WO1998039667A1

WO1998039667A1 - Method of, and satellite navigational receiver for, determining a geographical location

Info

Publication number: WO1998039667A1
Application number: PCT/IB1998/000113
Authority: WO
Inventors: Kenneth Ronald Whight; Andrew Thomas Yule
Original assignee: Koninklijke Philips Electronics N.V.; Philips Ab
Priority date: 1997-03-05
Filing date: 1998-01-29
Publication date: 1998-09-11
Also published as: GB9704497D0

Abstract

A method of determining a user position fix in a satellite navigation system, comprises receiving encoded data signals transmitted by a plurality of orbiting navigational satellite vehicles (11, 12, 13 and 14), said data signals including ephemeris information which can be used to determine the relative position of each satellite vehicle whose data signals are being received, decoding the received signals to provide base band signals and determining the user position fix by processing the base band signals using fixed point arithmetic to calculate the relative position of each satellite vehicle from the ephemeris information contained in the received data signals.

Description

DESCRIPTION

METHOD OF, AND SATELLITE NAVIGATIONAL RECEIVER FOR, DETERMINING A GEOGRAPHICAL LOCATION

Technical Field

The present invention relates to a method of, and satellite navigational receiver for determining a geographical location.

Background Art Equipments using a satellite based global positioning system (GPS), such as NAVSTAR, for determining a geographical location are becoming more widely used. The NAVSTAR GPS is described in NATO Standardisation Agreement STANAG 4294 "NAVSTAR global positioning system GPS system characteristics - preliminary draft" but a brief summary of the system is included here. The NAVSTAR GPS consists of a number of satellite vehicles in approximately 12 hour, inclined orbits of the earth, each satellite transmitting continuous positional information. Two positioning services are provided by NAVSTAR, the precise positioning service (PPS) which is reserved for military use and the standard positioning service (SPS) which is available for general use. The following description is confined to the SPS although some features are common to both systems. A user of the GPS receives the transmissions from those GPS satellite vehicles currently in view and calculates their correct positions. Details of these calculations, using an Earth-Centred, Earth-Fixed (ECEF) reference system are given in the STANAG document. By measuring the propagation time of the satellite vehicle transmissions and hence the distances from three satellite vehicles to himself, the user can make an accurate calculation of his position in three dimensions. To make a valid positional fix, the user needs to measure the propagation times to an accuracy of better than 100ns and to facilitate this the satellite vehicle signals each have timing marks at approximately 1μs intervals. However, each satellite vehicle's signals are synchronised to an atomic clock and the normal user of the system will not maintain such an accurate clock. As a result the user's clock is said to be in error (in other words, different from satellite vehicle time) by a clock bias C_B. By measuring the apparent satellite vehicle signal propagation times from four satellite vehicles rather than three, the redundancy can be used to solve for C_B and the three accurate propagation times required can be calculated. The ranges of the user from the satellite vehicles are equal to the signal propagation times multiplied by the speed of light c. Prior to correction for the user's clock bias C_B, the apparent ranges of the satellite vehicles are all in error by a fixed amount and are called pseudoranges. Figure 1 of the accompanying drawings shows a radio receiver 16 in a user's vehicle 15 receiving signals from four GPS satellite vehicles 11 , 12, 13 and 14. The four pseudoranges of the satellite signals are denoted R1 , R2, R3 and R4. The positions of the satellite vehicles and the user's vehicle are shown as three-dimensional coordinates whose origin is the centre of the earth. Figure 2 of the accompanying drawings shows the equations used by a GPS receiver to calculate the three dimensional coordinates and the clock bias C_B from a knowledge of four satellite vehicle positions and their respective pseudoranges.

The data transmitted by each satellite vehicle consists broadly of three sets of information, the ephemeris, the almanac and the clock correction parameters. The ephemeris consists of detailed information about the satellite vehicle's own course over a period of a few hours, the almanac consists of less detailed information about the complete satellite vehicle constellation for a longer period and the clock correction parameters allow the user to correct for the GPS satellite vehicle's own clock errors. The positions of the satellite vehicles are calculated from the GPS ephemeris data and the Keplerian Orbital Parameters which are used to describe the orbit of each satellite vehicle. The satellite vehicle transmissions consist of a direct sequence spread spectrum (DSSS) signal containing the ephemeris, almanac, and the clock correction information at a rate of 50 bits per second (bps). In the case of the SPS a pseudo random noise (PRN) signal which has a chip rate of 1.023MHz and which is unique to each satellite vehicle is used to spread the spectrum of the information, which is then transmitted on a centre frequency of 1575.42MHz. The PRN signal is known as a coarse/acquisition (C/A) code since it provides the timing marks required for fast acquisition of GPS signals and coarse navigation. The signals received at a user's receiver have a bandwidth of approximately 2MHz and a signal to noise ratio (S/N) of approximately -20dB. In addition, since the satellite vehicles are each moving at a speed in excess of 3km/s, the GPS signals are received with a Doppler frequency offset from the GPS centre frequency. As a result, a stationary GPS receiver has to be capable of receiving signals with frequencies of up to +4KHz from the GPS centre frequency, and a mobile receiver (as is usually the case) has to be able to receive signals over an even greater frequency range. To recover the data and measure the propagation time of the satellite vehicle signals, the GPS receiver must cancel or allow for the Doppler frequency offset and generate the C/A code relevant to each satellite vehicle. Initially, at least, this can be very time consuming since to despread the DSSS signals, the incoming and locally generated PRN codes must be exactly at synchronism. To find the PRN code delay the receiver must compare the locally generated code and the incoming code at a number of different positions until the point of synchronism or correlation is found. With a code length of 1023 chips this comparison can be a lengthy procedure. However, once the frequency offset and the PRN code delay for each satellite vehicle are known, tracking them is relatively easy.

As electronic navigation systems become more widely available and the receiving apparatus comprises hand held devices as well as vehicle borne apparatus, factors such as unit costs and running costs are assuming a greater importance. Implementing the receiving apparatus as integrated circuits helps reduce both types of cost factors but the currently used method for carrying out the high accuracy geometric calculations for a receiver to determine the positions of the satellite vehicles from the ephemeris data usually involves 64 bit double precision floating point numbers requiring a relatively large processor to do the calculations which is expensive from the points of view of the cost of the device, the time required to do the calculations and the electrical power consumed.

Disclosure of Invention An object of the present invention is to expedite the calculation of the positions of the satellite vehicles in a satellite navigation system.

According to one aspect of the present invention there is provided a method of determining a user position fix in a satellite navigation system, comprising receiving encoded data signals transmitted by a plurality of orbiting navigational satellite vehicles, said data signals including ephemeris information which can be used to determine the relative position of each satellite vehicle whose data signals are being received, decoding the received signals to provide base band signals and determining the user position fix by processing the base band signals using fixed point arithmetic to calculate the relative position of each satellite vehicle from the ephemeris information contained in the received data signals.

According to a second aspect of the present invention there is provided a satellite navigational receiver comprising signal receiving means for receiving encoded data signals from a plurality of orbiting navigational satellite vehicles, said data signals including ephemeris information which can be used to determine the relative position of each satellite vehicle whose data signals are being received, decoding means for decoding the encoded signals and base band signal processing means, characterised in that the base band signal processing means comprises a fixed point arithmetic processing means to calculate the relative position of each satellite vehicle from the ephemeris information contained in the received data signals.

The present invention is based on the realisation that by carrying out each step in the calculation at an appropriate, predefined level of accuracy, the processing time drops by an order of magnitude. Further the calculations can be done using 32 bit integer arithmetic which is far easier to implement on a 16 bit microprocessor than 64 bit double precision floating point arithmetic. Thus not only is the microprocessor cheaper but also the amount of electrical power is reduced, which is of particular importance with portable, battery powered apparatus, because the calculations are carried out quicker.

If desired all the distances may be represented by non-standard units, for example 1/64 of a metre, which will enable any value of interest to be stored as a 32 bit integer.

Brief Description of Drawings

The present invention will now be explained and described, by way of example, with reference to the accompanying drawings, wherein:-

Figure 1 is a diagrammatic representation of an electronic navigation system,

Figure 2 shows the four equations for use in a satellite navigation receiver in calculating three dimensional coordinates and the clock bias C_B, Figure 3 is a block diagram of the main components of a satellite navigational receiver made in accordance with the present invention, and

Figure 4 is a block diagram of the base band processing stage included in the receiver shown in Figure 3.

Mode for Carrying Out the Invention

As Figures 1 and 2 have been described in the preamble of this specification their description will not be repeated.

The satellite navigational receiver shown in Figure 3 comprises an antenna 20 generally implemented as a small metal patch which collects respective spread spectrum signals from the orbiting satellite vehicles (not shown). The signals are right-hand circularly polarised on a 1575.42MHz carrier. An rf front end 22 is coupled to the antenna 20. The front end is a comparatively simple analogue section including a local oscillator tuned to a local oscillator frequency which is used to frequency down-convert the received signals to a much lower IF. A base band processing stage 24 is coupled to the rf front end 22 and comprises digital circuitry necessary to decode the respective spread spectrum signals from the satellite vehicles in view and to process the information to calculate the positions of the satellite vehicles and then a user position fix.

The base band processing stage 24 is shown in greater detail in Figure 4. The stage 24 comprises custom hardware arranged as a number of parallel channels CH1 to CHn. Each channel is capable of tracking a signal transmitted by a single space vehicle. Outputs of the channels CH1 to CHn are coupled to a processor 26 which is implemented as an embedded microprocessor. The processor 26 controls the channels and performs the position or location calculation. The processor 26 may be implemented as a digital signal processor (DSP).

Referring back to Figure 3, a user interface 28 is coupled to an output of the stage 24 and is implemented in a manner to suit a particular application. In the case of a stand alone device, the user interface 28 will comprise some form of display and man/machine means, for example a keypad (not shown), whereby a user controls the device. However if the electronic navigation components 20, 22 and 24 are incorporated into a larger system, such as an automotive navigation system, the user interface 28 will comprise a connection to components in the larger system which has its own display and control devices.

The heart of a satellite navigation receiver lies in the baseband processing. The custom digital channel hardware CH1 to CHn (Figure 4) is used to acquire and track signals from a number of different satellite vehicles 11 to 14 (Figure 1) (one satellite vehicle per channel). As all the satellite vehicles have on-board atomic clocks for synchronising their signals with each other, it is possible for the receiver to determine its relative range to each satellite vehicle by measuring the relative time of arrival of characteristic parts of these signals. Referring to Figure 1 , if, for example, the signal from satellite vehicle 11 arrives at the receiver 16 1ms before the signal from satellite vehicle 12, then satellite vehicle 11 must be 300 km closer to the receiver 16 than the satellite vehicle 12. If a receiver is tracking signals from four different satellite vehicles it is possible for it to use this timing information to determine its position in three dimensions, as well as obtaining an accurate (sub microsecond) absolute time. If the position of four satellite vehicles (X,, Y_it Z,) and their relative ranges R; (where i has the values 1 to 4) can be determined, it is possible to express the actual range to the satellite vehicles, using an (as yet unknown) offset range

R₀, as:

R, - R₀ + ΔR and then use simple Pythagorean geometry to express the receiver's position (X_r, Y_r, Z_r):

- xf⁺ (Yi - Y_r)^{2 +} (z, - z_rf = Ri

Combining these two equations gives us four equations in four unknowns (X_r, Y_r, Z_r and R₀):

sj{X₂ - X_{r +} (Y₂ - Y_rf ₊ (Z₂ - Z_r = R₀+AR₂

Solving these gives us the position of the receiver and, as a consequence of determining the range offset (R₀) and bearing in mind the atomic clock accuracy of the satellite vehicle signals, a very accurate measure of the current time. However, in order to do all this, the receiver needs to know where the satellite vehicles are. This is achieved by processing orbital parameters termed ephemeris data broadcast by the satellite vehicles themselves. These parameters define the motion of a satellite vehicle as a function of time in terms of a Keplerian (elliptical) orbit with 2nd order corrections. By following the algorithm published in the STANAG document, referred to in the preamble of this specification, it is possible for the receiver to calculate the position of a satellite vehicle to an accuracy of a few meters at any specified moment of time. Double precision floating point arithmetic has traditionally been employed and requires considerable processing involving trigonometric functions.

The Applicant has found that by studying the physical significance of the equations in the ephemeris algorithm this has given an understanding to be formed of the errors introduced by any inaccuracies. This has lead to the conclusion that, rather than using double precision floating point arithmetic, a customised form of integer arithmetic can be employed without incurring a significant loss of accuracy.

The basic concept is to carry out every step of the calculation at the appropriate (pre-defined) level of accuracy. For example, parameters that are measurements of distance or position need a range of +/-30, 000, 000m (defined by the orbital radius of the satellite vehicles and a resolution of a fraction of a meter. Therefore if all distances were represented using units of 1 /64th of a meter, that is about 1.5cm, a 32 bit integer could be used to store distance values (the range of a 32 bit integer is approximately +/-2,000,000,000; 2,000,000,000 times 1/64 is 32,000,000). Another way of considering this matter is that the 32 bit integer is a 32 bit number with a fixed (binary) point after the 26th bit.

This concept can be expanded to every type of parameter used in the satellite vehicle position calculation (e.g. angles, times, etc) so that all values can be represented by a 32 bit integer, albeit with different implicit binary point positions. The following table of parameters lists their appropriate accuracy and binary point position.

Implementing the appropriate calculations then involves purely integer arithmetic with some additional binary shifts to handle conversion between numeric representations. Integer arithmetic and binary shifts are operations which can be carried out very efficiently on a typical 16 bit embedded microprocessor (e.g. the XA obtainable from Philips Semiconductors).

Simulations of the custom fixed point arithmetic show that the position errors introduced by using this form of representation are substantially less than 1m in all cases and average about 20cm. This is perfectly adequate for anything but very specialised electronic navigational applications.

Benefits of using the method in accordance with the present invention are:

1. Using fixed point arithmetic reduces the processing time which drops by an order of magnitude. 32 bit integer arithmetic is far easier to implement on a 16 bit microprocessor than (64 bit) double precision floating point arithmetic. Estimated clock cycle counts for an XA processor suggest that the basic arithmetic operations (add, subtract, multiply and divide) can all be carried out at least a factor of 10 quicker than the known performance of a typical floating point library. This means that the processor need not be clocked as fast (saving power) or more significantly, a smaller, cheaper, processor can be designed into the system. 2. Code size is reduced because floating point arithmetic routines are no longer needed. This leads to a saving in ROM size and thus a reduction in cost. Also there is a slight reduction in RAM usage as all temporary values can be stored in 32 bits rather than 64 bit. From reading the present disclosure, other modifications will be apparent to persons skilled in the art. Such modifications may involve other features which are already known in the design, manufacture and use of systems satellite navigational systems, receivers therefor and component parts thereof and which may be used instead of or in addition to features already described herein.

Industrial Applicability

Satellite navigation receivers.

Claims

1. A method of determining a user position fix in a satellite navigation system, comprising receiving encoded data signals transmitted by a plurality of orbiting navigational satellite vehicles, said data signals including ephemeris information which can be used to determine the relative position of each satellite vehicle whose data signals are being received, decoding the received signals to provide base band signals and determining the user position fix by processing the base band signals using fixed point arithmetic to calculate the relative position of each satellite vehicle from the ephemeris information contained in the received data signals.

2. A method as claimed in claim 1 , characterised in that all distances are represented in units comprising a fixed fraction of a metre.

3. A method as claimed in claim 2, characterised in that the fixed fraction comprises 1/64 of a metre.

4. A method as claimed in claim 3, characterised in that distances determined by processing the base band signals are stored as a 32 bit integer.

5. A method as claimed in claim 4, characterised in that the distances are computed using a 16 bit microprocessor.

6. A satellite navigational receiver comprising signal receiving means for receiving encoded data signals from a plurality of orbiting navigational satellite vehicles, said data signals including ephemeris information which can be used to determine the relative position of each satellite vehicle whose data signals are being received, decoding means for decoding the encoded signals and base band signal processing means, characterised in that the base band signal processing means comprises a fixed point arithmetic processing means to calculate the relative position of each satellite vehicle from the ephemeris information contained in the received data signals.