',
`
`PROVISIONAL APPLICATION COVER SHEET
`
`s~ a request for filing a PROVIS][ONAL APPLICATION FOR PATENT under 37 CPR 1.53(b )(2).
`__.
`-
`'°
`o-~
`.......
`N ....
`=
`-
`c:
`.
`-'i:n
`.......
`-..a
`e~
`......
`~ LAST NAME
`
`INVENTOR(s)/ APPLICANT(s)
`
`FIRSI'NAME
`
`MIDDLEINIT1AL
`
`RESIDENCE (CilY AND EITHER STAIB OR FOREIGN
`COUNfRY)
`
`Docket Number
`
`1005PROV
`
`Type a plus sign ( +)
`inside this box~
`
`+
`
`Dupray
`
`Dennis
`
`J.
`
`222 So. Marion Parkway
`Denver, Colorado 80209
`
`TITLE OF THE INVENTION (280 characters max)
`
`"ENGINEERING A WIRELESS LOCATION SYSTEM'
`
`CORRESPONDENCE ADDRESS
`
`~f~
`
`::;::-~
`:::J
`.• -:.:::4
`
`Dennis J. Dupray
`222 So. Marion Parkway
`Denver
`
`I ZIPCODE I 80209
`
`COUNfRY I United States of America
`
`ENCLOSED APPLICATION PARTS (check all that annlv)
`.;
`ZS<
`Number of Sheets a
`
`Number of Pages
`
`Small Entity Statement
`
`"""'
`;~~AIB I Colorado
`1;;~~
`tL Specification
`::'J Drawing(s)
`;;::-:r.:
`E'.i
`fUx
`A check or money order is enclosed to cover the filing fees
`~
`f';;~J
`...
`•""'
`The invention was made by an agency of the United States Government or under a contract with an agency of the United States Government.
`
`Other (mecifv)
`
`I
`
`METHOD OF PAYMENT (check all that annly)
`
`The Commissioner is hereby authorized to charge filing I
`
`fees and credit Denosit Account Number:
`
`PROVISIONAL
`FILING FEE
`AMOUNf($)
`
`$75.00
`
`181
`
`D
`
`No.
`
`Yes, the name of the U.S. Government agency and the Government contract number are:
`
`~~~~~~~~~~~
`
`Oennis J.
`ray
`222 So. Marion Parkway
`Denver, CO 80209
`303/863-2975
`
`"EXPRESS MAIL" MAILING LABEL NUMBER:
`DATE OF DEPOSIT: October 21, 1997
`
`IB914070839US
`
`I HEREBY CERTIFY THAT THIS PAPER OR FEE
`IS BEING
`DEPOSITED l.llTH THE UNITED STATES POSTAL SERVICE "EXPRESS
`MAIL POST OFFICE TO ADDRESSEE" SERVICE UNDER 37 CFR 1.10
`ON THE DATE
`INDICATED ABOVE ANO
`IS ADDRESSED TO THE
`ASSISTANT COMMISSIONER FOR PATENTS, WASHINGTON, D.C.
`20231.
`
`~
`
`TYPED OR PRIN'TE~ NAME:
`/
`),
`/
`
`CONSTANfE' ROBNETT
`i . . .
`
`A
`
`-+t--·
`s1GNATURE: "-·-. Ol,I\.. w=z ·v\.CV.P IJ,,1 j.:\..LLA 1
`
`•
`
`V'
`
`Apple, Inc. Exhibit 1042 Page 1
`
`

`
`, '
`
`VERIFIED STATEMENT (DECLARATION) CLAIMING SMALL ENTITY
`STATUS (37 CFR 1.9(f) and l.27(b)) - INDEPENDENT INVENTOR
`
`As a below named inventor, I hereby declare that I qualify as an independent inventor as defined in
`37 CFR l.9(c) for purposes of paying reduced fees under section 41(a) and (b) of Title 35, United States
`Code, to the Patent and Trademark Office with regard to the invention entitled "ENGINEERING A
`WIRELESS LOCATION SYSTEM," and described in the specification filed herewith.
`
`I have not assigned, granted, conveyed or licensed and am under no obligation under contract or law
`to assign, grant, convey or license, any rights in the invention to any person who could not be classified as
`an independent inventor under 37 CFR 1.9( c) if that person had made the invention, or to any concern which
`would not qualify as a small business concern under 37 CFR I.9(d) or a nonprofit organization under 37 CFR
`I.9(e).
`
`Each person, concern or organization to which I have assigned, granted, conveyed, or licensed or am
`under an obligation under contract or law to assign, grant, convey, or license any rights in the invention is
`listed below:
`
`[X]
`[ ]
`
`no such person, concern or organization
`persons, concern or organizations listed below*
`
`*NOTE: Separate verified statements are required from each named person, concern or organization
`having rights to the invention averring to their status as small entities. (3 7 CFR 1.27)
`
`[] INDIVIDUAL
`
`[] SMAIL BUSINESS CONCERN
`
`[] NONPROFIT ORGANIZATION
`
`I acknowledge the duty to file, in this application or patent, notification of any change in status
`resulting in loss of entitlement to small entity status prior to paying, or at the time of paying, the earliest of
`the issue fee or any maintenance fee due after the date on which status as a small entity is no longer
`appropriate. (37 CFR l.28(b))
`
`I hereby declare that all statements made herein of my own knowledge are true and that all statements
`made on information and belief are believed to be true; and further that these statements were made with the
`knowledge that willful false statements and the like so made are punishable by fine or imprisonment, or both,
`under section I 00 I of Title 18 of the United States Code, and that such willful false statements may
`jeopardize the validity of the application, any patent issuing thereon, or any patent to which this verified
`statement is directed.
`
`Date: /{? ... 4/-f//
`
`y
`Denver, Colorado 80209
`
`Apple, Inc. Exhibit 1042 Page 2
`
`

`
`''
`
`ENGINEERING A WIRELESS LOCATION SYSTEM
`
`Apple, Inc. Exhibit 1042 Page 3
`
`

`
`Property of lntellabs: Confidential
`
`Table of Contents
`
`1. HIGH LEVEL DESCRIPTION OF THE INTELLABS' LOCATION SYSTEM 3
`
`1.1 Major Features of the lntellab Wireless Location System
`
`1.2 The Wireless Network and Provisioning for the Intellabs' Location System
`
`1.4 lntellabs' Location Engine Software Architecture
`1.4.1 Basic Processing Steps of the Intellabs' Location Engine
`1.4.2 Detailed Descriptions of the Location Engine Processing Steps
`1.4.2.1 Receive and Filter Wireless Signal Measurements
`1.4.2.2 Generate Initial Location Hypotheses
`1. 4 .2.2.1 Theoretical Computations Related To Location Estimation Reliability
`1.4.2.2.2 Theoretical Computations Related To Location Estimation Accuracy
`1.4.2.3 Use past performance to both evaluate and adjust the initial location hypotheses
`
`1.5 Automated Location Data Collection
`
`1.6 MS Data Measurements, Parameters and Telemetry
`
`1. 7 Wireless Location Related Services
`
`1.8 Supplemental Wireless Location Technologies
`1.8.1 Location Base Stations
`1.8.2 Mobile Base Stations
`1.8.3 MS ASIC Modification
`1.8.4 Fast Application Programming Interfaces
`1.8.5 Distributed Antennas for location
`1.8.6 Location Using Home Base Stations
`
`i::l
`
`1.9 Utilization of Wireless Signal Measurements From Multiple CMRS
`
`3
`
`5
`
`6
`6
`8
`8
`9
`11
`12
`16
`
`17
`
`19
`
`20
`
`20
`21
`22
`22
`22
`23
`23
`
`24
`
`2. INTELLABS TECHNOLOGY AND TELECOMMUNICATIONS STANDARDS25
`
`3. PATENTS RELATED TO THE INTELLABS' LOCATION SYSTEM
`
`27
`
`2
`
`Apple, Inc. Exhibit 1042 Page 4
`
`

`
`Property of lntellabs: Confidential
`
`1. High Level Description of the lntellabs' Location System
`
`The Intellabs' Location System is essentially a software based solution for wireless
`location that utilizes the native wireless signaling capabilities of a wireless network for
`locating mobile stations. Only minor network enhancements are required to support the
`Intellabs' Location System for deploying network-wide location services. Further, the
`Intellabs' Location System will meet and exceed FCC phase 2 requirements and can adapt
`to satisfy additional location enhancement requirements. Thus, the Intellabs' Location
`System can cost effectively satisfy the governmental public safety mandates as well as
`provide a wireless location technology for driving a low cost wireless location industry
`providing location services that can be mass marketed.
`
`1.1 Major Features of the Intel/ab Wireless Location System
`In addition to being a software solution for wireless location, the Intellabs'
`Location System is :fundamentally a network centric approach that is consistent with
`modem telecommunications economic theory. That is, the lowest life cycle costs, along
`with best capital utilization, is achieved by locating capital expenditures at "natural" or
`high utilization points of concentration that enhance maintainability.
`
`Since the Intellabs' Location System provides wireless location capabilities using
`only the measurements of wireless signals communicated between mobile stations and a
`conventionally provisioned network of base stations, the same communication standard or
`protocol is utilized for location as is used by the network of base stations for providing
`wireless communications with mobile stations for other purposes such as voice
`communication and/or visual communication (such as text paging, graphical or video
`communications). The Intellabs' Location System is capable of being utilized for wireless
`location with wireless standards such as CDMA, TDMA, GSM, and AMP/NAMPS.
`However, it is likely that higher location accuracy will be provided by using the digital
`wireless protocols of CDMA and TDMA.
`
`A fundamental Intellabs design approach is to synergistically combine various
`location technologies so as to mitigate inherent limitations imposed by these various
`location technologies. In particular, multiple location technologies, or more precisely
`location computational models derived therefrom, can be activated or "fired" concurrently
`when attempting to locate a mobile station. Thus, multiple simultaneous mobile station
`location estimates can be generated for thereby enhancing location resolution beyond what
`any single location technology model can provide. This is discussed in some detail in the
`sections related to location estimating reliability and accuracy below. Examples of
`location computational models based on different location technologies are:
`(a) models using time difference of arrival and/or time of arrival in determining the
`distance of a mobile station from one or more network base stations;
`
`3
`
`Apple, Inc. Exhibit 1042 Page 5
`
`

`
`Property of lntellabs: Confidential
`
`(b) models using signal strength for estimating the distance of the target mobile
`station from the network base stations;
`(c) models using various adaptive learning techniques and/or pattern matching
`techniques for associating and matching wireless signal measurements
`geographical areas. For example, artificial neural networks and/or statistical
`pattern recognizer can be used;
`( d) models using base station coverage area extents and/or wireless signal
`characteristic contour maps similar to topographical maps for performing Venn
`diagram-like set theoretic computations such as unions, intersections, and set
`subtractions.
`
`Note that other types of types oflocation models can also be easily incorporated
`into the Intellabs' Location System and easily incorporated into the Intellabs' Location
`System. In fact, another fundamental Intellabs design approach is that such location
`computational models can be easily added, modified, and/or deleted. Thus, as better
`location computational models become available, these models can be readily used.
`Furthermore, models can be mixed and matched to customize the model set for a
`particular geographical area, thereby providing the flexibility to effectively address the
`wireless signaling peculiarities of virtually any environment. Thus, San Francisco may
`likely have a somewhat different set of models from Kansas City. Moreover, this
`flexibility of the Intellabs' Location System provides the ability to offer various levels of
`mobile station location estimating performance, and therefore, location estimating
`performance may be determined by the funds allocated by a community or municipality for
`wireless location. That is, it may be possible to generate additional revenues by providing
`wireless location estimating enhancements (such as adding additional location estimating
`models) for thereby obtaining a location estimation reliability and accuracy beyond some
`minimal standard (such as that specified by the FCC).
`
`Additionally, note that the Intellabs' Location System can utilize location
`computational models based on substantially any location technology that the wireless
`network is capable of supporting. Thus, if some portion of the wireless network is
`provisioned with additional components (hardware and/or software) that can provide
`supplemental location data, and the network is capable of transferring location related data
`from these components to the Intellabs' Location Center, then a location estimating
`computational model can be developed that utilizes this data For example, as is discussed
`briefly in subsequent sections, if the network includes or communicates with: distributed
`antenna systems, home base stations, mobile units (denoted "Mobile Base Stations") for
`detecting transmissions from a mobile station and thereby homing in on it, and/or low cost
`fixed location wireless receivers (denoted "Location Base Stations"), then location related
`data obtained from such "specialized" components can also be utilized for enhancing
`location estimating performance on an "as needed" basis. Note that the ability to utilize
`location related data from particularly the Location Base Stations, provides a low cost,
`quick strategy for enhancing wireless location in areas where enhanced location is desired
`by a community, but where additional base stations either are not economical, or are
`strenuously objected to by community groups.
`
`4
`
`Apple, Inc. Exhibit 1042 Page 6
`
`

`
`Property of Intel/abs: Confidential
`
`It is also important to note that the Intellabs' Location System is data driven in that
`when determining a mobile station location estimate the system typically derives such
`location estimates from a combination of both real time and archived location data.
`Relatedly, the Intellabs' Location System is also adaptive in that it can adapt automatically
`to changing wireless signal characteristics of the coverage area. That is, the Intellabs'
`Location System is able to adaptively select the archived location data to be used in
`subsequently derived mobile station location estimates. Further discussion of these
`features is provided in the section, "Automated Location Data Collection" below.
`
`Additionally, the Intellabs' Location System provides a :flexible, uniform, cost(cid:173)
`effective capability for distributing location information to, for example, the Public Safety
`Answering Points (PSAPs), and other wireless location requesting applications. The
`Intellabs' Location System also allows for additional wireless location services to be
`developed related, for example, to public safety that are not currently provided. In
`particular, new location services can now be provided that are related to a variety of new
`location-based services for public and private group safety, including family support
`functions. Note that these aspects of the Intellabs' location technology are not covered in
`detail here, but are covered in detail in other Intellabs' patent applications which are
`discussed in the section, "Patents Related to the Intellabs' Location System", and
`incorporated herein by reference.
`
`1.2 The Wireless Network and Provisioning for the Intel/abs' Location
`System
`Figure A illustrates a high level view of how the Intellabs' system architecture is
`incorporated into a conventional wireless network. As is well known in the wireless
`industry, basic wireless network components consist of the following:
`
`1.1) Mobile stations (having reference number 140 in Fig. A);
`1.2) Mobile switch centers (MSC having ref no. 112), optionally including a base
`station controller (ref. no. 174);
`1.3) Base stations (ref. no. 122);
`1.4) Transport facilities (ref no. 176) connecting base stations with MSCs;
`1.5) Trunk connections (ref. no. 179) and services from the MSC to the public
`telephone switching network (PSTN) (ref. no. 124).
`
`Various inter-network services may also be integrated with the wireless network
`such as advanced intelligent network (AIN) capabilities, some of these being universal
`and/or single number service, dual mode (i.e., hard hand-off between PCS and traditional
`cellular), roaming and related services. Accordingly, to provision such services
`commercial mobile radio service providers (CMRSs) may further utilize integrated and/or
`standalone service control points (SCPs, ref. no. 104) that are connected via signal
`transfer points (STPs, ref. no. 110) using the Signaling System 7 and/or other IS-41
`
`5
`
`Apple, Inc. Exhibit 1042 Page 7
`
`

`
`Property of Intellabs: Confidential
`
`signaling protocols over various data networks such as X.25, which connect MSCs with
`appropriate AIN and roaming infrastructures.
`
`Referring to Fig. A, note that, as mentioned above, the lntellabs' wireless location
`architecture is a network centric data driven architecture. Further, note that the Intellabs'
`location architecture primarily consists of the following three components:
`
`(2.1) a location "center" or system (ref. no. 142) having:
`(a) a Location Engine 139 which is the location data processor for the
`providing mobile station location estimates; and
`(b) various application programming interfaces (APis) between the Location
`Engine 139 and each of the following: the MSC, optional network
`components such as an SCP 104, and various wireless location
`application(s) 146.
`
`(2.2) a wireless data capture subsystem (not shown in Fig. A) for updating a
`geographical data base of archived location related wireless signal
`measurements with newly captured data. This is discussed in more detail in
`the section, "Automated Location Data Collection" below.
`
`It is important to note that the above two components (2.1) and (2.2) provide the
`entire core wireless location capability of the Intellabs' Location System. That is, the
`lntellabs' Location System is not dependent upon additional base station antennas or any
`non-standard wireless network components. However, as mentioned above, if such
`additional network components are available for supplying additional mobile station
`location related data, then the Intellabs' Location System can be easily modified to utilize
`this additional data for locating mobile stations.
`Physically the location center of (2.1) consists essentially of a high availability mid(cid:173)
`range UNIX computing platform, optionally hardened for security purposes, and made
`fault-tolerant if required. Depending upon location capacity performance requirements,
`this UNIX platform may be supplemented with multiple, low cost parallel processors.
`Note that the location center - MSC interface generally requires a l.5441\1B bandwidth
`transmission channel. However, the location center may be co-located with the MSC.
`Note that there is likely to be a location center per MSC.
`
`1.3 Intel/abs' Location Engine Software Architecture
`
`1.3.1 Basic Processing Steps of the Intellabs' Location Engine
`A high level module decomposition of the Intellabs' Location Engine is provided in
`Fig. H accompanying this description. At a high level, the lntellabs' Location Engine
`(LE) 139 performs the following steps when determining the location of a mobile station.
`
`4.1) Receive and filter wireless signal measurements: When samples of signal data
`measurements corresponding to wireless transmissions between a mobile station
`
`6
`
`Apple, Inc. Exhibit 1042 Page 8
`
`

`
`Property of Intellabs: Confidential
`
`to be located (also denoted the "target mobile station") and a wireless telephony
`infrastructure are received, the Location Engine initially activates the Signal
`Processing Subsystem 1220 to organize these samples and to construct one or
`more data "ensembles" or collective views of the samples. Note that each
`sample here corresponds to a single collection of wireless signal measurements,
`and that the number of samples taken before the data ensembles are constructed
`can vary from less than 10 to as much as 128 depending on, e.g., processing time
`constraints. In general, it is believed that the more samples taken, the more
`likely that the data ensembles will reflect the non-transient wireless signaling
`characteristics of the coverage area such as buildings, mountains, water bodies.
`Conversely, it is believed that the fewer the samples taken, the more likely that
`the data ensembles will contain transient wireless signaling characteristics such as
`signal reflections from vehicles, tree leaves, etc.
`Subsequently, these data ensembles are :filtered using analog and digital
`spectral :filtering techniques. One result of the processing performed here is that
`data collections of composite wireless signal characteristic measurements are
`generated wherein outliers in the data ensembles have been :filtered out.
`
`4.2) Generate initial location hypotheses: The composite wireless signal
`characteristic data collections created by the Signal Processing Subsystem 1220
`are provided to multiple mobile station location hypothesizing computational
`models (also denoted as "first order models 1224" and "location estimating
`models"). Each such model subsequently generates one or more initial location
`estimates of the target mobile station, and outputs a common data structure
`denoted a "location hypothesis" for each initial location estimate generated.
`That is, each location estimate generated has a corresponding unique location
`hypothesis containing the location estimate. In a typical location estimation
`process there will be multiple first order models activated for generating multiple
`simultaneous location hypotheses for the target mobile station .
`
`4.3) Use past performance to both evaluate and adjust the initial location
`hypotheses: The multiple simultaneous location hypotheses are likely to have
`different target mobile station location estimates, and the corresponding first
`order models are likely to have different historical location prediction
`performances. A confidence value will be assigned to each location hypothesis.
`Preferably, the confidence value for a location hypothesis is a probability relating
`to the likeliness that the corresponding first order model has generated correct
`location estimates in the past. The Location Engine may also derive additional
`location hypotheses from the location hypotheses generated by the first order
`models. That is, "adjuster modules" 1326 are used to derive location
`hypotheses that are related to the first order model generated location
`hypotheses, but the newly derived location hypotheses are believed to be either
`of higher reliability and/or of higher accuracy. The process for deriving the new
`location hypotheses utilizes archived location data in the Location Signature
`Data Base 1320, as is discussed further below.
`
`7
`
`Apple, Inc. Exhibit 1042 Page 9
`
`

`
`Property of Intellabs: Confidential
`
`4.4) Determine a "most likely" mobile station location estimate: The resulting
`location hypotheses, both generated by first order models and derived by adjuster
`modules, are supplied to the most likely location estimator 1344. This most
`likely location estimator uses the location estimates and the corresponding
`confidences from the location hypotheses to determine a most likely location
`estimate of the target mobile station, wherein essentially a summing of the
`confidence values is performed in areas where location estimates occur (and
`likely overlap).
`
`1.3.2 Detailed Descriptions of the Location Engine Processing Steps
`The subsections below provide additional elaboration of the above four steps ( 4 .1)
`- (4.4) for locating a mobile station.
`
`1.3.2.1 Receive and Filter Wireless Signal Measurements
`This step is performed by a subsystem denoted the "Signal Processing and Filtering
`Subsystem" (or simply the Signal Processing Subsystem). In particular, this subsystem
`receives samples of wireless signal characteristic measurements such as a plurality of
`relative signal strengths and corresponding signal time delay value pairs from each of a
`plurality of wireless network base stations. Basically, this subsystem:
`(a) determines, for each network base station detected by or detecting the target
`mobile station, the smallest time delay for wireless signals transmitted between
`the target mobile station and various of the base stations;
`(b) determines, for each network base station detected by or detecting the target
`mobile station, a histogram-like data model is built up from a time series of
`samples of measurements of wireless signal transmissions between the mobile
`station and the base station;
`(c) normalizes the histogram-like data models so that, for instance, the make and
`model of the target mobile station is unnecessary in all further processing;
`( d) filters the histogram-like data models using various combinations of filters in
`order to remove measurements of transient wireless signals that do not reoccur
`frequently enough to be assumed indicative of the wireless signal
`characteristics of the fundamental terrain or topography of the area about the
`target mobile station. For example, the histogram-like data model can be
`subjected to input cropping and various median filters employing filtering
`techniques such as convolution, median digital, Fast Fourier transform, Radon
`transform, Gabar transform, nearest neighbor, histogram equalization, input
`and output cropping, Sobel, Wiener, and the like.
`
`Accordingly, measurements obtained from (a) immediately above can be
`subsequently used by any first order model( s) desiring information related to a maximum
`range of the target mobile station from each of the base stations detected and/or detecting
`the target mobile station. Additionally, the histogram-like data models (also denoted as
`"location signatures" ) resulting from steps (b) through ( d) can be used by various types of
`
`8
`
`,,
`
`Apple, Inc. Exhibit 1042 Page 10
`
`

`
`Property of Intellabs: Confidential
`
`first order models that perform pattern recognition and/or associate location signatures
`with geographical locations of a target mobile station.
`
`1.3.2.2 Generate Initial Location Hypotheses
`The initial location hypotheses are generated by multiple first order models using
`different location technologies. Examples of some of the more general location
`technologies utilized by the Intellabs' Location System include models for locating mobile
`stations using:
`(a) base station coverage areas; e.g., computing a mobile station location using
`base station coverage areas with Venn diagram or set theoretic reasoning as to
`where a mobile station is located. In particular, both positive and negative
`detection of wireless transmissions between the mobile station and base
`stations is useful in this context;
`(b) triangulation and/ or trilateration; i.e., standard time difference of arrival
`(TDOA) and/or time of arrival (TOA) are utilized for estimating a mobile
`station's location. However, note that the computations here can be much
`different from the simple TDOA or TOA computations conventionally
`performed. For example, since the Intellabs' Location System is inherently
`architected to accommodate and resolve conflicts between multiple
`simultaneous location estimates, a number of, say, TDOA can be generated and
`effectively used. For instance, for every three base stations detecting the
`mobile station, a location estimate may be generated. Further, TDOA location
`estimates may be generated both for the forward wireless transmissions to the
`mobile station as well as for the reverse wireless transmissions from the mobile
`station to the base station network;
`( c) stochastic wireless signal pattern recognition; note that there are a number of
`sophisticated statistics packages commercially available that are capable of
`statistically examining collections of data and statistically identifying patterns
`within the data. By providing a location estimating model based on the
`associations between geographical locations and archived measurements of
`wireless signal patterns at these locations, an entirely different approach to
`estimating the location of a mobile station is provided. Moreover, as with (b)
`above, multiple location estimates are acceptable using, for example, the
`forward and reverse transmissions paths to provide different versions of a
`mobile station location estimating model; and
`( d) adaptive or trainable computational processes such as artificial neural networks
`(ANNs). Note that ANNs have been used effectively in other signal processing
`applications for, e.g., associating chemicals with their spectrographic
`signatures once such an ANN is properly trained. Moreover, it is believed that
`measurements of wireless signals between mobile stations and the base station
`network provide a particularly rich set of data upon which to train ANNs.
`Additionally, as with (b) and (c) above, a number of ANNs can be utilized in
`estimating the location of a mobile station. In fact, it is likely that a relatively
`large number ANNs may be utilized in the Intellabs' Location System, wherein
`
`9
`
`Apple, Inc. Exhibit 1042 Page 11
`
`

`
`Property of Intellabs: Confidential
`
`it is likely that at least two and as many as four or more such ANN based
`models may be simultaneously activated for generating location estimates of a
`mobile station.
`
`The Intellabs' location system also provides for the generation and utilization of
`location hypotheses by a first order model using signals from low cost "location base
`stations" that have a narrow range of coverage and that can be activated and deactivated
`by the Location Engine. Moreover, the Intellabs' Locatibn Engine also supports the
`utilization timing delay signals from distributed antenna systems for locating mobile
`stations indoors such as in high-rise buildings. Moreover, the Location Engine can also
`support and utilize communication with mobile devices used in locating and/or tracking a
`target mobile station. Such mobile devices, denoted "mobile base stations," can be
`particularly useful in locating a mobile station in an area of poor reception.
`
`Note that the Intellabs' Location Engine can utilize location hypotheses having
`geographical estimates indicating where the target MS is unlikely to be as well as where
`the target MS is likely to be. For example, for a first order model using the location base
`stations mentioned above, one or more location hypotheses may be generated from
`information indicating that the target MS is not detected by some of the location stations.
`
`Thus, there may be a relatively large number of location hypotheses simultaneously
`generated as estimates of a target mobile station, especially if there are different first
`model invocations for signal measurements related to the forward and to the reverse signal
`path. This is consistent with a fundamental strategy of the Intellabs' Location System; i.e.,
`take advantage of synergies between location hypotheses generated by computationally
`different first order models (preferably at least some of the first order models also using
`different input data). In particular, it is believed that in most wireless coverage areas
`significant increases in location accuracy and reliability can be obtained by utilizing this
`strategy in that a relatively large number of moderately reliable and moderately accurate
`first order models can generate target MS location hypotheses which can be combined to
`provide a highly reliable and highly accurate resulting target MS location estimate.
`
`As mentioned above, each of the first order models 1224 have default confidence
`values associated therewith, and these confidence values are probabilities. More precisely,
`such probability confidence values can be determined as follows. Assume there is a
`partition of the coverage area into subareas, each subarea being denoted a "partition area."
`For each partition area, activate each first order model 1224 with historical location data
`in the Location Signature Data Base 1320 (Fig. H), wherein the historical location data
`has been obtained from corresponding known mobile station locations in the partition
`area. For each first order model, determine a probability of the first order model
`generating a location hypothesis whose location estimate contains the corresponding
`known mobile station location. To accomplish this, assume the coverage area is
`partitioned into partition areas A, wherein each partition area A is specified as the
`collection of coverage area locations such that for each location, the detected wireless
`transmissions between the network base stations and a target mobile station at the location
`
`10
`
`Apple, Inc. Exhibit 1042 Page 12
`
`

`
`Property of Intellabs: Confidential
`
`can be straightforwardly equated with other locations of area A. For example, one such
`partition, Po, can be defined wherein each partition area A is specified in terms of three
`sets of base station identifiers, namely, (a) the base station identifiers of the base stations
`that can be both detected at each location of A and can detect a target mobile station at
`each location, (b) the identifiers for base station