JP5455542B2 - GPS receiver and method of processing GPS signals - Google Patents

Description

Related applications

This application is related to two patent applications filed by the same inventor on the same day as this application. The two applications are an improved GPS receiver using a communication link (Serial No. 08 / 612,582, filed on March 8, 1996), an improved GPS receiver with a power management function (March 8, 1996). Serial No. 08 / 613,966 of Japanese application).
This application is filed by the same inventor Norman F. U.S. Patent Application No. 60 / No. 60, dated provisional patent application entitled "Low Power, Sensitive Pseudo-Measurement Apparatus and Method for Global Positioning Systems" filed October 9, 1995 by Krasner. Insist on the right of.
Part of the disclosure of this patent specification includes what is subject to copyright protection. The copyright owner does not object to copying the patent document or patent disclosure as it is, as in the patent file or record of the Patent and Trademark Office, but otherwise holds all copyrights.

  The present invention relates to a receiver capable of determining satellite position information, and more particularly to a receiver applied to a global positioning satellite (GPS) system.

  A GPS receiver typically determines its position by calculating the relative time of arrival of signals transmitted simultaneously from multiple GPS (or NAVSTAR) satellites. As part of the message, these satellites transmit satellite position data called “ephemeris” data in addition to clock timing data. The process of searching for and acquiring GPS signals, reading the ephemeris data of multiple satellites, and calculating the position of the receiver from this data is time consuming and often takes several minutes. In many cases, this long processing time is unacceptable and significantly limits battery life in ultra-miniaturized portable applications.

  Another limitation of current GPS receivers is that their operation is limited when multiple satellites are clearly visible without obstructions and properly placed with good antennas to receive such signals. . Therefore, it is usually not usable in portable applications attached to the body, in areas obstructed by large foliage or buildings, or indoor applications.

  The GPS receiving system has the following two basic functions. In other words, (1) pseudoranges of various GPS satellites are calculated, and (2) the position of the receiving platform is calculated using the pseudoranges, satellite timing, and astronomical calendar (ephemeris) data. Pseudorange is simply the time delay measured between the received signal from each satellite and the local clock. Once acquired, the satellite astronomical calendar (ephemeris) and timing data are extracted from the GPS signal and tracked. As mentioned above, collecting this information usually takes a relatively long time (30 seconds to several minutes) and must be performed at a good received signal level to achieve a low error rate.

  Nearly all known GPS receivers use the correlation method to calculate pseudoranges. This correlation method is performed in real time and is often performed by a hardware correlator. The GPS signal includes a fast repetition signal called a pseudo-random number (PN) sequence. A code that can be used in consumer applications is called a C / A code and has a 1.023 MHz binary phase reversal or "chipping" rate and a repetition period of 1023 chips in a 1 millisecond code period. Chord sequences belong to a family known as gold chords. Each GPS satellite broadcasts a signal having a unique Gold code.

  For each signal received from any GPS satellite, the correlation receiver multiplies the received signal by the stored replica of the appropriate gold code contained in its local memory after the decrementing process to baseband, and then the presence of the signal. The product is integrated, ie, a low-pass filter, to obtain an index. This process is called a “correlation” operation. By sequentially adjusting the relative timing of this stored replica with respect to the received signal and observing the correlation output, the receiver can determine the time delay between the received signal and the local clock. The first determination of the presence of such output is called “capture”. When acquisition occurs, the process enters a “tracking” phase that adjusts the local reference clock by a small amount to maintain a highly correlated output. The correlation output during the tracking phase can be regarded as a GPS signal with the pseudo-random code removed, or in general terms “despred”. This signal has a narrow band, which is the same standard as the 50 bit / second binary phase-shifted data signal superimposed on the GPS waveform.

  The correlation acquisition process is very time consuming, especially when the received signal is weak. To improve acquisition time, most GPS receivers use multiple correlator devices (usually up to 12) that can search for correlation peaks in parallel.

  Some previous GPS receivers used FFT technology to determine the Doppler frequency of the received GPS signal. The receiver despreads the GPS signal using conventional correlation operations and provides a narrowband signal, typically having a 10 kHz to 30 kHz band. The resulting narrowband signal is then Fourier analyzed using an FFT algorithm to determine the frequency of the carrier. Determining the carrier in this way also provides an indication that the local PN reference is adjusted to the correct phase of the received signal and accurately measures the carrier frequency. This frequency can then be used for receiver tracking operations.

  Johnson's US Pat. No. 5,420,592 considers using the FFT algorithm to calculate pseudoranges at a central processing location rather than a mobile unit. According to that method, a snapshot of data is collected at a GPS receiver and then transmitted over a data link to a remote receiver where it undergoes FFT processing. However, the method disclosed here calculates only one forward and inverse fast Fourier transform (corresponding to four PN periods) to perform the correlation set.

As will be apparent from the following description of the invention, multiple FFT operations can be performed with special pre-processing and post-processing operations to increase sensitivity and increase processing speed.
(the term)

  In this patent, the terms correlation, convolution, and matched filtering are frequently used. The term “correlation” when applied to two series of numbers is to multiply the corresponding members of the two series by terms and then sum the series. This is sometimes referred to as “serial correlation” and produces a single numeric output. In some environments, a series of correlation operations are performed on consecutive groups of data.

The term “convolution”, when applied to two sequences of numbers, is the same as is commonly used in the art, filtering a second sequence of length m with a filter, and impulses of length n Equivalent to corresponding to the first sequence having a response. The result is a third series of length m + n-1. “Matched filtering” refers to a convolution, ie, a filtering operation, in which the above-described filter has an impulse response that is a complex conjugate that is time-reversed with the first sequence. The term “fast convolution” refers to a series of algorithms that compute convolution operations in an efficient manner.

  Although some authors use the terms correlation and convolution in an interchangeable manner, in this patent the term correlation always refers to the serial correlation operation described above.

  One embodiment of the present invention provides a method for determining the location of a remote GPS receiver by sending GPS satellite information, such as Doppler, from a base station to a remote unit or mobile GPS unit via a data communication link. The remote unit then uses this information and the received GPS signal from the view satellite to calculate the satellite pseudorange. The calculated pseudorange is then transmitted to the base station where the position of the remote unit is calculated. Various embodiments of an apparatus capable of performing this method are also described.

  Another embodiment of the present invention provides a GPS receiver having an antenna that receives a GPS signal from a view satellite and a multiplier that reduces the RF frequency of the received GPS signal to an intermediate frequency (IF). The IF signal is digitized and stored in memory for later processing by the receiver. This process is typically a programmable digital signal processor that, in one embodiment of the present invention, performs the necessary instructions to perform a fast convolution (eg, FFT) operation on the sampled IF GPS signal to obtain pseudorange information. It is executed using These operations typically also include pre-processing (before fast convolution) and post-processing (after fast convolution) of the stored version of the GPS signal or processing or stored version of the GPS signal.

  Yet another embodiment of the present invention provides a GPS receiver power management method and also provides a GPS receiver having a power management function. By receiving a GPS signal from the view satellite, buffering the signal and then turning off the GPS receiver, power consumption is reduced over prior art systems. Other power management functions are also described.

The invention is illustrated by way of example and non-restrictive accompanying drawing figures, in which reference numerals indicate similar elements.
FIG. 2 is a block diagram of the main components of a remote or mobile GPS receiver system that uses the method of the present invention, showing the data link that exists between the base station and the remote unit. FIG. 6 is a block diagram of an alternative GPS mobile unit. FIG. 6 is a block diagram of another alternative GPS mobile unit. 2 is two alternative forms of the RF and IF portions of a receiver that is an embodiment of the present invention. 2 is two alternative forms of the RF and IF portions of a receiver that is an embodiment of the present invention. 4 is a flowchart of the main operations (eg, software operations) performed by a programmable DSP processor according to the method of the present invention. Fig. 5 shows signal processing waveforms at various processing stages according to the method of the invention. 1 shows a base station system according to one embodiment of the present invention. 2 shows a base station system of an alternative embodiment of the present invention. In accordance with one aspect of the present invention, a GPS mobile unit with a local oscillator correlation or calibration function is shown. 5 is a flowchart illustrating a power management method for a mobile unit according to an embodiment of the present invention.

  The present invention relates to an apparatus and method for calculating movement, i.e. the position of a remote object, in a way that results in remote hardware that can operate at very low received signal levels with very little power consumption. That is, the power consumption is reduced while the receiver sensitivity is increased. This is made possible by implementing a remote reception function, as shown in FIG. 1A, and sending Doppler information from a separately located base station 10 to a remote or GPS mobile unit 20.

It should be understood that pseudoranges can be used to calculate the remote unit's geographic location in various ways. There are three examples:
1. Method 1: By retransmitting the satellite data message from the base station 10 to the remote unit 20, the remote unit 20 can combine this information with the pseudorange measurements and calculate its position. See, for example, US Pat. No. 5,365,450, incorporated herein by reference. Normally, the remote unit 20 performs a calculation of the position of the remote unit 20.
2. Method 2: Remote unit 20 collects satellite ephemeris data from GPS signals received in a common manner commonly practiced in the art. This data is typically valid for 1 to 2 hours and can be combined with pseudorange measurements to complete the position calculation, usually at a remote unit.
3. Method 3: The remote unit 20 may transmit a pseudorange to the base station 10 via the communication link 16, and the base station may combine this information with the satellite ephemeris data to complete the position calculation. it can. See, for example, US Pat. No. 5,225,842, incorporated herein by reference.

  In approaches (ie, methods) 1 and 3, the base station 10 and the remote unit 20 are sufficient to resolve the time ambiguity associated with the repetition rate of the GPS pseudorange code, looking at all satellites in common. It is assumed that they are placed as close as possible. This is satisfied when the range between the base station 10 and the remote unit 20 is 1/2 times the speed of light multiplied by the PN repetition period (1 millisecond), ie about 150 km.

  To illustrate the invention, it is assumed that Method 3 is used to complete the position calculation. However, upon review of this specification, those skilled in the art will appreciate that various aspects and embodiments of the present invention can be used with the three methods and other approaches described above. For example, in a variation of Method 1, satellite data information, such as data representing the satellite's astronomical calendar (ephemeris), is transmitted from the base station to the remote unit, and this satellite data information is calculated by the present invention from the buffered GPS signal. In combination with the pseudorange, the latitude and longitude of the remote unit (and often altitude) can be provided. It will be appreciated that the location information received from the remote unit can be limited to latitude and longitude, or can be comprehensive information including latitude, longitude, altitude, speed and orientation of the remote unit. Furthermore, the local oscillator correction and / or power management aspects of the present invention can be used in this variation of Method 1. Further, Doppler information can be transmitted to the remote unit 20 and used in the remote unit 20 in accordance with aspects of the present invention.

  In method 3, base station 10 commands remote unit 20 to perform a measurement via a message sent on data communication link 16 as shown in FIG. 1A. The base station 10 also sends Doppler information to the satellites in view within this message. This is a form of satellite data information. This Doppler information is typically in the form of frequency information, and the message typically also identifies identification data or other initialization data for a particular satellite in view. This message is received by a separate modem 22 that is part of the remote unit 20 and stored in a memory 30 coupled to a low power microprocessor 26. Microprocessor 26 handles data information transfer between remote unit processing elements 32-48 and modem 22, and also controls power management functions within remote receiver 20, as will be apparent from the discussion below. Typically, the microprocessor 26 will place most or all of the remote unit 20 hardware in a low power state or power outage, except when performing pseudorange and / or other GPS calculations, or when an alternative power source is available. Set to state. However, the receiver portion of the modem must at least periodically power up (to full power) and determine whether the base station 10 has sent a command to determine the location of the remote unit.

  The above Doppler information has a very short duration because the accuracy required for such Doppler information is not high. For example, if accuracy of 10 Hz is required and the maximum Doppler is about ± 7 kHz, an 11-bit word is sufficient for each satellite in view. If eight satellites are in view, 88 bits are required to specify all such Dopplers. Using this information eliminates the need for remote unit 20 to search for such Dopplers, thus reducing its processing time by a factor of ten or less. Using Doppler information further allows the GPS mobile unit 20 to process GPS signal samples more quickly, which tends to reduce the time that the processor 32 must receive full power to calculate location information. This alone reduces the power consumed by the remote unit 20 and contributes to improved sensitivity. Additional information may also be sent to the remote unit 20, such as an epoch of GPS message data.

  The received data link signal can use a high precision carrier frequency. The remote receiver 20 uses an automatic frequency control (AFC) loop to lock onto this carrier, thereby further calibrating its reference oscillator, as shown in FIG. 6 described below. A message transmission time of 10 milliseconds and a reception S / N ratio of 20 dB usually allows an accuracy of 10 Hz or more for frequency measurement via AFC. This generally exceeds the accuracy sufficient for the requirements of the present invention. This feature also improves the accuracy of position calculations performed conventionally or using the fast convolution method of the present invention.

  In one embodiment of the invention, communication link 16 is a narrowband radio frequency communication medium that is commercially available, such as a two-way pager system. This system can be used in embodiments where the amount of data transmitted between the remote unit 20 and the base station 10 is relatively small. The amount of data required to transmit Doppler and other data (eg, initialization data such as the ID of a satellite in view) is relatively small, and the amount of data required for position information (eg, pseudorange) is also relatively small. Therefore, a narrow band system is sufficient for this embodiment. This is different from systems that need to send large amounts of data in a short period of time, which systems may require higher bandwidth radio frequency communication media.

  When remote unit 20 receives both GPS processing instructions and Doppler information (eg, from base station 10), via the battery and power regulator and power switch circuit 36 (and controlled power lines 21a, 21b, 21c and 21d), Microprocessor 26 activates RF / IF converter 42, analog / digital converter 44 and digital snapshot memory 46, thereby providing sufficient power to these components. As a result, the signal from the GPS satellite received via the antenna 40 is reduced to the IF frequency and then digitized. A continuous set of such data corresponding to a duration of typically 100 milliseconds to 1 second (or longer) is then stored in snapshot memory 46. The amount of data stored increases the data that can be stored in the memory 46 (to obtain better sensitivity) if power savings are not as important as gaining better sensitivity, so that power savings are more sensitive than sensitivity. When important, the microprocessor 26 controls the data storage amount to be reduced. Usually, sensitivity is more important when the GPS signal is partially disturbed, and power saving is less important when abundant power sources (eg, automotive batteries) are available. Addressing this memory 46 for storing this data is controlled by the field programmable gate array integrated circuit 48. The reduction of the GPS signal is accomplished using a frequency synthesizer 38 that provides a local oscillator signal 39 to the converter 42, as discussed further below.

  Throughout this time (while filling the snapshot memory 46 with digitized GPS signals from satellites in view), the DSP microprocessor 32 can remain in a low power state. The RF / IF converter 42 and the analog / digital converter 44 are typically turned on for a short period of time sufficient to collect and store the data necessary for the pseudorange calculation. When the data collection is complete, the converter circuit is turned off or otherwise reduced in power via the controlled power lines 21b and 21c (while the memory 46 continues to receive sufficient power), so that the actual Does not contribute to further power consumption during pseudorange calculation. The pseudorange calculation is performed, in one embodiment, using a general purpose programmable digital signal processing IC 32 (DSP), as illustrated by the TMS320C30 integrated circuit from Texas Instruments. The DSP 32 is brought into an active power state by the microprocessor 26 and circuit 36 via a controlled power line 21e before performing such calculations.

  This DSP 32 differs from other DSPs used in certain remote GPS units in that it uses general purpose and programmable ICs rather than specialized custom digital signal processing ICs. In addition, the DSP 32 allows the use of a Fast Fourier Transform (FFT) algorithm, which makes pseudoranges very fast by quickly performing large amounts of correlation operations between locally generated references and received signals. Can be calculated. Usually, such a correlation 2046 is required to finish searching for an epoch for each received GPS signal. The fast Fourier transform algorithm can search for all such positions simultaneously and in parallel, thus increasing the required computational process by 10 to 100 times over conventional approaches.

  When DSP 32 finishes calculating pseudoranges for each of the in-view satellites, in one embodiment of the present invention, this information is sent to microprocessor 26 via interconnect bus 33. At this point, the microprocessor 26 can put the DSP 32 and memory 46 back into a low power state by sending appropriate control signals to the battery and power conditioning circuit 36. Microprocessor 26 then uses modem 22 to send pseudorange data to base station 10 over data link 16 for final position calculation. In addition to the pseudorange data, a time tag can be sent to the base station 10 at the same time indicating the time elapsed from the time when the data was first collected in the buffer 46 to the time when the data link 16 sends the data. This time tag improves the base station's ability to perform position calculations. This is because the GPS satellite position can be calculated when data is collected. Alternatively, Method 1 above allows DSP 32 to calculate the location of the remote unit (e.g., latitude, longitude or latitude, longitude and altitude) and send this data to microprocessor 26, which is also modem 22 This data is relayed to the base station 10 via In this case, position calculation is facilitated by maintaining the elapsed time from the time the DSP received the satellite data message to the time the buffer data collection started. This increases the ability of the remote unit to perform position calculations. This is because the GPS satellite position can be calculated at the time of data collection.

  As shown in FIG. 1A, the modem 22 transmits and receives received messages on the data link 16 using another antenna 24 in one embodiment. It is understood that the modem 22 includes a communication receiver and a communication transmitter that are alternately coupled to the antenna 24. Similarly, the base station 10 can use another antenna 14 to send and receive data link messages, so that the base station 10 can continuously receive GPS signals via the GPS antenna 12. it can.

  In a typical example, DSP 32 position calculations are expected to take no more than a few seconds depending on the amount of data stored in the digital snapshot memory 46 and the speed of the DSP or DSPs.

  From the above discussion, it is clear that the remote unit 20 need only start up a circuit with high power consumption for a short period of time unless the position calculation command from the base station 10 is frequent. In at least many situations, such an instruction is expected to result in the remote device being activated to a high power consumption state within only about 1% of the time.

  This in turn allows the battery to operate 100 times longer than would otherwise be possible. Program instructions required to perform power management operations are stored in EEPROM 28 or other suitable storage medium. This power management strategy is applicable to various power availability situations. For example, if the prime mover is available, the position determination is performed continuously.

  As indicated above, the digital snapshot memory 46 captures records corresponding to relatively long periods of time. When this large block of data is efficiently processed using a fast convolution method, the low reception level signal of the present invention is processed (for example, when reception is poor due to partial obstruction by buildings, trees, etc.). Contribute to ability. All pseudoranges of visible GPS satellites are calculated using the same buffered data. This improves performance over GPS receivers that track continuously in situations where the amplitude of the signal is changing rapidly (such as in an urban disturbance state).

  The slightly different implementation shown in FIG. 1B eliminates the microprocessor 26 and its peripherals (RAM 30 and EEPROM 28), and its functionality is contained within a more complex FPGA (Field Programmable Gate Array) 49. Additional circuitry takes over. In this case, the FPGA 49, that is, the low-power device, works to wake up the DSP 32 a chip when it detects activation from the modem 22 through the interconnect 19. The interconnect 19 connects the modem to the DSP 32a and the FPGA 19. When the DSP chip 32a wakes up, it transmits and receives data directly to and from the modem. The DSP 32a also performs power control operations through the interconnect 18 that is coupled to the battery and power regulator and switch 36 to provide power on / off commands to the circuit 36. The DSP 32a selectively turns on or lowers the power of various components through a power on / off command given to the circuit 36 by the interconnection unit 18 by a power management method as shown in FIG. . Circuit 36 receives such a command and selectively powers (or reduces power) various components. The circuit 36 wakes up the DSP 32 a through the interconnect 17. The circuit 36 selectively provides power to the various components by selectively switching power through the power lines 21a, 21b, 21c, 21d and 21f of the selected controller. Thus, for example, to supply power to converter 42 and converter 44, power is supplied to those converters via lines 21b and 21c. Similarly, power to the modem is supplied through a controlled power line 21f.

  Low frequency crystal oscillator 47 is coupled to memory and power management FPGA 49. In one embodiment, the memory and power management FPGA 49 includes a low power timer that includes a low frequency oscillator 47. When the timer of the FPGA 49 expires, the FPGA 49 sends a wake-up signal to the DSP 32a through the interconnect 17, and the DSP 32a then turns on other circuits by providing a power on / off command to the battery and power regulator and power switch circuit 36. Can be awakened. Other circuits are powered through controlled power lines 21a, 21b, 21c, 21d and 21f under the control of circuit 36 for positioning operations (eg, determining pseudo-range or position information such as latitude and longitude). The After the positioning operation, the DSP 32A resets the FPGA timer and reduces power to itself and the circuit 36 also reduces power to other components in the manner shown in FIG. The battery or batteries are expected to supply power to the total power control circuit through a control power line controlled by the memory and power management FPGA 49 and DSP 32a. Rather than directly reducing power by controlling a power line (such as 21b) to a component, the power consumed by the component is configured (as in the case of DSP 32a via interconnect 17 in FIG. 1B). It is also anticipated that it can be reduced by sending a signal to the element to reduce power, or by waking up to enough power, which often causes components such as integrated circuits to control the power state of the component This is possible if it has an input and the component has the internal logic necessary to control power consumption (eg, logic that reduces power to the various logic blocks of the component). Memory and power management FPGA 49 controls and manages the memory, such as address operations when data is stored from the converter 44 to the memory 46, or when the DSP component 32a is reading data from the memory 46. The FPGA 49 can also control other memory functions such as memory refresh as needed.

  FIG. 1C illustrates another embodiment according to the present invention of a GPS mobile unit that includes many of the same components as the GPS mobile unit shown in FIGS. 1A and 1B. The GPS mobile unit shown in FIG. 1C includes a plurality of batteries 81 and a power regulator 77 coupled to receive power from an optional external power input 83 and solar cell 79. The power regulator 77 supplies power to all the circuits under the control of the control power line managed by the DSP chip 32a and the memory and power management FPGA 39 shown in FIG. 1C. The solar cell 79 can charge these batteries using conventional charging techniques. In addition to charging the battery, the solar cell 79 can also supply power to the GPS mobile unit. In the embodiment shown in FIG. 1C, FPGA 49 provides a wake-up signal to DSP chip 32a via interconnect 75. This signal causes the DSP chip to return to a sufficient power state and perform various functions described for the DSP chip 32a. The DSP chip can also be activated to a sufficient power state via an external command from a modem 22 that is directly coupled to the DSP via the interconnect 19.

  FIG. 1C also illustrates a feature of the present invention in which a GPS mobile unit can sacrifice sensitivity for power savings. As described herein, the sensitivity of the GPS mobile unit can be improved by increasing the amount of interfering GPS signals stored in memory 46. This is done by capturing and digitizing more GPS signals and storing this data in memory 46. Thus, increasing the buffering increases the power consumption but also improves the sensitivity of the GPS mobile unit. The mode with the increased sensitivity can be selected by the power mode switch 85 of the GPS unit, which is coupled to the bus 19 and gives a command to the DPS chip 32a to enter the high sensitivity mode. Alternatively, this power mode switch 85 increases power storage by sending commands to the DSP 32a chip, reducing the capture snapshot of the GPS signal and thereby reducing the amount of GPS signal stored in the memory 46. The sensitivity can also be lowered. It will be appreciated that this power mode selection is made with the signal transmitted from the base station to the modem 22 and the modem can then communicate this command to the DSP chip 32 a via the interconnect 19.

  A representative example of a mobile GPS unit RF / IF frequency converter and digitization system is shown in FIG. 2A. An input signal of 1575.42 MHz passes through a band limiting filter (BPF) 50 and a low noise amplifier (LNA) 52 and is sent to the frequency conversion stage. The local oscillator (LO) 56 used in this stage is phase locked to a 2.048 MHz (or its harmonic) temperature compensated crystal oscillator (TCXO) 60 (via PLL 58). In the preferred embodiment, the LO frequency is 1531.392 MHz, which is 2991 × 0.512 MHz. The resulting IF signal is now centered at 44.028 MHz. This IF is desirable because low cost components are available near 44 MHz. In particular, surface acoustic wave filters (SAW) that are widely used for television applications can be easily obtained. Needless to say, other band limiting devices can be used instead of SAW devices.

  The received GPS signal is mixed with the LO signal by the mixer 54 to generate an IF signal. This IF signal passes through a SAW filter 64 to precisely limit the bandwidth to 2 MHz, and is then sent to an I / Q reducer 68, which converts the signal to near baseband (usually centered at 4 kHz). . The frequency of the local oscillator of this reducer 68 is obtained from the TCXO 60 at 2.048 MHz in the form of the 43rd harmonic of 1.024 MHz, ie 44.032 MHz.

  The I / Q reducer 68 is typically marketed as an RF component. This usually consists of two mixers and a low-pass filter. In such a case, the IF signal and the LO signal are supplied to the input port of one mixer, and the same IF signal and the LO signal shifted by 90 ° phase are supplied to the input port of the other mixer. The output of the two mixers is low-pass filtered to remove feedthrough and other distortion products.

  As shown in FIG. 2A, amplifiers 62 and 66 can be used before and after the band limiting operation, if desired.

  The two outputs of I / Q reducer 68 are sent to two matched A / D converters 44 that sample the signal at 2.048 MHz. An alternative implementation replaces the A / D converter 44 with a comparator (not shown), each of which outputs a binary (1 bit) series of data according to the polarity of the incoming signal. It is well known that this approach results in a loss of about 1.96 dB of receiver sensitivity relative to the multilevel A / D converter. However, the use of a comparator can save significant costs for the A / D converter and further reduce the memory requirements in the subsequent snapshot memory 46.

  An alternative realization of the reducer and A / D system is shown in FIG. 2B. This uses a band sampling method. The TCXO 70 to be used has a frequency of 4.096 MHz (or a harmonic thereof). The output of the TCXO can be used as a sample clock to the A / D converter 44 (or comparator), which serves to convert the signal to 1.028 MHz. This frequency is the difference between the eleventh harmonic of 4.096 MHz and the input IF frequency of 44.028 MHz. The resulting 1.028 MHz IF is about a quarter of the sample rate, which is known to be nearly ideal for minimizing sampling type distortions. Compared to the I / Q sampling of FIG. 2A, this signal sampler gives 1 channel data instead of 2 channels, but is twice as fast. The data is effective at an IF of 1.028 MHz. Next, I / Q frequency conversion to approximately 0 MHz is realized by digital means in the subsequent processing described below. The devices of FIGS. 2A and 2B are comparable in cost and complexity, and component availability often determines the preferred approach. However, it will be apparent to those skilled in the art that other receiver configurations can be used to achieve similar results.

  To simplify the following discussion, the following uses the I / Q sampling of FIG. 2A and the snapshot memory 46 is assumed to contain two channels of digitized data at 2.048 MHz.

  Details of the signal processing performed in the DSP 32 can be understood with the help of the flow diagram of FIG. 3 and the diagrams of FIGS. 4A, 4B, 4C, 4D and 4E. It will be apparent to those skilled in the art that the EPROM 34 has machine code or other suitable code for performing the following signal processing. Other non-volatile storage devices can also be used. The purpose of the process is to determine the timing of the received waveform relative to the locally generated waveform. In addition, to achieve high sensitivity, a very long portion of such a waveform is processed, typically 1 millisecond to 1 second.

  To understand the processing, first, each received GPS signal (C / A mode) is composed of a high-speed (1 MHz) repetitive pseudo-random number (PN) pattern made up of 1023 symbols, usually called “chips”. You can see that This “chip” is similar to the waveform shown in FIG. 4A. In addition, low speed data transmitted from the satellite at 50 baud is added to this pattern. All this data is received with a very low S / N ratio, measured in a 2 MHz band. If the carrier frequency and all data rates are known to very high accuracy and there is no data, the data-to-noise ratio is greatly improved by adding consecutive frames to each other and greatly reducing the data. be able to. For example, there are 1000 PN frames in a 1 second period. The first such frame can be consistently added to the next frame, the result added to the third frame, and so on. As a result, the duration is a signal of 1023 chips. Now the phase of this sequence can be compared with the local reference sequence and the relative time between the two can be determined to establish a so-called pseudorange.

  The above process must be performed separately for each satellite in the field of view from the same set of received data stored in the snapshot memory 46. This is because GPS signals from different satellites generally have different Doppler frequencies and different PN patterns.

  The above process becomes difficult because the carrier frequency becomes unknown above 5 kHz due to signal Doppler uncertainty and this amount is added due to uncertainty in the local oscillator of the receiver. This Doppler uncertainty is removed in one embodiment of the invention by sending such information from the base station 10 that simultaneously monitors all GPS signals from satellites in view. Accordingly, Doppler search is avoided at the remote unit 20. Local oscillator uncertainty is also significantly reduced (possibly up to 50 Hz) by performing AFC operations using base-to-remote communication signals, as shown in FIG.

  The presence of 50 baud data superimposed on the GPS signal still limits the consistent addition of PN frames over a 20 millisecond period. That is, up to 20 frames can be consistently added before further processing gain is hampered by the inversion of the data code. Additional processing gain can be achieved by frame matched filtering and addition of absolute values (or squares of absolute values), as detailed in the following paragraphs.

  The flowchart of FIG. 3 starts from step 100 and initializes a GPS processing operation with a command from the base station 10 (referred to as “fixed command” in FIG. 3). This command involves sending the Doppler shift of each satellite in view and the ID of that satellite via communication link 16. In step 102, the remote unit 20 calculates the local oscillator drift by frequency locking to the signal transmitted from the base station 10. An alternative is to use a very good temperature compensated crystal oscillator for the remote unit. For example, digitally controlled TCXOs, so-called DCXOs, can currently achieve an accuracy of about 0.1 / million, that is, an error of about 150 Hz on the L1 GPS signal.

  In step 104, the remote unit microprocessor 26 powers up the receiver front end 42, analog to digital converter 44, and digital snapshot memory 46, and K PN frames of C / A code. Data snapshots are collected for the duration, where K is typically 100 to 1000 (corresponding to a duration of 100 milliseconds to 1 second). When a sufficient amount of data has been collected, the microprocessor 26 turns off the RF / IF converter 42 and the A / D converter 44.

  The pseudorange of each satellite is calculated as follows. First, in step 106, a corresponding pseudo random number code (PN) code is searched from the EPROM 34 for any GPS satellite signal to be processed. As briefly discussed, the preferred PN storage format is actually the Fourier transform of this PN code sampled at a rate of 2048 samples per 1023 PN bits.

  The data in the snapshot memory 46 is processed in blocks of N consecutive PN frames, i.e. 2048N complex samples (N is typically an integer in the range of 5-10). Similar operations are performed on each block, as shown in the bottom loop (steps 108-124) of FIG. That is, this loop is executed a total of K / N times for each GPS signal to be processed.

  In step 108, the block of 2048N data words is multiplied by a complex exponent that removes the Doppler effect of the signal carrier and the drift effect of the local oscillator of the receiver. To illustrate, consider the Doppler frequency transmitted from the base station 10 and the local oscillator offset corresponding to feHz. When the data is multiplied in advance, the function e−j2πfenT, n = [0, 1, 2... 2048N−1] + (B−1) × 2048N, where T = 1 / 2.048 MHz is In the sampling period, the block number B is in the range of 1 to K / N.

  Next, in step 110, adjacent groups of N (usually 10) frames of data in the block are added together. That is, samples 0, 2048, 4096,... 2048 (N−1) are added together, then 1, 2049, 4097,... 2048 (N−1) +1 are added together, and so on. . At this point, the block contains only 2048 complex samples. An example of a waveform generated by such an addition operation is shown in FIG. 4B in the case of four PN frames. This addition operation can be regarded as a preprocessing operation prior to the high-speed convolution operation.

Next, in steps 112-118, each of the averaged frames is subjected to a matched filtering operation, the purpose of which is relative between the received PN code contained within the block of data and the locally generated PN reference signal. Is to determine the time. At the same time, the Doppler effect on the sampling time is also corrected. These operations are typically greatly speeded up in one embodiment by using fast convolution operations, such as the fast Fourier transform algorithm used in the method of performing cyclic convolution as described herein. .
To simplify the story, the above Doppler correction is initially ignored.

  The basic operation performed is to compare the data of the processed block (2048 complex samples) with a similar reference PN block stored locally. The comparison is actually performed by multiplying the reference element corresponding to each element of the data block (in a complex number) and summing the results. This comparison is called “correlation”. However, individual correlations can only be performed at a certain start time of the data block, and there are 2048 possible locations that are likely to be better aligned. The set of all correlation actions for all possible starting positions is called a “matched filtering” operation. In the preferred embodiment, a complete matched filtering operation is required.

  Other times of the PN block can be tested by cyclically shifting the PN reference and performing the same operation. In other words, when the PN code is p (0), p (1)... P (2047), the cyclic shift by one sample is p (1) p (2)... P (2047) p (0). It becomes. This variant sequence is tested to determine if the data block contains a PN signal starting with sample p (1). Similarly, a data block can start with samples p (2), p (3), etc., each tested by cyclically shifting the reference PN and rerunning the test. It is clear that the complete set of tests requires 2048 × 2048 = 4,194,304 operations, each of which requires complex multiplication and addition.

  Using Fast Fourier Transform (FFT), a more efficient and mathematically equivalent method can be used, which only requires about 12 × 2048 complex multiplication and twice the number of additions . In this method, in step 112, the FFT is performed on the data block and the FFT is performed on the PN block. The FFT of the data block is multiplied by the complex conjugate matrix of the reference FFT at step 114 and the result is inverse Fourier transformed at step 118. The resulting data thus obtained is 2048 in length and includes a set of data block and PN block correlations at every possible location. Each forward or reverse FFT operation requires P / 2 log2 P, where P is the size of the data to be sent (assuming an FFT algorithm with a radix of 2). In the case of the problem, since B = 2048, each FFT requires an 11 × 1024 complex multiplication. However, if the PN sequence FFT is pre-stored in the EPROM 34 as in the preferred embodiment, there is no need to calculate the FFT during the filtering process. The sum of the complex numbers is multiplied by the forward FFT and the inverse FFT, and the product of the FFT is (2 × 11 ÷ 2) × 1024 = 24576, which is a 171 times saving over the direct correlation. FIG. 4C shows a waveform generated by this matched filtering operation.

  The preferred method of the present invention uses a sample rate such that 2048 samples of data are taken during the PN period of 1023 chips. This allows the use of a 2048 FFT algorithm. A power two or four FFT algorithm has been found to be usually much more efficient than other sizes (and 2048 = 211). Therefore, the sampling rate selected in this way greatly improves the processing speed. The number of FFT samples is preferably equal to the number of samples in one PN frame so that proper circular convolution can be achieved. That is, as described above, this state allows the data block to be tested against all cyclic shift versions of the PN code. If you choose the size of the FFT to span a different number of samples than the length of one PN frame, use a set of alternative methods known in the art as "overlap save" or "overlap addition" be able to. This approach requires approximately twice as much computation as the approach described above for the preferred implementation.

  It will be clear to those skilled in the art how to modify the above process using different FFT algorithms of different sizes and different sample rates to perform fast convolution operations. There is also a set of fast convolution algorithms that have the property that the number of calculations required is proportional to Blog2B rather than B2 required for direct correlation. Many such algorithms are described in standard references such as H.264. J. et al. Listed in Nussbaumer, “Fast Fourier Transform and Convolution Algorithms” (New York, Springer-Verlag, C1982). Important examples of such algorithms are the Agarwal-Cooley algorithm, the split nesting algorithm, the recursive polynomial nesting algorithm, and the Wingrad Fourier algorithm, the first three for convolution and the latter for performing the Fourier transform. use. These algorithms may be used instead of the preferred method described above.

  Next, the time Doppler correction method used in step 116 will be described. In the preferred implementation, due to the Doppler effect on the received GPS signal and local oscillator instability, the sample rate used may not correspond exactly to the number of 2048 samples per PN frame. For example, it is known that Doppler shift can cause a delay error of ± 2700 nanoseconds / second. In order to correct this effect, the block of data processed as described above needs to be shifted in time to correct this error. As an example, if the size of a block to be processed corresponds to 5 PN frames (5 milliseconds), the time shift for each block can be as much as ± 13.5 nanoseconds. The time shift due to local oscillator instability is smaller. These shifts can be corrected by time shifting successive blocks of data by multiples of the time shift required for a single block. That is, if the Doppler time shift per flock is d, the block is time shifted by nd (n = 0, 1, 2,...).

  Generally, these time shifts are sample fragments. If these operations are directly executed using a digital signal processing method, a non-integer signal interpolation method is used, which increases the calculation burden. An alternative approach, the preferred method of the present invention, incorporates processing into the function of the fast Fourier transform. It is well known that a time shift of d seconds is equal to the Fourier transform of a function multiplied by e−j2πfd, where f is a frequency variable. Therefore, the time shift is the e-j2πnd / Tf (n = 0, 1, 2,..., 1023) and e-j2π (n-2048) d / Tf (n = 1024, 1025, ... 2047), where Tf is the duration of the PN frame (1 millisecond). This correction only increases the processing time associated with the FFT process by about 8%. The correction is split into two halves to ensure continuity of phase correction over 0 Hz.

  When the matched filtering operation is completed, the absolute value of the complex number of the block or the square of the absolute value is calculated in step 120. Both options work in much the same way. This operation removes the 50 Hz data phase inversion (as shown in FIG. 4D) and the remaining low frequency carrier error. Next, 2048 sample blocks are added to the sum of the previous blocks processed in step 122. Step 122 can be viewed as a post-processing operation following the fast convolution operation performed in steps 112-118. This continues until all K / N blocks have been processed, as shown by the decision block in step 124, where one block of 2048 samples remains, which calculates the pseudorange. FIG. 4E shows the resulting waveform after the addition operation.

  The pseudo distance is determined in step 126. Search for peaks above the locally calculated noise level. When such a peak is found, its time of occurrence relative to the start time of the block represents the pseudorange associated with the particular PN code and associated GPS satellite.

  In step 126, an interpolation routine is used to find the peak position to an accuracy much greater than that associated with the sample rate (2.048 MHz). The interpolation routine depends on the previous band filtering used for the RF / IF portion of the remote receiver 20. A good filter yields a roughly triangular peak with a base width equal to 4 samples. In this state, after subtracting the average amplitude (to remove the DC baseline), the two largest amplitudes can be used to more accurately determine the peak position. Let the sample amplitudes be Ap and Ap + 1, where Ap.gtoreq.Ap + 1, there is no loss of generality, and p is an indicator of peak amplitude. Thus, the position of the peak with respect to the position corresponding to Ap is obtained by the equation: peak position = p + Ap / (Ap + Ap + 1). For example, if Ap = Ap + 1, it can be seen that the peak position is p + 0.5, that is, between the indices of the two samples. In some cases, band filtering can round the peak, and 3-point polynomial interpolation is more appropriate.

  In the above processing, the local noise criterion used for threshold determination can be calculated by averaging all data of the final block averaged after removing some such maximum peaks.

  Once the pseudorange is known, processing continues in a similar manner for the next satellite in view at step 128 until all the satellites have been processed. Once all satellites have been processed, the process continues at step 130 where pseudorange data is sent over communication link 16 to base station 10 where the final position calculation of the remote unit is performed (using method 3). And). Finally, step 132 waits for a new command to put most of the remote unit 20 circuitry in a low power state and perform another positioning operation.

Next, the signal processing described above and illustrated in FIG. 3 will be outlined. An antenna on the remote GPS unit is used to receive GPS signals from one or more GPS satellites in the field of view at the remote GPS unit. This signal is digitized and stored in the remote GPS unit buffer. After storing this signal, the processor performs preprocessing, fast convolution processing, and post-processing operations. These processing operations involve the following.
a) The stored data is divided into a series of adjacent blocks whose duration is equal to a multiple of the frame period of the pseudo random number (PN) code included in the GPS signal.
b) Perform a pre-processing step in each block that generates a compressed block of data having a length equal to the duration of the pseudo-random code period by consistently adding successive sub-blocks of data. The duration of the sub-block is equal to 1PN frame. This adding step is to add the corresponding number of samples of each sub-block together.
c) Perform a matched filtering operation for each compressed block. This uses a fast convolution technique to determine the relative timing between the received PN code contained within the block of data and the locally generated PN reference signal (eg, a pseudo-random sequence of GPS satellites being processed). .
d) A pseudo-range is determined by executing an absolute value square operation with the product generated by the matched filtering operation, and blocks of absolute value square data are added to each other to generate a peak. Post-processing is performed by combining the data into one block.
e) Finding the peak position of the data of the single block with high accuracy using the digital interpolation method. Here, the position is the distance from the start of the data block to the peak, and the position represents the pseudorange of the GPS satellite corresponding to the pseudorandom sequence being processed.

  Usually, the fast convolution technique used to process the buffered GPS signal is Fast Fourier Transform (FFT), where the result of the convolution is the product of the forward transform of the compressed block and the forward transform representation of the pseudorandom sequence stored in advance. To generate a first result, and then perform the inverse transformation of the first result to collect the result. Also, the effect of the time delay induced by Doppler and the time error induced by the local oscillator is the complex FFT with phase-adjusted phase to the number of samples corresponding to the forward FFT of the compression block and the delay correction required for the block. A function multiplication is inserted between the forward and inverse fast Fourier transform operations to correct for each compressed block of data.

  In the above embodiment, the processing of the GPS signal from each satellite occurs not in parallel but in order with the passage of time. In an alternative embodiment, the GPS signals of all satellites in the field of view can be processed together in parallel in time.

  The base station 10 has a common field of view for all satellites of interest and is sufficiently close to the remote unit 20 to avoid ambiguity associated with the C / A PN code repetition period. . A distance of 90 miles meets this standard. The base station 10 also has a GPS receiver and is in a good geographical location that continuously tracks all satellites in view with high accuracy.

  Some embodiments of the described base station 10 use data processing components, such as a base station computer to calculate location information such as the latitude and longitude of the mobile GPS unit, but each base station 10 May simply relay received information, such as pseudoranges from a mobile GPS unit, to one or more central locations that actually calculate latitude and longitude. In this way, the cost and complexity of these relay base stations can be reduced by removing the data processing unit and its associated components from each relay base station. The central location will include a receiver (eg, a telecommunications receiver) and a data processing unit and related components. Further, in certain embodiments, a base station may be a satellite that sends Doppler information to a remote unit, thereby being virtual in that it emulates a base station in a transmitting cell.

  5A and 5B show two embodiments of a base station according to the present invention. In the base station shown in FIG. 5A, the GPS receiver 501 receives a GPS signal through the GPS antenna 501a. The GPS receiver 501 may be a conventional GPS receiver, and typically provides a time reference signal that is timed with respect to the GPS signal, and also provides Doppler information for satellites in the field of view. The GPS receiver 501 is coupled to a tuned local oscillator 505 that receives the time reference signal 510 and phase locks to this reference. This adjusted local oscillator 505 outputs to the modulator 506. Modulator 506 also receives Doppler data information signals and other satellite data information signals 511 from each satellite in the field of view of the GPS mobile unit. The modulator 506 modulates Doppler information and other satellite data information with the local oscillator signal received from the adjusted local oscillator 505 to provide a modulated signal 513 to the transmitter 503. The transmitter 503 is connected to the data processing unit 502 via the interconnection 514. The data processing unit controls the operation of the transmitter 503 so that satellite data information such as Doppler information is transmitted to the GPS mobile unit via the transmitter antenna 503a. In this way, the GPS mobile unit can receive Doppler information, the source of which is a GPS receiver 501, which can be used to calibrate the local oscillator of the GPS mobile unit, as shown in FIG. An accurate local oscillator carrier signal is also received.

  The base station as shown in FIG. 5A also includes a receiver 504 connected to receive communication signals from a remote or GPS mobile unit via a communication antenna 504a. It is understood that antenna 504a is the same antenna as transmitter antenna 503a in that one antenna acts as both transmitter and receiver in a conventional manner. The receiver 504 is connected to the data processing unit 502. The processing unit may be a conventional computer system. The processing unit 502 also includes an internal connection 512 that receives Doppler information and other satellite data information from the GPS receiver 511. This information can be used to process pseudorange information received from the mobile unit via the receiver 504 and other information. The data processing unit 502 is connected to a display device 508 such as a conventional CRT. The data processing unit 502 is also connected to a mass storage device 507 including GIS (Geographic Information System) software used to display the map on the display 508 (eg, Atlas GIS by Strategic Mapping, Inc., Santa Clara, Calif.). The Using the map on the display, the position of the moving GPS relative to the map displayed on the display can be displayed on the display.

  The alternative base station shown in FIG. 5B includes many of the same components as FIG. 5A. However, instead of acquiring Doppler information or other satellite data information from the GPS receiver, the base station of FIG. 5B does not acquire Doppler information or other satellite data information 552 acquired from the remote communication link or wireless link as usual. Including sources. This Doppler information and satellite information is transmitted to the modulator 506 through the interconnection 553. Another input of modulator 506 shown in FIG. 5B is an oscillator output signal obtained from a reference quality local oscillator, such as a cesium reference local oscillator. This reference local oscillator 551 provides a precise carrier frequency that modulates Doppler information and other satellite data information. This information is transmitted to the mobile GPS unit via the transmitter 503.

  FIG. 6 shows an embodiment of the GPS mobile unit of the present invention that uses a precise carrier frequency signal received through a communication channel antenna 601 similar to antenna 24 of FIG. 1A. The antenna 601 is connected to a modem 602 that is similar to the modem 22 of FIG. 1A. The modem 602 is connected to an automatic frequency control circuit 603 that is locked to a precise carrier frequency signal sent from the base station described herein in accordance with one embodiment of the present invention. Automatic frequency control circuit 603 provides output 604. This is usually locked to the precision carrier frequency. Comparator 605 compares this signal 604 with the output of GPS local oscillator 606 via interconnect 608. The result of the comparison performed by the comparator 605 is an error correction signal 610 given to the frequency synthesizer 609. In this manner, the frequency synthesizer 609 provides a higher quality calibrated local oscillation signal to the GPS reducer 614 through the interconnect 612. The signal supplied to interconnect 612 is similar to the local oscillator signal provided to converter 42 by interconnect 39 in FIG. 1A, and converter 42 connects to GPS antenna 613 to receive GPS signals. This is the same as the GPS decrementer 614. In an alternative embodiment, the result of the comparison performed by the comparator 605 is output as error correction to the DSP component 620 similar to the DSP chip 32 shown in FIG. 1A via the interconnect 610a. In this case, the error correction signal 610 is not supplied to the frequency synthesizer 609. The automatic frequency control circuit can be implemented using several conventional techniques, such as a phase locked loop or a frequency locked loop or block phase estimator.

  FIG. 7 illustrates a specific power management sequence according to one embodiment of the present invention. It will be appreciated that many methods are known in the art for reducing power. This can be done by slowing down the time given to the timed synchronization component, completely shutting off power to a particular component, or turning off a particular component circuit and not cutting off the rest. is there. For example, it will be appreciated that phase-locked loops and oscillator circuits require start-up and stabilization time, so the designer can decide not to completely (or not) turn off these components. The example shown in FIG. 7 begins at step 701 where various components of the system are initialized and put into a low power state. Periodically or after a predetermined time, the communication receiver of the modem 22 returns to the full power state and determines whether a command is transmitted from the base station 10. This occurs at step 703. If a request for location information from the base unit is received at step 705, the modem 22 alerts the power management circuit at step 707. At this point, the communication receiver of modem 22 is turned off for a predetermined time, or turned off and then periodically turned on again, indicated as step 709. The communication receiver can remain in full power at this point without turning off the power. Next, at step 711, the power management circuit restores the GPS receiver portion of the mobile unit to full power by increasing the power of the converter 42 and the analog / digital converter 44. If the frequency oscillator 38 is also powered off, the component will power up at this point, return to full power, and stabilize over time. Next, at step 713, a GPS receiver including components 38, 42 and 44 receives the GPS signal. This GPS signal is buffered in the memory 46 that has returned to the full power state as well when the GPS receiver returns to the full power state at step 711. When the collection of snapshot information is complete, the GPS receiver returns to the low power state at step 717. This typically reduces the power of the converters 42 and 44 and the memory 46 maintains a full power state. Next, at step 719, the processing system returns to the full power state, but in one embodiment, the DSP chip 32 is also supplied with full power. However, it will be appreciated that if the DSP chip 32 has a power management function, as in the embodiment shown in FIG. 1C, the DSP chip 32a will normally return to the full power state at step 707. In the embodiment shown in FIG. 1A where microprocessor 26 performs power management functions, a processing system such as DSP chip 32 may return to a full power state at step 719. In step 721, the GPS signal is processed by the method of the present invention as shown in FIG. Next, when the GPS signal processing is complete, the processing system is in a low power state as indicated at step 723 (unless the processing system also controls the power management function as described above). Next, in step 725, the communication transmitter of the modem 22 returns to the full power state in order to send the processed GPS signal back to the base station 10 in step 727. When the transmission of the processed GPS signal such as pseudorange information or latitude and longitude information is finished, the communication transmitter returns to the low power state in step 729, and the power management system delays for a predetermined period in step 731. wait. After this delay, the communication receiver of modem 22 returns to the full power state to determine if a request is being transmitted from the base station.

  Although the method and apparatus of the present invention has been described with reference to GPS satellites, it is understood that the teachings are equally applicable to positioning systems that use psuedolite or a combination of satellites and pseudolites. A pseudolite is a ground-based transmitter that broadcasts a PN code (similar to a GPS signal) modulated with an L-band carrier signal that is substantially synchronized with GPS time. Each transmitter is assigned a unique PN code so that the remote receiver can be identified. Pseudolites are useful in situations where GPS signals from orbiting satellites are not available, such as tunnels, mines, buildings, or other enclosed areas. The term “satellite” as used herein includes pseudolites or pseudolite equivalents, and the term GPS signal is intended herein to include signals such as GPS from pseudolites or pseudolite equivalents.

  In the foregoing discussion, the present invention has been described with reference to its use in the United States Global Positioning Satellite (GPS) system. However, it is clear that these methods are equally applicable to similar satellite positioning systems, and in particular the Russian Glonass system. The Glonass system differs from the GPS system in that it primarily distinguishes radiation from different satellites from each other by using slightly different carrier frequencies rather than using different pseudorandom codes. In this state, almost all the circuits and algorithms described above can be applied, however, when processing new satellite emissions, the data is preprocessed using a different exponential multiplier. This operation can be combined with the Doppler correction operation of box 108 of FIG. 3 without requiring additional processing operations. In this situation, only one PN code is required, so block 106 is omitted. The term “GPS” as used herein includes such alternative satellite positioning systems, such as the Russian Glonass system.

  1A, 1B and 1C show a plurality of logic blocks (eg, 46, 32, 34, 26, 30, 28 in FIG. 1A) that process digital signals, some or all of these blocks being a single It should be understood that while integrated into an integrated circuit, the programmable nature of the DSP portion of such a circuit can remain. Such an implementation is important for applications where power is very low and price is an issue.

  It will also be understood that one or several of the operations of FIG. 3 can be performed with physically embedded logic while maintaining the programmable nature of the DSP processor to improve overall processing speed. I want. For example, the Doppler correction capability of block 108 can be implemented with dedicated hardware that can be placed between the digital snapshot memory 46 and the DSP IC 32. All other software functions of FIG. 3 may be executed by the DSP processor in such a case. Also, several DSPs may be used together in one remote unit to increase processing power. Also, consider the time between collection of each frame set while collecting (sampling) multiple sets of frames of GPS data signal and processing each set as shown in FIG.

  An example embodiment of the present invention has been constructed that demonstrates the operation of the methods and algorithms described herein and further demonstrates the sensitivity improvement possible using the methods and algorithms. The demonstration system consists of a GPS antenna and an RF demultiplier from GEC Plesssey Semiconductors, Gage Applied Sciences, Inc. Followed by a digitizer buffer board. The antenna and reducer perform functions 38, 40, 42 and 44 of FIG. 1A, and the digitizer buffer performs functions 44, 46 and 48 of FIG. 1A. Signal processing was performed on an IBM PC compatible system using a Pentium® microprocessor running on a Windows® 95 operating system. This emulated the functions of the DSP chip 32 and the memory peripheral 34. The Doppler information of the satellites in view was supplied to the signal processing software as an input to the signal processing routine, emulating the functionality of the modem and microprocessors 22, 24, 25, 26.

  This demonstration system algorithm was developed using the MATLAB programming language. Numerous tests were performed on raw GPS signals acquired in various jamming situations. The test verified that the sensitivity performance of the demonstration system was much better than several commercial GPS receivers tested at the same time. Appendix A provides a detailed list of the MATLAB machine code used for this test and is an example of the fast convolution operation of the present invention (eg, FIG. 3).

  In the foregoing specification, the invention has been described with reference to specific illustrative embodiments. However, it will be apparent that various modifications and changes may be made without departing from the broad spirit and scope of the invention as set forth in the appended claims. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.

Appendix A

10: base station 16: data link 20: mobile or remote unit 32: general purpose programmable DSP chip 40: GPS antenna 42: RF / IF converter 44: analog / digital converter 46: digital snapshot memory 49: FPGA ( Field programmable gate array)
56: Local oscillator 58: PLL
501: GPS receiver 603: automatic frequency control circuit 613: GPS antenna 614: GPS decrementer 616: analog / digital converter 618: digital snapshot memory

Claims (8)

A method for determining the position of a satellite positioning system (SPS) mobile unit comprising :
Receiving at the SPS mobile unit a data signal containing satellite data information from a base station of a cellular communication system via a cellular communication link without using a relay satellite;
Receiving a SPS signal by a SPS antenna from a satellite within the field of view of the SPS mobile unit after activating a reducer coupled to the SPS antenna of the SPS mobile unit;
After stopping the downconverter coupled to said SPS antenna, in said SPS mobile unit, calculating at least one pseudorange based on said SPS signal satellites in view,
Transmitting the at least one pseudorange from the SPS mobile unit to the base station via the cellular communication link to determine the position of the SPS mobile unit within the base station;
The satellite data information includes Doppler information of satellites in the field of view of the SPS mobile unit without including satellite ephemeris information of the satellites in field of view;
The method, wherein the step of calculating the at least one pseudorange is performed using the Doppler information without using the satellite ephemeris information.   The method of claim 1, wherein the satellite data information is received from a location server and sent to the SPS mobile unit by the base station. The method of claim 1, wherein the at least one pseudorange is used to determine the position of the SPS mobile unit within the base station in combination with the satellite ephemeris information of a satellite within the field of view. The method of claim 1, wherein the at least one pseudorange is sent by the base station to a location server to determine the position of the SPS mobile unit in combination with the satellite ephemeris information of satellites in the field of view. . An antenna for receiving SPS signals from satellites that are in view of the S PS mobile unit,
A reducer coupled to the antenna;
A power management unit connected to the reducer to activate / deactivate the reducer;
A receiver for receiving at the SPS mobile unit a data signal including satellite data information from a base station of a cellular communication system via a cellular communication link without using a relay satellite;
A satellite positioning system (SPS) mobile unit includes a processing unit for calculating at least one pseudorange based on SPS satellite signals within said field of view,
The satellite data information includes Doppler information of satellites in the field of view of the SPS mobile unit without including satellite ephemeris information of the satellites in field of view;
The SPS signal is received by the power management unit activating a reducer coupled to the antenna;
The calculation of the at least one pseudorange is performed using the Doppler information without using the satellite ephemeris information after the power management unit stops the reducer ,
The receiver is configured to transmit the at least one pseudorange from the SPS mobile unit to the base station via the cellular communication link to determine a position of the SPS mobile unit within the base station. Said satellite positioning system (SPS) mobile unit. The SPS mobile unit according to claim 5 , wherein the satellite data information is received from a location server and sent by the base station to the SPS mobile unit. 6. The SPS mobile unit according to claim 5 , wherein the at least one pseudorange is used in combination with the satellite ephemeris information of a satellite in the field of view to determine the position of the SPS mobile unit within the base station. 6. The SPS of claim 5 , wherein the at least one pseudorange is sent to a location server by the base station to determine the position of the SPS mobile unit in combination with the satellite ephemeris information of satellites in the field of view. Mobile unit.

Images