Assured Position, Navigation, and Timing for the United States

It is shown that the issue associated with assuring Position, Navigation, and Timing is not one of building backup systems to augment the GPS, but, rather, one of designing and building a reliable overall PNT architecture. A continuously functioning Positioning, Navigation, and Timing is a matter of national security and should be recognized as such by the incoming Administration..

I – Introduction-

Much has been said, recently, about the need for adding a nationwide system to compensate for the shortcomings of the Global Positioning System(GPS). It will be shown below that what I define as discrete system thinking is likely to lead to disappointing performance and high cost. We need to discard the idea of one system replacing another, and, rather, develop ideas that will guarantee adequate, and continuously improving Positioning, Navigation, and Timing capabilities for the nation independent of the components that comprise the architecture.

II – The Current Situation –

Since the Global Positioning System became fully operational in April 1995, the nation has relied more, and more, on it to provide essential position and timing functions throughout the United States. Some claim that the order for civilian use of GPS was mandated by an order from President Reagan following the Soviets shooting down a Korean airliner, KAL007 after it strayed into Soviet airspace as the result of a navigation error. President Reagan’s decision was announced at a daily White House press briefing on 16 September 1983, but there appears to be no record of an executive order to that effect unless it was part of a classified order which was never released.

The press statement also proclaimed that the system would become operational in 1988, seven years before Full operational capability was declared in 1995. President Clinton declared GPS to be a dual-use system to be made available to civil users on March 29, 1996, and, at the same time, announced the addition of a new civil signal on the L2 frequency.

On May 1, 2000, it was announced that the clock dither called selective availability (SA) which had been used by the military to limit GPS accuracy to, approximately, 100 meters for civil users would be discontinued. At the time that it was discontinued, selective availability, practically, did not achieve its goal of completely limiting civilian accuracy. At little additional cost receivers could be bought equipped for the detection of U. S. government-supplied signals, such as WAAS and the Coast Guard differential system, that were effective in removing the effects of SA. Thus, the government developed, and funded, systems that defeated its goals. Certainly a bizarre situation. After SA disappeared, though ordinary civil users experienced a noticeable jump in accuracy, and the use of the GPS signals throughout the world increased rapidly. Some government systems, such as the SA defeating signal associated with WAAS and the Coast Guard system, became redundant

Today many essential services, which make our nation a modern society, depend on GPS. Timing synchronization for cellphone networks depends on the signals. No GPS, no cellphones! Timestamps for financial transactions and phase locking of national electric power networks also find the system essential. Most precision aircraft instrument approaches throughout the country, which facilitates operations in poor weather, are GPS-based using the FAA Wide Area Augmentation System(WAAS). As of April 2015, there were 3543 such approaches in the United States, and the FAA is no longer installing self-contained instrument landing systems. The WAAS system is more reliable than the self-contained system and does not have some significant failings, such as sensitivity to snow reflections. Without WAAS, 983 airports in the country would have no precision approach. One should keep this fact in mind when proposing backup systems for GPS.

The remainder of the world has similar situations. Some, such as the Europeans, Russians, and the Chinese have responded by designing and fielding similar satellite systems under their national control, but, as yet, those systems have produced neither the performance nor the reliability of GPS.

In a way, the universality of uses of GPS has taken the government of the United States by complete surprise. Certainly, there were no plans for it to happen, and there is no coherent policy throughout the government bureaucracies which defines a way of dealing with it.

It is no wonder that many government agencies are now attempting to distinguish themselves as the primary agency for deciding what to do about assuring the future of essential Positioning, Navigation, and Timing (PNT) capabilities. The problem, now, is to make the suggested policies constructive, rather than destructive.

III – Real and Perceived Problems and Issues –

A legitimate concern is that the satellite signal received at the surface of the earth is very weak and susceptible to natural interference or intentional jamming. In fact, the signal is as strong as it needs to be to produce acceptable results under normal circumstances. The signal structure is designed to permit processing of the signal to increase the effective power by a factor of about 20, 000. Still not very strong, but adequate for detection using modern electronic circuits. The detection circuitry requires some margin in effective power to operate properly, and a jamming signal in the form of noise at the power level of the received satellite signal would deny accurate time and position information.

Observations, such as the one above have led to a sky is falling attitude in many government organizations. The usual claim is that the signal is so weak and our dependence on it so strong, that we need to create a backup system that can replace GPS if it is unavailable. That claim will be explored below, and it will be shown that a full, completely equivalent system is impractical and unlikely, but there are measures that can be taken to minimize the effects of GPS unavailability. Several issues and possibilities must be considered:

1. What are the threats?

2. What needs to be protected?

3. What kind of protection is needed?

4. How much protection is needed?

5. What is the optimum way to provide protection?

6. Can protection be combined with PNT performance improvement and modernization?

Each of these items is discussed below:

1. What are the threats?

Perhaps the most permanent possibility for loss of signal is either an attack on the system or extensive component failure. An attack is unlikely. The satellites, themselves, are designed to operate in orbital radiation belts. Even a nuclear weapon attack on the constellation would require that each satellite is attacked individually. Thus, the current constellation would require, at least, ten nuclear weapons to reduce performance. The existing satellites have, also, been exposed to intense solar storms without failure. Other satellites have failed during storms. Anti-satellite weapons, which use kinetic effects to destroy satellites have been tested by, at least, two nations against satellites in low earth orbit (LEO).

Intercepting a GPS satellite at an altitude of 20,000 km in medium earth orbit (MEO) is a much more challenging task. The time required for intercept will be long enough to permit retaliation by the United States. It is possible, of course, for rival nations to develop stealthy interceptors and deceptive ways of launching them. Therefore, effective space situational awareness must be maintained [1]. It is possible that enough satellites could fail spontaneously because of, say, component failure, but the experience is that the satellites can operate, on average, ten years, or more, without failure. Six, or seven would have to fail, simultaneously, to degrade performance, noticeably.

Therefore, it is reasonable to assume that failure of the space segment of the system is unlikely enough to be ignored. Of course, the assumption is made that the United States government will launch enough satellites often enough to maintain constellation performance. This is an acquisition issue, rather than a technology issue.

It is, also, possible that the ground stations, which control the satellites, could be damaged, or destroyed, by terrorist, or cyber, attack. However, control systems are isolated and encrypted. The two stations, themselves, a master and an alternate, are located on well-protected military bases and the computers are backed up regularly. Also, without controls, the system will degrade slowly, not instantaneously, allowing time for repair.

Therefore, the control issue should be examined and enhanced, if necessary. The control stations must communicate with the satellites over large distances. Therefore, the vulnerability of the communication links must be investigated. The vulnerability of the satellite systems to cyber and electronic warfare attack is, also, of concern. A world map showing the ground-based system components is shown in fig. 1. The component locations are spread throughout the world. Complete physical protection may be a challenge.

In some cases, such as in urban canyons or indoors the GPS signal may disappear or at least become too weak for the loss to be compensated by processing gain. Also, reflections from buildings and natural objects can generate multipath errors which degrade accuracy.

Finally, intentional jamming and spoofing must be considered. Jamming is a real and a serious threat because of the weak satellite signals, and it will be considered in detail below. Spoofing, on the other hand, has received considerable press coverage, and the results can be spectacular, but it is, relatively, easy to prevent if receivers are properly designed. Spoofing will be considered further, below. In some cases, jamming is intentional. It may be generated by, say, a truck driver who is attempting to prevent his/her company from tracking routes and stops. Or, it may be the result of a serious terrorist attack.

GPS jamming devices can be found for sale on the internet, and jamming appears not to incur criminal penalties, a situation which should be changed by Congress, given that GPS jamming can be life-threatening. Substantial fines were levied against a Chinese company that was selling jammers on the internet, but there is no evidence that any portion of the fine has been collected yet. South Korea has experienced GPS jamming that, almost certainly, was generated in North Korea. This is, perhaps, the only known case of intentional jamming by a rogue nation.

2. What needs to be Protected?

Certainly, some GPS receivers are more critical to the national infrastructure, than others. To protect these receivers, one is likely to be willing to spend more money, and effort, on protection than, say, on a GPS navigator used in a private automobile which gives directions to grandma’s house. Some receivers, especially those which provide timing information, may be fixed at a point, and protection of location information is unnecessary. For an airliner on an instrument approach, though, location may be the important information. Protection methods may differ in difficulty, and cost, for varying situations.

Protection of military weapon delivery is being ignored, here, to some extent, but the protection methods to be described will be similar for the military case, although one may be willing to incur higher costs for this application.

3. What kind of protection?

As mentioned above, the most serious threats to the use of GPS come from signals which mask, or interfere with, the satellite signals, or which provide false, or misleading, information. The remainder of this paper will concentrate on dealing with these problems.

IV – Providing protection and improving performance

A significant amount of work has been done on the problem of detecting and locating GPS ja 2]. There are, also commercial devices, one of which is the Signal Sentry 1000 manufactured by the Harris Corporation, which are said to detect and locate jammers, even very low power ones.

Thus, a rational path to be taken by the U. S. government would be to place detectors at critical points around the country with reports transmitted to a central point. Then, trained first responders could be dispatched to find and disable the devices, possibly, arresting the users. The maximum jamming would then be limited to no more than an hour, or so.

Most of the attention, though, has been focused on establishing a backup system in case GPS signals are not available. The system most often identified for this purpose is enhanced LORAN, or eLORAN, a significantly improved, and modernized, version of the LORAN system of World War II. Given the inertia one finds in a typical user community, however, one cannot expect many consumers to purchase eLORAN add-ons to their GPS receivers knowing that the capability will, probably, never be used. It will be shown below, though that while eLORAN does not provide PNT solutions as accurate as GPS, and it is, therefore, not a perfect substitute for GPS, the two systems integrated and working together can provide PNT solutions more accurate than either eLORAN or GPS, standing alone.

V – What is an un-jammable GPS receiver?

Even, given that most GPS jammers sold online have powers of only a few milliwatts, it is still necessary to prepare for the case of terrorist action, or an extremely massive solar storm. I will define, therefore, an unjammable GPS receiver as one that can withstand a jammer generating noise signals that cover the GPS signal bands, having a power of one kilowatt at a distance of one kilometer from the receiver. Such a jammer is no longer inexpensive, and it is easily detected, and located. This specification seems to be an upper limit. It is likely that many, maybe most, receivers can function safely and accurately with somewhat less protection if the cost is less and safety is not compromised.

VI – System Performance
Standard performance characteristics of the systems are given below.

1. eLORAN

The standards proposed for eLORAN are: Position accuracy 8-20 meters more than 99 percent of the time, and Timing accuracy, with respect to UTC, is 50 ns[3]. It is said that the position accuracy the system for aviation purposes may require more extensive mapping than does its use for maritime operations. The error has been shown to depend on altitude [4]

1.1 eLORAN Message Transmission

The eLORAN system has a message channel which can transmit data at a rather slow rate in the range of 20-50 bits per second (bps)[5].

The LORAN system has been used for data transmission at much higher rates. Liang, et al.[6] have proposed a modulation system said to produce a transmission rate of 2560 bps if there are as many as 20 group repetition intervals (GRI) [5] per second. A message transmission experiment was done in Alaska to transmit WAAS data in regions where the WAAS message satellites were not visible[7]. Data rates somewhat higher than 290 bps were obtained.

The data rates were limited, apparently, by the need to maintain the standard LORAN signal while transmitting. In the discussion, which follows, that limitation will not be observed. The system to be described below can be described as navigation and data multiplexed system. The LORAN signal will be transmitted in the usual form on the eLORAN frequency of 100 kHz, but the time during which the navigation signal is transmitted will be limited to, approximately, fifty percent of the available time and the remaining fifty percent will be used for data. The point is to retain the LORAN format while using the LORAN frequency and bandwidth simultaneously for transmitting data at a high rate. It is believed that, even if some receiver and transmitter redesign is required, the result will be worth it.

Although the details must be specified, one possibility is, 10 kHz, or higher. The signal should remain within the 20-kHz bandwidth of the eLORAN channel. It should, also, not require a major redesign of the eLORAN transmitters, or antennas. If there are problems with international conventions, those can be negotiated away. The point is to generate a system design that will assure the existence of accurate and reliable PNT. Most of the modifications are directed toward the eLORAN system because it is a ground based system. While it may be possible to make some changes in the GPS system by specifying software changes from the control station, those changes are likely to be very limited in scope.

2. GPS

Surveys done by the FAA [8] at many sites throughout the United States gave a 95 percent vertical accuracy of 4.684 meters, a horizontal accuracy of 3.351 meters, and a time transfer accuracy of 18 nanoseconds with a mean of 5 ns and a standard deviation of 8 ns.

Jamming protection is determined by the processing gain mentioned above in sec. III. The gain of 43 dB is determined by the reduction in bandwidth determined by the correlation time of the process of matching the known pseudo-random code to the signal. The navigation data transmitted by the satellite must be read from the signal by the receiver. The data are transmitted at a rate of 50 Hz, and, therefore, a correlation time of 20 ms is the longest permissible. The ratio of the 1MHz civil signal bandwidth to the 50 Hz data rate is 20,000, or 43 dB. The process of extending the correlation time, and, thus, reducing bandwidth can be continued to give increased processing gain, but one, eventually, is required to consider the effects of receiver motion during the correlation time. Hence, the reason for the sharing of the channel in 200 ms slices. It is customary when increasing the correlation time to invoke the existence of an inertial system to give a solution for sensor motion during the correlation time. In this case, it will be the eLoran that provides motion information during the correlation time. I am told that the eLoran zero crossing time, which is used for rating, can be measured with an accuracy of 100 ps[9]

VII – Protection Through System Integration

Now the process for integrating the eLORAN and GPS systems in a way that will enhance resistance to jamming will be described. The parameters of the legacy system will be used as representative, even though changes to be made in the future may influence details of the suggested process. The timing suggestions should, therefore, be considered as suggestive, not accurate in detail. The process requires fundamental changes in the eLORAN system, but the changes should be easy and inexpensive to generate. The suggestions will apply to all GNSS systems, so there will be an incentive for worldwide acceptance. Otherwise, the system will become an American system, only.

First, consider the power levels involved will be described. The satellite power received at the surface of the earth will be taken as -157 dBW. The received power from a one-kilowatt jammer at one kilometer is -63.85 dBW. The jam to signal ratio, J/S, is 93.2 dB. Normally, a margin of, say, 10 dB would be required to assure signal tracking would be required, but the margin will be reduced for the process to be described.

If one looks at the eLORAN signal structure, it is seen that pulses are impressed on a carrier of frequency 100 kHz with a bandwidth of 20 kHz. Therefore, if the signal were used as a communication channel, a transmission rate of 10 kHz could be practical. Since the limit to correlation times is determined by the necessity of detecting the navigation message, it is proposed that the navigation messages for all GPS satellites that may be in view of a user of a particular eLORAN chain be transmitted over the eLORAN channel. A complete navigation message is 37,500 bits, and 3.75 seconds per satellite would be required. The total number in view throughout the eLORAN signal area could be as large as 16 satellites. Allowing some bits for identification and timing, a reasonable time could be 4 seconds per satellite, for a total maximum time of 64 seconds. The time required for a set of eLoran pulses can be as long as 200 ms. Thus, it seems reasonable to transmit satellite navigation data for 200 ms following the LORAN pulses. At least 2000 bits of the navigation message could be transmitted during this period. The number of bits is more than enough to transmit three satellite’s ephemeris information, for example.

It is proposed, too, that each eLORAN receiver. The position of the station receivers can be measured to an accuracy of a few millimeters, and the range from the station to each satellite can be transmitted to the control station. The control station can then calculate an extremely accurate ephemeris for each of the satellites in view. The precision ephemeris can be transmitted over the eLORAN channel, and the GPS error using the precision ephemeris will be less than one meter. An accuracy near one centimeter should be possible, eventually. Therefore, the accuracy of the GPS position will be noticeably improved by the integrated system. Timing also improves since time is a component of the GPS solution.

For operation in a jamming situation, a processing gain improvement of 93.2 dB is required to generate the unjammable receiver. Given that the eLORAN solution provides a location accuracy of about 10 meters’ accuracy, one can use the precision ephemeris to identify an accurate starting point for the correlation delay, even for a receiver just beginning to attempt satellite acquisition. One will, typically, have as many as fifteen satellites in view, with two signals tracked from each. Since the signals are generated coherently in the satellites, their powers will add, if properly combined, to give an effective increase o, at least 10, or 10dB. The protection needed, then, becomes 83 dB, or a factor of 2 × 108. Generating that much protection by increasing correlation time would require, for a 1MHz signal, a time of 120 seconds. During this time, the eLORAN signal functions as a differential signal with a GPS calibration at each end of the interval. This may be acceptable, but for many applications, the interval may be too long.

To address the problem of long correlation times, one can consider the use of nulling antennas. The controlled radiation pattern antenna, or CRPA, can be made to generate nulls in the direction of a jammer by a factor of 30 dB, or more. The cost, however, can become quite high. A considerably less expensive solution using commercially available parts has been demonstrated to give jammer nulls of 20 dB [10]. The use of an antenna of this type will reduce the correlation time factor to 63 dB, an interval of only 0.5 seconds. If only the L5 signals at 10 MHz chip rate are used, the time becomes 50 ms. That is possible if the satellites in view are members of block IIF or III. Thus, one can generate an un-jammable GPS receiver with an error of ess than one meter by integrating the eLORAN and GPS systems. In cases where the GPS signals become very weak from natural causes, the method can still be used, and the worst case is a default to the eLORAN performance.

VIII – Conclusions

The system described here will compensate for all interference and weak signal situations. The usual performance will be sub-meter accuracy and a worst case of 20 meters. Changes in the eLORAN signal structure will be required, but it appears that the minor system changes required will be well-worth the cost and effort.

This paper is not intended as a detailed description of the integrated system. The intent is only to show that the parameters for an assured PNT system with accuracy better than the current GPS accuracy are favorable. There are those, though, who understand both systems well enough to collaborate in the process of generating an integrated system of higher accuracy and greatly increased jam resistance. For the integrated system, the whole is, indeed, greater than the sum of the parts.

President-Elect Trump should prioritize guaranteeing adequate and continuously improving Positioning, Navigation, and Timing capabilities that are necessary to secure the U.S. civilian nd military critical infrastructures.

References

[1] McCall, G. and Darrah, J., Air and Space Journal, 30(2), 6 (2014).
[2] Brown, A., etProc. ION GPS ’99, Nashville, TN, September 1999. [3] Enhanced Loran Definition Document, version 0.112 January 2007.
[4] Johnson, G., et al., US Coast Guard Academy Report 06320-8103, 2006.
[5] Hedwige, A., et al., International Loran Association Report ILA-40, November 2011.
[6] Liang, Q., et al., Proc. of the 8th International Conference on Wireless Communications, Networks, and Mobile Computing, Shanghai, 21-23 September 2012.
[7] Lo, S. et al., WAAS Performance in the 2001 Alaska Flight Trials of the High-Speed Loran Data Channel, IEEE Position Location and Navigation Symposium, 2002 IEEE.
[8] Federal Aviation Administration, GPS Product Team Report number 86, July 31, 2014. [9] Charles Schue, private communication.
[10] Chen, et al., Proc. of the 2013 International Tech. Mtg of the ION; San Diego, CA, 2013.
~~~

* Dr. Gene H. McCall, is Laboratory Fellow at Los Alamos National Laboratory and an Affiliate Research Professor, Desert Research Institute, Reno Nevada. Among his many assignments, Dr. McCall served as the Chief Scientist with Air Force Space Command at Peterson Air Force Base, Colo. He is also a member of the Advisory Board of the ACD’s Economic Warfare Institue.