GEOPRIV Working Group J. Polk INTERNET-DRAFT Cisco Systems Obsoletes: 3825 (if approved) J. Schnizlein Category: Standards Track ISOC Expires: June 21, 2010 M. Linsner 17 December 2009 Cisco Systems M. Thomson Andrew B. Aboba (ed) Microsoft Corporation Dynamic Host Configuration Protocol Options for Coordinate-based Location Configuration Information draft-ietf-geopriv-rfc3825bis-04.txt Status of This Memo This Internet-Draft is submitted to IETF in full conformance with the provisions of BCP 78 and BCP 79. This document may contain material from IETF Documents or IETF Contributions published or made publicly available before November 10, 2008. The person(s) controlling the copyright in some of this material may not have granted the IETF Trust the right to allow modifications of such material outside the IETF Standards Process. Without obtaining an adequate license from the person(s) controlling the copyright in such materials, this document may not be modified outside the IETF Standards Process, and derivative works of it may not be created outside the IETF Standards Process, except to format it for publication as an RFC or to translate it into languages other than English. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet- Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt. The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. This Internet-Draft will expire on June 21, 2010. Polk, et al. Standards Track [Page 1] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 Copyright Notice Copyright (c) 2009 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents in effect on the date of publication of this document (http://trustee.ietf.org/license-info). Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Abstract This document specifies Dynamic Host Configuration Protocol Options (both DHCPv4 and DHCPv6) for the coordinate-based geographic location of the client. The Location Configuration Information (LCI) includes latitude, longitude, and altitude, with resolution or uncertainty indicators for each. Separate parameters indicate the reference datum for each of these values. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.1. Conventions . . . . . . . . . . . . . . . . . . . . . . 3 1.2 Resolution and Uncertainty . . . . . . . . . . . . . . . 4 2. DHCP Option Format . . . . . . . . . . . . . . . . . . . . . . 4 2.1 DHCPv6 Option . . . . . . . . . . . . . . . . . . . . . 5 2.2 DHCPv4 Option . . . . . . . . . . . . . . . . . . . . . 6 2.3 Latitude and Longitude Fields . . . . . . . . . . . . . 8 2.4 Altitude . . . . . . . . . . . . . . . . . . . . . . . . 11 2.5 Datum . . . . . . . . . . . . . . . . . . . . . . . . . 12 3. Security Considerations. . . . . . . . . . . . . . . . . . . . 13 4. IANA Considerations. . . . . . . . . . . . . . . . . . . . . . 14 5. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 15 6. References . . . . . . . . . . . . . . . . . . . . . . . . . . 15 6.1. Normative References . . . . . . . . . . . . . . . . . . 15 6.2. Informational References . . . . . . . . . . . . . . . . 15 Appendix A. Calculations of Resolution . . . . . . . . . . . . . . 16 A.1. LCI of "White House" (Example 1) . . . . . . . . . . . . 16 A.2. LCI of "Sears Tower" (Example 2) . . . . . . . . . . . . 19 Appendix B. Calculations of Uncertainty . . . . . . . . . . . . . 20 B.1 LCI of "Sydney Opera House" (Example 3) . . . . . . . . 20 Appendix C. Changes from RFC 3825 . . . . . . . . . . . . . . . . 24 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 25 Polk, et al. Standards Track [Page 2] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 1. Introduction The physical location of a network device has a range of applications. In particular, emergency telephony applications rely on knowing the location of a caller in order to determine the correct emergency center. The location of a device can be represented either in terms of geospatial (or geodetic) coordinates, or as a civic address. Different applications may be more suited to one form of location information; therefore, both the geodetic and civic forms may be used simultaneously. This document specifies Dynamic Host Configuration Protocol (DHCPv4) [RFC2131] and DHCPv6 [RFC3315]) options for the coordinate-based geographic location of the client, to be provided by the server. "Dynamic Host Configuration Protocol (DHCPv4 and DHCPv6) Option for Civic Addresses Configuration Information" [RFC4776] specifies DHCP options for civic addresses. The geodetic coordinate options defined in this document and the civic address options defined in [RFC4776] enable a DHCP client to obtain its location. For example, a wired Ethernet host might use these options for location determination. In this case, the location information could be derived from a wiremap by the DHCP server, using the Circuit-ID Relay Agent Information Option (RAIO) defined (as Sub- Option 1) in RFC 3046 [RFC3046]. The DHCP server could correlate the Circuit-ID with the geographic location where the identified circuit terminates (such as the location of the wall jack). The options defined in this document have limited applicability for mobile hosts. Typically DHCP clients refresh their configuration in response to changes in interface state or pending lease expirations. As a result, when a mobile host changes location without subsequently completing another DHCP exchange, location configuration information initially obtained via DHCP could become outdated. An important feature of this specification is that after the relevant DHCP exchanges have taken place, the location information is stored on the end device rather than somewhere else, where retrieving it might be difficult in practice. 1.1. Conventions used in this document The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119]. Polk, et al. Standards Track [Page 3] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 1.2. Resolution and Uncertainty The DHCP options defined in this document include fields quantifying the resolution or uncertainty associated with a target location. No inferences relating to privacy policies can be drawn from either uncertainty or resolution values. As utilized in this document, resolution refers to the accuracy of a reported location, as expressed by the number of valid bits in each of the Latitude, Longitude and Altitude fields. In the context of location technology, uncertainty is a quantification of errors. Any method for determining location is subject to some sources of error; uncertainty describes the amount of error that is present. Uncertainty might be the coverage area of a wireless transmitter, the extent of a building or a single room. Uncertainty is usually represented as an area within which the target is located. In this document, each of the three axes can be assigned an uncertainty value. In effect, this describes a rectangular prism. When representing locations from sources that can quantify uncertainty, the goal is to find the smallest possible rectangular prism that this format can describe. This is achieved by taking the minimum and maximum values on each axis and ensuring that the final encoding covers these points. This increases the region of uncertainty, but ensures that the region that is described encompasses the target location. The DHCPv4 option format defined in this document supports both resolution and uncertainty parameters. Version 0 of the DHCPv4 option format defined in this document includes a resolution parameter for each of the dimensions of location. Since this resolution parameter need not apply to all dimensions equally, a resolution value is included for each of the 3 location elements. The DHCPv6 option format as well as version 1 of the DHCPv4 option format utilizes an uncertainty parameter. Appendix A of this document provides examples showing the calculation of resolution values. Appendix B provides an example demonstrating calculation of uncertainty values. 2. DHCP Option Format This section defines the format for the DHCPv4 and DHCPv6 options. These options utilize a similar format, differing primarily in the option code. Polk, et al. Standards Track [Page 4] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 2.1. DHCPv6 Option The DHCPv6 [RFC3315] option format is as follows: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Option Code (TBD) | OptLen (16) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | LatUnc | Latitude + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Lat (cont'd) | LongUnc | Longitude + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Longitude (cont'd) | AT | AltUnc | Altitude + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Altitude (cont'd) | Datum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Code: GEOCONF_GEODETIC (8 bits). OptLen: Option Length (8 bits). This option is fixed size, the value of this octet will always be 16. LatUnc: Latitude Uncertainty (6 bits). Latitude: Latitude (34 bits). LongUnc: Longitude Uncertainty (6 bits). Longitude: Longitude (34 bits). AType: Altitude Type (4 bits). AltUnc: Altitude Uncertainty (6 bits). Altitude: Altitude (30 bits). Datum: Datum (8 bits). Polk, et al. Standards Track [Page 5] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 2.2. DHCPv4 Option The DHCPv4 option format is as follows: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Code 123 | Length | LatUnc | Latitude + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Latitude (cont'd) | LongUnc | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Longitude | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AType | AltUnc | Altitude + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Alt.(cont'd) |Ver| Res |Datum| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Code: 8 bits. The code for the DHCPv4 option (123). Length: 8 bits. The length of the DHCPv4 option, in octets. For versions 0 and 1, the option length is 16. LatUnc: 6 bits. When the Ver field = 0, this field represents Latitude resolution. When the Ver field = 1, this field represents Latitude uncertainty. Latitude: a 34 bit fixed point value consisting of 9 bits of integer and 25 bits of fraction. Latitude SHOULD be normalized to within +/- 90 degrees. Positive numbers are north of the equator and negative numbers are south of the equator. LongUnc: 6 bits. When the Ver field = 0, this field represents Longitude resolution. When the Ver field = 1, this field represents Longitude uncertainty. Longitude: a 34 bit fixed point value consisting of 9 bits of integer and 25 bits of fraction. Longitude SHOULD be normalized to within +/- 180 degrees. Positive values are East of the prime meridian and negative (2s complement) numbers are West of the prime meridian. AType: Altitude Type (4 bits). AltUnc: 6 bits. When the Ver field = 0, this field represents Altitude resolution. When the Ver field = 1, this field represents Altitude uncertainty. Polk, et al. Standards Track [Page 6] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 Altitude: A 30 bit value defined by the AType field. Ver: The Ver field is two bits, providing for four potential versions. This specification defines the behavior of version 0 (originally specified in [RFC3825]) as well as version 1. The Ver field is always located at the same offset from the beginning of the option, regardless of the version in use. Res: The Res field which is 3 bits, is reserved. These bits have been used by [IEEE-802.11y], but are not defined within this specification. Datum: 3 bits. The Map Datum used for the coordinates given in this Option. 2.2.1. Version Support 2.2.1.1. Client Version Support DHCPv4 clients implementing this specification MUST support receiving responses of versions 0 and 1. Since this specification utilizes the same DHCPv4 option code as [RFC3825], the option format does not provide a means for the client to indicate the highest version that it supports to the server. 2.2.1.2. Server Version Selection A DHCPv4 server that provides location information cannot provide options with both version 0 and version 1 formats in the same response. This is not useful since receiving two copies of the same Option (either in the same response or a separate response) causes a DHCPv4 client to replace the information in the old Option with the information in the new Option. A server uses configuration to determine which version to send in a response. For example, where a mixture of version 0 and version 1 clients are expected, the server could be configured to send version 0 or version 1 depending on configuration (possibly making the choice based on information such as the client MAC address). Where few version 0 clients are expected, the server could be configured to send only version 1 responses. Version 0 options will provide resolution, while version 1 options will provide an area of uncertainty. An RFC 3825 DHCPv4 client that receives a version 1 option, as defined in this document, will either reject the Option or will not understand the additions to the Datum field and will misinterpret the Polk, et al. Standards Track [Page 7] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 LongUnc, LatUnc, and AltUnc values. If the RFC 3825 DHCPv4 client does not reject the option and utilizes the location data it will most likely assume a datum and interpret the LongUnc/LatUnc/AltUnc values as significant digits and apply them to the Latitude, Longitude, and Altitude values. The resultant location value will be in error up to a full degree of latitude and longitude, and a full increment of altitude. This results in a version 0-only client either not obtaining location information (with no ability to indicate to the server that version 1 was unsupported), or misinterpreting the option. Therefore, in situations where some DHCPv4 clients are known to support only version 0, by default the DHCPv4 server SHOULD send a version 0 response. It is also RECOMMENDED that DHCPv4 client implementations support version 1, so the versioning capability added by this document does not cause errors interpreting the latitude, longitude and altitude values. Moving forward, clients not understanding a datum value MUST assume a World Geodesic System 1984 (WGS84) [WGS84] datum (EPSG [EPSG] 4326 or 4979, depending on whether there is an altitude value present) and proceed accordingly. Assuming that a less accurate location value is better than none, this ensures that some (perhaps less accurate) location is available to the client. 2.3. Latitude and Longitude Fields The Latitude and Longitude values in this specification are encoded as 34 bit, twos complement, fixed point values with 9 integer bits and 25 fractional bits. The exact meaning of these values is determined by the datum; the description in this section applies to the datums defined in this document. New datums MUST define the way that the 34 bit values and the respective 6 bit uncertainties are interpreted. This document uses the same definition for all datums it specifies. Latitude values MUST be constrained to the range from -90 to +90 degrees. Positive latitudes are north of the equator; negative latitude are south of the equator. Longitude values SHOULD be normalized to the range from -180 to +180 degrees. Values outside this range are normalized by adding or subtracting 360 until they fall within this range. Positive longitudes are east of the Prime Meridian (Greenwich); negative longitudes are west of the Prime Meridian. When encoding, latitude and longitude values are rounded to the Polk, et al. Standards Track [Page 8] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 nearest 34-bit binary representation. This imprecision is considered acceptable for the purposes to which this form is intended to be applied and is ignored when decoding. 2.3.1. Latitude and Longitude Resolution In the version 0 DHCPv4 Option, the Latitude, Longitude and Altitude fields are each preceded by an accuracy sub-field of 6 bits, indicating the number of bits of resolution. The resolution sub- fields accommodate the desire to easily adjust the precision of a reported location. Contents beyond the claimed resolution MAY be randomized to obscure greater precision that might be available. When encoded within the version 0 DHCPv4 Option, the LatUnc value encodes the number of high-order Latitude bits that should be considered valid. Any bits entered to the right of this limit should not be considered valid and might be purposely false, or zeroed by the sender. The examples in Appendix A illustrate that a smaller value in the resolution field increases the area within which the device is located. A value of 2 in the LatUnc field indicates a precision of no greater than 1/6th that of the globe (see the first example of Appendix A). A value of 34 in the LatUnc field indicates a precision of about 3.11 mm in Latitude at the equator. When encoded within the version 0 DHCPv4 Option, the LongUnc value encodes the number of high-order Longitude bits that should be considered valid. Any bits entered to the right of this limit should not be considered valid and might be purposely false, or zeroed by the sender. A value of 2 in the LongUnc field indicates precision of no greater than 1/6th that of the globe (see the first example of Appendix A). A value of 34 in the LongUnc field indicates a precision of about 2.42 mm in longitude (at the equator). Because lines of longitude converge at the poles, the distance is smaller (better precision) for locations away from the equator. 2.3.2. Latitude and Longitude Uncertainty The latitude and longitude uncertainty fields are encoded as 6 bit, unsigned integer values. These values quantify the amount of uncertainty in each of the latitude and longitude values respectively. A value of 0 is reserved to indicate that the uncertainty is unknown; values greater than 34 are reserved. A point within the region of uncertainty is selected to be the encoded point; the centroid of the region is often an appropriate choice. The value for uncertainty is taken as the distance from the selected point to the furthest extreme of the region of uncertainty on that axis. Polk, et al. Standards Track [Page 9] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 The following figure shows a two-dimensional figure that is projected to each axis. In the figure, "X" marks the point that is selected; the ranges marked with "U" is the uncertainty. ___ ___________ ^ | / | | | / | | | / | U | / | | | ( | V | | | --X | X | | | `---------. | | | | | | | | | - `-------------------------' |---------X---------------| |<------U------>| Uncertainty applies to each axis independently. The amount of uncertainty can be determined from the encoding by taking 2 to the power of 8, less the encoded value. As is shown in the following formula, where "x" is the encoded integer value: uncertainty = 2 ^ ( 8 - x ) The result of this formula is expressed in degrees of latitude or longitude. The uncertainty is added to the base latitude or longitude value to determine the maximum value in the uncertainty range; similarly, the uncertainty is subtracted from the base value to determine the minimum value. Note that because lines of longitude converge at the poles, the actual distance represented by this uncertainty changes with the distance from the equator. If the maximum or minimum latitude values derived from applying uncertainty are outside the range of -90 to +90, these values are trimmed to within this range. If the maximum or minimum longitude values derived from applying uncertainty are outside the range of -180 to +180, then these values are normalized to this range by adding or subtracting 360 as necessary. The encoded value is determined by subtracting the next highest whole integer value for the base 2 logarithm of uncertainty from 8. As is shown by the following formula, where uncertainty is the midpoint of the known range less the lower bound of that range: Polk, et al. Standards Track [Page 10] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 x = 8 - ceil( log2( uncertainty ) ) Note that the result of encoding this value increases the range of uncertainty to the next available power of two; subsequent repeated encodings and decodings do not change the value. Only increasing uncertainty means that the associated confidence does not have to decrease. 2.4. Altitude The altitude is expressed as a 30 bit, fixed point, twos complement integer with 22 integer bits and 8 fractional bits. How the altitude value is interpreted depends on the type of altitude and the selected datum. New altitude types and datums MUST define the way that the 30 bit value and the associated 6 bit uncertainty are interpreted. Three altitude types are defined in this document: unknown (0), meters (1) and floors (2). Additional altitude types MUST be defined in a Standards Track RFC. 2.4.1. No Known Altitude (AT = 0) In some cases, the altitude of the location might not be provided. An altitude type of 0 indicates that the altitude is not given to the client. In this case, the altitude and altitude uncertainty fields can contain any value and MUST be ignored. 2.4.2. Altitude in Meters (AT = 1) If the altitude type has a value of 1, the altitude is measured in meters. The altitude is measured in relation to the zero set by the vertical datum. 2.4.3. Altitude in Floors (AT = 2) A value of 2 for altitude type indicates that the altitude value is measured in floors. This value is relevant only in relation to a building; the value is relative to the ground level of the building. In this definition, numbering starts at ground level, which is floor 0 regardless of local convention. Non-integer values can be used to represent intermediate or sub- floors, such as mezzanine levels. For instance, a mezzanine between floors 4 and 5 could be represented as 4.1. Polk, et al. Standards Track [Page 11] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 2.4.4. Altitude Resolution When encoded within the version 0 DHCPv4 Option, the AltUnc value encodes the number of high-order Altitude bits that should be considered valid. Values above 30 (decimal) are undefined and reserved. If AT = 1, an AltUnc value 0.0 would indicate unknown altitude. The most precise Altitude would have an AltUnc value of 30. Many values of AltUnc would obscure any variation due to vertical datum differences. The AltUnc field SHOULD be set to maximum precision when AT = 2 (floors) when a floor value is included in the DHCP Reply, or when AT = 0, to denote that the floor isn't known. An altitude coded as AT = 2, AltRes = 30, and Altitude = 0.0 is meaningful even outside a building, and represents ground level at the given latitude and longitude. 2.4.5. Altitude Uncertainty Altitude uncertainty uses the same form of expression as latitude and longitude uncertainty. Like latitude and longitude, a value of 0 is reserved to indicate that uncertainty is not known; values above 30 are also reserved. Altitude uncertainty only applies to altitude type 1. The amount of altitude uncertainty can be determined by the following formula, where x is the encoded integer value: uncertainty = 2 ^ ( 21 - x ) This value uses the same units as the associated altitude. Similarly, a value for the encoded integer value can be derived by the following formula: x = 21 - ceil( log2( x ) ) 2.5. Datum The datum field determines how coordinates are organized and related to the real world. Three datums are defined in this document, based on the definitions in [OGP.Geodesy]: 1: WGS84 (Latitude, Longitude, Altitude): The World Geodesic System 1984 [WGS84] coordinate reference system. Polk, et al. Standards Track [Page 12] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 This datum is identified by the European Petroleum Survey Group (EPSG)/International Association of Oil & Gas Producers (OGP) with the code 4979, or by the URN "urn:ogc:def:crs:EPSG::4979". Without altitude, this datum is identified by the EPSG/OGP code 4326 and the URN "urn:ogc:def:crs:EPSG::4326". 2: NAD83 (Latitude, Longitude) + NAVD88: This datum uses a combination of the North American Datum 1983 (NAD83) for horizontal (latitude and longitude) values, plus the North American Vertical Datum of 1988 (NAVD88) vertical datum. This datum is used for referencing location on land (not near tidal water) within North America. NAD83 is identified by the EPSG/OGP code of 4269, or the URN "urn:ogc:def:crs:EPSG::4269". NAVD88 is identified by the EPSG/ OGP code of 5703, or the URN "urn:ogc:def:crs:EPSG::5703". 3: NAD83 (Latitude, Longitude) + MLLW: This datum uses a combination of the North American Datum 1983 (NAD83) for horizontal (latitude and longitude) values, plus the Mean Lower Low Water (MLLW) vertical datum. This datum is used for referencing location on or near tidal water within North America. NAD83 is identified by the EPSG/OGP code of 4269, or the URN "urn:ogc:def:crs:EPSG::4269". MLLW does not have a specific code or URN. All hosts MUST support the WGS84 datum (Datum 1). New datum codes can be registered in the IANA registry (Section 4) by a Standards Track RFC. 3. Security Considerations Where critical decisions might be based on the value of this GeoConf option, DHCP authentication as defined in "Authentication for DHCP Messages" [RFC3118] and "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)" [RFC3315] SHOULD be used to protect the integrity of the DHCP options. Since there is no privacy protection for DHCP messages, an eavesdropper who can monitor the link between the DHCP server and requesting client can discover this LCI. To minimize the unintended exposure of location information, the LCI Polk, et al. Standards Track [Page 13] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 option SHOULD be returned by DHCP servers only when the DHCP client has included this option in its 'parameter request list' (section 3.5 [RFC2131]). When implementing a DHCP server that will serve clients across an uncontrolled network, one should consider the potential security risks. 4. IANA Considerations IANA has assigned a DHCPv4 option code of 123 for the GeoConf option defined in this document. Assignment of a DHCPv6 option code is requested. The GeoConf Option defines two fields for which IANA maintains a registry: The Altitude (AT) field and the Datum field (see Section 2). The datum indicator MUST include specification of both horizontal and vertical datum. New values for the Altitude (AT) field are assigned through "Standards Action" [RFC5226]. The initial values of the Altitude registry are as follows: AT = 1 meters of Altitude defined by the vertical datum specified. AT = 2 building Floors of Altitude. Datum = 1 denotes the vertical datum WGS 84 as defined by the EPSG as their CRS Code 4327; CRS Code 4327 also specifies WGS 84 as the vertical datum Datum = 2 denotes the vertical datum NAD83 as defined by the EPSG as their CRS Code 4269; North American Vertical Datum of 1988 (NAVD88) is the associated vertical datum for NAD83 Datum = 3 denotes the vertical datum NAD83 as defined by the EPSG as their CRS Code 4269; Mean Lower Low Water (MLLW) is the associated vertical datum for NAD83 Any additional LCI datum(s) to be defined for use via the DHCPv4 or DHCPv6 Options defined in this document MUST be done through a Standards Track RFC. This document defines the Ver field for the DHCPv4 Option, with values as follows: 0: Implementations conforming to [RFC3825] 1: Implementations of this specification Any additional Ver field values to be defined for use with the DHCPv4 Polk, et al. Standards Track [Page 14] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 Option MUST be done through a Standards Track RFC. 5. Acknowledgments The authors would like to thank Patrik Falstrom, Ralph Droms, Ted Hardie, Jon Peterson, and Nadine Abbott for their inputs and constructive comments regarding this document. Additionally, the authors would like to thank Greg Troxel for the education on vertical datums, as well as Carl Reed. 6. References 6.1. Normative References [EPSG] European Petroleum Survey Group, http://www.epsg.org/ and http://www.epsg-registry.org/ [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", RFC 2131, March 1997. [RFC3046] Patrick, M., "DHCP Relay Agent Information Option", RFC 3046, January 2001. [RFC3118] Droms, R. and W. Arbaugh, "Authentication for DHCP Messages", RFC 3118, June 2001. [RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C. and M. Carney, "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", RFC 3315, July 2003. [WGS84] US National Imagery and Mapping Agency, "Department of Defense (DoD) World Geodetic System 1984 (WGS 84), Third Edition", NIMA TR8350.2, January 2000, https://www1.nga.mil/PRODUCTSSERVICES/GEODESYGEOPHYSICS/ WORLDGEODETICSYSTEM/Pages/default.aspx and http://www.ngs.noaa.gov/faq.shtml#WGS84 6.2. Informational References [GeoShape] Thomson, M. and C. Reed, "GML 3.1.1 PIDF-LO Shape Application Schema for use by the Internet Engineering Task Force (IETF)", Candidate OpenGIS Implementation Specification 06-142, Version: 0.0.9, December 2006. Polk, et al. Standards Track [Page 15] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 [IEEE-802.11y] Information technology - Telecommunications and information exchange between systems - Local and metropolitan area networks - Specific requirements - Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) specifications Amendment 3: 3650-3700 MHz Operation in USA, November 2008. [NENA] National Emergency Number Association (NENA) www.nena.org NENA Technical Information Document on Model Legislation Enhanced 911 for Multi-Line Telephone Systems. [RFC3825] Polk, J., Schnizlein, J. and M. Linsner, "Dynamic Host Configuration Protocol Option for Coordinate-based Location Configuration Information", RFC 3825, July 2004. [RFC4119] Peterson, J., "A Presence-based GEOPRIV Location Object Format", RFC 4119, December 2005. [RFC4776] Schulzrinne, H., "Dynamic Host Configuration Protocol (DHCPv4 and DHCPv6) Option for Civic Addresses Configuration Information", RFC 4776, November 2006. [RFC5139] Thomson, M. and J. Winterbottom, "Revised Civic Location Format for Presence Information Data Format Location Object (PIDF-LO)", RFC 5139, February 2008. [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an IANA Considerations Section in RFCs", RFC 5226, May 2008. Appendix A. Calculations of Resolution The following examples for two different locations demonstrate how the Resolution values for Latitude, Longitude, and Altitude (used in the version 0 DHCPv4 option) can be used. In both examples, the geo-location values were derived from maps using the WGS84 map datum, therefore in these examples, the Datum field would have a value = 1 (00000001, or 0x01). A.1. Location Configuration Information of "White House" (Example 1) The address was NOT picked for any political reason and can easily be found on the Internet or mapping software, but was picked as an easily identifiable location on our planet. Postal Address: White House 1600 Pennsylvania Ave. NW Polk, et al. Standards Track [Page 16] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 Washington, DC 20006 Standing on the sidewalk, north side of White House, between driveways. Latitude 38.89868 degrees North (or +38.89868 degrees) Using 2s complement, 34 bit fixed point, 25 bit fraction Latitude = 0x04dcc1fc8, Latitude = 0001001101110011000001111111001000 Longitude 77.03723 degrees West (or -77.03723 degrees) Using 2s complement, 34 bit fixed point, 25 bit fraction Longitude = 0xf65ecf031, Longitude = 1101100101111011001111000000110001 Altitude 15 In this example, we are not inside a structure, therefore we will assume an altitude value of 15 meters, interpolated from the US Geological survey map, Washington West quadrangle. AltUnc = 30, 0x1e, 011110 AT = 1, 0x01, 000001 Altitude = 15, 0x0F00, 00000000000000000000000001111100000000 If: LatUnc is expressed as value 2 (0x02 or 000010) and LongUnc is expressed as value 2 (0x02 or 000010), then it would describe a geo-location region that is north of the equator and extends from -1 degree (west of the meridian) to -128 degrees. This would include the area from approximately 600km south of Saltpond, Ghana, due north to the North Pole and approximately 4400km south-southwest of Los Angeles, CA due north to the North Pole. This would cover an area of about one-sixth of the globe, approximately 20 million square nautical miles (nm). If: LatUnc is expressed as value 3 (0x03 or 000011) and LongUnc is expressed as value 3 (0x03 or 000011), then it would describe a geo-location area that is north from the equator to 63 degrees north, and -65 degrees to -128 degrees longitude. This area includes south of a line from Anchorage, AL to eastern Nunavut, CN, and from the Amazons of northern Brazil to approximately 4400km south-southwest of Los Angeles, CA. This area would include North America, Central America, and parts of Venezuela and Columbia, except portions of Alaska and northern and eastern Canada, approximately 10 million square nm. If: LatUnc is expressed as value 5 (0x05 or 000101) and LongUnc is expressed as value 5 (0x05 or 000101), then it would describe a Polk, et al. Standards Track [Page 17] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 geo-location area that is latitude 32 north of the equator to latitude 48 and extends from -64 degrees to -80 degrees longitude. This is approximately an east-west boundary of a time zone, an area of approximately 700,000 square nm. If: LatUnc is expressed as value 9 (0x09 or 001001) and LongUnc is expressed as value 9 (0x09 or 001001), which includes all the integer bits, then it would describe a geo-location area that is latitude 38 north of the equator to latitude 39 and extends from -77 degrees to -78 degrees longitude. This is an area of approximately 9600 square km (111.3km x 86.5km). If: LatUnc is expressed as value 18 (0x12 or 010010) and LongUnc is expressed as value 18 (0x12 or 010010), then it would describe a geo-location area that is latitude 38.8984375 north to latitude 38.9003906 and extends from -77.0390625 degrees to -77.0371094 degrees longitude. This is an area of approximately 36,600 square meters (169m x 217m). If: LatUnc is expressed as value 22 (0x16 or 010110) and LongUnc is expressed as value 22 (0x16 or 010110), then it would describe a geo-location area that is latitude 38.896816 north to latitude 38.8985596 and extends from -77.0372314 degrees to -77.0371094 degrees longitude. This is an area of approximately 143 square meters (10.5m x 13.6m). If: LatUnc is expressed as value 28 (0x1c or 011100) and LongUnc is expressed as value 28 (0x1c or 011100), then it would describe a geo-location area that is latitude 38.8986797 north to latitude 38.8986816 and extends from -77.0372314 degrees to -77.0372296 degrees longitude. This is an area of approximately 339 square centimeters (20.9cm x 16.23cm). If: LatUnc is expressed as value 30 (0x1e or 011110) and LongUnc is expressed as value 30 (0x1e or 011110), then it would describe a geo-location area that is latitude 38.8986797 north to latitude 38.8986802 and extends from -77.0372300 degrees to -77.0372296 degrees longitude. This is an area of approximately 19.5 square centimeters (50mm x 39mm). If: LatUnc is expressed as value 34 (0x22 or 100010) and LongUnc is expressed as value 34 (0x22 or 100010), then it would describe a geo-location area that is latitude 38.8986800 north to latitude 38.8986802 and extends from -77.0372300 degrees to -77.0372296 degrees longitude. This is an area of approximately 7.5 square millimeters (3.11mm x 2.42mm). In the (White House) example, the requirement of emergency responders Polk, et al. Standards Track [Page 18] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 in North America via their NENA Model Legislation [NENA] could be met by a LatUnc value of 21 and a LongUnc value of 20. This would yield a geo-location that is latitude 38.8984375 north to latitude 38.8988616 north and longitude -77.0371094 to longitude -77.0375977. This is an area of approximately 89 feet by 75 feet or 6669 square feet, which is very close to the 7000 square feet requested by NENA. In this example, a service provider could enforce that a device send a Location Configuration Information with this minimum amount of resolution for this particular location when calling emergency services. A.2. Location Configuration Information of "Sears Tower" (Example 2) Postal Address: Sears Tower 103rd Floor 233 S. Wacker Dr. Chicago, IL 60606 Viewing the Chicago area from the Observation Deck of the Sears Tower. Latitude 41.87884 degrees North (or +41.87884 degrees) Using 2s complement, 34 bit fixed point, 25 bit fraction Latitude = 0x053c1f751, Latitude = 0001010011110000011111011101010001 Longitude 87.63602 degrees West (or -87.63602 degrees) Using 2s complement, 34 bit fixed point, 25 bit fraction Longitude = 0xf50ba5b97, Longitude = 1101010000101110100101101110010111 Altitude 103 In this example, we are inside a structure, therefore we will assume an altitude value of 103 to indicate the floor we are on. The Altitude Type value is 2, indicating floors. The AltUnc field would indicate that all bits in the Altitude field are true, as we want to accurately represent the floor of the structure where we are located. AltUnc = 30, 0x1e, 011110 AT = 2, 0x02, 000010 Altitude = 103, 0x00006700, 000000000000000110011100000000 For the accuracy of the latitude and longitude, the best information available to us was supplied by a generic mapping service that shows a single geo-loc for all of the Sears Tower. Therefore we are going to show LatUnc as value 18 (0x12 or 010010) and LongUnc as value 18 (0x12 or 010010). This would be describing a geo-location area that Polk, et al. Standards Track [Page 19] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 is latitude 41.8769531 to latitude 41.8789062 and extends from -87.6367188 degrees to -87.6347657 degrees longitude. This is an area of approximately 373412 square feet (713.3 ft. x 523.5 ft.). Appendix B. Calculations of Uncertainty The following example demonstrates how Uncertainty values for Latitude, Longitude, and Altitude (used in the DHCPv6 Option as well as the version 1 DHCPv4 option) can be calculated. B.1 Location Configuration Information of "Sydney Opera House" (Example 3) This section describes an example of encoding and decoding the geodetic DHCP Option. The textual results are expressed in GML [OGC.GML-3.1.1] form, suitable for inclusion in PIDF-LO [RFC4119]. These examples all assume a datum of WGS84 (datum = 1) and an altitude type of meters (AT = 1). B.1.1. Encoding a Location into DHCP Geodetic Form This example draws a rough polygon around the Sydney Opera House. This polygon consists of the following six points: 33.856625 S, 151.215906 E 33.856299 S, 151.215343 E 33.856326 S, 151.214731 E 33.857533 S, 151.214495 E 33.857720 S, 151.214613 E 33.857369 S, 151.215375 E The top of the building 67.4 meters above sea level, and a starting altitude of 0 meters above the WGS84 geoid is assumed. The first step is to determine the range of latitude and longitude values. Latitude ranges from -33.857720 to -33.856299; longitude ranges from 151.214495 to 151.215906. For this example, the point that is encoded is chosen by finding the middle of each range, that is (-33.8570095, 151.2152005). This is encoded as (1110111100010010010011011000001101, 0100101110011011100010111011000011) in binary, or (3BC49360D, 12E6E2EC3) in hexadecimal notation (with an extra 2 bits of leading padding on each). Altitude is set at 33.7 meters, which is 000000000000000010000110110011 (binary) or 000021B3 (hexadecimal). The latitude uncertainty is given by inserting the difference between Polk, et al. Standards Track [Page 20] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 the center value and the outer value into the formula from Section 2.3.1. This gives: x = 8 - ceil( log2( -33.8570095 - -33.857720 ) ) The result of this equation is 18, therefore the uncertainty is encoded as 010010 in binary. Similarly, longitude uncertainty is given by the formula: x = 8 - ceil( log2( 151.2152005 - 151.214495 ) ) The result of this equation is also 18, or 010010 in binary. Altitude uncertainty uses the formula from Section 2.4.4: x = 21 - ceil( log2( 33.7 - 0 ) ) The result of this equation is 15, which is encoded as 001111 in binary. Adding an Altitude Type of 1 (meters) and a Datum of 1 (WGS84), this gives the following DHCPv4 form: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Code (123) | OptLen (16) | LatUnc | Latitude . |0 1 1 1 1 0 1 1|0 0 0 1 0 0 0 0|0 1 0 0 1 0|1 1 1 0 1 1 1 1 0 0. +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ . Latitude (cont'd) | LongUnc | . .0 1 0 0 1 0 0 1 0 0 1 1 0 1 1 0 0 0 0 0 1 1 0 1|0 1 0 0 1 0|0 1. +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ . Longitude (cont'd) | .0 0 1 0 1 1 1 0 0 1 1 0 1 1 1 0 0 0 1 0 1 1 1 0 1 1 0 0 0 0 1 1| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AType | AltUnc | Altitude . |0 0 0 1|0 0 1 1 1 1|0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1. +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ . Alt (cont'd) | Datum | .1 0 1 1 0 0 1 1|0 1 0 0 0 0 0 1| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ In hexadecimal, this is 7B104BBC 49360D49 2E6E2EC3 13C00021 B341. The DHCPv6 form only differs in the code and option length portion. B.1.2. Decoding a Location from DHCP Geodetic Form If receiving the binary form created in the previous section, this section describes how that would be interpreted. The result is then represented as a GML object, as defined in [GeoShape]. A latitude value of 1110111100010010010011011000001101 decodes to a value of -33.8570095003 (to 10 decimal places). The longitude value Polk, et al. Standards Track [Page 21] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 of 0100101110011011100010111011000011 decodes to 151.2152005136. Decoding Tip: If the raw values of latitude and longitude are placed in integer variables, the actual value can be derived by the following process: 1. If the highest order bit is set (i.e. the number is a twos complement negative), then subtract 2 to the power of 34 (the total number of bits). 2. Divide the result by 2 to the power of 25 (the number of fractional bits) to determine the final value. The same principle can be applied when decoding altitude values, except with different powers of 2 (30 and 8 respectively). The latitude and longitude uncertainty are both 18, which gives an uncertainty value using the formula from Section 2.3.1 of 0.0009765625. Therefore, the decoded latitudes is -33.8570095003 +/- 0.0009765625 (or the range from -33.8579860628 to -33.8560329378) and the decoded longitude is 151.2152005136 +/- 0.0009765625 (or the range from 151.2142239511 to 151.2161770761). The encoded altitude of 000000000000000010000110110011 decodes to 33.69921875. The encoded uncertainty of 15 gives a value of 64, therefore the final uncertainty is 33.69921875 +/- 64 (or the range from -30.30078125 to 97.69921875). B.1.2.1. GML Representation of Decoded Locations The GML representation of a decoded DHCP option depends on what fields are specified. Uncertainty can be omitted from all of the respective fields, and altitude can also be absent. In the absence of uncertainty information, the value decoded from the DHCP form can be expressed as a single point. If the point includes altitude, it uses a three dimensional CRS, otherwise it uses a two dimensional CRS. The following GML shows the value decoded in the previous example as a point in a three dimensional CRS: -33.8570095003 151.2152005136 33.69921875 If all fields are included along with uncertainty, the shape Polk, et al. Standards Track [Page 22] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 described is a rectangular prism. Note that this is necessary given that uncertainty for each axis is provided idependently. The following example uses all of the decoded information from the previous example: -33.8579860628 151.2142239511 -30.30078125 -33.8579860628 151.2161770761 -30.30078125 -33.8560329378 151.2161770761 -30.30078125 -33.8560329378 151.2142239511 -30.30078125 -33.8579860628 151.2142239511 -30.30078125 128 Note that this representation is only appropriate if the uncertainty is sufficiently small. [GeoShape] recommends that distances between polygon vertices be kept short. A GML representation like this one is only appropriate where uncertainty is less than 1 degree (an encoded value of 9 or greater). If altitude or altitude uncertainty is not specified, the shape is described as a rectangle using the "gml:Polygon" shape. If altitude is available, a three dimensional CRS is used, otherwise a two dimensional CRS is used. For Datum values of 2 or 3 (NAD83), there is no available CRS URN that covers three dimensional coordinates. By necessity, locations described in these datums can be represented by two dimensional shapes only; that is, either a two dimensional point or a polygon. If the altitude type is 2 (floors), then this value can be represented using a civic address object [RFC5139] that is presented alongside the geodetic object. Polk, et al. Standards Track [Page 23] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 Appendix C. Changes from RFC 3825 Technical changes: -04: Added Appendix B providing an example relating to uncertainty. Added Section 2.3.1 on Latitude and Longitude resolution and Section 2.4.4 on Altitude resolution. Added definition of Resolution to Section 1.2. -03: Clarified potential behavior of version 0 clients receiving a version 1 option and added recommendations for clients and servers. -02: Added Section 1.2 introducing uncertainty and resolution concepts. Added Section 2.1 defining DHCPv6 option format. -01: Within Section 2.1, split Datum field from RFC 3825 into three fields: Ver, Res and Datum fields. Explained that the Ver field is always located at the same offset. Added Section 2.2 relating to Version Support. -00: None Editorial changes: -03: Changed "DHC" to "DHCP" in some usages. Clarified relationship of resolution and uncertainty to privacy. Changed all uses of the LoRes/LaRes/AltRes terminology to LongUnc/LatUnc/AltUnc, and clarified when these parameters were used to encode resolution vs. uncertainty. -02: Reorganized Sections 1 and 2. -01: Added references to IEEE 802.11y, RFC 3825. -00: Changed boilerplate. Added B. Aboba as editor. Re-positioned Appendix A and Acknowledgments sections. Changed reference numbers to names, added reference to RFC 5226 (since RFC 3825 was missing a reference to RFC 2434, now obsolete), updated references (and URLs). Updated author affiliations and email addresses. Changed references to "the appendix" to Appendix A. Added Appendix B listing changes. Polk, et al. Standards Track [Page 24] INTERNET-DRAFT DHCP Option for Coordinate LCI 17 December 2009 Authors' Addresses James M. Polk Cisco Systems 2200 East President George Bush Turnpike Richardson, Texas 75082 USA USA EMail: jmpolk@cisco.com John Schnizlein Technology Program Manager Internet Society 1775 Wiehle Avenue Suite 201 Reston, VA 20190-5108 USA USA EMail: schnizlein@isoc.org Marc Linsner Cisco Systems Marco Island, FL 34145 USA USA EMail: marc.linsner@cisco.com Martin Thomson Andrew PO Box U40 Wollongong University Campus, NSW 2500 AU EMail: martin.thomson@andrew.com Bernard Aboba Microsoft Corporation One Microsoft Way Redmond, WA 98052 USA USA EMail: bernarda@microsoft.com Polk, et al. Standards Track [Page 25]