FI85916C - FOERFARANDE FOER ATT AVLEDA LOKALISERINGSRELATERAD DATA FRAON SIGNALER MED SPRIDNINGSSPEKTRUM. - Google Patents

Description

1 85916

A method for obtaining positioning information from spectra transmitted from a spectrum

Divided by application 830619 5

The present invention relates to a method for obtaining positioning information from spectral propagated signals having suppressed carriers as a result of modulation by mutually orthogonal spreading codes.

Some systems for radio positioning take advantage of the direction of the radiation pattern of the transmitting or receiving antenna. Other systems incorporating the present invention do not rely on the orientation of any antenna. The present invention belongs to a general class of systems in which the location of a receiving antenna is determined by measuring the phase or group delay difference, or both, between signals coming from two or more different transmitting antennas, the location of which is already known. If the two transmission sources are synchronized, or if the deviation of the two transmitters from synchronization is known independently, then the difference between the group delays of signals from the two sources at the receiving location determines that the receiver is positioned three-dimensionally with a special rotational hyperboloid. If similar measurements made at the same receiving location from signals from several suitably positioned transmitters are combined, then the location of the receiving location 30 can be unambiguously determined from the intersection of the respective hyperboloids.

Techniques for determining the relative positions of different locations, one relative to another, from measurements of phase or group delay differences between radio signals received simultaneously at these locations are also known in the art and are collectively referred to as 2,85916 geodetic techniques using radio interferometry.

The antennas at different locations are thought to form an interferon tr in and a relative position vector extending from one antenna to another is called the baseline vector of interferometry. The baseline or relative position vector between two antennas can usually be determined with less uncertainty than the position of the rubber antenna, because many potential sources of error tend to affect the measurements at both antennas in approximately the same way, resulting in a reversal of the difference between the two antennas. Microwave radio interferometric geodesy technology is known to provide an unparalleled combination of accuracy, velocity, and range for determining the relative position or in-15 interferometer baseline vectors. Such a determination may be based on measurements of either the group delay difference or the phase difference, or both, between the signals received at the ends of the baseline vector. Phase measurements are inherently more accurate than group delay measurements, but the interpretation of phase measurements is more complex due to their inherent integer period ambiguity. A general presentation of interferometric measurement techniques and related interpretation problems is given in Charles C. Counselman III's article ot-25, "Radio Astrometry", published in Annual Reviews of Astronomy and Astrophysics, Vol.

14 (1976) pages 197-214. A large collection of relevant technical publications appears in Conference Publication 2115 of the National Aeronautics and Space Administration under the heading "Radio Interferometry Techniques for Geodesy". Geodesy using radio interferometry has been practiced with radio signals received from a variety of sources, including natural sources such as quasars and artificial sources such as NAVSTAR Global Positioning System (GPS) satellites.

As is well known, the country is currently orbited by about six ···. GPS satellites. The orbits of the satellites can be determined • · • * · 1 1 3 85916 with an accuracy of about two meters. These satellites transmit radio signals with wavelengths close to 19.0 cm and also 24.4 cm. Provided that the ambiguity of the integer-5 periods of the interferometric phase observations of these signals can be solved correctly, the baseline vector extending from one antenna to another can be determined interferometrically with an uncertainty much smaller than the wavelengths of GPS transmissions. The three baseline determinations, each with a baseline length of the order of 10 to 100 m, were shown by interferometric phase measurements of GPS signals to be accurate within about 1 cm in Eos (Transactions of the American Geophysical Union), Vol. 62, page 260, April 28.

According to a report published in 1981, by Charles C.

15 Counselman III, S.A. Gourevitch, R.W. King, T.A. Herring, I.I. Shapiro, R.L. Greenspan, A.E.E. Rogers, A.R. Whitney and R.J. Cappallo. The method used in these interferometric baseline determinations was based on the known technique of direct cross-correlation of signals at a central location 20, which signals were received separately, but simultaneously, at both ends of each baseline.

. U.S. Patent 4,170,776 describes a system for measuring changes in a baseline vector between a pair of ground locations using signals transmitted from GPS satellites, in which the radio signals received at both locations are accurately timed and then transmitted over telephone lines to a central location where approximately real-time phase comparison is performed by cross-correlating the two signal groups. The system described in Patent 30 includes "bowl reflector type" receiving antennas. Because the radio frequency of the GPS signal is low relative to the background noise level, and because the bandwidth of the GPS signal far exceeds the bandwidth of the telephone line, the signal-to-noise ratio of the power transmitted from both locations over the telephone line is small. Mainly useful for increasing this signal-to-noise ratio. le level in this system, 4 85916 antennas with large collection areas are used "bowl type" * · • · · • · • · · ·. Another important reason for using such antennas is that they are directional so that signals arriving at the antenna other than directly from the desired source are rejected.

5 Systems for measuring baseline vectors using other types of signals from orbiting satellites are also known.

An article by Charles C. Counselman III and Irwin I. Shapiro, entitled "Miniature Interferometer 10 Terminals for Earth Surveying" (MITES) and published in Bulletin Geodesique, Volume 53 (1979) pages 139-163, describes the proposed system as a baseline. for measuring signals using multi-frequency radio signals transmitted from orbiting satellites, in which system the phases of the received signals are determined separately at both ends of the baseline. That is, a signal received at one location is not cross-correlated with a signal received at another to determine the phase difference between the two signals. 20 In order to solve the phase ambiguity, the MITES system always relies on a combination of a group of measurements performed at ten frequencies, which frequencies are suitably placed between 1-2 GHz. Unfortunately, as far as is known, there are currently no satellites orbiting the earth that transmit such signals.

Systems for determining relative position using signals transmitted from sources other than artificial satellites are also known. An example of such a system using moon-based transmission is also disclosed in U.S. Patent No. 4,170,776.

Systems are also known for measuring either only one location or relative position using signals from sources other than orbiting satellites.

35 For example, W.O. In an article by Henry entitled "Some Developments in Loran," published in the Journal of Geophysical Research, Vol. 65 pages 506-513, «« • · • · λ · 5 85916 February 1960, describes a system for determining a position (such as a ship at sea) using signals from ground-based (fixed) transmitters. The system, known as the Loran-C navigation system, uses several tu-5 thousand-kilometer-long chains of synchronized transmitters placed on the ground with all transmitters using the same carrier frequency, 100 kHz, and each transmitter being amplitude modulated by a unique, periodic pulse pattern. This pattern, which includes 10 amplitude signal inversions, allows the receiver to distinguish signals from different transmitters. A suitable combination of observations from more than one pair of transmitters can provide an determination of the location of the receiver on the ground.

Another example of this type of system is the Omega system described in Pierce's article entitled "Omega" published in IEEE Transactions on Aerospace and Electronic Systems, Vol.

AES-1, No. 3, pages 206-215, December 1965. The Omega system measures the phase differences 20 of the received signals and not the group delays, as in principle in the Loran-C system. Because the frequencies used in both the Loran-C and Omega systems are very low, the accuracy of position measurements with these systems is quite poor compared to the said satellite systems.

25 The prior art also includes other methods for determining location and target location. Iin Using the Positioning System. The normal method described, for example, in J.J. In an article by Spilker Jr. in Navigation, Volume 25, No. 2 (1978), pages 121-30 146 and further in several other articles in the same issue of this publication, the group delay of coded modulation of GPS signals is based on measurements of differences between In principle, this method is a hyperbolic localization method and is substantially similar to LORAN. The bandwidth of about 10 MHz of GP S-modulation limits the accuracy of group delay measurement and thus positioning by the normal method to several tens of 6 85916 centimeters «· • · · • · • * * • · · · 1 ·« ·. An accuracy of the order of one centimeter is potentially achievable using carrier phase measurements, as described, for example, in J.D.

Bosslerin, C.M. Goadin and P.L. In Bender's article ot-5 from his pig "Using the Global Positioning System for Geodetic Positioning," published in Bulletin Geodesique, vol. 54, No. 4, page 553 (1980). However, all published methods using the GPS carrier phase have the disadvantage that they require information about code modulation and its use, which code modulation may be secret or require cross-correlation of signals received at different locations, or require the use of large antennas to increase received signal noise and reflections. -15 or otherwise suffers from more than one of these disadvantages. The present invention does not have any of these disadvantages.

In particular, the present invention does not require information about codes that modulate GPS carriers, does not require cross-correlation of a signal received at another location with a signal received at another location, and does not require the use of large or strongly directional receiving antennas.

It is an object of the present invention to provide a method for determining a location by radio.

Another object of the present invention is to provide a method. to measure a baseline vector between a pair of points radio; e * i using interferometry.

Yet another object of the present invention is to provide a method for determining a ground point pair, such as a baseline vector between measurement marks, using two-way band-suppressed carrier radio signals transmitted from orbiting satellites of a global positioning system.

Another object of the present invention is to provide a method for determining a baseline vector between two measurement signals using radio signals from φ m * · 7 85916 orbiting satellites in the Global Positioning System, the determination involving measuring the phases of the carriers included in the signals received by both measurement signals.

Yet another object of the present invention is to provide a technique for processing phase information derived from radio signals received from different directions at two ground locations to determine a relative position.

Yet another object of the present invention is to provide a method for measuring the powers and carrier phases of radio signals received from Global Positioning System satellites without knowledge of the encoded signals modulated by the carriers in the transmitters of those satellites.

Yet another object of the present invention is to provide a method for determining a baseline vector between two points by measuring the phases of radio signals received at both points without cross-correlating a signal received at one point with a signal received at another point, without storing a signal received at either point and otherwise from one location to another. .

Another object of the present invention is to provide a method for locating a radio without requiring the use of a directional antenna.

The objectives described above are achieved by a method similar to that described at the outset, wherein in the method of claim 1, a signal combination comprising potentially interfering continuous wave components in addition to the signals propagated from the spectrum is received at the antenna; filtering from the signal combination the continuous wave components which would be involved as an interfering signal in a continuous wave component reconstructed from the signals propagated by the spectrum, which then reconstructing the continuous wave component from the spectra of the spread signals; and deriving information from the reconstructed component.

In the drawings, in which the same reference numerals represent the same parts: Fig. 1 shows a system for determining a baseline vector by radio interferometry associated with GPS satellites in accordance with the principles of the invention; Fig. 2 is a block diagram of another interferometric field terminal shown in Fig. 1; Fig. 3 is a block diagram of the antenna arrangement shown in Fig. 2; Fig. 4 is a block diagram of the receiver unit shown in Fig. 2; Fig. 5 is a block diagram of the digital electronics unit shown in Fig. 2; Fig. 6 is a block diagram of the signal modifier shown in Fig. 5; Fig. 7 is a block diagram of one of the correlator modules of the correlator arrangement shown in Fig. 5; Fig. 8 is a block diagram of one of the numerical oscillator modules of the numerical oscillator arrangement 25 shown in Fig. 5; Fig. 9 is a block diagram of the Kent terminal computer shown in Fig. 2.

The present invention is directed to the technique between a pair of dots in 30 countries, such as measurement marks; to measure a baseline vector by radio interferometry using two-sided, carrier-suppressed radio signals transmitted by NAVSTAR Global Positioning System (GPS) orbiting satellites. The Tek-: 35 ankle involves measuring the phase of the carriers included in the signals received at each station and then processing the phase information obtained from both stations to determine a baseline vector. One advantage of the technique is that it measures carrier phases without using accepted information about the encoded signals used in satellites to modulate carriers. Another advantage is that it does not require the transmission of received signals either in real time or as the transfer of recordings from two locations to a common location. Another advantage is that it does not require the use of large or 10 very directional antennas. Another advantage is that it is relatively immune to errors caused by radio wave scattering or reflection near the receiving antennas.

Although the invention will hereinafter be described in particular in connection with the use of GPS satellites, it is to be understood that certain features thereof are not limited to use in connection with such satellites and may be useful in connection with signals received from other sources.

20 As is well known, NAVSTAR Global Positioning System (GPS) satellites orbit the earth at an altitude of about 20,000 km and transmit signals in the 1575.42 MHz centered frequency band known as the "LI" band and the 1227.60 MHz centered frequency band. - 25 in a band known as the "L2" band. The signals are modulated to generate approximately symmetrical lower and upper side bands with the carrier fully suppressed.

For both bands, a signal received from a particular satellite at a particular location can be viewed - ·; as a function of time: s (t) = m (t) cos (2πί ^ + φ) + n (t) sin (2πί ^ + φ)::: 35 where m (t) and n (t) are modulating functions , both over time

*> J

real-valued functions, f is the nominal carrier frequency -... cheese for the band 1575.42 MHz L1 and 1227.60 MHz L2, m · 1 · «· · ίο 8591 6 and φ is the phase of the received carrier in radians, which is unknown and to be determined. Both modulating functions m (t) and n (t) are pseudo-random functions of time with a mean of zero. These two functions are mutually 5 orthogonal. Each of the functions used to modulate the L1 carrier for any satellite is also orthogonal to the corresponding function used for any other satellite, although the same m (t) or n (t) function or both can be used to modulate both L1 and L2 a carrier wave. The bandwidths m (t) and n (t) of the two functions differ by a factor of 10, with m (t) having a narrower bandwidth and n (t) having a wider bandwidth. L1 usually has both m (t) and n (t) signal components present and L2 has only n (t) 15 components present when m (t) is set to zero or "off". The power spectrum density m (t): 11a, which corresponds to a modulating signal known in the GPS literature as the "clear / acquisition" code, is proportional to the function 20 2 sin (πΓ / 1.023 MHz) UF / 1, 023 MHz) 2 where F represents the modulation frequency. This function has 25 half the width at about 450 kHz, or about half of its maximum frequency. That is, the value of the function is about 0.5 when F = ± 450 kHz, while the value is a unit when F = 0. Power spectral density n (t): 11a, which corresponds to a modulating signal known in the GPS literature as the "exact code" 30 or "as a P-code" is proportional to the expression Y; sin2 (ttF / 10.23 MHz) (ttF / 1 0.23 MHz) 2 • · · '· 1:35 Thus, half the width at half the maximum power spectral density of n (t) is about 4.5 MHz.

• · 1 1 · m 1 · »P · • · 1 11 8591 6

For the signal L1, 1575.32 MHz, the root mean square n (t) is usually half the value of m (t), i.e. <n2 (t)> = 0.5 <m2 (t)> 5 satellites in unusual states where the ratio of the root mean square or power ratio is different from 0.5, in particular a value of zero is possible.) Thus, the ratio of the power spectral density of n (t) to the power spectral density-10 of m (t) is usually about 0.5 to f · 10 = 0,05 for values of F close to zero such that if a band-pass filter fitted to the spectrum of m (t) is centered on the carrier frequency L1, about 90% of the power at the output of this filter comes from the m (t) signal component and less than 10% is pe-15 derived from the n (t) component. For simplicity, the remainder of this explanation therefore assumes that the GPS L1 signal has no n (t) components and has a simpler form 20 s (t) = m (t) cos (2wf0t + <j))

The generally received carrier phase, φ, is a slowly changing function of time, so that in reality - · the received carrier frequency is obtained as an algebraic sum- Mana ”7 f = fQ + (2π) -1 (dφ / dt) where f is the nominal carrier frequency and dφ / dt is φ: η::: 30 time derivatives. By "slowly changing" is meant that (2π) ^ (άφ / dt) is very small compared to fQ and m (t) bandwidth. The main reason for the φ: η time variation is

Doppler shift that can cause f to deviate from fQ by about ± 4.5 kHz.

• '35 · ...' The received signal s (t) does not contain a separate thermal spectrum component at the carrier frequency, because the average of m (t) • 12 85916 is zero. Thus, the carrier is completely suppressed and the power spectral density function of the L1 signal s (t) is the same as the power spectral density function of the modulation m (t) shifted to the carrier frequency f received from the baseband. Since m (t) is a real-time function, its power spectral density is . Thus, the power spectral density of s (t) has an even symmetry with respect to the carrier frequency f and is called a two-band spectrum. The portion of this power spectrum that already corresponds to frequencies greater than f is called the upper sideband, the portion corresponding to the lower frequencies is the lower sideband. The slight asymmetry between the upper and lower sidebands due to Doppler "elongation", up to about three parts per million, is not significant here. In accordance with the present invention, an antenna is located at both ends of the baseline vector. The signals received by both antennas are separated by upper and lower sideband components. These separated components are filtered, converted to a one-bit digital format, and then multiplied, and their input is digitally analyzed by correlation with the quadrature outputs of a local oscillator to determine the carrier power and phase of the two-sided signal received from each satellite to this local oscillator. The Doppler shift is used to separate the carriers of different satellites, so the powers and carrier phases of several satellites are measured simultaneously -__ '· and the measurement results are numerical data are obtained with each measurement mark. Measurements are made in real time for each signal without comparison to signals received at another location and hour. ; without encoding any of the encoded signals that modulate the GPS carriers. Data from measurements taken simultaneously, but independently with two measurement signals once per second over a sufficiently long period of time, such as about 5,000 seconds, are then processed together to determine a baseline projector extending from one signal to another. The explanation presents two processing methods. In both methods, an "ambiguity function" is calculated, which is a function of the measurement data and the default value of the baseline vector 5. The vector state of £ 5 is systematically examined to find the unique value of b that maximizes the calculated function. This value of b is taken as the desired determination of the unknown baseline vector S.

Referring now to Figure 1, there is shown a system 11 for determining a baseline vector S in accordance with the present invention. The baseline vector S, hereinafter also referred to as the "baseline", is the relative position vector of the second measurement mark SM-2 with respect to the second measurement mark SM-1. The baseline extends from the measurement mark SM-1 at the starting point or at the other end of the baseline SM-2 at a baseline endpoint or at one end System 11 comprises two intelligent interferometer field terminals 13-1 and 13-2 at each end of one baseline and a computer which may be structurally and functionally incorporated into and be part of one of the terminals 13. be a separate unit 15, as shown.

The system requires certain numerical information from external sources for its normal operation. It also requires some means for transferring numerical data between the computer 15 and both terminals 13 before and after or (optionally) during performing baseline measurements.

:: * - Before the measurements to determine the baseline: 30 have started, data from the first data memory 17, which also represent the orbits of several GPS satellites, from which. ; Two of the satellites, labeled GPS-1 and GPS-2, .... are shown for illustrative purposes, are entered into a computer 15 together with approximate data 35 representing the locations of the measurement marks SM-1 and SM-2, which are · obtained from another from the data memory 19. The latter data could, for example, represent the locations of measurement marks ". to within a few kilometers. From this information of satellites • · · 14 8591 6 orbit and measurement locations, computer 15 generates in tabular form a forecast of The 42 MHz signal will have when it is received at each measurement signal, and the computer 15 will also generate a table prediction of the power level of the signal received from each satellite by each signal, the predicted power is zero if the satellite is below the horizon above e due to the angular dependence of the gain of the receiving antenna (marked) and, to a lesser extent, the angular dependence of the transmitting antenna (satellite). Predicted frequency offset and power tables for the time slot comprising the predicted measurement-15 time slot for all GPS satellites expected to be visible at each measurement signal are now transmitted by any known means, such as a telephone line or radiotelephone link, and fed to a special interferometer field terminal 13 in the computer memory, which terminal will be placed or may already be placed on the measurement mark. Alternatively, frequency and power prediction tables can be generated on a computer inside the interferometer field terminal.

Doppler frequency predictions are calculated based on formulas. ". · 25 well known in the art. Errors in such predictions are of the order of 1 Hz per error kilometer of the presumed position of the measurement mark. An additional error in the frequency prediction ·; · due to an error in satellite orbital extrapolation is: usually of the order of 1 Hz or less The frequency prediction errors for several Herzs are tolerable within the scope of the present invention.The received power of the predictions need not be very accurate, errors of several decibels would be tolerable because these predictions are not used for any very critical purpose. is mainly used to allow the field terminal computer to check whether the desired • * • «4 is 8591 6 signal or some random signal is received.About sacrificing some reliability, the power prediction tables could be waived.

The interferometer field terminal 13, located at 5 measurement marks, now receives 1575.42 MHz signals from several satellites up to seven, but in no case from less than two satellites simultaneously. In order to accurately determine the baseline to be measured, it is essential for the terminals at both ends of the baseline to monitor the satellites simultaneously.

The electronic circuits (described below) at each terminal distinguish the upper and lower sideband components of the received signals and, using Doppler shift predictions, analyze these sideband components to determine the power and phase of the carrier included in the signal received from each satellite. Information about these power and phase configurations is stored in the field terminal and possibly returned to the host computer 15 by any conventional means.

The data from these two interferometric field terminals 13-1 and 13-2 must be processed together to obtain an accurate determination of the baseline vector.

It should be noted that the means for remote transmission or transmission of data are not necessary for the operation of this system. Terminals 13-1 and 13-2 can be physically moved to the same location as the computer 15 and there ": The prediction tables can be moved from the computer to the terminals 13. The terminals 13 containing the tables in memory can then be carried on the measurement marks SM-1 and SM-2. , where the satellites are monitored 30. After these observations are completed, the terminals 13 can be carried back to the computer 15 at a location where the carrier phase information can be transferred from both terminals to the computer for processing.

Referring now to Figure 2, there are shown the main components of an inter-ferrometer terminal 13, also referred to as a "field terminal". Each field terminal 13 has an anti-21 · · • · · ie 8591 6 tennis arrangement 21 which is connected to the electronic equipment 23 by a coaxial cable 25.

Each antenna arrangement 21 includes an antenna 27 and a preamplifier apparatus 29. The antenna is located 5 on the measurement mark SM, and the position of the phase center of the antenna 27 with respect to the measurement mark SM must be precisely known.

Antenna 27 receives 1575.42 MHz radio signals transmitted by GPS satellites. The received signals are amplified by a preamplifier 29 and are fed via a coaxial cable 25 to a receiver unit 31 included in the electronic equipment 23, which receiver unit 31 includes a sideband isolator 33, a receiver power circuit 34 and an oscillator circuit 35.

In the sideband separator 33, the upper sideband portion of the signal, comprising the portion of the signals received from all satellites combined having a radio frequency range above 1575.42 MHz, is separated from the lower sideband portion corresponding to radio frequencies below 1572.42 MHz. To perform this separation, the sideband isolator 33 uses a 1575.42 MHz reference signal supplied by the oscillator circuit 35.

The receiver unit 31 supplies three signals. . in analog form to a digital electronic unit 37. · One analog signal, designated ··· * u (t), represents the upper sideband component of the received radio frequency signals shifted to the baseband. The second analog signal, designated l (t), •: - represents the lower sideband component also shifted to the baseband. Both of these signals contain effects from all visible satellites. The third. : The signal that is input to the digital electronics ···. to unit 37, is a sinusoidal signal having a frequency of 5.115 MHz which is the output of a freely oscillating, stable quartz crystal oscillator in the oscillator circuit 35 * '"*** 35.» »» «· i7 of the same oscillator The output of 8591 6 is multiplied by a fixed integer multiplier 308 within the oscillator apparatus to obtain a reference frequency of 1575.42 MHz used by the sideband separator.The accuracy of the frequencies generated by the oscillator apparatus 35 is typically about one in 9 8 parts of 10, although the accuracy is one. 10 would be tolerable.

In the digital electronics unit 37, all three analog inputs are converted into a digital 10 logic signal. These digital signals are processed under the control of the field terminal computer 39 to generate carrier power and false information. The digital electronic equipment 37 is connected to the field terminal computer 39 via a bidirectional data bus 41. The field terminal computer may be a Digital Equipment Corporation (DEC) model LSI-11/2 microcomputer, the data bus 41 may in this case be the "Q" bus of the DEC.

The carrier phase information is stored in the memory of the field terminal computer 39 until it is desirable to transmit this information to the central computer 15 for processing. As mentioned, the central computer 15 can be eliminated and the processing performed in one of the field terminals 39. The phase information can also be written by the field computer. . 39 out to a data storage medium such as a magnetic tape cassette or disk (not shown). Data can also be transmitted ti »* •» directly via an electronic connection or via a modern and telephone connection or by any conventional means.

Referring now to Figure 3, the components of the antenna arrangement 21 are shown in more detail. The arrangement 21 includes an antenna 27 which, as mentioned, is: constructed so that its phase center can be positioned 1 - * accurately with respect to the measurement mark. The 1575.42 MHz radio signals received by the antenna 27 are fed to a preamplifier circuit 35, which serves to increase their power level sufficiently to overcome the attenuation of the coaxial cable 25, which connects the antenna arrangement. 21 to the receiving unit 31 and the receiver unit 31 in the input-amplifier to overcome the generated background noise.

In the amplifier circuit 29, the signals received from the antenna 27 are first filtered by a bandpass filter 43 having a bandwidth of about 50 MHz centered at 1575.42 MHz. The function of the filter 43 is to prevent the receiver equipment 31 from being overloaded by strong random signals that may occur outside the GPS signal band. The output of the bandpass filter 43 is fed to a passive diode limiter 45 which protects the low noise amplifier 47 from burning under the influence of strong signals, such as those that could emanate from high power radars. The Vä-15 carbon noise amplifier 47 is a conventional gallium arsenide channel transistor (FET) amplifier with a noise pattern of about 2dB.

DC power to the low noise amplifier is supplied via a coaxial cable 25 connected to the pre-20 amplifier apparatus 29 from the receiver unit 31 via a high-frequency fuse 49 and a voltage regulator 51. Capacitor 53 connects the low noise amplifier 47 to the high frequency output cable 25 while shutting off direct current from the amplifier.

Referring to Figure 4, the components of the receiver unit 31 are shown in more detail. The receiver unit 31 includes a receiver power supply 34, a sideband isolator 33 and an oscillator circuit 35. The receiver power circuit 34 provides a DC power oscillator circuit 30, a sideband isolator 33 and coaxial time through the game 25 for low noise operation of the antenna arrangement 21. The oscillator circuit 35 generates a reference frequency of 1575.42 MHz for the sideband separator 33 and a reference frequency of 5.115 MHz for the digital electronic unit 37. The sideband separator 33 separates the signals received at the 1575.42 MHz centered and up and down 19 8591 6 high frequency band, as separate upper and lower side band components in the baseband.

The receiver power circuit 34 includes a controlled DC power supply 61 and in addition a storage battery 63. The Pa-5 Risto 63 occur. Thus, the frequency stability of the oscillator is maintained, the clock time setting is not lost, and the data stored in the computer's memory is not lost.

The oscillator 65 in the oscillator circuit 35 is a quartz crystal oscillator, such as the Frequency and Time Systems 15 (FTS) model 1001, which generates a 5.115 MHz output level.

O

one part of the tolerance of 10 or less. The stability of the FTS model 1001 is about one part of 10 ^ per day 12 and one part of 10 at intervals of 1-100 s and is therefore more than suitable for this purpose. The oscillator 65 generates two identical outputs, one going to the digital electronics unit 37 and the other to the 1575.42 MHz synthesizer 67 in the oscillator circuit 35.

The 1575.42 MHz synthesizer 67 includes a voltage controlled transistor oscillator (VCO) 69 which oscillates at 393.855 MHz, which is equal to 77 times 5.115 MHz. The phase of this oscillator is stabilized; With respect to the 5.115 MHz reference phase using a phase-shifted loop assembled from VCO 69, circuit 71, 30 divider 73, phase-to-frequency error detector 75, and loop filter 77. Part of the output power of VCO 69 is switched: / ·; connected 71 to the input of a frequency divider 73, which divider is assembled from conventional emitter-switched logic (ECL) integrated circuits which divide by llr and then by ... 35. The output of divider 73 is the "variable" input and the output of the oscillator 65 at 5.115 MHz is the "reference input::: input" to a conventional ECL integrated circuit-frequency detector 75. , such as Motorola Type nto MC12040 The output of detector 75 is low pass filtered in loop filter 77 to generate a control voltage applied to VCO 69. , which controls the sideband isolator 33.

Signals received at 1575.42 MHz in the center-10 band received from the antenna arrangement 21 via the coaxial cable 25 to the input 83 of the sideband isolator 33 are coupled by a DC blocking capacitor 85 through a bandpass filter 87 and amplified by an input amplifier 89. The DC power 25 via a high frequency fuse 91 from the receiver power circuit 34.

A high frequency power splitter or "hybrid" 93, a 1575.42 MHz local oscillator-quadrature hybrid 95, two 20 double-smoothed mixers 97 and 99, and a wideband video-frequency quadrature hybrid 101 in a sideband separator 33 demodulator ". Such demodulators are described, for example, in Alan E.E. In Rogers 'article • »···' in Proceedings of the IEEE vol. 59 (1971), pages 1617-1618. Its operation can be described here as follows.

Let fQ denote the frequency of the reference signal which: · - the oscillator circuit 35 supplies to the sideband isolator 33.

: 30 Nominally f is 1575.42 MHz, which is the same as the nominal carrier frequency of the GPS satellite "L1" transmissions,. : before the (first-order) Doppler transition. Thus, the outputs 102 and 103 of the quadrature hybrid 95 can be written sin 2n £ t and cos 2v £ t, respectively. These outputs **: 35 lots, which are in the quadrature phase, are "local oscillator inputs" to mixer 97 and 99, respectively. the two mixers • · • · · • · 21 85916 are identical. The baseband outputs of the mixers are correspondingly identical except for the phase shift of π / 2 radians. ("In the baseband, we mean a frequency range closer to zero than f, which corresponds to the difference between input frequency 5 and fQ.") The direction of this phase shift, above or below, depends on whether the frequency of the input signal is above or below fQ. Thus, it is possible to select either the upper sideband (higher input frequency) or the lower sideband and discard the opposite sideband by shifting the output phase of the second filter by more tt / 2 radians and then either by summing or subtracting (depending on which sideband is desired) the two mixers. outputs.

A quadrature hybrid 101 with two inputs 109 and 111 and two outputs 105 and 107 performs this π / 2 phase shift and summation / subtraction. The upper output 105 of the hybrid 101 is obtained as the arithmetic sum of the upper input 109 and the lower input 111, both inputs being delayed in phase by an amount dependent on frequency, but the phase shift of the lower input is greater than the upper input constant tt / 2 rad. The lower output 107 is obtained as the arithmetic difference 25 of the same two differently phase-shifted inputs 109 and 111, the difference being taken in the upper minus lower direction. Defined tt / 2 radian (one quarter period) phase difference

is accurately maintained for all frequencies between f and. . HP

at least fLp, where fHp m 19 kHz is much smaller than f ^ p <450 kHz and fLp is approximately equal to the single-bandwidth of the GPS "C / A" modulation m (t), as previously discussed. The structure of a quadrature hybrid having these properties is given in the above - mentioned Rogers article ·. glycol.

. Now the outputs of the quadrature hybrid 101 are amplified 35 by identical video amplifiers 113 and 115 and filtered by high-pass filters 117 and 119 and low-pass filters 121 and 123. Filters 117 and 119 are identical 22 8591 6 high-pass filters with a low-frequency cut-off frequency fAp. The purpose of the high-pass filters is to eliminate the DC components of the mixer outputs and any low-frequency spectral components that have the same or lower frequency than the maximum Doppler shift level that the GPS satellite signal could have.

It is desirable to limit all such components, as they could otherwise interfere with the subsequent determination of the phase of the received Doppler-shifted carrier 10 in the digital electronics and the field terminal computer. Such potentially interfering signals could include low frequency "flicker" noise generated in the mixers themselves or may follow mixer imbalance and (undesirable) low frequency amplitude or phase variations of the 1575.42 MHz reference signal, or any pre-mixers. Another potential source of low frequency interference is the hum or ripple of the power supply output voltages or currents. Another source could be an interfering continuous wave signal at a frequency close to fQ.

Low-pass filters 121 and 123 are identical sub-pass filters with a bandwidth of fTr), which is the same, ·. ·. than the bandwidth of one sideband of m (t). The response of each filter. ·· *, 25 men as a function of frequency is adapted to match the power spectral density of m (t). The purpose of these filters is to prevent noise and interference outside the bandwidth of m (t). Note that the wideband GPS "P-code" modulation signal n (t) would normally be a source of interference here. The majority of about 80% of the power from n (t) is discarded by these low-pass filters. This degree of rejection is sufficient to ensure that the "P-code" interference has a negligible effect, however, we find that if the '- narrowband m (t) modulation were turned off at 35 GPS satellites, then the wideband n (t) modulation would no longer represent an undesirable interfering signal , that: would become the desired signal Such a change in the structure of the GPS signal 23 8591 6 could be adapted by increasing the bandwidths of the low pass filters 121 and 123 by a factor of 10 to match them to the new signal.

The output u (t) from the low-pass filter 121 represents the lower side component of the original signal s (t) modified down and filtered down, and the output l (t) from the low-pass filter 123 represents the lower sideband. It should be noted that the spectrum of u (t) is shifted up in frequency and the spectrum of l (t) is shifted in frequency down relative to the spectrum of the original modulation m (t) by an amount (f-fQ) equal to the frequency f of the actual received carrier. and the difference between the frequency oscillator f of the local oscillator. (If the carrier Doppler shift (f-fQ) is negative then the spectrum of u (t) is shifted downwards and l (t) upwards.) The level of this shift is assumed to be less than fjjp and much smaller than f ^ p- This the assumption is satisfied if the frequency shift is mainly due to a Dopp-ler shift that ej can never exceed 5 kHz, provided that f is set to approximately lilr 20 io kHz. Any side-setting of the reference crystal oscillator 65 at the desired 5.115 MHz frequency also causes (308 times greater) spectral shift of u (t) and l (t). Usually, however, such a shift is very much smaller than fHp.

25 In addition to the offset j of the upper and lower sideband outputs u (t) and l (t), there is a frequency-dependent, mixed phase shift at both outputs due to the quadrature hybrid 101. However, for Rogers' (cited above) particular quadrature hybrid structure, this phase shift is too small. important. Similarly, a bandpass filter 87 and high and low pass filters. . The additional phase shifts caused by 117, 119, 121, and 123 are insignificant if conventional filter structures are used. Each of these effects also tends to undo-35, given the difference between the terminals in the next data processing. The cancellation is not accurate because the two filters are never exactly alike, and also; Doppler shifts in different parts are different at any one time 24 8 5 91 6. However, the residual effects are insignificant, as shown by direct calculations and confirmed by a real experiment.

Referring now to Figure 5, there is shown a block diagram of a digital 5 for its electronics unit 37. The digital electronics unit 37 includes a signal modulator 125, a correlator arrangement 127 comprising a group of seven identical correlators, a numerical oscillator arrangement 129 comprising a corresponding set of seven identical 10 numerical oscillators, and a real time clock 131 with a correlator arrangement 127, a correlator arrangement 127, a numerical oscillator on the data bus 133 to each other and to the field terminal computer 39. The first function of the signal modifier 125 is to convert the 15 analog upper sideband signal u (t), the analog lower sideband signal 1 (t) and the analog 5.115 MHz blue signal to each binary "digital" or "logic" signal. which is suitable for processing with conventional transistor-transistor logic (TTL) circuits. 20 The signal converter 125 produces only two outputs.

The second is a binary, TTL logic-level, periodic square waveform with a frequency of 10.23 MHz produced by doubling the input frequency of 5.115 MHz. This 10.23 MHz output acts as a "clock signal" to control the timing of all subsequent digital circuits. This clock signal is divided by 1023 (= 3 x 11 x 31) in a real-time clock 131 to obtain one pulse per 100 microseconds, further divisions by successive factors 10 thus result in a complete decimal representation of time in seconds -4 ·. ·. * The last 30 significant digits represent units 10 s.

The time is always readable in this format via data bus 133. Correlator arrangement 127, numerical oscillators -: * ♦ *: market arrangement 129 and field terminal computer 39 all ’. the functions are controlled by the real-time clock 131 via the data bus 133.

J 35 The second "digital" output of the signal converter 125 is derived from the analog inputs u (t) and l (t) and is a binary, TTL logic level, non-periodic waveform. This output is produced by a TTL-exclusive non- or logic gate with two inputs: one input represents the u (t) input character and the other the l (t) character. Thus, the gate output is "true" IT or 5 binary 1) if and only if the analog u (t) and l (t) signals have the same sign.

Figure 6 shows a block diagram of a signal processor 125. The analog signal u (t) is input to a comparator 135, the output of which is TTL logic level true when 10 u (t) is positive and false when u (t) is negative. This TTL logic signal is input as a second input to the TTL exclusive non- or gate 137. The analog signal 1 (t) is similarly input to a comparator 139, the output of which is input to the second exclusive non- or gate 137 15 input. The sinusoidal 5.115 MHz signal obtained from the crystal oscillator 65 is input to a conventional analog frequency doubling circuit 141, the output of which is input to a third comparator 143 to produce a 10.23 MHz TTL level square wave output. The 10.23 MHz 20 output is also used as a "clock" input to flip-flop 145, which performs sampling and holding of the output of gate 137. Thus, the output of flip-flop 145 is an exclusive non- or function of the signs u (t) and 1 (t) taken uniformly. at a frequency of 10.23 x 10 6 per second and a maintained sampling interval. It is well known in the art of radio interferometers, such as J.M. Moran has presented in an article published in Methods of Experimental Physics, Vol. 12, Part C, pages 228-260, that the time binary function UHL has a Fourier transform 30 or "spectrum" which is a good approximation of the Fourier spectrum of both the phase and the relative amplitude of the analog input u (t) 1 (t). The accuracy of the approximation depends on analog signals that are random in nature ’. and normal. Also, the correlation coefficient between these two inputs must be much smaller than 1. (In fact, the noise "equalizes" the nonlinearities of the comparators. The exclusion non- or gate 137 can be considered as a multiplier with values of + 1 and 26 85 91 6 for each input. -1.) These conditions are well met in the present system. Thus, in the following, the logic plane flip-flop 145 is considered to simply represent the input u (t) l (t).

5 The UHL "input" from the signal modulator 125 is applied in series to each identical correlator in the correlator arrangement 127.

Before describing the structure of the correlator arrangement 127, the principles of its operation will be briefly explained.

10 In each correlator, the product of u (t) l (t) is correlated with the binary approximations of the sine and cosine functions over time, developed by the equivalent of seven numerical oscillators. The frequency of the oscillator is controlled by the field terminal computer 39 according to the time-15 of the real-time clock 131. At any given time, the oscillator frequency is set to twice the predicted Doppler frequency shift of one of the 1575.42 MHz carriers transmitted by the satellites. One oscillator and one correlator are associated with each of the 20 visible satellites up to a maximum of seven satellites. (In principle, if more than seven satellites were ever visible, more numerical oscillators and correlators could be used in the system. In practice, seven is enough.) If the predicted Doppler-25 shift is close enough to the actual Doppler shift, then the correlator outputs accurately measure the power and phase of the particular satellite signal for which the prediction was made and is not significantly affected by the presence of signals from other satellites with different Doppler shifts.

As a mathematical representation, the operation of a single numerical oscillator and its associated correlator is described as follows: As a function of time t, the predicted Doppler frequency shift of the satellite carrier 131 is given by f (t). The value of the function f (t)

P P

* ... * is interpolated from a table of pre-calculated values previously stored in the memory of the field terminal computer.

a · · 27 8591 6

The numerical oscillator generates two functions of time: cos / “2φ (t)] and sin / 12φ (t) J in quadrature phases, where φ (t) represents the predicted phase, which is a function of time.

P

The function 4> p (t) is initially equal to zero at time tQ, 5 when the numerical oscillator starts to oscillate and at any subsequent moment φ (t) is obtained as an integral

P

t φ (t) = 2tt / f (t ') dt' p t p o

10 where f (t ') represents the instantaneous value of f as the current het-P P

as t '. A factor of 2it is necessary if, as usual, the frequency f is measured in units of cycles per unit time and

step φ is assumed to be measured in radian units pitch-P

than as episodes.

15 Now a correlator operating between moments tQ and t ^ forms the values a and b from its inputs f \ x {t) l (t) J,

cos At) J and sin £ 2 $ At) J formulas P P

t-i a = / u (t) l (t) cos / 2φ (t) Jdt t p 20 ro and t b = / ^ u (t) l (t) sin / 2φ (t) ydt.

P

The integration interval t ^ -t is one second and shows-_ ···. The integrated integrations are performed every second. At each one-second pulse from the real-time clock, the values of the integrals are selected in the memory registers, the integrations are reset to zero, the numerical oscillator is restarted, and a new integration cycle begins. Thus, at the end of each second 30, the correlator supplies outputs a and b representing the input u (t) l (t) cos / ^ p (t) J and the input u (t) l (t) sin / '2φp, respectively. t) 7 time averages in the interval of one: 1 · 1: second. These outputs represent the correlations of the input u (t) l (t) with the cosine and sine functions. 35 During a one-second interval, the oscillator frequency “··· 'f (t) is updated every 0.1 second by a computer,: Y: synchronized with 0.1-second" pulses "from the real-time 28 85 91 6 clock. This update is necessary because the satellite's Doppler shift changes due to the satellite's motion with respect to the ground terminal and the change in relative velocity projection along the line of sight at a rate of 5, which can be a substantial fraction of 1 Herz per second.

Now the correlator outputs a and b can be combined to obtain an estimate of the signal power and phase of the particular satellite for which the prediction f (t) was made.

P

Determine a complex number c whose real part is yh-10 as a and whose imaginary part is equal to b. That is, c = a + jb 15 where j is the square root minus one. Thus c ~ C <m2> <exp / 2 j (φ-φ) J> tr where C is a positive real constant scaling factor, 2 20 <m> is the time mean square of the GPS modulation function m (t) over the integration time period tQ to t ^ and <exp / 2j (φ-φ0) _7> is the time-Γ average of the complex exponential function exp / l2 j (φ-φ) J over the same time interval. Provided that the difference, (φ “Φρ) / between the phase 25 φ = φ (-b) of the received GPS carrier signal and the corresponding prediction φ = φ (t) is not •» · Γ Γ ____: vary for a substantial part of the period during the integration time, then The level of C is approximately proportional to the average received power 30 [c | Ξ (a2 + b2) 1/2 C <m2> and the angle of c is approximately equal to twice the average phase difference (φ-φ):, ···. 35 le tan (b / a) 2 <(φ-φ)>

* · P

»» »1 · • · · •» • · · 29 8591 6

Note that for b and a, the angle of c is similarly determined by modulo 2π radians. Thus, the difference (φ-φρ) is defined as modulo π radians.

In order to accurately determine the power of the received signal and the carrier-5 phase (modulo ir) from a and b according to these formulas, two conditions must be met: first, as mentioned, the actual phase, φ (t), must differ from the predicted phase , Φρ ^) by an amount that changes by much less than one period during an integration time of one second, secondly, the correlator output-to-noise ratio obtained from the formula

SNRc = l2 / llll / 4IIBeffTint | 1/2 F

15 1 <1'2> <BeffTint) 1/2 F

must be much larger than one, where is the effective bandwidth of the signals u (t) and 1 (t), about 5 x 105 Hz, T ^ nt is the integration time equal to one second and 20 F is that proportion in u (t) and the power at l (t) derived from the GPS m (t) signal, not the noise. The coefficient (2 / π) takes into account the loss of correlation between u (t) and l (t) caused by the analog-to-digital conversion performed by the comparators of these signals in the sig-25 signal modifier. The coefficient (π / 4) takes into account its loss, ____; associated with the use of square wave approximation with sine and cosine functions in a correlator. Input B __Τ. . square ·: err mt root is equal to about 700. As a result, is valid; ratio 30

SNR = 350 · F

. . c ::: The proportion F of the power of each sideband from the GPS satellite depends on the gain of the receiving antenna and the noise pattern of the receiving system.

For the "MITES" antenna and the reception system described above and for the satellite's elevation angle of more than 20 °, it is experimentally known that F exceeds about 0.03.

As a result, the SNR is £ 10 c 5 which is sufficient for accurate power and phase measurements. The normal distribution of the noise of the complex quantity c in each part, real and imaginary, is given by the formula 10 oc ~! C | / SNRc

The former condition for the accuracy of power and phase measurements, namely that (φ-φ) does not vary

P

during the one-second integration time, is the same as the condition that the difference between the actual received carrier frequency f and the local reference frequency f does not differ from the predicted (numerical oscillator) frequency f by a substantial 1 Hertz fraction. This condition is satisfied in the present system by supplying feedback control to the frequency of the numerical oscillator to keep this frequency close to the actual received carrier frequency. This control is performed by a simple program executed by the field terminal computer 39. The description of this program follows.

The complex number c formed by the correlator outputs a and b k at the end of the one-second integration time interval is denoted c (t ^), where t ^ represents the time in the middle of this time interval. To the frequency (k + 1) of the numerical oscillator: for this time interval, a corrective increment V · is added: K · / [σ (^) ο * (^ _ 1)] / 2π Hz where K is a positive equal constant, ] ... 35 represents the angle of the complex number enclosed in parentheses, and cMt ^ -j) is the complex conjugate of the complex number c of the penultimate (k-1) time interval.

• · • * * 31 8591 6 The principle of operation of this program can be understood from the following example: If the frequency prediction is too low, say 0.1 Hz, then the angle of c advances 0.1 cycles in 1 second and the complex number c (t ^) c * ( t ^ _ ^) gets an angle of 5 (+0.1) x (2tt) radians (plus some zero-mean noise). In this case, increasing the positive increment reduces the negative error level of the frequency prediction from 0.1 Hz to (1 — K) x (0.1 Hz).

The value of K must be greater than zero or no reduction in the frequency prediction error will result from the feedback supply. The value must be less than 1 or the feedback is due to a delay in entering the correction. The exact value is not critical and the optimal value can be determined experimentally. The present system uses a ni-15 cell value of 0.5.

Another important effect of this frequency feedback is that the frequency of the numerical oscillator is "pulled" towards the actual received carrier frequency from the original frequency, which can be as much as 20 several Hertz above or below. This "locking" procedure is well known in the field of phase or frequency tracking feedback loops, as shown, for example, in Floyd M. Gardner's book, Phaselock Techniques, published by John Wiley & Sons, Inc., New York,. 25 1966.

_: The significance of the "locking" procedure for the present system is that the location of the pre-information measurement mark does not need to be a few kilometers more accurate.

· - The potential detrimental 30 side effect of the “locking” procedure in the present system is that of a numerical oscillator that is assumed to follow a certain; satellite, may instead be pulled to a different satellite frequency if the frequency of the latter is close to the frequency of the former and if the signal of the latter is strong compared to the signal of the former. Of such cases, · in order to limit possible subsequent damage, the field terminal computer program includes a device that limits 32 8591 6 to its cumulative increment, which can be summed up in the pre-frequency forecast to exactly about 10 Hz. Since the difference between the frequencies of satellites typically changes by about 1 Hz per second, the result is that only about 10 se-5 measurement data, or less than 1% of the measurement data obtained in the field, may become invalid by tracking the wrong satellite. Experience shows that this percentage is insignificant.

Referring now to Figure 7, there is shown a block diagram of a korre-10 generator module 149, which is one of seven identical such modules in the correlator arrangement 127. All seven modules have the same input UBL, which is the output of the signal processor 125 HBL. Each module 149 also receives an "cosine" input-15 input and a "blue" input from a corresponding set of seven numerical oscillator modules. The UBL input and cosine input go to an exclusive non- or gate 151, the output of which is an input to a "clocked" digital counter 153. The UBL input and a blue input go to a second exclusive non- or gate 155, the output of which is an input to a second counter 157. 153, 157, the contents are stored in the respective output buffers 159, 161 by the pulse-25 of the real-time clock 131 in the digital electronics apparatus 37, and the counters are then reset to zero. At a frequency of 10.23 MHz, which is controlled by a "clock" signal by a signal modifier. of 125, each counter 153, 157 is added by one if ··; and only if its input from the associated exclusive non- or gate 151, 155 is "true". Thus, at the end of each one to 30 second interval, the contents of the output buffer 159, 161 indicate the number of times between zero and: 10 230,000 that the UBL and cosine / blue inputs: ***: have been matched during the previous 1 second. When the contents of the counter output buffers 159, 161 are also connected to the data bus 133, through which the field terminal information machine reads the contents every second. Each counter / latch »► • · * 33 8591 6 can be a single integrated circuit, such as the LS7060 32-bit device manufactured by LSI Systems Inc.

The value of a previously determined by the cross-correlation between fu (t) l (t) J and cos / 2φ (t) J is obtained by field Γ 5 in the terminal computer 39 by subtracting 5 115 000 from the output of the "cosine" counter and dividing the result by 5,115,000. A value of 6 is similarly obtained by subtracting 5,115,000 from the output of the "sine" counter and dividing the result by 5,115,000. (Thus, the unit level of a or b represents a complete correlation / u (t) l, respectively. (t) with 7 'and a cosine or sine function. Before these results are stored in the memory of the field terminal computer 39, each number can be truncated to as few as 4 bits to save memory space.)

Referring now to Figure 8, there is shown a block diagram 15 of one of seven identical numerical oscillators in a numerical oscillator arrangement 129, each of which 163 provides a "cosine" and "sine" input to a single correlator module 149. Each numeric oscillator 163 includes a binary phase. -20 frequency register 169, binary adder 171, exclusive no or gate 173, inverter 175 and frequency divider 177.

Phase register 167 and frequency register 169 each have 32 bits and adder 171 is a 32-bit adder.

The binary number included in the phase register 167 represents the phase of the oscillator output at any time, with the most significant bit representing half a period, the next most significant representing a quarter period, and so on. The binary number included in the frequency register 169 in the same way - · 30 represents the frequency of the oscillator, in which case the most significant bit in this case has a value of 155,000 Hz, which is the same *. * ·: As 1/66 of the 10.23 MHz "clock" signal from the section • · · v: from the signal processor 125. The adder 171 sums the numbers contained in the back ____: cheese 169 and the phase registers 167. The sum, ·. · # 35 ma is loaded into the phase register 167, replacing the previous input once in the period of the output of the divider 177, which divides the 10.23 MHz "clock" signal by a fixed factor 33.

A *: Phase register 167 is thus accurately updated at a frequency of 34 8591 6 310 000 times per second. The amount by which the phase proceeds with each update is obtained from the contents of the frequency register 169. The frequency register 169, as mentioned, is updated 10 times per second via the data bus 133 by the field terminal computer 39. (Both negative and positive frequencies are represented by the contents of the frequency register using a conventional two-complement method. According to this invention, a negative binary number is formed by complementing each bit and summing one. The largest positive number is accordingly shown to have zero as the most significant bit and all bits as ones. is one, it indicates that the number is negative.)

The sine output of the numerical oscillator 163 is obtained from the inverter 175, which inverts the most significant bit of the phase register 167. Blue output has a value of one when the phase is between zero and a plus half cycle and a value of zero when the phase is between stretch and the side (which is the same as the period should be between minus half period and zero sequence). The cosine output of the numerical oscillator 163 is taken from an exclusive non- or gate 173, the inputs of which are the most significant and next most significant bits of the phase register. The cosine output has a value of one when and only when the phase is plus or minus one quarter period from zero.

Referring now to Figure 9, there is shown a block diagram of a field terminal computer 39. The computer comprises a central processing unit (CPU) 181, program memory 183 and data memory 185, an external bidirectional data port 187 connected to the operator terminal 189, and an external bidirectional data port 191 connected to a modulator-demodulator (modern) 193, which in turn is connected to a telephone , to a radiotelephone or other telecommunications link 195. The parts of the computer 39 are connected to each other by a data bus 133, which also connects the computer 39 to other parts of the field terminal (see Figure 5).

- * CPU 181 may be Digital Equipment Corporation (DEC) model LSI-11/2 (part number KD11-GC), program memory 35 8591 6 183 may be 32 KB programmable read only memory, such as DEC part number MRV11-C, data memory 185 may be 32 KB direct access read-write memory, such as DEC part number MXV11-AC, two external bidirectional data ports (187 5 and 191) may be RS-232 serial data ports included in MXV11-AC, operator terminal 189 may be DEC model VT-100 or any series ASCII terminal that, like the VT-100, can be connected to the MXV11-AC's RS-322 serial data interface or any other suitable external data port device to the computer, the modern 193 can be any conventional RS-232 compatible device and it can be completely eliminated if, as mentioned, the field terminal computer 39 is connected directly to the base computer 15. The data bus 133 may be an LSI-11 Q bus.

15 The real-time clock 131, numerical oscillator arrangement 129, and correlator arrangement can be connected to the Q-bus by constructing them on conventional circuit boards directly connected to the "backplane" card edge connectors of the LSI-11 computer system. Such circuit boards are available from DEC equipped with special integrated circuits that can handle all data traffic between the Q-bus and special interferometer terminal circuits constructed on the boards.

The measurement data 25 stored in the memory 185 of the field terminal computer 39 comprises a time series of complex numbers for each detected satellite up to seven, with one such number being received every second. This data is obtained at intervals of about 5,000 seconds, during which time at least two satellites are always monitored, with an average of at least four satellite-30 paths being monitored. At time t i. The complex value of the satellite is denoted A ^ (t), V: where the level of this complex number is proportional to the measured ____ power of the signal received from this satellite at this time, the proportionality factor being arbitrary, ··· 35 but the same for all satellites and where the complex number; the angle is equal to twice the carrier phase measured for the same satellite during the same period, comparing the phases of each of the 36 8591 6 satellites to the same local reference oscillator signal, namely the 1575.42 MHz signal generated by the field terminal 13-1 oscillator circuit 35.

5 The complex data A ^ (t), i = 1, ..., 7 are applied by a field terminal computer 39 to a correlator adapter spindle 127 from the outputs of seven correlators 149 a and b as follows, i.

10 A ^ t) = / fa (t) + jb (t) 7 βχρ / ^ φρ (ΐ) J

where a (t) and b (t) represent the normalized outputs a and b, respectively, for a one-second "integration" or calculation interval centered around time t, j is minus 15 per square root and 2 ja (t) is i.

P

carrier phase in duplicate at time t. Note that the complex number A ^ (t) is the same as the complex number c derived from i. the output of the correlator multiplied by the term βχρ / ^ φ {t) J. The angle of A. represents the (double) phase of the received carrier compared to the (double) local comparison phase of 1575.42 MHz, while the angle of c is compared to the phase of this reference oscillator and the phase of the numerical oscillator (double). ).

For the purposes of this explanation, it is assumed that the data group {A ^ (t)} is that developed by the field terminal 13-1 at the starting point of the baseline vector. On the other. field terminal 13-2, which is a field terminal at the endpoint of the baseline vector and which monitors the same satellites at the same time as the first terminal, produces data corresponding to A ^ (t), designated B ^ t). Same •: satellites are monitored because both terminals were given forecast data from the same mainframe computer 15, *: **: which numbers satellites 1-7 in only one way. The observations in both terminals are practically simultaneous, because the clocks of the two terminals were synchronized immediately before the observations and the clock frequencies differ by * * * “· 4» · 37 8591 6 insignificant amounts. (The principal effect of the frequency difference between the crystal oscillators controlling the frequency of the clocks is to change the phase difference between the 1575.42 MHz comparisons.) It is irrelevant if at a given moment the known satellite is only visible from one terminal and not from another. In this case, the level of either A ^ (t) or Bi <t) will simply be zero or almost zero.

The steps performed by the central computer 15 to complete the determination of the interferometer baseline vector 10, providing the power and phase measurement data collected from the two field terminals 13-1 and 13-2 located at the ends of the baseline vector, will now be described.

The first step in processing the A ^ (t) and B ^ (t) data in the central computer is to multiply the A ^ (t) complex conjugate-ti, denoted by A ^ (t) B ^ itjs. input

Si (t) = A * (t) B ± (t) is the angle ZS ^ (t), which is equal to twice the difference between the measured phases of the carrier signals received from the satellite to the two terminals, each phase being measured at the corresponding terminal locally. with respect to the reference oscillator. Accordingly, the angle S '(t) is related to the difference between the phases of the local oscillators and to the baseline vector between the terminals with the theoretical dependence ^ Si (t) = A <J> L0 + (Anf ^ / c) S ·! (t) 30 where Δφ represents the phase difference of the local oscillators, is i. the received frequency of the satellite, approximately equal to 1575.42 MHz, c is the speed of light, S is the baseline vector, and s ^ t) is the unit vector i. viewed at time t from the center of the baseline vector. (This dependence 35 gives the angle ZS ^ (t) in radians and not in periods. Since the frequency f ^ is defined in periods and in radians per second the coefficient 2tr must be included. The reason that 38 8591 6 is 4π and not 2π is that both terminals measure double dependence of the received signal.) This dependence is an approximation insofar as it ignores second-order parallax-5 side effects, multipath, relativistic effects, noise, etc. of the propagation medium. These minor effects are omitted here for clarity. The error associated with omitting these effects is equal to about 1 cm of baseline error with a baseline length of less than about 1 km. / Lu-10 without listening to the noise, which is completely random, it is possible to create a model for the effects omitted above /, in order to obtain an even more accurate theoretical representation of Si (t). This model creation is described, for example, in I.I. In Shapiro's article entitled "Estimation of astrometric 15 and geodetic parameters from VLBI observations", published in Methods of Experimental Physics, vol. 12, part C, pages 261-176, 1976 ..7 Theoretically the plane of S is obtained 20 Is.I = C * G2 (cosO.) Where C is constant and G is the directional power gain of the receiving antenna written as a function of the cosine of the i. . G is assumed to be independent of 25 azimuths and is normalized so that the power received by an isotropic antenna matched by its ring polar foundation is equal to 1. For a MITES antenna structure, G (cos0) a (1.23) · (1 + cos0) 2 · sin2 ((3tt / 4) cos0), 0 ° <θ <90 °. 30 G (cos0) st 0, 9O ° <0: The value of this function is approximately 2.46 in zenith (0 = 0), it has one maximum of about 3.63 when Θ ä 40 °, the unit value when Θ ai 72 ° and approaches 0 as Θ approaches - ··· ': 35 90 °.

:: The next step in processing the measurement data obtained from the interferometer terminals is to sum the complex numbers S ^ (t) over 39 85916 i to obtain the sum S (t) for each measurement moment t S (t) = l S. (t) i = 1 x 5 where the sum covers all satellites observed at time t.

The next step in processing the measurement data is to select the default value b for the baseline vector b and calculate from this value b the time function S (t), which theoretically represents the value-10 that S (t) should have if the actual baseline value S were the same as the default value b : n S (t) = Σ | (t) · | (t) * exp [-j 4irb * s ^ (t) / λ ^] 15 where is the radio wavelength corresponding to the received carrier frequency. In other words, = c / f ^. The method for selecting the value of b is described below. Note that the theoretical function S (t), in contrast to the function S (t) derived from the measurements, does not include a term to represent the phase difference of the local oscillator-20 rin. The scaling constant C is also omitted.

Next, the level of S (t) is multiplied by the level of S (t) and the product of these levels is summed over all measurement moments to obtain a value of R (b) that depends on both b and naturally. ”25 measurements: R (b) = E | S (t) · | S (tn) i where t ^ represents l out of about 5,000 measurement moments.

-:; 30 The function R (b) is called the "ambiguity function".

The next processing step is to repeat the calculation of R (b) _ ·; for several values of b and determine the specific value of b for which the function R (b) has the largest value. This value of b is the desired determination of the baseline vector S.

35 The default value b of the baseline vector is originally selected to be the same as the best estimate of b available from independent information about the location of the measurement marks, 40 85 91 6 such as the locations obtained by identifying the landmarks * on the map. The maximization of Rlb) with respect to b is performed by examining a three-dimensional volume centered on

A

b is the original value of and is large enough to cover the inaccuracy of the al-5 original estimate. In the search, each point of a continuous-period three-dimensional network is examined to locate the single point where R (5) is at its maximum. The distance between the points in the grid is initially one meter. The volume extending two meters from the maximum point of R (b) is then examined by examining the grid at 20 cm intervals. The maximum of R (b) is found in this finer network. The network spacing is then halved and the linear dimension of the network is also halved and the search is repeated. This halving process is continued until the mesh spacing is less than 1 mm. The value of 15 b that finally maximizes R (b) is taken as the desired baseline vector S determination. Using n = 5 as the number of satellites, the determination of the baseline vector can be obtained by the method of the present invention with an accuracy of about 5 mm at both coordinates 20 for a baseline length of about 100 meters.

The method described above for processing measurement data from an interferometer terminal pair to determine a baseline vector between terminals represents a special case of the general method described by Charles C.

! 25 In an article by Counselman and Sergei A. Gourevitch entitled "Miniature Interferometer Terminals for Earth ·: · *: Surveying: Amgiquity and Multipath with Global Positioning *: · System", published in IEEE Transactions on Geoscience and Remote Sensing, Vol. GE-19, No. 4, pages 224-252, October 30, 1981.

In another measurement data according to the present invention. . in the embodiment of the processing method, the ambiguity function R (b) is also formed from the measurement data and the baseline **] "default value b, however, the function generation method 35 is different. In this embodiment, as in the previous * 85916 embodiment, the complex conjugate of A ^ (t) is multiplied by Bi (t) to obtain the complex input Si (t):

Si (t) = A * (t) Bi (t) 5 where (t) is a complex number representing i. Measurements of the signal received from the satellite to the second interferometer terminal at time t, the level of A ^ (t) being proportional to the received power and the angle ^ Ai (t) being 10 twice the carrier phase with respect to the local oscillator of the terminal and B ^ (t) is like A ^ t) except that it is derived from another terminal at the other end of the baseline vector.

Next, S ^ (t) is multiplied by a given complex exponential function of the baseline vector 15 default value b, and the product is then summed over all satellites observed at time t to obtain the sum S (t), which is a function of time and default value b 20 S (t) = Σ Si (t) exp [-j4irb * si (t) / λ ±] where s ^ (t) is the unit vector i. in the direction of the satellite at time t and i. the wavelength of the signal received from the satellite. (Note that if b is equal to £, then the angle of each term in the direction i is equal to Δφ, ~ regardless of i.) - “· Next, take the plane of S (t). and it is nested over all observation times to obtain the function R (b): 30 R (b) = Σ | S (t,) l 1 1. . where t ^ is 1. from about 5,000 measurement times.

Finally, the value of b that maximizes R (b) '· * * is searched for using the same search procedure described in the original *: **: 35 data processing method. This value of b is the desired baseline vector & determination.

This latter embodiment is computationally »« *; more effective than the method described first.

Claims (15)

A method for deriving location-related data from scattering spectrum signals having attenuated carrier paths followed by modulation by means of mutually orthogonal spectrum spreading codes, characterized in that in the method is received (21) antenna (27) a signal combination comprising potentially interfering continuous path components alongside the spread spectrum signals; from the signal combination, continuous path components are filtered (31) which would be present as an interfering signal in a continuous path component reconstructed from the spread spectrum signals, whereupon a continuous path component is reconstructed (125) from the spread spectrum signals; and from the reconstructed component (127) data is derived. 2. A method according to claim 1, wherein the spread spectrum signals are simultaneously transmitted at the same frequencies from the respective satellite, characterized in that in the filtering phase of the interference, the continuous path interference is also filtered on a narrow frequency band comprising a maximum Doppler displacement of a selected frequency caused by the relative movement between the satellites and the antenna. Method according to claim 1 or 2, characterized in that in the noise filtering phase (93, 79, 99, 95, 35, 101) the received signals are changed on a base frequency band, and the narrow frequency band is filtered (117, 119) on the base frequency band. 4. A method according to claim 3, characterized in that in the interference filtering phase (121, 123) further frequencies are filtered outside the broad frequency band, which essentially contains signals whose spectrum is expanded by a certain modulation code. 46 8 5 91 6 5. A method according to claim 1 or 2, characterized in that the selected frequency is a damped carrier with the average frequencies. 6. A method according to claim 5, characterized in that during the reconstruction phase, a number of continuous road components are reconstructed. 7. A method according to claim 6, characterized in that the same selected frequency belongs to a plurality of continuous road components. 10 8. A method according to claim 6, characterized in that the same spectrum-spreading modulation code includes a plurality of reconstructed components. 9. A method according to claim 1 or 2, characterized in that in the reconstruction phase (135, 137, 139, 145, 143) the phase and frequency of the signals whose spectrum is extended by a certain modulation code are doubled, so that these signals are reconstructed. is independent of the data obtained with respect to the information content of the modulation code. Method according to claim 9, characterized in that in the filtering and reconstruction phases (33) the received signals are further divided into. . first and second spectrum components, representing substantially different portions of their spectrum; and the spectrum components are cross-correlated (127) to reconstruct the spread spectrum signals. Method according to claim 10, characterized in that in the filtering phase (117, 121) the continuous wave components of the first spectrum component belonging to the selected frequency are filtered (119, 123) and the continuous wave components of the the second spectrum component belonging to the selected frequency. 35 4? 8591 6 12. A method according to claim 11, characterized in that in the filtration phase of the spectrum components (117, 119), the narrow frequency band containing a maximum Doppler displacement of the selected frequency caused by the selected frequency band is filtered further. the relative motion between the satellites and the antenna, and signals outside the broad frequency band are filtered (121, 123), to which bands essentially belong to signals which have been extended by a certain modulation code. 10 13. A method according to claim 10, characterized in that the first and second spectrum compartments are the upper and lower lateral band of the attenuated medium frequency carrier path included in the spread spectrum signals. 15 14. A method according to claim 9, characterized in that in the reconstruction phase, continuous road components which are included in the signals received from each satellite are reconstructed, whereby to derive data (127, 129) -20 in response to the different satellites from the reconstructed signals on the basis of frequency differences in the received signals caused by the Doppler offset. 15. A method according to claim 9, characterized in that in the reconstruction phase, a second combination is formed, which simultaneously includes continuous road components which are included in the signals received from the respective satellite, thereby further forming ( 129) a series of predicted signals whose frequencies belong to the frequencies received from the respective satellite, each predicted series is correlated (127) with a second combination signal to isolate the reconstructed components belonging to the respective satellite; and from each isolated component data is derived.