EP1051637A1 - Receiver for ds-cdma signals - Google Patents

Abstract

A receiver for direct sequence code division multiple access (DS-CDMA) signals particularly for GPS signals includes a plurality of correlators. The correlators include a multiple delay and summing circuit (29) which produces mutually offset versions of the received signal which are summed and then correlated with the spreading code. Alternatively an element (39) takes the output of the code NCO (26) and produces from it multiple offset versions of the spreading code summed together which are correlated with the received signal. Correlation will be achieved if any one of the multiple offset received signals correlates with the locally generated spreading code or if any one of the offset spreading codes correlates with the received signal.

Description

RECEIVER FOR DS-CDMA SIGNALS

The invention relates to a receiver for spread spectrum direct sequence code division multiple access signals (DS-CDMA). The invention particularly, but not exclusively, relates to such receivers for receiving GPS signals. In this respect the term GPS is intended to include not only the US Global Positioning System but also the Russian Global Satellite System (GLONASS) and any other equivalent systems which may be established in the future. The US GPS system is a time of arrival positioning system which uses a nominal constellation of 24 low earth orbit satellites to provide a position fix anywhere on the earth's surface. The satellites broadcast their position and timing information using a direct sequence spread spectrum signal. Currently two frequencies are used in the full system but for low cost civilian use the carrier is at 1575 MHz . The satellites' orbits are designed such that provided that there is a clear view of the sky at least five satellites will be in view anywhere at any time. The minimum number of satellites required to be in view for a full position fix is four. These are needed to resolve the three unknown spacial dimensions and the ambiguity between the receiver clock and satellite clocks.

All of the satellites broadcast on the same frequency using different pseudo random noise (PRN) codes. Thus the down conversion of all the signals can be performed in parallel using a single front end stage. The separation into multiple channels can be carried out at baseband and is normally done in the digital domain.

The GPS signal is a spread spectrum DS-CDMA signal. In order to retrieve the data from this signal it must be correlated with a copy of the PRN code that was used to spread the signal on transmission. This code is known, but the frequency at which the code is being received and the exact timing of the code are not known.

The standard technique for finding the GPS signal is to perform a two dimensional search through the carrier frequency and code phase spaces. This involves choosing a possible frequency and shifting the code through the one thousand and twenty three possible phases while monitoring for a correlation product. If one is not found the code phase shifting must be repeated for a slightly different frequency until a correlation is located. The total process may take many seconds to complete depending on how close to the actual values of frequency and code phase the original choice was. Various algorithms are used to determine the best order and pattern in which to search.

The time to first fix and time to re-acquire are both critical measures of the performance of a consumer GPS system since they are the parameters most apparent to the user. Thus the basic issue is the fastest location of a spread spectrum signal which prior to despreading is below the noise floor. GPS receivers normally comprise a number of channels for example eight or twelve which can acquire different satellites in parallel. Thus the time taken to acquire signals from several satellites will not be significantly longer than that requires to acquire the signal from one satellite. Quite clearly this technique could be expanded to reduce the time necessary to acquire the signal from a single satellite. For example it is already been said that having chosen a possible frequency the code is stepped through the one thousand and twenty three possible locations while monitoring for correlation product. Quite clearly if one thousand and twenty three correlators were provided in each channel and each was offset by one chip period from the others then for each frequency it would only require the time taken for the PRN code to be transmitted as for detection of that code. However this would require a large increase in the amount of circuitry and therefore be expensive.

US Patent No. 5600670 discloses a GPS receiver in which the channels of the GPS receiver system are dynamically allocated during acquisition mode to hierarchically chain a slave channel module to a master channel. Each channel circuit includes two correlators, where each correlator receives a digitized received GPS signal and a delayed local PRN reference code signal to provide an output signal to an accumulator. During acquisition, the system provides two correlator channels where each channel includes two correlators for a total of four correlators. A PRN reference code signal is progressively delayed to provide a sequence of progressively delayed PRN reference code signals, where one of the sequence of progressively delayed PRN reference code signals are applied to each of the correlators. Each correlator channel also includes a digital mixer which receives a digitized GPS IF signal and a digital carrier reference signal and which provides the digitized received GPS signal. This is similar to, but not as extravagant of hardware as the solution mentioned in the previous paragraph in that only a limited number of delayed versions of the PRN reference signals are used depending on how many free correlators i.e. those not being presently used for tracking or acquisition, are available in the receiver.

It is an object of the invention to increase the effective speed of the code search and therefore the time to first fix and time to re-acquire the satellite signal without requiring an excessive amount of circuitry per channel. The invention provides a receiver for direct sequence spread spectrum code division multiple access (DS-CDMA) signals, said receiver comprising at least one correlation channel; wherein the correlation channel includes means for generating a plurality of versions of the spreading code mutually offset by an integer of the chip period, means for summing the plurality of spreading codes, and means for correlating the received signal with the summed spreading codes.

The invention further provides a receiver for direct sequence spread spectrum code division multiple access (DS-CDMA) signals, said receiver comprising at least one correlation channel; wherein the correlation channel includes means for generating a plurality of delayed versions of the received signal, each delayed version being delayed by a integer multiple of the chip period, means for summing the plurality of delayed versions of the received signal, and means for correlating the summed received signals with the spreading codes. The invention is based on the realisation that in the first case a correlation will be formed if the received signal corresponds with any of the offset spreading codes. Thus several code positions can be searched simultaneously. In a similar manner if a plurality of delayed versions of the received signal are summed and correlated within the spreading code then on despreading a correlation will be formed if the de-spreading code matches with any one of the delayed components. Thus in either case several code positions can be searched simultaneously. It will be appreciated that there is no conceptual difference between the correlation of a delayed signal with the spreading code and the correlation of the signal with an advanced code. The hardware requirements for the two solutions will of course be somewhat different. The integer multiple may be one. In this case a block of possible code spaces will be checked simultaneously. Thus for example if there are ten summed delayed received signals or ten summed offset codes then the code space search times will be reduced by a factor of approximately ten. There will of course be some additional time to find the correct code correlation in that what will have been obtained is the fact that the correct correlation is within the group which have been checked together. It will then be necessary to determine which member of that group produced the actual correlation. The alternative of a multiple chip delay would result in separated areas of code space being searched simultaneously. There is no distinction in principle between these two implementations except that the former would probably require less hardware or less complex software in order to sort out which member of the group produced the correct correlation.

The number of versions of the spreading code or the number of delayed versions of the received signal may be dependent on the actual or predicted strength of the received signal.

The signal to noise ratio will depend on the position of the satellite that is transmitting the signal to be acquired and may differ by up to 7dB between strong and weak signals. Basic search algorithms previously used make no distinction between the strong and weak signals and in order to acquire weak signals they use a longer time than is necessary to acquire strong signals. Thus by attempting a correlation between a greater number of delayed versions of the signal and the spreading code or between a greater number of offset spreading codes and the signal when the signal to noise ration is high less time is used for acquiring the strong signal than is necessary for acquiring a weaker signal.

The invention further provides a GPS receiver comprising a receiver as claimed in any preceding claim, said receiver having a plurality of reception channels for locking on to signals from a plurality of satellites, and means for calculating the position of the receiver by processing the signals received from the satellites.

The invention has particular application in GPS receivers where the time to acquire the position (which requires signals to be obtained from at least four satellites; assuming that no other position information is available) is a parameter which is frequently of great importance to the user. Whilst in most cases the signals will be acquired from the satellites in parallel it is necessary to obtain all four signals before the position calculation can be made. Further when acquiring a satellite for the first time any performance gain in sweeping the code space more quickly will be further multiplied by the number of frequencies which have to be searched.

The above and other features and advantages of the invention will be apparent from the following description, by way of example, of an embodiment of the invention with reference to the accompanying drawings in which:-

Figure 1 shows in block schematic form a GPS receiver in which the invention may be embodied,

Figure 2 shows in block schematic form an embodiment of a correlator channel incorporating the invention, and Figure 3 shows an embodiment of a delay and summing circuit suitable for incorporation in the correlator channel of Figure 2.

Figure 1 shows in block schematic form a typical GPS receiver which comprises an aerial 1 which is connected to the input of a low noise amplifier 2 to amplify the incoming signal. This signal is passed through a band pass filter 3 which has its pass band centered on the frequency of 1575 MHz. The output of the band pass filter is fed to a first input of a mixer 4 whose second input is fed from a local oscillator 5. The resultant output is fed to a first intermediate frequency band pass filter 6 which has each pass band centered on a frequency of 42 MHz. The output of the band pass filter 6 fed to a first input of a second mixer 7 whose second input is fed by the local oscillator 5 through a frequency divider 8. The output of the mixer 7 is fed through a low pass filter 9 whose output is connected to the input of an analogue to digital converter 10. The output of the analogue to digital converter 10 is fed to a bank of correlators 1 1. A microprocessor 12 controls the operation of the correlators and receives output signals from them. The microprocessor 12 also calculates from the data received from the correlators the position of the receiver.

All of the satellites broadcast on the same frequency using different PRN codes. Thus the down conversion of all the signals can be performed in parallel using a single front end stage and the separation into multiple channels can be carried out at baseband, and as in the case of the embodiment described is normally done in the digital domain, although this is not essential.

The signal appears at the aerial 1 as a very weak wide band signal. This is first amplified using the low noise amplifier 2. The 1575 MHz signal is then mixed down to a more manageable intermediate frequency. At this point it is still an analogue spread spectrum signal. It is then converted into a digital signal by the analogue to digital converter ten for application to the bank of correlators 1 1. Generally one correlator channel is required for each satellite to be simultaneously tracked, although time multiplexing of channels is possible. The acquisition and tracking of the correlation is controlled by the microprocessor 12 which will also perform the position calculation. The bank of correlators 1 1 and microprocessor 12 form the baseband part of the system which performs several functions. The first is to correlate with the DS-CDMA signals from the satellite vehicles. To do this it has to search for the correct frequency and code phase, lock on to the DS-CDMA signal, and obtain accurate timing information from the local code phase. Secondly, it has to down load data from the DS-CDMA signal concerning satellite positional information, such as almanac and ephemeris data, and other data such as satellite vehicle health, atmospheric conditions. Thirdly, it has to calculate pseudo ranges using timing data from the tracking loops and downloaded data to give the range from each satellite vehicle. Fourthly, it has to calculate the position using an appropriate model of the globe to calculate the position and velocity in the required reference system.

As indicated above a major problem is locking on to the satellite signals and in order to do this a search has to be made through both frequency and code space. The invention is particularly concerned with reducing the time taken for the code search.

Figure 2 shows in block schematic form a correlator channel. As shown in Figure 2 an input 20 which receives the signal produced by the analogue to digital converter ten is connected to first inputs of first and second mixer circuits 21 and 22. The second input of the mixers 21 and 22 are fed with the outputs of a carrier numerically controlled oscillator (NCO) 23. The two outputs of the numerically controlled oscillator 23 are in phase quadrature with each other. The output of the mixers 21 and 22 are fed to first inputs of further mixers 24 and 25. A code numerically controlled oscillator (NCO) 26 again produces an output which is fed to the second inputs of the mixers 24 and 25. The outputs of the mixers 24 and 25 are fed to respective integrate and dump registers 27 and 28 whose outputs are fed as data outputs to the microprocessor 12. A control output from the microprocessor 12 is fed to the code and carrier NCOs 23 and 26, to control their output frequencies.

If the elements 29 and 39 shown dotted are omitted then what has been shown so far is a standard correlator architecture for a GPS receiver.

The element 29 comprises multiple delay and summing units and Figure 3 shows such a unit suitable for use as the element 29 in the position between the input 20 and the two mixers 21 and 22. As shown in Figure 3 the unit has an input 30 which is connected to the signal input 20 and an ouput 31 which is connected to the first inputs of the mixers 21 and 22. The input 30 is connected to a first input of a summing circuit 33 and to the inputs of a number of delay circuits 34-1 , 34-2 to 34-N. The outputs of the delay circuits 34-1 to 34-N are connected to further inputs of the summing circuit 33 and the output of the summing circuit 33 is connected to the output 31.

At the input 30 the GPS signal has been converted to a digitised baseband signal. At this point it is a pseudo-random signal from which data can only be retrieved by correlation with the original spreading code. The delay and summing circuit 29 therefore, generates a new signal prior to correlation by summing copies of the original signal which have been delayed by integer multiples of the bit period of the PRN code, also known as a chip period. On despreading a correlation will be formed if the despreading code matches with any one of the delayed components. Thus several code positions can be searched simultaneously.

For each additional delayed component the signal to noise ratio will be reduced but the time required to search the full 1 ,023 positions of code space will also be reduced. Basic signal to noise ratio calculations indicate that up to ten summed components of the signal may be used for strong signals, giving up to a factor of ten reduction in code space search times.

It has been stated that if ten delayed versions of the signals are summed then the speed of acquisition can be increased by a factor of up to ten. This factor will, however, typically be less than ten since although the search through the code space will be ten times as fast the number of false correlations is likely to increase due to the reduced effective signal to noise ratio. Another factor which will reduce the advantage obtained is that all that has been obtained by this procedure is the knowledge that the correct code position is one of the ten which had been checked simultaneously. It is, of course, then necessary to identify which of those ten signals is the correct code position. Clearly, this can be done simply by going through each one of the ten positions in turn. There may, however, particularly with binary related numbers, be ways of speeding the search for the correct position. For example if there are eight mutually delayed versions of the signal then a knowledge that one of the eight is the correct position enables the actual position to be obtained with another three passes of the correlator. The first takes a group of four and identifies which group of four the correct position lies in. The next pass takes a group of two out of the four which have been identified as containing the correct position and the third pass takes one of the two remaining which contain the correct signal. Thus it can be determined in three passes of the correlator which is the correct delay position. There are of course other search strategies which may be more efficient for different numbers of summed signals or summed spreading codes.

The delay between the components which are summed can be a single chip period or a multiple of the chip period. A single chip period delay would result in block of code space being simultaneously correlated allowing a faster code sweep speed. A multiple delay would result in separated areas of code space being searched simultaneously. There is no conceptual distinction between these two implementations but the former implementation may be simpler to implement. There is no conceptual difference between the correlation of a delayed signal with a spreading code as described above and the correlation of the signal with an advanced code. Thus the element 39 which takes the output of the code NCO 26 and produces from it multiple offset versions of the spreading code summed together could be placed between the code NCO 26 and the mixers 24 and 25. Thus the second implementation is to produce a composite despreading code formed by summing copies of the original despreading code with delayed copies of itself.

The despreading code is formed using a shift register. Therefore the composite code can be produced by tapping various points of the shift register rather than by adding delay elements. This may enable a reduced hardware requirement. The digital baseband GPS signal would then be despread using the composite despreading code. The ultimate result is the same as for the first implementation.

In either case the number of components to be summed may be selectable. This would allow fast searches for known strong signals and slower searches for unknown or weak signals where the reduction in signal to noise ratio would more seriously effect the number of false correlations and the possibility of failing to detect true correlations and may result in a longer time to acquisition than if fewer or no multiple components were used to attempt correlation.

While the specific embodiment is described with reference to a receiver for GPS signals the invention has a more general application to receivers for DS-CDMA signals and will give the same advantages with respect to acquisition of the transmitted signal. Thus the invention may be applied in general communications receivers and the claims appended hereto are intended to include within their scope such receivers unless specifically restricted to GPS receivers. Although in the specific embodiment the baseband signal processing has been carried out in the digital domain this is not essential, but may be advantageous when separate ICs are used for the front end and the baseband signal processing, and appropriate analogue circuitry could be used in each correlator channel, in which case the ADC 10 would not be required.

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 and use of radio receivers, particularly but not exclusively GPS radio receivers, and component parts thereof and which may be used instead of or in addition to features already described herein. Although claims have been formulated in this application to particular combinations of features, it should be understood that the scope of the disclosure of the present application also includes any novel feature or any novel combination of features disclosed herein either explicitly or implicitly or any generalisation of one or more of those features which would be obvious to persons skilled in the art, whether or not it relates to the same invention as presently claimed in any claim and whether or not it mitigates any or all of the same technical problems as does the present invention. The applicants hereby give notice that new claims may be formulated to such features and/or combinations of such features during the prosecution of the present application or of any further application derived therefrom.

Claims

1. A receiver for spread spectrum direct sequence code division multiple access (DS-CDMA) signals, said receiver comprising at least one correlation channel, wherein the correlation channel includes means for generating a plurality of versions of the spreading code mutually offset by an integer multiple of the chip period, means for summing the plurality of spreading codes, and means for correlating the received signal with the summed spreading codes.

2. A receiver for spread spectrum direct sequence code division multiple access (DS-CDMA) signals, said receiver comprising at least one correlation channel, wherein the correlation channel includes means for generating a plurality of delayed versions of the received signal, each delayed version being delayed by an integer multiple of the chip period, means for summing the plurality of delayed versions of the received signal, and means for correlating the summed received signals with the spreading codes.

3. A receiver as Claimed in Claim 1 or Claim 2 in which the integer multiple is one.

4. A receiver as Claimed in any preceding claim in which the number of versions of the spreading code or the number of delayed versions of the received signal is dependent on the actual or predicted strength of the received signal.

5. A GPS receiver comprising a receiver as claimed in any preceding claim, said receiver having a plurality of reception channels for locking on to signals from a plurality of satellites, and means for calculating the position of the receiver by processing the signals received from the satellites.