
`.
`
`Appendix A:
`
`SLAM Code For the Conventional
`Simulation Model
`
`Appendix 8:
`
`SLAM Code For the Trunked
`Simulation Model
`.
`.
`,
`._.
`
`Appendix C:
`
`Frequencies of Number of Transmissions
`Per Message For Eadh Channel Manitored .
`
`Appendix D:
`
`SLAM Output From the Conventional Kodel
`
`t
`
`.
`
`.
`
`.
`
`Appendix E:
`
`SLAM Output Fram the Trunked Model
`
`.
`
`Bibliography
`
`Vita
`
`Page
`
`A?
`
`A7
`
`4?
`49
`56
`59
`60
`63
`
`65
`
`63
`
`65
`69
`
`71
`
`73
`
`?3
`73
`75
`
`76
`
`92
`
`107
`
`111
`
`125
`
`139
`
`1&1
`
`iv
`
`Page 6 0f155
`
`PETITIONER‘S EXHIBIT 1017
`
`
`
`4flmaflf ngxhbingn
`
`).\
`
`mam
`
`Figure
`
`Conventional Land Mobile Radio Model
`
`Predicted and Measured Hair Time
`(Conventional Model)
`
`Trunked System Model (Part 1)
`
`Trunked System Model (Part 2)
`
`Flow of Entities in the Modified
`
`Trunked System Model
`
`.
`
`,
`
`.
`
`.
`
`.
`
`.
`
`,
`
`.
`
`Q
`
`Predicted and Measured Channel Queue Length
`{modified Trunked System Mofiel)
`.
`,
`.
`4
`,
`
`.
`
`e
`
`*
`
`.
`
`.
`
`Predicted and Measured Channel Queue Length
`as the Number of Hessages Per Transmissicn
`is Varied (Modified Trunked System Model)
`
`.
`
`.
`
`«
`
`i
`
`.
`
`Percent of Callers Obtaining a Channel
`Within 1 Second
`
`Wait Time Unti} 80% of Callers Obtain
`a Channel
`
`Wait Time Until 90% of Callers Obtain
`a Channel
`.
`.
`.
`.
`.
`.
`,
`.
`.
`A
`.
`,
`,
`
`Wait Time Until 98% of Callers Obtain
`a Channal
`
`Delay in Obtaining a Channel For Trans-
`missions Other Than the First One in a
`
`Message as a Function of Parameter RU
`
`Frequency of Messages By Number of Trans—
`missions For the Security Police Net
`.
`
`Frequency of Hessages By Number of Trans«
`misaions For the Motorpool Net
`.
`
`Frequency of Messages By Number of Trans~
`missions For the Base Supply & Distribution
`C Net
`
`‘10‘
`
`11.
`
`12.
`
`13.
`
`1a.
`
`15.
`
`Page
`
`18
`
`24
`
`26
`
`28
`
`33
`
`37
`
`38
`
`51
`
`52
`
`53
`
`64
`
`107
`
`108
`
`108
`
`Page 7 0f155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`16.
`
`17.
`
`19.
`
`13.
`
`Frequency cf Hessagas By Number of Trans-
`missions For the Fire/Crash Net
`
`Frequency of Messages By Number of Trans~
`missions For the Civil Engineers Channal
`1 Net
`.
`.
`.
`.
`\
`
`Frequency of Messages By Number 0f Trans-
`missions Far the Civil Engineers Channel
`2 Net
`
`Frequency of Kessages By Number of Transmis-
`sions For the Specialist Dispatch/POL/Base
`Oyerations flat
`,
`.
`,
`.
`.
`.
`.
`
`Page
`
`109
`
`109
`
`110
`
`110
`
`vi
`
`PETITIONER’S EXHIBIT 1017
`Page 8 0f155
`w M“,+WWWWW~WW
`
`
`
`
`
`Lifi£ 2i Iflhl§§
`
`Table
`
`II.
`
`III.
`
`IV,
`
`Predicted and Measured Wait Time
`(Conventional Model)
`
`Predicted and Measured Channel Queue Length
`(Modified Trunked System Model)
`
`Predicted and Measured Channel Queue Length
`as the Number of Transmissions Fer Ressage
`is Varied (Modified Trunked System Model)
`
`Number of Messages Noted During various Times
`of the Day (Measured During Weekdays For
`3107.5 Seconds of the Rout)
`
`Measured Characteristics of Transmission
`
`Length and Time Between Transmissions
`(Within a Message)
`.
`.
`g
`.
`.
`
`VI.
`
`Measured Characteristics of the Number of
`
`Transmissions ?er Message
`
`VII.
`
`VIII.
`
`IX.
`
`XI.
`
`XII,
`
`Channel Load Used in the Computer Models
`to Simulate Normal Conditions
`
`Fleet and Net Inputs Used in the Computer
`Simulation Models to Compare the
`Conventional and Trunked Systems .
`
`Comparison of 7 Channel/7 Fleet Trunked
`Model
`(With RD Set
`to 0) and the
`Corresponding Conventional Hodel
`
`Comparison of the Effects of Division
`Into Sub~f1eets of the Original Seven
`Fleet Trunked System (Time For 98% of
`Callers to Obtain a Channel)
`
`Comparison Between a Prioritized Trunked
`System and a Similar System With Priorities
`Set to the Same Value
`
`Effects of a Change in Mean Transmission
`Length of the Security Police Fleet on
`the ConVentional and 4 Channel Trunked
`Models (Time For 981 of Callers to Obtain
`a Channel)
`
`vii
`
`Page
`
`23
`
`36
`
`38
`
`$2
`
`an
`
`45
`
`as
`
`$8
`
`55
`
`58
`
`60
`
`61
`
`Page 9 0f155
`w. .M “WM". -W.~
`
`PETITIONER'S EXHIBIT 1017
`
`
`
`Page
`
`XIII. Effects of a Change in Mean Messages/Hour
`of the Security Folice Fleet on the Con~
`ventional and a Channel wrunked nodels
`(Time For 981 of Callers to Obtain a Channel)
`
`.
`
`.
`
`.
`
`.
`
`.
`
`61
`
`XIV. Effects of a Change in Mean Transmis-
`sicns/Message of the Security Police Fleet
`an the Conventional and & Channel Trunked
`Mbéels (Time For 981 of Callers to Obtain
`a Channel) 4
`J
`.
`.
`.
`.
`e
`.
`,
`.
`,
`.
`t
`,
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`62
`
`XV. Effects of a Change in Standard Deviation
`ofi Eransmissions/Message of the Security
`Police Fleet on the Conventional andih
`Channel Trunked Models (Time For 982-of
`Callers to Obtain a Channel)
`.
`.
`.
`.
`.
`
`.
`
`XVI.
`
`’Results of an Increased Load on the
`Security Felice Fleet
`.
`,
`.
`.
`.
`.
`
`.
`
`‘
`
`.
`
`.
`
`.
`
`.
`
`s
`
`.
`
`a
`
`.
`
`.
`
`.
`
`,
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`i
`
`.
`
`62
`
`6?
`
`67
`
`XVII, Rgsukts of an Increased Load on the
`.Fire/Crash Fleet
`.
`‘
`.
`.
`.
`.
`.
`.
`fl
`
`,
`
`§
`
`XVIII. Resu1t$,c£ an Increased Lead on the
`SecuritvaQIIQe and Fire/CraSh Fleets
`
`.
`
`.
`
`XIX‘ Overall Message Delay on the Trunked
`System With an Extra Security Palice
`Fleet Added, Compared With the Normal
`7 Fleet System (Time For 90% of Callers
`( to Obtain a Channel)
`.
`,
`.
`,
`.
`.
`,
`.
`.
`
`.
`
`.
`
`,
`
`,
`
`.
`
`.
`
`;
`
`.
`
`.
`
`.
`
`.
`
`.
`
`4
`
`68
`
`,
`
`.
`
`,
`
`70
`
`XX. Overall Message Eelay on the Trunked
`System With an Extra Fire/Crash Flset
`Addad, Compared With the Normal
`7
`Fleet System (Time For 902 of Callers
`to thain a Channel)
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`A
`
`.
`
`.
`
`.
`
`7O
`
`XXI. Overall Message Delay on the Trunked
`System with Extra Security Police and
`Fire/Crash Fleets Added, Compared with
`the Normal 7 Fieet System (Time For
`90% of Callers to Obtain a Channel)
`
`.
`
`.
`
`,
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`71
`
`viii
`
`Page 10 0f155
`
`PETITIONER‘S EXHIBIT 1017
`
`
`
`AFI'r/GE/ENG/sm. 11
`
`.‘M
`if
`
`/,'3
`
`J W
`
`"
`
`Trunked land mdbile’radio systems, currently being developed by
`
`several companies, allow many groups 9f land mobilg radio (LMR) users
`
`tn share a set of channals dynamically. reducing the total number of
`II
`
`channels needed to sfippnrt these groups. These systems also support
`/
`5fdynamic regroupingfi?’a reassigning individual users to different groups
`
`through 30ftware in the contrailing computer.
`
`flybrid trunked systems
`
`(KISS) have the added advantage of being able.
`
`in the event of
`
`tontrolling system failure,
`
`to default to certain channels, adding a
`
`degree of robustness to the system. HTSs seem to be an answer to many
`
`of the Air Force’s intra~base communications needs. These needs
`
`include the ability to support an ever increasing number of users with
`
`a minimal
`
`increase in allocated channels, a very high level of system
`
`reliability under extremely adverse conditions, and an ability to
`
`manage users under a variety of contingencies (base attack, aircraft
`
`crash, etc.)
`
`In order to derermiua the number oi channelS a HTS will
`
`
`
`require for a specific facility, information about traffic loading. and
`I
`4’
`
`how the system reacts to ity is needed. wwM"
`
`‘
`
`”
`
`»
`
`"WW
`
`This paper discusses a computer model of existing LMR networks on
`
`Wright Patterson Air Force Base (VPAFB), and a model of a possible
`
`crunked system for the base. Data was collected from off the air
`
`monitoring of LMR nets, and was used to dntetmine numerical V81u85 for
`
`various parameters. These values were input
`
`to the computer models to
`
`determine the time required for a user to obtain a channel while
`
`Page 11 0f155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`tra£fic load and (£0: the trunked model) user grouping were varied to
`
`simulate varicus conditions.
`
`A 5 (1 data, 4 voice) channel HTS was faund to adequately support
`
`W?AFB, eVen with a loss of one repeater and an increase in LMR traffic.
`
`With proper usat grnuping,
`
`trunked system yerformance is shown to be
`
`suparior to the existing conventinnal system while using fewer
`
`channels,
`
`Page 12 of 155
`
`PETITIONER'S EXHIBIT 1017
`
`
`
`A COMPUTER SIMULATION ANALYSIS OF CONVENTIONAL AND TRUNKED
`
`LAND MOBILE RADIO SYSTEMS AT WRIGHT PATTERSON AIR FORCE BASE
`
`mem
`
`W L
`
`and Mobile Radios {LMRs) (also called "walkie»talkies” or
`
`"brick5") are small, hand held rgdios used by police, fire departments.
`
`and other organizations desiring portable, rapid communications.
`Because of the LMR'S decreasing cost and increasing availabllity, many
`
`organizations on Air Force bases now have, or want, their own LMR
`
`network (net). Because of this,
`
`the Air Force now faces the problem of
`
`obtaining allocation'of a larger number of channels from the Federal
`
`Communications Commission (FCC) and host nations.
`
`Trunked LMR systems reduce this problem by allowing users to Share
`
`a set of channels dynamically.
`
`In one type of Lrunked system, all of
`
`the radios are originally tuned to a digital channel monitored by a
`
`computer ariven central controller,
`
`If a user, a fireman for exampleX
`
`wants to talk with his department, he keys the radio, which sends a
`
`digital signal to the central controller.
`
`The controller examines the
`
`set of allocated voice channels and.
`
`if it finds one not currently in
`
`use, it sends a digital signal to every raélo on the fireman's net
`
`(called "fleet" in trunked systems) rewtuning them to the channel.
`
`When the fireman de~keys his radio all the radios in the fleet re»tune
`
`back to the digital channel, Normally this whole procedure occurs so
`
`Page 13 0f155
`
`PETITIONER‘S EXHIBIT 1017
`
`
`
`mW‘JLL‘Swu '1
`
`quickly the user dcesn’t notice any difference frcm a canventional
`
`system. Bowever, if all of the voice channels are in use, other users
`
`trying to get a channel are queued on a priority basis by the con-
`
`troller.
`
`Trunked systems have several advantages over conventional systems:
`
`1. As mentioned above,
`
`the primary advantage is in requiring
`
`fewer channels to Satisfy more users. This is based ofi the observation
`
`that transmiasions usually take place on a ccnvenCional net for only'a
`
`small percentage of time.
`
`2.
`
`Individual radios in a trunked system can be reallocated to
`
`different fleets based on programs stored in the cantral controlleri
`
`This has great advantages on an Air Force base, particularly during
`
`cantingencies when individuals are performing different missions,
`
`reporting chains are changed, and same conventional LMR nets would
`
`becoma saturated.
`
`3. Assuming compatibility between Air Force trunked systems,
`
`deployed units can communicate with other units at their new lucation.
`
`For example: national guard units deployed overseas can integrate
`
`their LMR system with that of their best basa.
`
`4.
`
`Indiviéual radios can be "turned off" of a system. This is an
`
`advantage in situations such as a hostage scenario where the hostage’s
`
`captured radio can be taken off of the fleets used by the rescue force
`
`and, if desired, assigned to its own fleet for use by the negotiating
`
`team.
`
`Hybrid trunked systems are trunked LMR systems with the added
`
`advantage that, if the central controller goes down, radios automati-
`
`z
`
`7";
`
`PETITIONER'S EXHIBIT 1017
`Page 14 0f155
`1
`1_“%M,”yM._~mM.h_u_a_...uaaanunmnaaa-anuuuauuaun-I-IhIfl~w-fl*'Hfl*-""““““'”""”"”'*““‘""J
`
`
`
`
`
`cally re-tune to preallocated channels. This is vital in the military
`
`environment, where loss of one element of the system shouldn‘t com~
`
`pletely eliminate communications.
`
`The 1842 Electronics Engineering group, Scott AFB. 11 is develop~
`
`ing‘Air Force requirements far the hybrid cranked LMR systems described
`
`above and needs data to determine thé number af channels necessary ta
`
`provide reliable communications in a contingency situation. They would
`
`like to have a computer model developed which wiil simulate a trunked
`
`system and determine its performance characteristics during various
`
`contingencies.
`
`WWW
`
`The objective of this thesis is.t0 design and build a computer
`
`simulation model of a cranked system for a specific Air Force base,
`
`determine appropriate values for input parameters for both day to day
`
`and contingency operations, and use the model
`
`to determine the number
`
`of channels needea to provide the base LMR users with a reasonable time
`
`to access a channel.
`
`AW
`
`Qang;§I Mgfifilfi»
`
`A computer model of a éonventional LHR system
`
`was built as a baseline fer measuring performance differences between
`
`it and the trunked model,
`
`In a conventional system there are two
`
`possibln reasons a user would have to wait for a channel:
`
`1) someone
`
`else on the user's net is already talking, or 2) someone on another net
`
`(sharing the channel) is talking.
`
`The computer model measures these
`
`Page 15 0f155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`0
`\
`-
`l
`0
`#
`~ A
`I
`'
`.J. a 5“!“ .1.*vam.m‘.._3m.‘) 1.x“.- r.‘ v
`
`'
`
`conditions for a given load and presents curves of the percent of
`
`transmissions delayed vs.
`
`the amount of time they are delayed,
`
`On a trunked system, delays in granting a user a channel can be
`
`due to somebody else talking on the same fleet. al& of the voice
`
`channels being in use, and mechanical flelay in the system'iwbich
`
`includes delay in accessing the controller on the digital channol and
`
`delay in the controller itself).
`
`The computer model of the trunked
`
`system assumes a constant mechanical dolay anfl measures the other two
`
`delay conditions for a given load, Like the conVentional model,
`
`the
`
`results are plotted as tho percent of transmisaions delayed vs.
`
`the
`
`amount of time they are delayed»
`
`Both computer models were built using SLAM II, a FORTRAN based
`
`simulation tool (7zvii).
`
`The models were verified by setting the input
`
`parameters to match simple mathematical models and comparing results‘
`
`ggllgggign gfi aggg1 Data was collected from off the air monitor-
`
`ing of nats in use at Hright~Patterson Air Force Basa (WPAFB).
`
`The
`
`data was used to determine, for each not,
`
`the number of messages per
`
`hour,
`
`the mean transmission length,
`
`the mean time batween transmissions
`
`(within a message}, and the mean number of transmissions per message.
`
`(Usually a conversation over LMRS consists of several
`
`transmissions
`
`making up a message.
`
`For example, a dispatcher asks for a police
`
`officer's location,
`
`the officer tells him, and the dispatcher responds.
`
`This is considered one message and consistfi of three transnissions:
`
`one by the police officer and two by tho dispatcher.)
`
`The data was
`
`also used to verify the legitimacy of the various distributions used in
`
`the computer models.
`
`Page 16 of 155
`
`PETITIONER’S EXHIBIT 1017
`
`1
`~ 1......m
`
`
`
`
`
`
`
` “‘un‘iw—E .-.¢:-u:;,«’~’l»-u«'x.~ 9.“ .:~- (u- .Mn '4 . ‘
`
`.1..
`
`a???“
`
`,L,.—‘.
`
`Normal Qanfiguzatign Emma.
`
`The data collected by off the air
`
`monitoring was put into the computer models and they were set up to
`
`simulate the existing conventional system, and a hypothetical
`
`trunked
`
`system, at WPAFB.
`
`The models were run for various loads, and for
`
`different numbers of channels in the trunked model.
`
`The curves
`
`obtained were then,compared to determine how many channels a trunked
`
`system would need to provide performance comparable to the existing
`
`system‘
`
`antingengx Syng, Various contingencies were also examined.
`
`Contingencies can affect an LMR system in at.least three ways:
`
`1.
`
`In certain circumstances,
`
`load might increase disproporv
`
`tionately for a few nets (or fleets).
`
`For example. an automatic fire
`
`alarm going off in a hospital stateroom might cause increased activity
`
`on the fire net,
`
`the hospital net, and the security police net, but
`
`would not affect the load-on other nets at all.
`
`2. On a computer controlled trucked system, fleets might be
`
`reellocated during certain contingencies, Host notably, if the base_is
`
`located in an area that could become a war zone, contingency plans
`
`probably call for reallocating resources (manpower and equipment) from
`
`non—essential functions to areas vital to the base's wartime mission.
`
`3. Certain contingencies might affect the LMR system itself.
`
`For
`
`example. a fire in the room housing a repeater would not only increase
`
`traffic load, but might
`
`take the repeater off the air.
`
`These situations were examined with the trunked model.
`
`Page 17 of 155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`
`
`Wm
`
`There appears to be no published data on call inter~arrival dis-
`
`tribution and call length distribution specifically taken from Air
`
`Force LMR nets,
`
`The assumption was made that these distributions,
`
`in
`
`general, are similar to commercial nets as described in the literature
`
`review. This assumption was checkeé to some extent through off the air
`
`monitoring of WPAFB new (see Chapter V).
`
`In off the air monitoring of QPAFB nets to determine mean call
`
`WW“
`
`inter-arrival times and mean call lengths,
`
`the statistical fluctuation
`
`over periods of time greater than several days was assumed,to be
`
`negligible. This was necessary due to the time constraints of the
`
`research.
`
`The nature of the LMR users on WPAFB led to an assumption that
`
`traffic intensity is fairly constant throughout the éay, and equal or
`
`heavier (depending on the specific user) during daytime than at night.
`
`This assumption was checked through off the air monitoring {see
`
`Chapter V).
`
`The Air force will require an adjustable 0 tor6 second "drop out“
`
`time for its hybrid trunked systems (16). Drop out time is an inten—
`
`tional delay in releasing a channel after a user do-keys, and allows a
`
`user to complete a transmission if he inadvertently de-keys for a
`
`moment. This is not modelled in the simulation and the effects on the
`
`measured results are assumed to be negligible.
`
`(Actually,
`
`the simula»
`
`tion models a trunked system with a drop out
`
`time set to 0 seconds,
`
`Any other drop out
`
`time would require modifying the trunked computer
`
`model.)
`
`Page 18 of 155
`
`PETITIONER'S EXHIBIT 1017
`
`
`
`_'
`.‘V
`H ,.~‘
`J~-,\z.wu;...A_.—'».Anigh“ ‘n'v'w ”v m 1.» ..x.-,
`
`.~ ,
`
`hum-Ac ‘
`
`"
`‘3 » w»¢o.-a~.,¢u ‘.wfi.
`
`W A
`
`VAX/VHS computer system owned by the Air Force Inatitute of
`
`Technology (AFIT) was used to run the simulation models, Data was
`
`collected using a Realistic PRO-2004 programmable scanning rece$ver and
`
`recorded on a Realistic VSC—QOOO variable speed cassette tape recorder.
`
`both cwned by the researcher.
`
`Page 19 of 155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`.
`-.
`,
`-
`.
`, w-
`
`’
` " sum»: '-
`M‘Mggvk “.5133“: “.‘aLum
`
`Wu“... .,‘___,, w~ ‘
`
` 1
`
`i
`
`mwm
`
`WW
`
`Reeves (8:3) discusses several
`
`trunking schemes. One of these,
`
`the simplest in terms of hardware required,
`
`includes a repeater for
`
`aach channel and a number of mobile (or portable) radics, assigned to
`
`specific netsi
`
`Each radio automatically scans through the channels,
`
`stopping when it finds a.signal indicating a call is about to start on
`
`the channel for that radio's net.
`
`A radio making a call finds an idle
`
`channel and sends a signal indicating which net tha radia belongs to
`
`and telling other radies on the net to monitor that channel.
`
`Another technique (8:3) involves connecting a camputar driven
`
`controller to tha repeaters and breadcasting an idle tone on an unused
`
`channei. Each mobila radio Scans the channels until it finds the tone.
`
`When a call is made,
`
`the contrullar has the channel’s repeater send a
`
`signal indicating which net is involved. Radios not on that net then
`
`continue scanning until they find the idle tone again, which the
`
`central controllgr has moved to another idle channel.
`
`A third technique discussed by Reeves, and described by Thro
`
`(11:302), usas a computer to contrai the repeaters, as with the system
`
`previously discussed, but uses one of the channels exclusively for
`
`signalling. When radios are idle,
`
`they monitor the signalling channel.
`
`When a call is made,
`
`the calling radio sends a digital signal to the
`
`central contrnller,
`
`indicating which fleet the raaio is an.
`
`The
`
`central controller then sends a digital signal OVer the signalling
`
`channel telling each radio in the fleet to tune to an idle chrfihcl,
`
`8
`
`
`
`Page 20 of 155
`
`WM...,.".._.__._.V.Mm..,w,.4~
`
`.
`
`.
`
`PETITIONER'S EXHIBIT 1017
`‘
`
`
`
`'3‘! ‘.-
`
`"&§a~-’.x£'.s .. .x .’
`
`v
`
`‘
`
`._
`
`.9
`
`h, lssw'4.«
`
`When the call is over, each radio rehtunes back to the signalling
`
`channel and continues monitoring. This technique gives the system fast
`
`access time and good reliability.
`
`mmmumm
`
`As
`
`in the civilian sector.
`
`the Air Force faces an increasing
`
`number of LHR users (about 30 nets on one base, tor example)
`
`(l;K~2—l)
`
`and a limited number of channels available for their use.
`
`In addition,
`
`the Air Force requireo a robust system capable of withstanding harsh
`
`conditions while performing roliably.
`
`The ability to inter~not
`
`(transfer a radio from one not or fleet to another) is also highly
`
`desirable, as is the ability to deploy radios from one location to
`
`another and use them with an existing system at the new location;
`
`An
`
`Air Force Communications Command {AFGC) technical report (12:7)
`
`examined several conventional and trunked LMR systems based on theae
`
`requirements and concluded a hybrid trunked system would beat meet Air
`
`Force needs.
`
`As explained in the report,
`
`the hybrid trunked system operates
`
`like the trunked system with a central controller and dedicated
`
`signalling channel as described above, with the added advantage of
`
`allowing each radio to operate in a conventional modo if the central
`
`controller is disabled.
`
`Air Force specifications for hybrid trunked portable radio
`
`transceivers (15), hybrid trunked mobile transceivers (1&), hybrid
`
`trunked control station transceivers (13), and trunked system central
`
`aontroller equipment (16) are currently being written‘
`
`Page 21 0f155
`
`PETITIONER’S EXHIBIT 1017
`
`~§
`
`g-
`
`L.”
`
`
`
`Wfimmmm
`
`Zdunek describes an existing hybrid crooked system built by
`
`notorola Inc. for use in the United States (1?) and a similar proposed
`
`system for use in the United Kingdom {18). Both of these systems can
`
`support between 5 and 20 channels and any of the four highest in
`
`frequency can be used as the data channel.
`
`Since the radios automati»
`
`cally scan until they find the data channel,
`
`there is protection
`
`against system failure should the data channel's repeater fail:
`
`the
`
`controller simply picks another channel and the radios quickly find it.
`
`Each channel consists of two frequencies, one used as an inbound link
`
`from the broadcasting radio to the repeater, and the other used as the
`
`outbound link from the repeater to the radios in the fleet. These are
`
`often referred to as the ”inbound channel“ and "outbound channel” in
`
`the literature, even though both make up the channel.
`
`Motorola's trunked system can operate so either the whole message
`
`is assigned a channel, or each transmission is asoigned a channel,
`
`which may, or may not, be tho same channel used in the last transmis»
`
`sion.
`
`Zdunek shows better performance is realized with the transmis—
`
`sion trunked mode (17:195).
`
`The transmission trunked mode is easy to implement, because a
`
`transmission is indicated to the central controller through the push to
`
`talk (PTT) switch on the transmitting radio,
`
`A transmission starts
`
`when the radio's user keys the PTT switch and ends when the PTT switch
`
`is de~keyed.
`
`A desirable modification to this scheme is to allow a
`
`small amount of ”drop out" time aftor dewkeying. This gives the
`
`broadcasting radio's user a chance to complete a transmisaion if he
`
`10
`
`Page 22 of 155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`inadvertancly de—keys for a mementx
`
`The Air Force will require a drop
`
`out
`
`time of 0 to 6 seconds (adjustable thraugh the central controller)
`
`(16).
`
`On a busy system channels might not always be immediately
`
`available, and this might cause a delay in the middle of a message on a
`
`transmission trunked system, This condition is very undesirable, and
`
`is taken care of with a ”recent user" queue which gives fleets complet~
`
`ing a transmission recently first priority in cbtaining a newly
`
`available channel.
`
`The Air Farce will require a quaue allowing recent
`
`users to reméin in it for between 0 and 90 seconds (adjustable through
`
`the central caatroller) and operating on a lasc-in-first—oux discipline
`
`(16}.
`
`In the Motorola system, when the uaer keys the PTT switch on his
`
`radio,
`
`the radio senda a 78 bit digital signal to the central con-
`
`troller via the 3600 BPS inbound signalling channel (17:198). The
`
`‘
`
`'
`
`radio coordinates these signals in time with received signals from the
`
`central controller, 30 the 78 bit Signal always begins at the start of
`
`a fixed length time slot (18:14). There is a chance two or more radios
`
`may try to send signals at the same time, and, because these signals
`
`are synchronized in time with the signals caning from the outbound
`
`signalling channel (the scheme is a modification of slotted ALOHA)
`
`the
`
`usable capacity of the inbound channel is about 1/(38) # 0.123
`
`of the
`
`total capacity on a fully loaded system (where e is the base of the
`
`natural logarithm) (1?:197).
`
`A fully loaded system,
`
`in this case,
`
`is a
`
`20 channel system with 3000 radios making an average of One call Qach
`
`an hour.
`
`On a fully loaded system,
`
`taking into account the usable
`
`11
`
`Page 23 of 155
`
`WWW,“
`
`PETITIONER'S EXHIBIT 1017
`
`1
`
`
`
`”~40: W .. ...~...._’-.w .
`
`.V
`
`capacity/total capacity ratio, a total capacity of 34 slots/second is
`
`raquired for the inbound channel (17:197).
`
`When the central controller receives a request for a voice
`
`channel, it checks and, if a channel is available, a digital signal is
`
`sent over the outbound signalling channel telling all of the radios on
`
`the requesting radio’s fleet (including the requesting radio itsolf) to
`
`re«tune to the available channel.
`
`In the Hotcrola system, a 3600 BPS
`
`handshaking signal is-sent over the outbouno voice channel until the
`
`requesting radio re-tunes, recognizes the signal, and responds over the
`
`inbound voice channel with an 1800 Hz tone. Both the radio and the
`
`Controller continua to send subaaudible signals over the voice channel
`
`for the duration of the transmission (digital data from the central
`
`controller ano.a constant tone from the radio) (18:14-15). On the
`
`proposed.United Kingdom trunked system, access time,
`
`the time betweon
`
`the channel request and achieving the voice channel,
`
`is estimated to
`
`take about 660 mseo when a channel
`
`is available (18:13).
`
`For the Air
`
`Foroe system, a 350 msec access time will be required (16).
`
`When the user finishes a transmission. be fie~keys the PTT switch,
`
`and, after the appropriate drop out interval, his radio re—tunes to the
`
`signalling channel.
`
`The other radios on the fleet detect the transmis-
`
`sion is over and also re~tune to the signalling channel.
`
`The central
`
`controller detects the transmission is over and assigns the channel
`
`to
`
`another user as necessary,
`
`mm
`
`The obvious drawback to trunked systems is that a channel may not
`
`always be available when needed.
`
`If nineteen users,
`
`from nineteen
`
`12
`
`Page 24 0f155
`
`PETITIONER’S EXHIBIT 1017
`
`
`
`
`
`different fleets, are using a twenty channel system (nineteen voice
`
`channels and one signalling channel) at a given time, other users will
`
`have to wait
`
`to obtain a channel.
`
`(When they attempt
`
`to make a call,
`
`they are said to be "blocked".)
`
`It is important for trunked system
`
`designers to be able to predict, for a specific system with a certain
`
`number of channels, what the probability of this Occurring will be.
`
`Also of interest is the average wait time for a blocked user, and the
`
`wait
`
`time cumulative distribution function (GDP).
`
`Another issue is whether users tend to talk longer on trunked
`
`systems than on conventional sharad repeater systema (systems in which
`
`two or more distinct user groups share a common frequency).
`
`The
`
`concern is, where users on-a conventional system can hear each other
`
`ané may have a natural channel discipline (short, concise,
`
`transwissions).
`
`trunked users, not being able to hear other floats, may
`
`tend to transmit longer (11:305).
`
`Many analyses have been done on these issuES, using at least three
`
`different approaches;
`
`evaluation of systems already in operation,
`
`mathematical modelling, and computer simulation.
`
`Davis and filtchell
`
`(2:345) point out that in LMR systems the
`
`traffic statistically has large inherent fluctuations.
`
`They show the
`
`measurement of mean traffic loads on existing systems can be inaccurate
`
`and an unreliable predictor.
`
`Two General Electric systems in Chicago. one crunked and one
`
`conventional, with shared repeaters. and both supporting commercial
`
`users, were analyzed using automatic recording equipment (8:4).
`
`No
`
`significant differences in transmission
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
