GB2382007A

GB2382007A - Composite pulse transmitter and receiver

Info

Publication number: GB2382007A
Application number: GB0127124A
Authority: GB
Inventors: Ewan Lindsay Frazer
Original assignee: Thales Research and Technology UK Ltd
Current assignee: Thales Research and Technology UK Ltd
Priority date: 2001-11-12
Filing date: 2001-11-12
Publication date: 2003-05-14
Anticipated expiration: 2021-11-12
Also published as: GB0127124D0; GB2382007B

Abstract

A composite pulse transmitter combines two Ultra Wide Band (UWB) pulses separated by a time interval t chosen such that the spectral components of the pulses are attenuated in the composite signal, thus reducing interference. The time separation can equal zero ( t =0) or less than a pulse length, and may consist of a phase multiple of 180{ or 360{. For addition of like pulses at a given frequency F,<F> t =(2k+1)/2F</F>. Receiving means combines the received signal with another (expected, composite or gating) pulse, thus rejecting the same frequencies which the transmitter avoided. Gates may be replaced by a correlation.

Description

COMPOSITE SIGNAL The present invention relates to a method of and apparatus for the creation of a composite signal and/or the reception and deconstruction of such a signal. In particular, the present invention relates to ultrawideband (UWB) signals.

A stream of short pulses sent by a transmitter is a common form of Ultra-Wideband (UWB) signal. UWB technology has been used in applications such as communication, ranging, radar and localisation, particularly in difficult radio environments.

In known systems, UWB pulses may be equally spaced in time and some of them inverted in a known manner or, alternatively, the timing between the pulses may be

changed in some predefined but primarily random way.

Sometimes filtering or shaping of the pulses is also applied.

With such current systems there are problems with some types of UWB transmissions interfering with other services, in particular GPS navigation services at 1.2GHz

and 1. 6GHz, and cellular phone services at 1. 8GHz to ?. 2GHz.

The present invention aims to alleviate the interference problems associated with known systems.

Accordingly, in a first aspect, the present invention provides a method of creating a composite signal including the step of combining two pulses wherein the pulses are separated by a time "T", with "T" being selected such that one or more spectral components of one or both of the pulses is attenuated in the resulting composite signal.

In this way, the spectral region of attenuation can be selected so that the composite signal provides reduced interference with one or more other services, such as

those mentioned earlier. Preferably the composite signal is an ultra-wideband, (UWB) signal.

The pulses used are not necessarily identical i. e. they may differ from each other in terms of amplitude and/or duration and/or shape and/or spectral composition.

In some examples, the pulses will be of roughly the same energy and shape as each other.

Where it is stated that the pulses are separated by a time"T", the time"T" may be measured between any appropriate point of each of the pulses. For example, the selected point may be the mid-point of a pulse.

Alternatively, e. g. in the case of a unipolar pulse, the point selected may be a point of maximum amplitude.

Further, e. g. in the case of a bipolar pulse, the point selected may be a zero crossing point of the pulse.

Finally, the appropriate point for one pulse may be the same as or different from the appropriate point selected for another pulse.

The method may include more than two pulses, e. g. in order to provide multiple areas of spectral attenuation,

or better attenuation characteristics in the single chosen area.

The first produced composite signal may in turn be combined with further pulse (s) in order to produce one or more further composite signals. Effectively, the method may be repeated with the composite signal taking the place of one or more of the pulses and being combined with a further pulse or composite pulse. In the case where only two pulses are used, the pulses may be combined by addition or subtraction. In the case where there are more than two pulses, the combination may be by addition alone, subtraction alone or a combination of addition and subtraction. In either case, one or more of the pulses may be amplified or attenuated prior to combination.

The time difference"T"between a pair of pulses may be zero, or sometimes less than a pulse length or in other examples sometimes more than a pulse length. A value greater than pulse length simplifies the hardware required and therefore reduced power consumption. Pulse length may be defined as the length of the part of the pulse which is above a chosen energy or amplitude level,

and the level may be selected as appropriate according to the embodiment. In some examples, the pulses which are to be combined will be of roughly the same energy and shape.

The time spacing"T"may be chosen in relation to the frequency or frequencies of one or more of the pulses so as to produce one or more areas of spectral attenuation in the resulting composite signal. Such regions of spectral attenuation may be adjacent each other, thereby determining the width of a larger region of attenuation.

A suitable value or values of"T"may be selected so as also to provide desired signal characteristics in spectral areas other than the area (s) of attenuation.

For example, the value (s) of"T"may be chosen in order to improve spectral flatness at frequencies other than those in the region of attenuation. This, or other desired effects, may be achieved by selecting some or all of the value (s) of""c"in a pseudo-random way.

In some embodiments of the invention, where there are more than two pulses, the respective values of"T" may each be a multiple of a selected time value.

Tn some embodiments, where pulses are combined by addition, the time "#" between pulses or composite pulses of like energy creates spectral zeros at odd multiples of frequency f (i. e. f, 31, : Jf, 7f : etc) where fez (time T in seconds and frequency f in Hertz).

Where pulses or composite pulses are combined by subtraction, a time"T"between the pulses or composite pulses creates spectral zeros at even multiples of frequency f (i. e. Of, 2f, 4f, 6f etc.) where f-1/2 T as before.

As will therefore be appreciated, for any given frequency F, there are several values of"T"which can be used to provide a spectral zero at frequency F. For the addition of like pulses, such values are T = (2k+1) / 2F, where K = an integer i. e. 0,1, 2 etc. Similarly, for subtraction of like pulses, suitable values are T = 2k/2F, where K is an integer (0,1, 2...).

In a further aspect, the present invention provides an apparatus for creating a composite signal, the apparatus including means for combining two pulses, wherein the pulses are separated by a time T"which is selected in order to provide a region of spectral attenuation in the composite signal. The means for combining the pulses may also be arranged to carry out any or all of the steps described above.

The method for the creation of composite signal described above can also be used to attenuate or remove unwanted spectral components from any given signal, for example a received signal which includes one or more areas of interference. Therefore, in some embodiments, one of the pulses described above can be considered to be the composite signal including interference to be attenuated and the further pulse (s) are selected as previously described in order to attenuate the composite signal in the area (s) of interference.

Similarly, in a further aspect, the present invention provides apparatus for receiving a signal, including means for combining the signal with an expected pulse or composite pulse, or a gating pulse derived from

it, by a multiplying or gating means. Alternatively, the signal is combined with a composite pulse, or gating pulse, which is produced from the expected pulse or composite pulse by combination as described previously.

As before the time spacing"T"between the pulses or composite pulses is selected to provide one or more regions of spectral attenuation in the received signal.

The means for combining may also be arranged to carry out any or all of the steps described above.

In one embodiment of a receiver, the time difference value"T"can he considered to be a phase at a given frequency, being a multiple of 180 degrees or 360 degrees depending upon whether there is adding or subtracting.

The principle is that the receiver, if it gates the same pulses times as the transmitter (or multiplies the received signal by the same composite template), will reject the same frequencies as the transmitter avoided, i. e. it is a filter. However, in addition, it is possible to reject further frequencies which the transmitter knew nothing of by gating or multiplying with a further composite pulse constructed by combining the timing gates or pulses of the transmitter spaced by another"T", (or set of"T"s) chosen to reject those new frequency (ies).

For example, if the transmitter did nothing clever and just transmitted a single pulse, then the receiver could: a. gate at those places and also those places separated

by"T"from them in order to reject those frequencies related to"T". (i. e. two combined pulse trains spaced by "1""). b. Same as a. , but multiply by the pulse template rather than a rectangular"gate". This variant could add the trains together at the spacing, but to keep the system simple, we would try to arrange "1"" so that there was no significant overlap, and addition could be traded for a simple rectangular gate with no gate overlaps as per a.

By way of example, embodiments of the present invention will now be described with reference to the accompanying drawings in which: Fig. 1 shows a typical UWB pulse for use with the present invention; Fig. 2 is a schematic diagram of a pulse transmitter according to an aspect of the present invention; Fig. 3 is a schematic diagram of a pulse receiver in accordance with a further aspect of the present invention;

Fig. 4 is a spectral diagram of a composite pulse which has been constructed according to an embodiment of the present invention ; Fig. 5 is a spectral diagram of a further composite pulse which has been constructed in accordance with an embodiment of the present invention; Fig. 6 is a graph showing part of the time wave form of the composite pulse of Fig-5 ; and Fig. 7 shows a further composite pulse constructed in accordance with an embodiment of the present invention- A pulse for use in the present invention may be unipolar or bipolar. It is usually short in duration (typically, but not restricted to, the range of 0. 1 to 10 nanoseconds). Figure 1 shows a typical UWB pulse x-axis is time (arbitrary units), y-axis is amplitude. This pulse is nominally infinite in duration, but the energy reduces very rapidly away from the peaks of the pulse.

Figure 2 shows a composite pulse transmitter according to the invention. In Figure 2 the timing controller (1) controls the time spacing"T"between pulses produced by the pulse generator (2) and optional

pulse generator (3) in accordance with the invention. It can also optionally control the amplitude of the pulses using multipliers (4) and (5). Any number of pulse generators (2) and (3) may be used; if the pulses to be generated do not overlap, then only one per pulse shape is required. The pulses are combined by addition or switching in sum unit (6) before being passed to the transmitter (7). In its simplest form transmitter (7) may perform no additional function, but usually there will also be amplification of the signal. The transmitter output is passed to an antenna (8) or alternatively an electrical connection such as a wire. The signal then passes through a medium, to the receiver (s). The medium is typically air or ground or building materials or a combination, although UWB can also be carried by conductive or other transport media such as cables or optical fibres.

Operation of the timing controller and pulse generator (s) will now be described.

1. In embodiments of this invention, composite pulses are constructed by combining an initial pulse or pulses in a specific manner.

2. The shape of the initial pulse or pulses may be any required, however it is/they arc usually short in duration. The energy contained rn an initial pulse may be predominantly consistent from instance to instance.

This can be achieved by making the shape consistent

f7 from instance to instance. The pulse can be unipolar or bipolar and a bipolar pulse can be created by subtracting two time spaced unipolar pulses.

3. A composite pulse can be formed by adding (or subtracting) two initial pulses with a time difference "T"between them."T"can be zero, sometimes less than the pulse length or more usually more than the pulse length. Normally the pulses which are to be combined will be of roughly the same energy and shape.

4. A further composite pulse can be formed by adding (or subtracting) composite pulses with another time difference "T" between them. "T" can be zero, less than the composite pulse length or more than the composite pulse length. This process can be repeated using different composite pulses, different values of"T"and a different choice of whether to add or subtract the pulses. By choosing these values the characteristics of the composite pulse can be controlled. Again, the

composite pulses will be of roughly the same energy and shape.

5. Pulses or composite pulses of the same or different shape may be added or subtracted with arbitrary amplitude and with arbitrary values to create a desired pulse shape if no other special characteristics are required. For example, this can be used to adjust the peak to mean amplitude of a composite pulse.

6. Where an addition is performed, a time"T"between pulses or composite pulses of like energy creates spectral zeros at odd multiples of frequency"f" (i. e. f, 3f, 5f, 7f etc), such that f=1/2't'. (Time"T"is in seconds and frequency"f"is in Hertz).

7. Where a subtraction is performed between pulses or composite pulses of like energy, a time", u" between the pulses or composite pulses creates spectral zeros at even multiples of frequency"f" (i. e. Of, 2f, 4f, 6f, 8f etc), such that f=1/2#.

8. For any given frequency"F", there are therefore several values of"T"which provide a spectral zero at "F". For addition of like pulses these are T= (2k+l)/2F, k=0, 1, 2-... For subtraction of like pulses these are T=2k/2F, k=0, 1, 2...."k"is therefore an integer.

Negative values of"k"are equivalent to positive

values of"k", since it is arbitrary whether a time spacing is measured forwards or backwards in time."k" is therefore an integer.

9. If the duration of the pulses combined is less than "T", then the shape of the pulses need not be similar to achieve the spectral zeros.

10. By choosing at least some values of F which are close spectrally (for example 1. 572, 1.574, 1.576, 1. 578GHz), a region of spectral attenuation can be produced; several candidate values of"T"can be calculated using the above equations. The larger the

value of", c", the narrower the spectral zero in frequency percentage terms, so normally "e" is kept small by keeping k small. By varying"k" (and hence "T"), between composite pulses the spectral shape can be improved at frequencies other than those at which the spectral attenuation is required. Further, the performance of some UWB application (and in particular localisation) are affected if too small a value of "T" is chosen. There is therefore a compromise to whether "T"should be small or large, but it is nonetheless usually constrained to discrete values in this invention determined by"F"and"k".

11. If a region of spectral attenuation is placed at a frequency, then harmonics of it are also attenuated.

This is one method by which multiple areas of spectral attenuation to be created. For example, a region at 200MHz will result in attenuation regions at multiples of 200MHz, and in particular 1.2GHz, 1. 6GHz, 1.8GHz and 2GHz, these frequency regions being in use by other sensitive services.

12. Another method of providing attenuation in multiple bands (in this case they do not have to be harmonically related), is to place the spectral zeros"F"in the two bands separately using suitable values of"T". Normally one would create a composite pulse for the band which had the greater bandwidth requiring attenuation, and then use this composite pulse as the initial pulse for creating composite pulses to provide the attenuation in the next band. This can be repeated for as many bands as are required. If the values of"T"are chosen so that the pulses have an energy in their overlap which is sufficiently small that the desired spectral characteristics are still adequate if the pulses are not added in this region, then a single pulse generator is sufficient to generate the composite pulse.

13. The composite pulses can be further combined using the different values of "1"" and "k", but with similar values of"F"to yield a system. This system can hence carry data, or simply vary its parameters randomly or otherwise to improve the spectral shape away from the spectral attenuation region.

14. Information can also be carried using the other parameters such as"k", and whether addition or subtraction is performed.

15. Where the val ues of "1"" must be formed from a discrete set of time values, such as are produced by a clock, the nearest available value of "1"" can be used, or simply the available sequential values of"T"can be used to avoid wasteful duplication. The spectral zeros may vary from their optimum position, but a useful region of attenuation can usually still be obtained The same process as described above (Points 1-15) can be used at a receiver to reject frequencies by replacing the pulse generator with a gate or correlative process.

Figure 3 shows a receiver apparatus in which the signal is received by an antenna or other connection (10)

and passed to the receiver (11). In its simplest form (11) may perform no additional function, but usually the signal is also amplified, sometimes with adjustment to a consistent amplitude. The timing controller (12) according to the invention controls the timing and amplitude of signals gated into the combiner (17). The order of components multiplier (13) and gate (15) may be reversed if desired. Items multiplier (14) and gate (16) are optional. Gates (15) and (16) may be replaced by a correlation with the expected pulse shape. The gated pulses are passed to processor (18) for further processing as usual.

This receiver process provides the same spectral zeros according to the equations above, although for the case of a gate, the gate time would need to be short relative to the spacing "1"" if the process is not to be degraded. This process can be used to reject interference, whether the transmitter has used the same timing pattern"T"or not. This is an independent invention, since it can be applied separately as well as in combination to the transmission of pulses. The receiver can use the composite pulse of the transmitter as its initial pulse gate timing from which to construct

a further composite pulse gate timing which rejects the interference at the spectral zeros defined by the additional gate timings "#". The values of"T"and the number of composite pulse gate timings may be varied from time to time in order to track intermittent and dynamic interference.

The receiver may also multiply by the expected pulse shape (i. e. it may use correlation), rather than use a gate.

The receiver may, in preference or additionally, pass information to the transmitter to adjust the number and values of "1"" in order effectively to reduce the interference at the receiver.

Both in the transmitter and in the receiver, the adjustment of the gain of the pulses to be combined can be derived by measurement or calibration in order to achieve a better gain match and hence a better rejection of the unwanted spectral components.

The transmitter and receiver need not be restricted to a single antenna each, or to one form of polarisation.

Pulses or composite pulses can, instead of being combined at the transmitter, can be transmitted on separate channels (for example antennas of different polarisation), and be combined at the receiver.

Figure 4 shows how construction of two pulses into a composite pulse, followed by construction of a further composite pulse from the first composite pulse can result in a method to reduce the power transmitted in two bands occupied by other services. In this example the pulses are closely spaced and may overlap. Combining two pulses twice is equivalent to combining four pulses.

Both addition and subtraction have been used for Figure 4 as a method of combining pulses. This yields an efficient method of attenuating the transmission in two frequency bands. Here the time interval between pulses is short in order to achieve a wide bandwidth of attenuation. The attenuation relative to the peak is shown dashed and is 20dB; the attenuation value can be readily improved by adding further spectral zeros, for example as a sum of 8,16 or more pulses. Normally these spectral zeros are chosen to be spaced out in the region to be attenuated. To achieve better attenuation, the

frequency spacing of the spectral zeros is smaller near the edge of the region than at the centre of the region.

In Figure 5 shows a further composite pulse construction in which a much narrower band of attenuation has been achieved by using a larger interval between pulses. A marker (dashed) has been included to show 30dB attenuation (relative to the UWB mean) within this GPS band at 1. 57542 GHz. With the larger interval"T"between the pulses, there are also a greater number of values of "T"which can be used ; if these are used to construct longer sequences, then the spectrum becomes more flat and noise llke, which is an advantage.

Figure 6 shows a composite of 16 pulses of Fig. 5 in the form of part of the associated time waveform ; the pulses can be seen to be well spaced but at irregular intervals. Here composite pulses are combined four times, this being the equivalent of combining 16 of the initial pulses; this can be seen in Figure 6. In this case the pulses have been spaced so that there no two pulses are close together. This can be an advantage with certain pulse controller and generator apparatus which need time between pulses.

Further, if the attenuation region in Figure 5 is placed instead at 200MHz or 400MHz, then attenuation regions will also occur at harmonic multiples of these frequencies; this can be used to protect the GPS bands at 1.2GHz and 1.6GHz. An example of a harmonically repeating region of spectral attenuation is shown in Fiure 7. In this embodiment of the invention is that by combining more than one pulse at different spacings, that the region of attenuation can be better controlled. In this case a region of attenuation over a greater frequency bandwidth has been achieved.

If the 4 pulses from Figure 4 are used as the initial pulse for Figure 5, then all regions of attenuation from both figures are retained. These attenuation regions need not be harmonically related. The pulse sequence resulting contains 64 pulses of which some overlap. By using the alternative values of"k", multiple instances of different pulse sequences can be created in order to improve the spectral flatness and reduce the spacing of the spectral lines which make up the spectrum.

The time spacing of each block of 64 pulses (whether each composite of 64 pulses is similar or different) can be

spaced differently in time in order to achieve an improvement to the spectral characteristics, either in flatness or spectral line spacing.

It is intended that variations and modifications such as would be readily apparent to the skilled person, may be made to the embodiments described herein without departing from the scope of the present invention disclosed herein.

Claims

CLAIMS 1. A method of creating a composite signal including the step of combining two pulses wherein the pulses are separated by a time"T", with"T being selected such that one or more spectral components of one or both of the pulses is attenuated in the resulting composite signal.
2. A method according to Claim 1 wherein the composite signal is an ultra-wideband (UWB) signal.
3. A method according to Claim 1 or Claim 2 wherein the pulses are substantially the same energy and shape as each other.
4. A method according to any of the above claims wherein a composite signal is in turn combined with further pulse (s) in order to produce one or more further composite signals.
5. A method according to any of the above claims wherein the step of combining is by addition alone, subtraction alone or a combination of addition and subtraction.

<Desc/Clms Page number 24>
6. A method according to any of the above claims wherein one or more of the pulses is amplified or attenuated prior to combination.
7. A method according to Claim 4 wherein the values of "#" for each step are each a multiple of a selected time value.
8. A method according to Claim 1 wherein one of the pulses is itself a composite signal including interference to be attenuated and the other pulse or composite pulse is selected in order to attenuate the composite signal in at least one of the area (s) of interference.
9. An apparatus for creating a composite signal, the apparatus including means for combining two pulses, wherein the pulses are separated by a time", u" which is selected in order to provide a region of spectral attenuation in the composite signal.
10. An apparatus for receiving a composite signal, including means for combining the signal with one or more pulses or composite pulses wherein the time spacing ""

<Desc/Clms Page number 25>

between the pulses or composite pulse is selected to provide one or more regions of spectral attenuation in the received signal.
11. An apparatus according to claim 10 in which the pulses or composite pulses are chosen to correspond to either the composite signal, or a further composite signal created by combining the pulse or composite pulses to provide one or more regions of spectral attenuation in the received signal.
12. An apparatus according to claims 9,10 or 11 including means for deriving a gating signal from the pulses, which is then used in place of the pulses or composite pulse in the means for combining.