GPS – where is it taking you?
Joseph Mendoza CDS/Muery Services 3411 Magic Dr. San Antonio, TX 78229 U.S.A. Abstract Global Positioning System technology is moving at a breakneck pace. As a result, it seems that everyone has a GPS receiver mounted to the car dashboard, hiking backpack- and even the handlebar of the paperboy’s bicycle. As the cost-to-accuracy gap of GPS closes, what impact will this have on industry grade equipment, methodologies, and cost? This paper will address how GPS works and explain industry terminology (geekspeak) such as PDOP and multi-pathing. It will explain how accuracy levels are determined and the methodologies used to achieve these, such as differential GPS versus corrected signal post-processing applications. Additionally, this paper will discuss how the latest advances in GPS technology impact field data collection, outlining how GPS works and the latest technology trends, as well as discussing when and if GPS is appropriate. Further, a comparison of GPS methodologies and accuracy targets will be included to best evaluate what method will meet the requirements of the end-user without breaking the budget. In short this paper will take you on a whirlwind tour of the GPS world at its current state and what the not-too-distant future will hold for you and your paperboy. Introduction Ever since the first man drew lines in the ground to represent rivers, mountains and lakes, man has been intent on referencing himself to the physical world he lives in. In time, the need, or perhaps more to the point, the desire, to more accurately reference himself to the physical world has driven methods and technology to new levels. With the advent of the Global Positioning System (GPS), man now has an accurate and powerful tool at his fingertips. GPS products now seem to be in use everywhere in a variety of applications as diverse as engineering and surveying to the leisure activities of the weekend backpacker and fisherman. The GPS market is certainly a growing and vast arena that seems to come up with new and varied applications all the time. This prompts the question: where is GPS taking us and how will it affect you? Before addressing that question, let’s look at the history and development of GPS. Note: In an effort to succinctly explain GPS, many scientific details have been left out of this paper. What remains is a general overview with the aim to introduce you to the exciting – and dare I say cool – world of GPS. Brief history, pre-space to satelites As early as 1903, the Russian scientist Konstantin Tsiolkovskiy had proven mathematically the feasibility of using the reactive force that lifts a rocket to eject a vehicle into space above the pull of the earth's gravity. Twenty years later, Romanian-born Hermann Oberth had independently worked out similar formulas. Neither man built a usable rocket to demonstrate the validity of his theories, nor had they so much as mentioned an unmanned artificial satellite. However the groundwork had been laid for just such work. (Green, Constance McLaughlin, and Milton Lomask, 1970) An American by the name of Robert Goddard had a similar vision, and while engaging in postgraduate work at Princeton University before World War I, Goddard demonstrated in the laboratory that rocketry propulsion would function in a vacuum. By 1918 Goddard had successfully developed a solid-fuel ballistic rocket, and by 1926 had successfully launched a rocket propelled by gasoline and liquid oxygen. In 1937 he launched a rocket that reached an altitude of 9,000 feet. Goddard was making great progress, but his work was not followed except by a small community of rocket enthusiasts. (Green, Constance McLaughlin, and Milton Lomask, 1970) In 1943, the Nazi "buzz" bombs and the supersonic "Vengeance" missile – the "V-2s" that rained on London during 1944 and early 1945 – awakened the entire world to the use of rockets as weapons. Soon a good many physicists and military men began to study the work of Robert Goddard with attention. (Green, Constance McLaughlin, and Milton Lomask, 1970) With the practical use of rockets now established, the post-war nations of the United States and the Soviet Union soon turned their attention to two fronts in rocketry: the development of intercontinental ballistic missiles and the development of a rocket capable of launching a satellite into space. The Space Age The Soviet Union launched the first man-made space vehicle, Sputnik (meaning “space companion”), in 1957, and thereby launched us all into the space age along with it. On January 31, 1958, the United States followed suit with the launch of Vanguard 1. (Green, Constance McLaughlin, and Milton Lomask, 1970) Prior to the launch of the Russian satellite, scientists had experimented with bouncing radio waves off of the moon. Therefore, they were eager to study and experiment with a manmade satellite and test their theories concerning tracking via radio waves. By studying the orbit of Sputnik, scientists discovered that it could indeed be tracked by its radio signal. This led to the concept that man could also determine his position on the earth by reading the signal from a satellite or space vehicle given that the precise orbit of the satellite was known. Several programs then began to be discussed in serious terms. The U. S. Navy’s NRL Naval Center for Space Technology (NCST) conceived of the TIMATION (TIMe/navigATION) program in 1964. This program was designed to provide the basis for a navigation system with three-dimensional coverage (longitude, latitude, and altitude) throughout the world. (U.S. Navy, Naval Research Laboratory, 2001) In 1973 the TIMATION program was merged with the Air Force's 621B program to form the Navigation Signal Timing and Ranging Global Positioning System or NAVSTAR GPS program (originally named the Navigation Technology program). The NAVSTAR GPS program is funded and controlled by the U.S. Department of Defense (DOD), and its primary function is to support military operations throughout the world. The (NAVSTAR) GPS program was officially declared fully operational July 17, 1995. (U.S. Navy, Naval Research Laboratory, 2001) The three major segments of GPS - Space, Control, and User The Space Segment An original constellation of 24 satellites in six orbital planes (four in each plane) are used to send coded satellite signals that can be processed in a GPS receiver, enabling the receiver to compute position, velocity and time. The satellites are spaced 60 degrees apart and are positioned at an altitude of 20,200 km (12,552 miles) with a 55-degree inclination. In addition to the 24 satellites in the constellation, three additional satellites are in orbit and will eventually replace older space vehicles. (Dana, Peter H., 2001) Control The Control segment consists of five Monitor Stations (located in Hawaii, Kwajalein, Ascension Island, Diego Garcia, and Colorado Springs), three Ground Antennas, (located at Ascension Island, Diego Garcia, and Kwajalein), and a Master Control Station (MCS) located at Schriever AFB in Colorado. The monitor stations passively track all satellites in view, accumulating ranging data. This data is processed at the MCS and incorporated into satellite orbital models. The updated orbital information, also called ephemeris data, is then transmitted to each satellite via the Ground Antennas and is sent with each satellite’s navigation message. (NASA – Jet Propulsion Laboratory, 2001) User Segment The GPS User Segment consists of the GPS receivers and us, the user community. The 24 satellites in their respective orbits provide the user with five to eight satellites visible any where on the earth to receive data. Satellite signal arrival times from at least four satellites are processed to estimate four quantities, position in three dimensions (X, Y, and Z) and GPS time (T). The receiver then computes position dimensions in Earth-Centered, Earth-Fixed X, Y, Z (ECEF XYZ) coordinates. (Dana, Peter H., 2001) The computations are based on simple principles of velocity x travel time = distance, which is somewhat like the old school math problem “If a train leaves Chicago traveling at speed of 30 miles an hour and travels for two hours, how far did it go?” (Trimble Navigation Limited, 2001) The signals are then processed, not unlike triangulation. One factor that complicates the situation is that the signals are traveling at the speed of light. So the difference between the arrival times of the signals is minute. Since the arrival times of the satellite signals are such a critical factor in calculating a position, each satellite is equipped with four atomic clocks, two cesium and two rubidium. Satellite clocks are monitored by Ground Control Stations and occasionally reset to maintain time to within one-millisecond of GPS time. This information is then transmitted to each satellite via the Ground Antennas and is sent with each satellite’s navigation message along with its ephemeris data. (Dana, Peter H., 2001) The Signal The satellites transmit information via two radio waves that can be picked up by GPS receivers. Each radio wave is modulated so that it can carry specific information. The modulated signal resembles random electrical noise, but since the signal is not random but coded (and therefore follows a pattern), it is referred to as a pseudorandom code. The radio waves, which carry the pseudorandom codes, are distinguished by the designations of L1 and L2, and each carries different information in its modulated code. L1 carries the Coarse Acquisition (C/A) Code used by the civilian sector (free of charge) and is also modulated to carry the Navigation Message and other satellite system parameter information. L2 carries the Precise Code (P-Code) used by the military. The P-Code is encrypted and can only be received by specific receivers equipped with key codes used to decipher the signal. (Dana, Peter H., 2001) The pseudorandom code for each satellite is distinct, which makes it easy for GPS receivers to distinguish between one satellite and another. In this way GPS receivers can tell exactly which satellites make up a given configuration. This is important since the signals are very weak. So a GPS receiver identifies one signal and, using built in almanacs, actually searches for signals from the other satellites it thinks should be in the configuration. Once it has identified all of the satellites in the configuration it then begins tracking their signals. The GPS receiver then mimics or mirrors the pseudorandom code for each of the satellites and compares the differences between its own code and the one received. It is able to do this because it knows the fluctuations in the pseudorandom code. It then matches up one known point in the signal received and its own and begins to make calculations. To illustrate how this works, imagine that a satellite was playing Iron Butterfly’s “In-A-Gadda-Da- Vida” from space. At the exact same time, you are sitting in a lounge chair also listening to “In-A-Gadda-Da-Vida”. As you listened to the version you are playing against the one from space, you would notice that the version from space was delayed slightly. This is because it takes some time to travel the distance from the satellite in space to your lazy-boy. To determine the distance, you could slow your version to match the one from space (which is hard to do on a 45rpm) until they were synchronized. Since the time shift between the two versions of “In-A-Gadda- Da-Vida” is equal to the travel time of the satellites version, we simply take the time shift between versions and multiply it by the speed of light and presto, we determine the distance traveled! (Trimble Navigation Limited, 2001) This same calculation is then made for each signal received and used to pinpoint a location. A word about Accuracy Several factors will affect the accuracy of your readings. Visibility or line of sight is crucial, since at least four satellites are needed to accurately locate your position. Buildings, mountains and even tree canopy can affect how many satellites you are “seeing” and may prevent you from “seeing” enough satellites to use in deciphering your location. Since the pseudorandom signal sent by satellites resembles electrical noise, receivers at times actually have trouble distinguishing the signal from “true noise” in space caused by solar flares or other naturally occurring events. Good receivers are better equipped to decipher the noise and filter it out. Nevertheless it is a distorting factor. (Dana, Peter H., 2001) An event known as multi-pathing may give you false readings via signal reflection. Multi-pathing occurs when a nearby object or surface is reflecting or bouncing the satellite signal to your receiver, making it think that it is the true line of sight reading. The reflected signal is received and is computed as a real signal and causes an effect similar to the ghosting of your TV screen. This occurrence is not likely to happen on the open sea but may be experienced in city locations where there are many surfaces. Multi-pathing can be hard to detect or even avoid, so good receivers are equipped to try and detect and then reject the reflected signal when multi-pathing occurs. Variances in the atmosphere may also cause distortions in your readings. Changes in temperature, pressure, and humidity affect the troposphere or the lower part of the atmosphere, which can delay readings. Imagine a glass of water with a spoon sitting in it. The portion of the spoon in the water appears distorted in relation to the portion out of the water. This is because light is slowed ever so slightly as it travels through the water distorting the image. This same effect occurs as the signal sent by a satellite is slowed as it passes through the water vapor in the air. In addition, delays can occur in the ionosphere, which consists of charged air particles 50 to 500 km in the atmosphere. These delays, although slight, are significant enough to effect calculating a good fixed coordinate. A term that you may hear in reference to GPS accuracy is GDOP or Geometric Precision of Dilution. GDOP is made up of other components such as PDOP (Position Dilution of Precision), HDOP (Horizontal Dilution of Precision), VDOP (Vertical Dilution of Precision), and TDOP (Time Dilution of Precision). Even though each of these can be calculated independently, they all make for a good or bad GDOP reading. (Dana, Peter H., 2001) To illustrate, you may be receiving signals from four satellites that all happen to be right on top of you or all in a straight line in front of you on the horizon. (This configuration is not possible but is used as an example to exaggerate the point). Therefore, even though you have four readings, the configuration of the satellites does not allow enough variance between the angles of the readings to obtain a good PDOP, which would require a better spread in the satellite configuration. Good receivers will automatically pick out the best satellites from a given constellation makeup to give you the best PDOP, which will make for a good GDOP. Another factor that cannot be overlooked is Dick Clark’s Law of Goofs and Blunders. This law states that not even GPS is immune to the occasional software glitch, hardware malfunction or good ole’ operator error, which might make for good video clips but bollixes your GPS data. In a perfect world, coordinates can be fixed from just three satellites. This would give us all the needed data to calculate accurately any given position on the earth. As we have just discussed, however, there are many factors that can delay, distort or mirror signals received giving our GPS units fits trying to filter out erroneous data. That is why signals from four satellites are used. The extra measurement helps our receivers to verify against the fourth signal how well it is computing our position. This increases our chances of obtaining an accurate measurement in fixing our position. Differential GPS The purpose of Differential GPS is to correct errors that may creep in due to numerous factors. It accomplishes this by taking satellite readings at a known fixed location. It then takes where the satellite tells it that it is located and compares that against where it knows it is located and computes an error calculation. That data is then passed onto the roving (or differential) receiver and used to correct for errors. This type of DGPS is known as “real time” DGPS and requires that the receiver be outfitted to receive and process this information. How well this process works is dependent on the quality of the receivers and the distance between the two points. The range can be anywhere between 30 to 200 kms. The concept is predicated on the fact that (to the degree possible) the atmospheric conditions are alike at each location. In instances where “real time” is not a critical factor, the fixed location readings are collected and processed by computer against the points taken by the roving receiver at a later time. This is known as “post-processing” and can only be done if the fixed location is taking readings at the exact time that the rover is also taking readings. If a field technician is taking readings and the fixed station happens to be down during that time, all of the readings taken will be useless since there is no fixed location information to be processed against the collected points. The user and his uses Military Support of military operations is the first and primary function of GPS as funded by the DOD. At one time, the DOD degraded GPS signals received by non-military personnel, thereby reducing accuracy. This effort was known as Selective Availability (SA) or anti-spoofing. The primary intent behind the implementation of SA was the belief that the signals could be easily used by enemy states against the US. (U.S. Navy, Naval Research Laboratory, 2001) However, on May 1, 2000, the President of the United States disabled SA with the intent of never using it again. (Interagency GPS Executive Board (IGEB) 2001) Of course, if circumstances warrant, it is conceivable that SA could be reactivated. Commercial The commercial sector is laden with GPS applications, and GPS receivers can be found in fire engines, police cars, airplanes, helicopters, boats, construction equipment, movie cameras, fleet vehicles, and farm equipment, just to name a few, and are used for surveying, construction, mapping, agriculture, vehicle theft tracking, digital information transfer security, environmental, archeological and other uses. Our transportation infrastructure, for instance, is heavily dependent upon GPS for tracking and navigation of airplanes, marine traffic, and land vehicles such as trains. Commercial Accuracy Targets The accuracy levels of commercial applications vary, although the disabling of SA has helped by making higher accuracy levels more obtainable in general. In comparison, the tracking of large ships in the ocean does not warrant the same accuracy as does surveying in pins and markers for an interstate bridge with costly pre-cast concrete sections. Therefore, accuracy is dependent upon use, and there is a direct correlation between level of accuracy and price. The greater the accuracy required, the more expensive the equipment, and also the more expensive the training and maintenance cost. Some commercial receivers can be purchased for a couple of thousand dollars with an accuracy from one to three meters (dependent upon whether the points are using correction data), while survey grade equipment may run $5,000 to $40,000 per receiver (and generally two are required) and have an accuracy of just a few centimeters for DGPS and less than a centimeter in static mode. (Dana, Peter H., 2001) Survey grade equipment can achieve this type of accuracy by a method called Carrier Phase Tracking. In the “In-A-Gadda-Da-Vida” illustration earlier, we discussed how the pseudorandom code signal from the satellite is matched to the code generated by the receiver to determine time traveled and calculate distance. By using the pseudorandom code in this manner, an accuracy of a few meters can be achieved. By use of pseudo random code correction DGPS methods (mapping grade), an accuracy of +-1 meter can be achieved. Survey grade GPS projects usually carry an accuracy requirement of better than a few centimeters. In order to accomplish this high accuracy, two receivers are used in the same manner as DGPS. But rather than comparing the pseudorandom code, surveyors compare the carrier (radio) wave itself. This results in higher accuracy since the carrier wave oscillates at a higher (or faster) frequency than the pseudorandom code. This means that in the same time frame, there are more points to match against the carrier wave and less dead time between those points. By means of this process, a higher accuracy can be achieved. This method is now commonly used along with other standard survey methods for the creation of landbase features with a high accuracy, such as those used by taxation offices, city planning departments and other high accuracy uses. (Dana, Peter H., 2001) GPS and GIS GIS data creation methods and software have significantly improved over the last ten years. Years ago, the standard method of data input and creation was a map or orthophoto registered to a digitizing table. Today, data is often geo-referenced to a vector land base or digital ortho, a method generally referred to as heads up digitizing. The trend now is to incorporate GPS data collected by field crews into the GIS system, which is facilitated by enhancements to GIS software that make integration of GPS data much easier. This means that many features are created in the field by technicians equipped with GPS units. This data is then sent to in-house technicians and made ready for delivery. This type of field data collection is used to collect various types of data such as highway signage, electrical structures, telephony equipment sites, and manholes. For capturing features with limited attributes, a data cap is used to store the GPS point and other related attributes. For features with a more robust attribute requirement, a computer pen-based unit is coupled with the GPS receiver and used to capture features and their associated attributed information. This latter method has the advantage that once a point is captured (assuming that you are using DGPS), it can be viewed with its associated land base backdrop for reference. The drawback is that a computer pen-based unit increases the cost to equip each technician by several thousand dollars. The additional cost that GPS equipment places on field data collection can make it a cost prohibitive endeavor. In addition to training costs and time, one also needs to consider the added effort that GPS places on a field technician resulting in reduced productivity, which may have an impact on the overall project schedule. To Use GPS or Not? Aside from these factors, which may be prohibitive, not all field data collection projects necessarily need to collect GPS data. Some companies simply do not have a need to place features with such precision. In truth, the addition of precisely located data may jumble an already working system. For instance, a company may be using a vector land base built upon USGS quad maps with a relative accuracy of +-50 feet (or an off-the-shelf land base) that has been modified over time, making it difficult to upgrade to a more accurate land base. The placement of highly accurate features into this system may be confusing and difficult to work with. For instance, let’s say that a cable company has installed a new pedestal, and a real world coordinate has been captured for the new feature. This pedestal is now added to the GIS system and placed according to its real world coordinate. The inconsistency of the land to the GPS-located pedestal may make it appear that the pedestal is on the opposite side of the street, or it may place it in an adjacent lot to the one in which it actually resides. The confusion caused by its seemingly improper placement may make it difficult to work with. In a situation like this, the decision may be not to use GPS at all. How Accurate it Accurate Enough? When used, the requirements of GPS-collected data for a GIS system vary. Generally, unless you are surveying, sub-meter and centimeter accuracy is not warranted. In most cases, one to three meters will suffice for most GIS applications today. For instance, field inventory of streetlight poles could easily be accomplished with an accuracy of one to three meters, depending on the accuracy of the land base. It stands to reason that a GPS point representing a streetlight can be navigated to within a few meters and then should be visible to the technician trying to locate it. Another factor to keep in mind is that the symbology used in a GIS system is usually substantially larger in scale than what is being represented and can change depending on the scale viewed. Many streetlight and utility poles are modeled with a symbol that is ten feet in diameter at its largest scale. Additionally, a precise absolute point may look good in GIS, but a series of precise absolute points may not. Therefore, precise GPS data may be moved in a GIS for cartographic reasons, which erodes the justification for a high accuracy reading in the first place. Practically speaking, the occupation of a pole site for GPS purposes is difficult at best. To truly occupy a point for a pole, you must be directly above the site (which is very difficult to do with a 30’ pole). So you must stand next to (or rather “near” in order to get a good GDOP) a pole to get a reading. This process in itself introduces an error of a couple of feet, which can then be exacerbated if a QC technician validating readings follows behind the next day and ends up on the other side of the pole since a better GDOP reading is being received there. Now there is a discrepancy of a meter and a half, just to start with…All of which makes it seem that a one-meter requirement may not be worth the trouble. Regardless, GPS data integration into GIS systems is not a passing fad. As land base and digital orthophoto precision increases, we will see GPS data of increasing accuracies continue to be used in these systems. To think that 1-3 meters accuracy is good enough is not taking into account what history has taught us. Just a few years ago +-30 meters seemed good enough, followed by +-10 meters then +-5 meters, and now we find this is not good enough. Since GPS accuracy for GIS is based upon the final or intended use of the GIS information, we will continue to see accuracies increase, as we expect more and more from our GIS systems. Recreational The recreational GPS user community is growing with new services and products being offered every day. Some of the common users are backpackers, fishermen, hunters and outdoorsmen all of who are now viewing a GPS receiver as standard equipment. Some vehicles now offer GPS tracking capabilities in conjunction with onboard digital maps that pinpoint your location and provide directions to your destination. A newer service being offered combines cell phone and GPS technology. This service tracks your vehicle so you can call for directions to your destination even if you don’t know where you happen to be at the moment. It can also detect if your airbags have been deployed or can dispatch emergency personnel as needed. If your vehicle is stolen, the service can aid police in locating your vehicle quickly. What is the future? The use of GPS is limited only by the imagination of man himself. Integration with existing devices is inevitable; perhaps one day soon all cars will come standard with GPS receivers. Who knows, one day we may have GPS nano-technology that you inject in your teenager, so you know exactly where your kid is when it’s ten o’clock. Integration of GPS receivers into PDAs and laptops is already occurring. It is a given that accuracy will continue to increase and prices will continue to fall, at least for the recreational user. Industry grade GPS units will continue to command a high price since these units are usually cutting edge. If technology has taught us anything, it is that if you want to be first, you are going to have to pay for it. We can be certain that GPS technology will continue to work its way into the average Joe’s life. Which brings us to our initial question. Where is GPS taking us? Your guess is as good as anyone’s, but when it comes to GPS one thing is for sure – when we get there we’ll know exactly where we are! References
| ||
|
|