The purpose of this article is threefold: To provide information about the current relationship between CATV and GPS; point out similar expressions and terms that are interchangeable; and provide information about policies and technical data which will influence the cable industry in the future. Because previous trade articles have been written on GPS history and theory, this paper will try to avoid repeating or recycling old material.

Introduction
GPS is a satellite-based radionavigation aid deployed by the U.S. Department of Defense that is primarily aimed to support the military. However, GPS is also made available for use by the general public and commercial entities to obtain accurate positioning.

In February 1993, the 18th "block-two" satellite was set into orbital plane B-1. There are now 22 block-one and block-two satellites in orbit out of a scheduled 24. Six launches are scheduled for 1993, which will complete the constellation and begin phasing out the older block-one satellites. For the first time in history, 24-hour three-dimensional coverage is possible. There are absolutely no lapses in two-dimensional coverage now.
Also, newly completed GPS algorithms are making constellation jumps, where a segment of the plotted course jumps to one side or the other (see Figure 1), a thing of the past. Jumps occur when a GPS receiver tracks different satellite constellations as they become visible. These jumps, when combined with selective availability (S/A) errors, can produce a messy course on a digital map. Multipath errors are still a problem, but they occur significantly less frequently and are usually single points instead of a segment.

Maintenance concepts
Let's back up and look at two concepts that are fundamental to CATV service: Demand maintenance (D/M) and preventive maintenance (P/M). D/M is reactive - which results in random pattern, like outages and priority service calls. It doesn't matter what part of town they occur, the course of action is to 'Drop what you're doing and proceed immediately to that new location." The other side of the coin is P/M, which is methodical and well planned. A tech could have 25 service calls, 20 amps to sweep, 30 leaks to fix, or 100 miles of plant to ride out, but his day is planned ahead of time, which is more efficient. A tech may have to respond to an outage, but he will return to his P/M duties.

GPS has two categories that correspond with the D/M and P/M formats: Real time and post processing. The idea of real time GPS is to supplement beepers and two-way radios with fleet management. The idea of post processed GPS is to enhance the productivity of P/M by streamlining the planning.

Surveyor accuracies
First, let's talk about how surveyors achieve accuracies of just a few centimeters. A sophisticated GPS receiver will measure not only the pseudoranges (code), but also the carrier phase. Utilizing differential techniques, the carrier phase yields centimeter precision compared to meter level precision offered by the pseudoranges. The carrier phase measurements are made ambiguous by an unknown integer number of carrier cycles - the so-called "carrier phase ambiguity."
This carrier phase ambiguity remains constant over time as long as the receiver is phase-locked to the incoming signal. No similar ambiguities exist for the pseudoranges. If we can correctly identify the carrier phase ambiguity number, we can convert the carrier phase measurements into very precise range measurements. This integer ambiguity resolution is the goal behind several different post processed procedures, including: static differential positioning; pseudo-kinematic surveying; stop-and-go surveying; and rapid static surveying. While a detailed discussion of these procedures is beyond the scope of this paper, the concept to remember is the difference between pseudorange (code) and carrier phase receiver abilities. The terms pseudorange and code are interchangeable in this application. All GPS receivers interpret code, but only high-end GPS receivers interpret code and carrier.

Time concepts
The next concept is the difference between GPS time and UTC (universal coordinated time). UTC time is measured in seconds and is referenced to the London time zone. This time zone is not to be confused with the absolute longitude coordinates 0 degrees 0 minutes 0 seconds, which runs north and south through London. This time zone encompasses the United Kingdom. GPS time is a product of a satellite's atomic clock, with nanosecond accuracy. A sophisticated GPS receiver would utilize this in time transfer technology, which is an entirely different field.

How does the FAA (Federal Aviation Administration) plan to use GPS? The answer falls into two categories: Accuracy and integrity. Civilian GPS, or SPS (Standard Positioning Service), has a stated accuracy of 100 meters (95 percent of the time). This is acceptable to the aviation community for en route and non-precision flight operations, but not for airport precision approaches. Three technical approaches for accuracy are under study - differential GPS; augmenting GPS signals with other navigation aids (top candidates are GLONASS [Russian GPS], Loran-C, and inertial navigation systems); and real-time kinematic carrier phase tracking.
The challenge here is to take this centimeter accuracy beyond the stationary mode and develop methods for dynamic civil aviation operations (ambiguity resolution on the fly). There is a new technique that shows great promise. It's called "widelaning the dual frequencies" and is solving the ambiguity resolution in 1 to 3 seconds.
GPS at this time cannot meet the FAAs' requirement for integrity. This is the ability of the system to provide timely warnings to the user when the system should not be used for navigation. These integrity requirements are 30 seconds for en route flight, 10 seconds for terminal areas and non-precision approaches, 6 seconds for certain precision approaches, and 1 to 2 seconds for precise approaches leading to actual runway touchdowns.

Minimum Operational Performance Standards (MOPS) set by a Washington-based commission, which affect aviation on a global scale, are calling for the 10-second integrity warning. Inmarsat is proposing an integrity channel via its 3rd-generation satellites beginning in 1995. It would uplink in C-band and downlink in L-band frequencies adjacent to GPS frequencies. This is called the Bent Pipe effect.

When one is dealing with something as critical as integrity, the question, "What if there is a total failure of the system?" must be asked. RAIM (receiver autonomous integrity monitoring), which would choose between other augmented systems seems to be the conventional answer. The GPS
process for aviation is evolving toward a GNSS (global navigation satellite system).

Horizontal accuracies
Horizontal accuracy is a subject that does apply to cable operators. Government-stated specifications for SPS (civilian) is 100 meters (95 percent of the time) with S/A on; PPS (precise positioning system) or military spec is 18 meters (95 percent of the time). This 95 percent spec is a huge cushion. Realistically, it's more like 99.5 percent. Every once in a while you'll see a huge jump that probably only the Dept. of Defense could explain. S/A does not affect the PPS. So, does the PPS represent what the civilian service would look like if S/A was turned off?

Remember, PPS is a product of the P code placed on both the L-1 and L-2 frequencies (see Figure 2). The dual frequencies allow the receiver to correct for ionospheric errors. The higher frequency will have a greater loss. The difference can be measured to determine the ionosphere loss. The C/A code is only on the L-1 frequency. Single frequency signal processing incorporates a mathematical model called the Klobucar model. This eliminates half the ionosphere errors but not all of them.

There's another error that is the product of S/A, called the "observables." These are the satellites that are in view for the individual GPS receivers when making differential corrections from a base station to a rover receiver. There may be, for example, eight satellites in the sky over an immediate area. The base station GPS receiver may see all eight while the rover may see only seven. This is because buildings and other obstructions may block the view of the rover. The percentage of observable error is the combination of how long an obscured satellite (for the field unit) is on the horizon and whether or not S/A is cranking on that satellite at that time, and whether a leak (event) occurs at that time.

Consequently, each receiver has its own unique navigation solution (prime course). This means that if (up to 100 meter error causing) S/A were removed, accuracy is slightly worse than 18 meters because half of the ionospheric error still exists for a single-frequency GPS unit.

Those who have ever seen a plotted course on a digital map with S/A working know that the goal is to be on the right street. Most streets are 300 to 400 feet apart. That meas accuracy of 150 feet or less would be sufficient to put the path on the right street. When you're dealing with the integrated navigation solutions (latitude and longitude) instead of the raw data (pseudoranges, range rates, etc.), S/A's effect is quasi-directional. The plotted course could be up to 300 feet above, below, left, or right of the actual prime course. Integrated differential corrections would produce a spec in the neighborhood of 25 meters (99 percent of the time because of the observables and multipath errors). Remember, these are worst case numbers and the norm would be 15 meters (street width accuracies).

There is a way to get this type of accuracy by using differential GPS receivers. The most accurate and expensive models produce 2- to 5- meter accuracy, but because they work with GPS raw data, they work in real-time only. The differential systems which utilize integrated solutions are more reasonably priced and work in the post processed realm. Consumers will buy the accuracy that applies to them. That's why it's important to understand all the facts.
Government policy re: S/A
There is another way to remove S/A, and that is to simply turn it off during peace time. S/A at one time was degrading GPS accuracies to 500 meters (95 percent of the ti-e). The Cold War threat of missiles being locked onto the signal was a real concern. In 1983 the Department of Defense reduced accuracy to 100 meters. Additional emphasis was placed on the civil use of GPS after the downing of flight KAL 007 by the Soviet Union. The Senate, after condemning the act, called for a speeded up timetable because GPS benefits public safety.

In this context, maybe the S/A error should be tied directly to defense conditions. Zero or minimum threat should equal no S/A. Medium threat would equal medium S/A, and so on.

Anti-spoofing
Anti-spoofing (A/S) is another product of the DOD that provides military protection against fake signals. A/S works by encrypting P code to Y code. P code is commercially used by civilians, but only military personnel have the Y code encryption keys (see Figure 2). It's important to understand the difference between S/A and A/S in the fact that A/S does not affect the average civilian GPS user. Most people familiar with GPS would say S/A is here to stay. However, there are many who believe the DOD is doing the American taxpayer a great disservice by operating S/A during peace time. CATV is now part of the long list of industries affected by the operation of S/A.

Conclusion
GPS technology promises some amazing things to different consumer groups. Unfortunately, some of these groups will have to wait for some answers. Fortunately, one of these groups isn't CATV. CED ((logo))