Culled satellite ephemeris information based on limiting a span of an inverted cone for locating satellite in-range determinations

[0001] This application claims priority from U.S. Provisional Application 60/618,606, filed Oct. 15, 2004, the entirety of which is expressly incorporated herein by reference.

[0002] 1. Field of the Invention

[0003] This invention relates generally to wireless and long distance carriers, Internet service providers (ISPs), and information content delivery services/providers and long distance carriers. More particularly, it relates to location services for the wireless industry.

[0004] 2. Background of Related Art

[0005] It is desired to accurately locate cell phones within a cellular network. While there are several techniques for determining location in a mobile device, a future generation of mobile phones may include a global positioning satellite (GPS) receiver chipset, thus having the ability to locate itself via GPS.

[0006] FIG. 10 depicts the conventional Global Positioning Satellite system including about 24 or more GPS satellites.

[0007] In particular, as shown in FIG. 10, the earth 200 is surrounded by approximately 24 GPS satellites 101–124, which each have their own rotational orbit about the earth 200. There are currently about 24 to 27 GPS satellites in the GPS network, each moving about the earth approximately 6 times each day.

[0008] Unfortunately, as the phone moves about the country, locations with respect to satellites change. Thus, GPS devices attempting to determine their position with respect to the earth 200 will only be able to communicate with a smaller number of the total GPS satellites at any one time.

[0009] The time required for lock in and location determination by a conventional GPS receiver in determining which of the GPS satellites in the GPS network takes several minutes, and as many as 5 or 6 minutes for a standard GPS receiver, which is longer than many if not most phone calls.

[0010] There is a need for a less cumbersome and more efficient technique for using GPS location information in a highly mobile and fast paced society.

[0011] In accordance with the principles of the present invention, a plurality of locating satellites are culled into a sub-plurality for communication by a plurality of wireless devices. A first inverted cone is defined above a first site, the first inverted cone having a first base angle. A second inverted cone is defined above a second site, the second inverted cone having a second base angle. If the second site is farther from an equator of Earth than the first site, then the second inverted cone is made to have a base angle larger than a base angle of the first inverted cone. If the first site is farther from the equator of Earth than the second site, then the first inverted cone is made to have a base angle larger than a base angle of the second inverted cone.

[0012] Any sites used as reference points other than the first or second will have inverted cones defined above them with base angles greater than, equal to, or less than the inverted cones above the first and second sites depending on whether those other reference points are farther from the equator, at the same distance from the equator, or closer to the equator than either the first or second site.

[0026] In a conventional GPS system, ephemeris data is provided to each GPS receiver to keep track of where each of the satellites in the GPS satellite network should be located in space. As part of the location determination process, each GPS receiver determines which of approximately 24 or more GPS satellites are to be used to determine GPS location. This determination requires a significant amount of real-time processing at the time of the location request, and a significant amount of time.

[0027] FIG. 1 shows a base station 150 of a wireless network (e.g., a cellular telephone network) determining which GPS satellites 101–124 are in a preferred group, e.g., being within a cone of space 217 with the longest dwell time within that space, in accordance with the principles of the present invention.

[0028] In particular, as shown in FIG. 1, of the twenty four or so GPS satellites 101–124 in the GPS system, only a sub-set 107–114 are visible to the base station 150 at any one time. Thus, based on ephemeris data, the satellites communicating with a subscriber or user within a service range of the base station 150 may be culled to only those visible, e.g., GPS satellites 107–114.

[0029] As a further culling, only those GPS satellites 109–113 currently within a cone of space 217 above the relevant base station 150 might be selected for communication with a receiver or subscriber within the service area of the relevant base station 150.

[0030] As an ultimate culling, a minimum set of GPS satellites 109–112 may be selected based on, e.g., being not only within an arbitrary cone of space 217 normal to the base station 150, but also projected to remain within that cone of space 217 for the longest period of time, i.e., having the longest dwell time. Thus, GPS satellite 113 may be eliminated or culled from the minimum set of GPS satellites as it has almost completely passed through the cone of space 217, and will have the shortest dwell time of all GPS satellites within the cone of space 217.

[0031] Ideally, the cone of space 217 will be defined sufficiently large to contain at least four GPS satellites at any one time. Alternatively, if fewer than the minimum GPS satellites are within the cone of space 217, then alternative selection criteria may be employed until sufficient number of GPS satellites enter the cone of space 217. For instance, a GPS satellite being closest to the cone of space may be utilized.

[0032] Updated ephemeris data is typically transmitted for one GPS satellite each second. Thus, for a GPS network of, e.g., 24 satellites, updated ephemeris data for all GPS satellites will have been transmitted after 24 seconds. If the network is larger, e.g., 27 GPS satellites, all ephemeris data will be transmitted after 27 seconds. Preferably, the satellites will be periodically culled in accordance with the principles of the present invention based on the provision of updated ephemeris data, e.g., once every 24 seconds, once every 27 seconds, etc.

[0033] In accordance with the principles of the present invention, the total number of available GPS satellites 101–124 is centrally culled for each service location, e.g., for each base station. The culling may be graduated, e.g., for each base station. The culling may be graduated, e.g., first down to those GPS satellites 107–114 that are visible, and then down to a preferred minimum group (e.g., four (4) GPS satellites) having the longest dwell time for use by the particular cell site, or the culling may be to some desired number of satellites all of which are annotated with an order of precedence indicator with which the mobile device can tell which is best, which is second best, etc. Of course, the culling may simply cull to a desired level (e.g., to a minimum of three or four GPS satellites within the cone of space and having the longest dwell time) without graduation or indication of precedence.

[0034] When needed, the selected GPS satellites for a particular region at any particular time of request will be passed to requesting mobile devices to help it determine its own location. For instance, each operative mobile unit is preferably periodically updated with a revised list of selected GPS satellites resulting from the culling of the GPS satellites, e.g., once after each new updated culled list of satellites is determined. The information provided to each subscriber upon request preferably contains the identity of those GPS satellites that are selected for communication. However, information provided in the reverse sense is also within the scope of the present invention (e.g., a list of GPS satellites NOT to communicate with).

[0035] A wireless network may generate a flush of updated culled GPS satellite information periodically, e.g., every 24 seconds. Note that based on the positions of the various GPS satellites 101–124, and in particular based on the positions of the selected GPS satellites 109–112 within the cone of space 217, the list of selected GPS satellites may or may not change.

[0036] Preferably, network traffic will be minimized by reducing or eliminating redundant GPS satellite information. Thus, in a preferred embodiment, GPS satellite list updating messages are sent only when a change in the list has occurred.

[0037] FIG. 2 shows a top view of the cone of space 217 as shown in FIG. 1.

[0038] In particular, as shown in FIG. 2, a cone of space 217 is viewed from space normal to a base station 150. On the ground, the base station 150 has a service region 151. The circular representation of the cone represents a plane cut across the cone of space 217.

[0039] Within the cone of space 217, GPS satellites 101–124 generally travel from one side to the other. Dwell time is determined based on a distance between the present location of the particular GPS satellite, and the exit edge of the cone of space 217, as well as the rate of speed of the GPS satellite.

[0040] A minimum dwell time may be defined and represented as a edge 218 shown in FIGS. 1 and 2.

[0041] A satellite velocity vector may be determined or predetermined. Satellite velocity vector as used herein refers to a vector whose magnitude is the velocity of the satellite and whose origin is the satellite's current position.

[0042] The satellite's velocity vector may be derived from the ephemeris data describing the satellite's orbit and current position. The satellite's velocity vector provides the satellite's velocity and direction of travel. The satellite's velocity and direction are used to compute the point of intersection of the satellite's orbit and the edge of the cone of visibility 217.

[0043] This computed point of intersection along the satellite's orbit defines the endpoint of an ellipsoid arc from the satellite's current position to the point of intersection. The satellite's velocity and orbital parameters—provided by ephemeris data—can be used to compute the time it will take the satellite to traverse that ellipsoid arc. This time value constitutes the satellite's dwell time within the cone of visibility.

[0044] It should be noted that GPS satellites have many different orbits, and almost never travel in precisely the same direction as other GPS satellites, as depicted by the various directions of the velocity vectors shown in FIG. 2.

[0045] In accordance with the principles of the present invention, a small group of GPS satellites with the longest “cone” dwell times will be selected and some of the other GPS satellites will be “culled”. The longest dwell time relates to the amount of time that a calculated GPS satellite vector will be within a respective cone of space 217 above a particular region of users, e.g., above a particular base station 150.

[0046] The cone of space 217 may be simply a cone of visibility above the base station 150, or may be more narrowly defined than visibility.

[0047] The resultant list of selected GPS satellites is preferably periodically and continually updated for any particular location, e.g., base station 150, as the GPS satellites 101–124 rotate about the earth. Updated selected GPS satellite lists are preferably communicated to the subscriber's mobile device (or other suitable application location) for their respective calculations of accurate location information. With the use of selected GPS satellites only by culling out unnecessary or unseen GPS satellites, the total time required for a mobile phone to locate itself can be reduced significantly, e.g., from minutes to just seconds.

[0048] FIG. 3 shows vector calculations for each GPS satellite 101–124, or for each of a preferred set of GPS satellites (e.g., those visible to the base station), in accordance with the principles of the present invention.

[0049] In particular, as shown in FIG. 3, conventional GPS ephemeris data is formatted as RINEX 2.10 data, which is provided by conventional GPS ephemeris data vendors to a Gateway Mobile Location Center (GMLC) and/or mobile position center (MPC). In the preferred embodiment, a Xypoint Location Platform (XLP) was used. In accordance with the principles of the present invention, received RINEX data is converted into an Earth Center (EC) position vector (i.e., a vector pointing to the satellite's current position whose origin is the center of the Earth). Then, using the Earth Center position vectors, the available GPS satellites are culled such that only desired GPS satellites are communicated with (i.e., those that will be in the cone of space 217 the longest).

[0050] In the disclosed embodiments, an Earth Center position vector is computed (or pre-computed) for every cell site 150 in a cellular network. The cell site's EC position vector can be subtracted from the GPS satellite's EC position vector to arrive at a vector that points from the cell site 150 to the particular GPS satellite 101–124. The resulting vector can be divided by its own magnitude to generate a unit vector that points from cell site 150 toward the particular GPS satellite 101–124. The cell site's EC position vector can similarly be divided by its own magnitude to generate a unit vector that points straight up from the cell site 150 (also pre-computed).

[0051] The dot product of the GPS satellite pointing unit vector and the vertical unit vector yields the cosine of the angle between the two vectors. The cosine of an angle of zero degrees yields the value 1.0. The resulting value of the equation “cosine (angle)” diminishes as the angle grows until the cosine of 90 degrees yields the value 0.0. The cosine of angles greater than 90 degrees yield negative results. This makes the cosine of the angle between the satellite pointing unit vector and the vertical unit vector particularly well suited for identifying whether or not the satellite is “visible”. An angular measurement defining a cone of space above the cell site (e.g., a “cone of visibility”) can be pre-computed as a function of the latitude at which the cell site is located. The cosine of this latitude derived angle can be saved. Any satellite whose dot product with the vertical unit vector yields a value greater than or equal to the saved cosine reference value can then be considered “visible”.

[0052] Thus, a rough culling of GPS satellites 101–124, e.g., to only those visible, or even better yet to only those most normal to a base station 150, certainly culling out all GPS satellites that aren't visible at all, and reducing the number of GPS satellites with which to communicate to some small number, each of which is annotated with its order of precedence. This helps keep the time needed to determine location short by providing “backup” satellites with which to communicate just in case local topography blocks the signal from a satellite with a longer dwell time (e.g., a satellite with a lower “order of precedence” value).

[0053] FIG. 4 shows an exemplary culled GPS satellite table 100, in accordance with the principles of the present invention.

[0054] In particular, as shown in FIG. 4, the selected group of satellites for any particular cell site may be maintained in a suitable database and/or other table 100, which may then be provided upon request to any particular mobile device within the service area of that particular cell site 150.

[0055] Thus, a small subgroup of GPS satellites having the longest dwell time with respect to a servicing cell site 150 are selected, and maintained in a culled satellite table 100 communicated to all active mobile subscribers (or other grouping of mobile users). Moreover, or alternatively, whenever a mobile device requires ephemeris data, it may request an update to a culled satellite table 100 containing the identity of the preferred satellites with which to use to determine location of the mobile device.

[0056] FIG. 5 shows an alternate example of a selected or culled GPS satellite information table, in accordance with the principles of the present invention.

[0057] In particular, as shown in FIG. 5, a database or table 100b may include information regarding all or most GPS satellites 101–124, with those that are selected for any particular base station 150 being appropriately indicated.

[0058] While the present invention is explained with reference to the use of as few as three (3) or four (4) and as many as 24 or 27 available satellites 101–124, the present invention relates to the use of any number less than all GPS satellites.

[0059] Moreover, while the present invention provides culling of visible satellites, and even to a minimum number of satellites, e.g., down to four from those visible satellites, the resultant number of satellites may be a number greater than or even less than 4, within the principles of the present invention. For instance, if only position is required, only three (3) GPS satellites are required. However, if altitude is also required, four (4) GPS satellites are the minimum required and thus the maximum culling level. Moreover, the use of more than approximately six (6) GPS satellites do not significantly improve the accuracy of the results.

[0060] If a mobile device is unable for some reason to communicate with one or more GPS satellites directed by the culled GPS satellite table or similar information, the mobile device may then attempt to achieve location information in an otherwise conventional manner, e.g., by attempting contact with all GPS satellites.

[0061] The core technology of culling locating satellites to a sub-plurality is disclosed in U.S. Pat. No. 6,650,288, co-owned with the present application. The following sets forth several significant advances to that core technology.

1) Compute Satellite Dwell Time Based on Ellipsoid Arc

[0062] FIGS. 6A and 6B show use of the calculation of an ellipsoid arc path actually traveled by each satellite to determine more accurately the length of time a locating satellite is within range of a particular location, in accordance with another embodiment of the present invention.

[0063] In particular, as described above, a satellite's path may be projected onto a plane that, cutting through the inverted cone over a reference point, forms a circle. As described, the satellite's speed could be used to approximate the satellite's dwell time within the cone. This is a simple function of computing the amount of time it would take the satellite to traverse the chord cutting across the circle as if it were actually traveling in a straight line.

[0064] However, a straight line in the plane cutting across the inverted cone would be an approximation. The satellite isn't actually traveling in a straight line, but rather in an ellipsoid arc.

[0065] In accordance with this embodiment, the dwell time is computed using the satellite's actual ellipsoid arc, as well as the position of the satellite with respect to the Earth (i.e., whether the satellite's altitude is increasing or decreasing) to improve the selection process.

[0066] As shown in FIG. 6A, the orbit of a locating satellite is typically elliptical. The depiction in FIG. 6A is somewhat exaggerated in ellipticity for ease of explanation. In the orbit, the point(s) at which the locating satellite is farthest from Earth is called the apogee of the orbit, while the point(s) at which the locating satellite is closest to the Earth is called the perigee of the orbit.

[0067] The present embodiment appreciates two aspects of the calculation of satellite dwell time that are not detailed in the prior art: the ellipsoid arc path of the locating satellite through the inverted cone (as depicted in FIG. 6B), and the fact that the acceleration of the locating satellite is decreasing as it moves on its orbit beyond perigee but toward apogee, causing a decrease in the speed of the locating satellite with respect to the Earth. As the locating satellite moves in its orbit between apogee and perigee, the gravitational pull of the Earth causes an increasing acceleration, and thus an increasing speed.

[0068] The math involved in calculating the ellipsoid art passing through any given cross section of any given inverted cone (e.g., as shown in FIG. 6B) is conventional, as is calculations of the acceleration and speed of the locating satellite on any given point of its orbit. It is the recognition and application of the actual path of the locating satellite as well as its actual speed, rather than use of a straight line approximation, that is important to the principles of the present embodiment of the invention.

2) Compute the Arc that Defines the Inverted Cone as a Function of Latitude

[0069] FIGS. 7A, 7B(1) and 7B(2) show adjustment of the computation of the arc that defines the inverted cone encompassing culled satellites as a function of latitude, or distance from the equator, in accordance with yet another embodiment of the present invention.

[0070] In particular, as described above, an angle may be defined that ultimately defines a cone above the reference point. In accordance with yet another embodiment, that angle defining the cone may be selected or computed as a function of the reference point's latitude.

[0071] FIG. 7A depicts a cross-sectional view of Earth, with an imaginary base station 707 located longitudinally on the equator. The angle A at the base of the inverted cone defining locating satellites within view is at a minimum.

[0072] The present embodiment establishes inverted cones with a larger angle (B and C, respectively) as compared to the angle A of the inverted cone at the equator. In this embodiment, the larger angles B, C of the inverted cones are established to be in a relationship to the latitude (e.g., the distance from the equator of the Earth) of the relevant base station.

[0073] Thus, the closer the relevant base station is to the equator, the more likely it is that a sufficient number of locating satellites will be visible (e.g., 4 satellites visible). The farther the relevant base station is from the equator, the fewer number of locating satellites would be visible within a same inverted cone.

[0074] Stated differently, the present embodiment establishes an inverted cone having a larger base angle at a base station farther from the equator to encompass the same number of preferred or minimum culled locating satellites as would a base station located more closely to the equator. This is depicted in FIGS. 7B(1) and 7B(2).

[0075] In particular, FIG. 7B(1) shows a cross section of the inverted cone established to define the culling of locating satellites for a base station 707. The base angle of the inverted cone has an angle A.

[0076] In comparison, as shown in FIG. 7B(2), a cross section of the inverted cone established to define the culling of locating satellites for a base station 708 located farther from the equator than was base station 707 shown in FIG. 7B(1). As depicted, at or near the equator the distribution of satellites within the cone of visibility will likely be fairly uniform. However, in locations significantly away from the equator, the distribution of visible satellites within the cone of visibility look tend to be nearer to one boundary or the other, depending on which hemisphere), rather than being uniformly distributed throughout the inverted cone. Thus, as shown in FIG. 7B(2), in a location away from the equator, the visible satellites tend to favor the edge of the inverted cone closest to the equator.

[0077] The present embodiment also appreciates that received signal strength from the locating satellites is best when the locating satellite is directly overhead of the base station (i.e., normal to the Earth at the location of the base station). Thus, the inverted cone of ‘visible’ locating satellites is made narrower when the base station is closer to the equator of the Earth, and made larger when the base station is farther from the equator of the Earth.

[0078] Thus, as appreciated by the present inventors, there are benefits to limiting the span of the inverted cone if one's latitude doesn't force a wider span just to see the GPS satellites.

3) Utilization of Cell Site Antennas as Reference Points

[0079] The embodiments thus far relate to the definition of a full inverted cone above a base station 150 (e.g., having 360 degree coverage). FIG. 8 shows the utilization of reference points for locating satellites, in accordance with a further embodiment of the present invention.

[0080] In particular, an inverted cone 217 is computed above given ‘reference points’ as an estimate for an inverted cone above the relevant wireless device 150. In accordance with a further embodiment, the reference point may be a wireless communications cell site 860 (e.g., cell phone base station antennas). Thus, cell sites 860 may be specifically used as reference points for culling the ephemeris information used to expedite Assisted GPS location determinations.

[0081] In order to enable the mobile handsets (i.e., cell phones, wireless enabled PDAs, etc.) to quickly locate themselves, it is desired to provide them with a good indication of with which satellites to communicate.

[0082] To give the handset the subset of satellites it should communicate with, a rough idea as to the location of all satellites is required. Such a rough idea allows the choice of a subset of the satellites that have a very high probability of being visible to the handset.

[0083] The parent patent (U.S. Pat. No. 6,650,288) refers to the “rough” or “approximate” positions as “reference points”. However, in accordance with this aspect of the invention, it is possible to use any point at all, as long as one is chosen that is easily and quickly picked, and that is also fairly close to the mobile handset.

[0084] Cell sites are particularly well suited for use as reference points for cellular phones because every mobile handset must communicate with a cell site in order to do even basic wireless functions, and all the cell sites have been surveyed such that their locations are known. Surveying the cell sites was necessary in the U.S. in order to provide Phase One (i.e. imprecise) location for enhanced 911 support so that data is readily available. Though Asia's and Europe's Emergency Services mandates are different than that of the United States, the cell sites in those areas are also surveyed so their locations are also available.

[0085] Since cellular phones connect to carrier's networks through cell sites, there is very little additional time necessary to use that cell site's location as a rough approximation of the handset's location for the purposes of choosing a subset of locating satellites.

[0086] In some cases, a cell site may have several sectors but be represented by only one location. In other cases, a cell site may represent only one receiver of a collection of receivers (see FIG. 8). In either case, the mobile handset will be communicating with one and only one of the cell sites, and that cell site will have a surveyed location that will quite nicely approximate the handset's location for the purposes of selecting a subset of locating satellites.

4) Provide More than Four (4) Satellite's Ephemeris Data with an Order of Precedence Indication

[0087] As described above, it is preferred that no more than four (4) satellite's ephemeris data would be delivered to a wireless handset. In this way, using too much Radio Frequency (RF) bandwidth is avoided, while providing the data necessary to achieve a good GPS fix for latitude, longitude, and altitude too.

[0088] Sometimes, more than just four (4) satellite's data is sent to the handset. Sometimes this is done so that the handset can perform its own satellite selection. Sometimes this is done to provide a fallback in case a particular handset is masked and cannot receive the signal of one or more of the primary selections.

[0089] FIGS. 9A(1), 9A(2) and 9B show the use of more than 4 culled satellites, providing the identities of additional satellites for the wireless device to use as backup satellites as necessary (e.g., when a culled, preferred satellite is masked and the wireless device cannot receive the signal of that primary selection), in accordance with still another embodiment of the present invention. Thus, a number ‘N’ (‘N’ being greater than 1 or 2 or 4 or whatever) of locating satellite's ephemeris data may be used, along with a preferred ‘order of precedence’, allowing a mobile device to use the desired minimum number of locating satellites (e.g., four (4) locating satellites) if available, but could fall back as necessary to the use of other locating satellites that are visible but still favorable.

[0090] FIG. 9A(1) shows an exemplary culled locating satellite table including a list of a preferred number of primary satellites (e.g., 4 preferred satellites). Additionally, the exemplary culled locating satellite table includes an identification of at least one or more secondary locating satellites.

[0091] Preferably, a priority is given to each of the secondary locating satellites such that the relevant mobile device will, as necessary, attempt to receive a signal from each of the secondary locating satellites in the prioritized order.

[0092] The mobile device may, depending upon the particular application, attempt contact with all secondary, tertiary, etc. locating satellites listed in the culled locating satellites table 100c. However, it is more preferable, to save network resources and time, that once a sufficient number of locating satellites have been achieved, the need to contact additional locating satellites becomes unnecessary and may be abandoned for that particular locating session.

[0093] FIG. 9A(2) shows an alternative method of culling satellites by assigning a priority to each satellite within a given single inverted cone, based on a remaining dwell time within the inverted cone, in accordance with another aspect of the present invention.

[0094] In particular, all dwell times are simply computed (i.e., durations within the inverted cone of visibility defined over a particular reference point), and then pick ‘N’ number of satellites. The number of satellites picked may be, e.g., five, or six, or nine, or twelve, or whatever, so long as there are more satellites picked then are necessary to determine location as desired.

[0095] These ‘N’ satellites are then annotated with an integer precedence value. The satellite with the longest dwell time would be annotated with a precedence value of ‘1’. The satellite with the second longest dwell time would be annotated with a precedence value of ‘2’. This would give the mobile handsets an easy indication as to which satellites to limit their position determination, and which satellites to attempt to utilize first, second, third, etc.

[0096] The assignment of precedence to satellites allows a mobile handset a viable fallback should one of the preferred four (4) satellites be temporarily obscured while attempting to determine the handset's location. In this case, the handset would timeout while attempting to receive one (or more) of the preferred satellites' signal, and simply move to the satellite in the list with the next lowest precedence indicator.

[0097] FIG. 9B shows an alternate exemplary culled locating satellite information table including visible, culled primary and secondary, and culled preferred (i.e., primary) GPS satellites, in accordance with yet another aspect of the present invention. Thus, a complete picture of all locating satellites that the relevant mobile device may possibly receive a locating signal from is provided to the mobile device, allowing the mobile device itself decide which locating satellites are to be communicated with, and in what order.

[0098] While the invention has been described with reference to the exemplary embodiments thereof, those skilled in the art will be able to make various modifications to the described embodiments of the invention without departing from the true spirit and scope of the invention.