GB2212352A - Deinterleaving of radiated signals - Google Patents

Description

2212%,s,52 0 AIR 85/J11 DEINTERT,EAVING OF RADIATED SIGNALS This invention

relates to deinti-_rleaviag of pulsed cmissions of radiation from a plurality of sources into a detection region and in particular to performing the deinterleaving in real time, with just a short processing delay.

In warfare, for example, it is known to perform so-called electronic reconnaissance by gathering infor-mation from signals radiated by communications or weapons systems, the signals being electromagnetic or possibly acoustic in form. it is expedient for descriptive purposes to consider only electromagnetic signals although the principles involved are applicable to all forms of signal radiation.

Each electromagnetic radiation source has an electromagnetic signature defined by parameters such as output frequency, that is, radiation wavelength in the electromagnetic spectrum, and in the case of radar type apparatus scan period and scan pattern and may emit pulses of radiation having a distinctive pulse repetition interval and/or pulse width.

Measurement of some or all of these parameters enables the emitter to be classified and with other measured parameters may be used to identify the particular emitter, possibly by comparing the determined parameters with those of known emitters held in a library of emitters carried by the apparatus.

Such classification procedure is however complicated by the fact that although such emitters generally operate over definite limited ranges of the electromagnetic spectrum the signals emitted by any are generally mixed, or interleaved, with those of other emitters and the process of separating from a received signal the separate emissions and relating them to their emitters is known as deinterleaving.

D 33 As indicated above it is in warfare where the extremes of classification complexity are reached in view of the large number of emitters to be found in and supporting the modern battlefield, including in particular guided weapons which rely upon detection of a target in a scanned region followed by traclking and targeE illumination.

Such weapons may be Purrierous 1 All\ 85/Jll and the electronic environment rpsulting from these and (-11'iici sources may be particularly ennfiispd for analv.-;',s.

At present and as foreseen for the immediale fiiiure, significant proportion of all such emitters have pulsed operation and furthermore, usually operate asynchonously to eac other with a very low pulse duty cycle.

However, such battlefield emitters defined limits of frequency, scan and other apart from installation in fast moving vehicles, may be considered as fixed in position.

operate within parameters, and An example of a typical pulsed-emission signal received by a suitable wide-band detector is shown in Figure 1 and comprises a train of pulse signals received over a period of time t to t 0 m The time between, amplitude of, and radiation frequency of, successive pulses is apparently random and it will be appreciated that investigation of the chronology of receipt will not determine easily even the number of interleaved signals detected.

Figure 2 shows that the received signal of Figure 1 in fact comprises asynchronous pulse trains (a) to (d), some or all of which may have the same radiation frequency, extending for a time period between t 0 and t m and it will be seen that in any particular situation the pulse trains may have irregular pulse repetition intervals and/or amplitudes and extend for different periods of time and may overlap each other in time.

In many operational situations, and particularly in a battlefield situation, it is desirable, and possibly essential for the reconnaissance including such signal deinterleaving to be performed on a continuing basis, in real time.

Signals have hitherto been deinterleaved in real time by two basic methods. A first operates on the basis of emitted frequency and emitter direction and the use of high resolution detectors, such as narrow-band swept superheterodyne receivers or multi-chanelled receivers for frequency measurement and/or complex antenna systems which often use phase measurements to provide angular accuracy of emitter direction.

Implementation = suclo a procedure requires complex aDDaratus which ls costly 01. 1 1 AIR 85/J11 to mp!ikifaetlii-.- and maintain in operation and subject to weight And POWP r consumption penalties. sophisticated algoritlii,.is based, for example, A second method uses on recording the time of arrival ot individual pulses for a time interval of several seconds and processing the potentially very large number of pulses received in this time in order either to generate a histogram of the pulses from which are determined arrival peaks or patterns or to compare the received pulse train against known emitter parameters in an attempt to recognise and isolate an individual pulse train. Such an algorithm, operating principally on time of arrival of pulse signals, requires storage and large computing capacity which is exacerbated by having to deal with shortterm variations, for example jitter, in signal radiation frequencies, pulse repetition interval and direction of source.

Furthermore, the recording and processing of signals is such that the radiation environment is only deinterleaved on the basis of an occasional sample and is not monitored in real time for reception of all possible emissions.

Electronic reconnaissance, for which deinterleaving is essential for complete indentification of sources within a region, is also practised in a simplified form by the use of so-called radar warning receivers or by laser warning receivers which are usually carried by vehicles and warn the operator of microwave or optical radiation incident thereon, enabling the vehicle to take evasive action of countermeasures.

Such receivers do little to identify the parameters of the source and whilst the warnings they give may be exceedingly useful in an electronically 'quiet' environment, in the above described battlefield conditions they are likely to be swamped with received radiation and serve no useful function.

It is an object of the present invention to provide a method of, and apparatus for, deinterleaving radiated signals in real time which is simpler and more readily implemented than known methods and apparatus.

According to a first aspect of the present invention a radiated pulse Sg-, ,a,s re _ c - - '.7; ij ', ---. - - W AIR 85/JH receiving apparatus comprisos defining at least one time frame for reception of raciiittd signals, defining. a set of ranges of possible angle of arrival of a received signal with respect to a predetermined reference direction related to the receiver, defining a set of ranges of possible radiation frequency of a received signal and relating values of said sets to define by each pair of set values the coordinate values of at least a notional matrix of storage locations, determining, for each signal pulse received in the time frame, in which range value of each set its parameters lie, storing a record of the reception in the matrix storage location corresponding to its associated set values, analysing at the end of the 'time frame the storage density of records in the matrix and identifying therefrom the existance of separate radiating sources distinguished by the range of frequencies and angles of arrival of signals.

According to a second aspect of the present invention apparatus for deinterleaving radiated pulse signals comprising receiver means operable to receive pulses in a set of frequency ranges and in a set of angle of arrival ranges and to determine for each pulse received at least in which range of values of frequency and angle of arrival its parameters lie, and processing means including storage means operable to store a record of each pulse received in a time interval in accordance with the set values of the pulse frequency and angle of arrival, and record analysis means operable after the time interval to determine numbers of stored records associated with different values of the sets and to identify set values for which record storage density peaks occur for both set values and relate these each to a radiation pulse emitter distinguished by the range of frequencies and angles of arrival of the set values.

An embodiment of the present invention will now be described by way of example with reference to the accompanying drawings-, in which:- Figure 1 is a representation of a received electromagnetic signal comprising a train of pulses represented by their amplitude and time of arrival within a time period t o aau from pulse train sj'LgLiLLb, 4 AIR 851/111, Figures 2(a) to (d are representations of four individual pulse tra-in signals comprising the signni of 1, Figure 3 is a block diagram of pulse signal deinterleaving apparatus according to the present invention, Figure 4 is a schematic plan view of the antenna array of the apparatus of Figure 3, Figure 5 is an illustration of a relationship between a parameter, such as frequency or angle of arrival of received signals and time of arrival divided into contiguous time ranges or frames for the purpose of illustrating time frame analysis, Figure 6 illustrates a frequency band broken into ranges of frequency and forming a set of such ranges, the extent of the ranges not all being the same.

Figure 7(a) is an isometric view of a notional three dimensional storage matrix defined for axes representing received pulse signal radiation frequency, angle of arrival and number of pulses with specified values of frequency and angle of arrival (pulse density), Figure 7(b) is a plan view of the storage matrix of Figure 7(a) in which the pulse density is represented by the number of dots in each location, and Figure 8 is a flow chart illustrating the procedural steps of record storage and processing by the apparatus of Fi-ure 3.

0 The embodiment described relates specifically to pulse signals the radiation frequency of which is in the microwave radar portion of the electromagnetic spectrum although it will be appreciated that operation is not limited to this form of radiation and this band of frequencies.

Referring to Figure 3 pulse signal deinterleaving apparatus comprise receiver means 11 including radiation sensing, or antenna, means 12 and a receiver 13.

The receiver 13 comprises an instantaneous frequency measurement receiver of known form, frequently referred to as an IFM receiver which is able to establish the frequency of radiation of any pulse received over a wide range of freauencies. lhe IFM receiver may operate on de-lay 14---- --nd AIR 85/J1i phase comparison techniques bu t preferably operates on a n-Cal led co-tizuOu's f; -t - - plp of quch;q receiver is the RWR5618 produced by Filtronic Components Limited.

The antenna means 12 comprises an array of antenna elements, with means to determine from the signals picked up thereby and the array configuration the angle of arrival of a pulse received.

The antenna array, shown in greater detail in plan view in Figure 4, comprises a set of four, but possibly more, directional antenna elements 14 1 '- 14 4 having their reception direction axes 15 1 ' 15 4 directed substantially radially from a reference axis 16. The sensor reception direction axis of each sensor, or a projection of it, lies in a sensor plane orthogonal to the reference axis and the angles in said sensor plane between said direction axes or their projections and a reference direction in the plane define angles of arrival of radiation sensed thereby.

The sensors, in this case the antenna elements, may for convenience, but not of necessity lie with their reception direction axes in the sensor plane and be symmetrically disposed so that the reception direction axes are mutually divergent.

Furthermore, it will be appreciated that the reference axis may also for convenience but not of necessity, be defined vertically such that the sensor plane lies in the azimuth plane and the angles of arrival of radiation correspond to azimuth bearings, and the main part of the following description assumes such disposition and orientation. The respective reception fields of the antenna elements are conveniently indicated in terms of the polar diagrams of reception sensitivity 17 1 17 4 superimposed upon the array plan and the reception fields of adjacent elements overlap in the azimuth sensor plane. Each antenna element feeds a separate receiver and it will be appreciated that for a signal received by two antenna elements simultaneously the relationship between the two received signal amplitudes is indicative of the direction of the source with respect to the axes of the two elements. Thus the angle of arrival may be determ ined to within a range cf azimuth angles - 7 1 AIR 85/J11 with respect to the array, the extent of the range, that i S accuracy of dettermination, being -- function of the!i,llllll,-r of antenna elements in the array f roin which comparative measurements may be made and the complexity of signal processing to obtain the directional information as well as the signal strength with respect to background and system noise.

The above outlined and other means by which a received pulse signal frequency and its angle of arrival are determined are well known and require no further description.

The receiver means produces a record of the reception of each pulse and is able to tag that record with such parameters as the frequency of the radiation, the angle of arrival, the time of arrival and amplitude of the pulse signal.

The remainder of the apparatus will be described in due course but it is convenient at this stage to outline the process by which the received pulse signals, or rather the records thereof formed by the receiver means, are deinterleaved.

As outlined above the interleaved pulse train signal shown in Figure 1 comprises a succession of individual pulses of electromagnetic radiation of varying frequency, amplitude and width, although in relation to the time scale of the Figure each pulse is sufficiently short to be perceived only as a line.

The individual pulse trains contributing to the received signal are illustrated for the same time interval t 0 to t m in Figures 2 (a) to (d) respectively, the pulse train signals also being represented as to their times of occurrance and duration by the lines A to D respectively on the time axis.

Figure 5 shows the occurrance of individual pulse trains A to D and further pulse trains E to H characterised by one parameter, such as their pulse frequencies or the pulse angle of arrival (AOA) plotted against time. It will be seen from the Figure that signals A, B and C overlap in time of reception although are separated in either frequency (or AOA), whereas signals B, D and C are received at separate times but from substantially the same AOA (or have substantially the same radiation frequency). Some of the signals, for instance signal L, are repeated, the repetition heina du e perhaps to r, AIR 85/JH inter-mittent transmission or the effects of spatially.,;canning a dircctional emitter, such as in radar -ppqrqt-11!;- The period of repetition may provide a further parameter for subsequent source analysis.

In accordance with the present invention operation is broken into storage and analysis of records for one, or preferably a succession of, contigious equal time frames t 0 to t n 3 t n to t 2n'... t (k-l)n to t (kn) the first of these intervals conveniently being known as t 0, the second as t n the third as t and so on. The parameters of the pulse 2n signals which are received in each time frame are analysed in the following time frame as illustrated schematically by the Figure.

0 It will be seen that providing the analysis takes no longer than one time frame period it can be conducted continuously as signals are received and, apart from a small processing delay of up to one time frame period, is effectively conducted in real time. The duration of each time frame may be less than one second reducing the number of received pulses for which data is recorded and the quantity of data subsequently to be processed.

For convenience the analysis is referred to in terms of the time frame of signal reception.

As indicated above for each pulse received in the time frame a record of the arrival is produced and stored for subsequent analysis.

To effect said storage and analysis the method of the present invention a set of ranges of possible pulse radiation frequencies and a set of ranges of possible azimuth angles of arrival are defined and each pulse received is stored in a form indicative of in which range its parameters lie, the presence of emitting sources having parameters of radiation frequency and angle of arrival within any appropriate range being indicated by reception of larger numbers of pulses associated with these ranges.

This defining of a set of ranges is illustrated graphically in Figure 6 wherein reception is accorded to all 1 ATR 85/JH pulse radiation frequencies witliii.i a baild extending continuously between upper and 1-ower limits of and' hand sub-divided into frequency ranges delimited by the boundary marks shown. The frequency ranges may all be of equal extent or as shown, may differ in extent, the ranges at more commonly used regions of the bands being smaller than in sparsely used regions. It will also be appreciated that the total frequency band may be discontinuous, with intervening regions ignored.

The angle of arrival (AOA) may similarly be divided into a set of AOA ranges, although the angle ranges are conveniently of equal extent.

Each set of ranges may be considered in terms of set values, each of which is indicative of a particular range of frequencies or angles of arrival, records of pulses received being identified by tagging with the set values in accordance with the appropriate ranges in which its parameters lie, or a real or notional matrix of storage locations may be defined by said pairs of set values, each location being equivalent to a repository of a record having the appropriate set values. 20 In this specification reference to a record by means of its storage matrix location is to be considered to be the same as to reference to it by means of its set values, and vice versa. In considering deinterleaving in accordance with the above it is convenient to consider a storage matrix for records of the received pulses as illustrated in Figures 7(a) and 7(b), one such storage matrix being defined for each time frame.

Referring to Figure 7(a) the storage matrix is created as a three dimensional storage matrix. A two dimensional or planar matrix of storage locations is formed with the coordinate values of each location being defined by the set values of the frequency ranges and AOA ranges. As stated above each received pulse results in a record tagged with the frequency and AOA and the tagged records are assigned to the appropriate matrix storage location in accordance with in which range their frequencies and AOA's are detected. The third axis of the matrix represents by the number of pulses stored in each location, the pulse density of the location.

AIR 851JH An alternative representation of the matrix is shown in Figure 7(b) which i -- -- -- s C' 11 t - i.111 Y P P1 ill v] PW o f th'. frequency/AOA plane showing the matrix location boundaries. In this representation, each of the tagged pulse records are represented by a dot so that the number of records in each location defining the third matrix axis, representing the pulse density, is directly visible as such a density.

It will be seen that the pulses tend to occur in groups around particular frequencies and AOA's which may reasonably be expected to correspond to individual sources although pulses resulting from noise may occur anywhere and several close groups may not be readily distinguishable.

At the end of the time frame a fresh storage matrix is created for pulses received in the next time frame and the recently completed matrix analysed.

The storage density of records in the matrix is determined by examining the number of records in each storage location and deriving therefrom any storage density peaks, with respect to neighbouring location, in the coordinate axes defined by the set values, that is, deriving two dimensional density peaks in the matrix plane.

Each such peak is assumed to be associated with an individual radiation source distinguished by, at least, the ranges of frequency and A0A. of the set values defining the matrix location of the density peak.

Any two dimensional density peaks which occur may be identified by a number of examination methods known per se but conveniently they may be determined by examining the matrix storage locations for each of successive set values of one coordinate of the matrix, say the AOA. For each set value of A0A. the matrix location associated with different set values of frequency range are examined.

It is convenient to eliminate small storage densities due to the reception of random noise pulses by defining a minimum threshold storage density for each matrix location. if none of the locations associated with the set values exceed che density threshold then the next A of frequency OA is 1 11 - AIR 85/J11 considered. If a density 'level in excess of the threshold is Lound, the u'ensi,,-,rs of each of the locations defined hy frequency set values are deter-mined in sequence and compared with the preceding one in order to determine whether the density is increasing towards a peak or is lower, and indicative that a peak may have been passed. Peaks determined in this way, and all associated with one value of AOA may be considered as one-dimensional density peaks.

When this procedure has been repeated for each set value of AOA the matrix locations of the identified one dimensional density peaks and the surrounding locations are further examined to identify locations o"f peaks in two dimension of the matrix, that is, with respect to both set values.

Referring to Figure 7(b) if one group of records are at locations defined axis and values by values A05A1)A 2' A 3 along the A0A.

F15F 2' F 3 along the frequency axis the locations may be identified as A 1 F13A 2 F 2" A 3 F 3 For each set value of AOA examined in turn one dimensional peaks may be found for F 2' that is at A 1 F 2' 20 A 2 F 2 and A 3 F 2 For each of the locations defined by these set values the record density locations, that is, is compared with that of surrounding for A 1 F 2 density comparision is made with locations A 0 F,) A 1 F1j A 2 F,) A 2 F 2' A 2 F 3 A 1 F 3' A 0 F 3 and A 0 F 2 If the record density is greater than in all of these, the location appears to be the location of a two dimensional peak and is recorded as a candidate density peak, subject to any further conditions which have to be satisfied. If, on the other hand, a comparison shows a higher record density in a neighbouring location, such as A 2 F 21 then the one dimensional peak A 1 F 2 is excluded from further consideration as a candidate peak.

The procedure is then repeated for peak A 2 F 2 and A 3 F 2 and which may result in A 2 F 2 being seem to be the matrix locations of a two dimensional density peak.

As the stored records in locations surrounding the one assumed to represent, by the set values defining t, the AIR 85/JH frequency range and AOA range of the source, those set values as tagged t C, t h. r! rccords t c place them i n those s ti r roi i n d i n F! locations inay be assigned to the location in which tile peak is defined, effectively moving these surrounding records into the peak location.

As indicated hereinbefore one of the difficulties encountered in known methods of deinterleaving is that caused by variations in AOA due to multipath reception or operating with a poor signal to noise ratio and/or variations in pulse radiation frequency due to source emitted frequency jitter, either unintentional or intentional, when 1requency agility is practised, or caused by noise in the receiver. It will be seen that processing the received pulse signal records in accordance with defined ranges as represented by the matrix location enables the association of received pulse records with particular sources on the basis of a statistical operation rather than on the basis of measured parameters of any individual received pulse.

It will be appreciated however that the effects of such possible variation in AOA and/or frequency do cause dispersion of the records of any particular group defining a source.

With reference to Figure 76) it may be expected that where there is no change of angle between the apparatus and an emitting source which also emits at a fixed radiation frequency the record grouping will be of the form shown at location A 2 F 2 described above. However if the source and/or apparatus moves during the time frame the record density may be distributed along the AOA axis. Similarly if the frequency changes, perhaps due to frequency agility, the record density may be distributed along the frequency axis. If a change in one affects the other, or both change independently, the distribution may extend along both axes. Such a distribution may not be even with a clearly defined peak and to remove the possibility of determining an excessive number of sources close together in the matrix, when the probability is of a single source, threshold conditions may be introduced which have to be sarisfied for a two dimensional density peak, which may -upon its t, 1 1 - 13 AP IR 85/J11 determination be considered only aS a 'candidate', before such candid2te pe a lt 5 qrf. rn ns- i d f, r P d n -- d i s t inguish inp-, a discrete source.

For example each candidate two dimensional peak may be required to be spaced from a neighbouring candidate peak by a predetermined number of matrix locations, (set value increments), for example two. Alternatively, or in addition, a pair of candidate peaks may only be considered discrete if one of the matrix locations separating them has a record density below a predetermined fraction, say one half, of at least one of the peak densities. The fraction may -be compared with the average density of the peaks.

If a plurality of candidate two dimensional density peaks are determined which fail the additional tests and fall to be considered as representing the same source, then a representative location, or set values for sourcefrequency range and AOA range may be defined between them, although this may be unnecessary for some forms of record utilisation, as will be considered hereinafter.

The deinterleaving procedure as described results in the determination of the number of discrete sources, classified by set values or ranges of radiation frequency and AOA, from which pulse emissions have been detected in the time frame and furthermore the records of the pulses attributed to each such source are available for further analysis as required in addition to the deinterleaving.

The method of deinterleaving by the process described above may be followed readily in practice by suitably storing, retreiving and comparing numbers of records in a number of ways, one of which will be described with further reference to Figure 3. In addition to receiver means 11 the apparatus comprises processing means shown generally at 20, this means comprising essentially a digital computer consisting of a CPU 21, ROM 22, or its equivalent, for storing the program and any constant values by which the computer performs the procedural steps deinterleaving operations, RAM 23 which provides working store for the CPU and also storaúe area which cor.ip-r].,;e a notLonal AIR 85/J11 record storage matrix, input means 24 by way of which records produced by the receivor mpans are applied to store storage area for recordal and output means 25 by which records and/or data appertaining to deinterleaved signal sources are output for further processing.

The processing means also includes timing means shown at 26 which defines the duration of the time frame for which pulse signals are received and recorded and also provides timing signals for operation of the computer.

As indicated hereinbefore the records produced by the receiver means aretagged with the radiati7on frequency and angle of arrival.

Throughout the time frame (or a first time frame t 0 to t 1) the computer operates as record storage means and defines the set values of frequency and AOA ranges which in turn define locations of the notional storage matrix and as each record is received it is further tagged with the set values associated with the range of frequencies in which its frequency tag lies and the range of AOA's in which its AOA tag lies. The record with these and original tags is stored in the RM.

It will be appreciated that each such record may be stored at any address in the RAM and that the storage matrix, although a useful functional concept need not exist as such.

The records may be stored in a readily addressed block in the RAM for ease and speed of access but may also be conveniently stored in successive address as they are received.

At the end of the time frame, as indicated by timing means 26, the computer functions as record analysis means essentially following the procedure described hereinbefore.

For each value of AOA the computer searches through the addresses of the RAM for all records tagged with that AOA set value and further examines them on the basis of the frequency set value which they were also tagged upon storage. Thus the 0 number of records tagged with the appropriate set values are counted which performs the function of counting the number of records, that is, determining the record density, in the storage waLrix locations defined by the st-t vallies. 7he records in AIR 85/J1i each location of the storage matrix are thus identified by, and loc.qted in, the RAM adresses in which the set values (--)f tlic, location are found.

Thus by examining the RAM for the set value tags the record density, as a function of these set values, is examined as for the storage matrix locations considered above and the set values corresponding to discrete two dimensional density peaks determined. These set values of course indicate the identity of a discrete source, or if additonal conditions have to be satisfied, candidate peaks.

As outlined above the set values representative of matrix locations surrounding one in wliich a peak density is determined may be amended to agree with those associated with a peak density such that in further processing substantially all the records attributed to a particular source can be accessed by means of the tagged set values.

Clearly the procedure of considering each two dimensional density peak only as a candidate until any conditions regarding separation in terms of set values and fall in density levels between peaks are satisfied may be readily implemented by counting the appropriate set values and record densities associated with them.

As described the apparatus represents a basic deinterleaving arrangement in which the existance of a discrete 25 source is indicated by its frequency and AOA set values. Clearly, once such a source is defined all the records associated with it may be read from the store in order to derive additional information from the data with which the record is tagged, such as a more accurate measure of frequency and/or AOA, the time of reception and the amplitude of the received pulse.

Figure 8 takes the form of a flow chart showing the procedure of record storage and processing by the computer in conformity with the above described procedure. The procedural steps are clearly identified within the flow chart which requires no further explanation.

The deinterleaving and availability of data thus far described is fur LhLrecv-pLiun of pulses and storage or tneir records in one time frame.

continuntisly defining AIR 85/Jll Preferably the apparatus functions a succession of contiguous time frames such that the computer operates in its dual role oLE record storing means and record processing means simultaneously, the records being stored for the current time frame but processed for the preceding one. This requires the definition of at least two notional storage matrices and the storage space available in the RAM must be large enough to store all the records for at least the two successive time frames and, depending upon the processing data after deinterleaving, may be required to store at least some of the -records for a larger number of time frames.

In order to provide satisfactory information about a source from which pulses have been detected it may be necessary, in an operation subsequent to deinterleaving, to examine the pattern of pulse emission. For example, each source may be expected to emit pulses in accordance with a pulse repetition interval which may be characteristic of the source type or its mode of functioning. Also, if pulses are emitted by a surveillance radar which is continuously scanning the region in which the apparatus is located then the received pulses will be expected to occur in groups, the spacing of which is both indicative of scanning per se and the rate of scan.

Alternatively if pulses are emitted by a tracking radar then reception of such pulses would be expected to be continuous.

Thus it may be desirable to determine whether the same source is indicated by detection of set values for two or more time frames to determine further information as to the nature of the source.

One exemplary use of deinterleaving apparatus as described is in a vehicle, such as a helicopter, for which the detection of such pulse emitting sources provides the means for taking appropriate action rather than enabling a full analysis of the signal environment to be carried out. Such appparatus may be considered as an enhanced form of the known radar warning receiver considered hereinbefore but which is effective in the AIR 85/JH electronically swamped environment and is small, fast and cheap enough to be used as a warning instrumpnr.

The apparatus as shown in Figure 3 further includes a PROM 27 of suitable form containing a library of data relating to parameters of known pulse emitters within the operating band of the apparatus, in particular such parameters as pulse repetition interval (PRI), scanning interval, radiation frequency and whether the emitter is associated, for example, with a type of weapon effective against the vehicle.

Thus as a result of deinterleaving, for each pulse radiation emitting source for which pulses are received in successive time frames the frequency set value is compared with the stored library sources to determine whether the source is a known type of emitter and if so, the times of reception of the pulses, as tagged to the records associated with that set value, are compared as to PRI with the possible library source or sources in order to determine the nature of the source. For instance, if the source is associated with a formidable weapon and/or is badly placed with respect ot the vehcile, as determined by the AOA, the vehicle may take avoiding action whereas if the vehicle is in a superior position it may seek out and attack the source.

In such an apparatus it will be appreciated that if a source is indicated as threatening at a particular AOA or frequency then it may be immaterial that there are two sources so close as to be deinterleaved by the simple and coarse criteria of the frequency and angle ranges defining the matrix locations as a single source. Similarly, if two neighbouring density peaks record of storage indicate a single source spread in frequency or angle the precise values attributable to these may not be important, particularly the frequency range, provided the probable emitter can be recognised from the library data.

From the above example it will be seen that the apparatus may be used on the move or in a movable vehicle and it will be appreciated that this rotational motion in correlating uiiit, rciiiL iiiae frames.

may present problems with the signals received in AIR 85/J11 The AOA -1s determind with respectrefercnee direction in the sensor plane def-,ned and fixnd in to Che anienna means, and to accommodate changes in apparatus orientation in the azimuth plane a heading indicator 28 is provided which, possibly with the assistance of an analog-to-digital converter 29, provides a digital representation of the orientation of the apparatus that is the reference direction with respect to an external reference direction for example, a terrestrial or inertial reference direction such as North. The AOA as determined by the reception means is then referred to the external reference direction by addition or subtraction of the apparatus heading as the pulse record is tagged. The storage matrix is then defined with the AOA relative to the external reference direction such that when the signals are deinterleaved the sources are distinguished by the set values as a function of frequency and heading with respect to the external reference direction.

Similarly the sensor plane may be inclined with respect to the azimuth plane and the directional measurements obtained further if referred by projection and the azimuth plane for relating to external that is, environmental references.

it will be appreciated that even where the deinterleaving apparatus is fixed in position deinterleaving may be effected with the A0A. referred to an external reference direction rather than an arbitrary reference direction in the apparatus and the heading indicator may be omitted after the relationship between the apparatus reference and external reference directions is determined.

it helicopter, fitted with will also be appreciated that a vehicle, such as a for which such apparatus is useful may already be a suitable heading and/or roll indicator from which signals may be taken.

In such apparatus in which only the continuance of pulse reception from a source for more than one small number of time frames is of interest then after the records have been stored for more than the appropriate number of time frames they may be erased freeing the me-niory to be used for StOriLlg records i of a later time frame. As indicated above the number of time f r amic r for Wilich records require stnrage in depends upon the nature of AIR 85/JH complete form processing which takes place subsequent to the deinterleaving and is beyond the consideration of the present invention.

The deinterleaving apparatus described herein with reference to Figure 3 is concerned with the reception of pulse signals at the microwave frequency portion of the electromagnetic spectrum.

The apparatus is operable at any other parts of the spectrum for which radio frequency emissio'ns are detectable by a suitable antenna system. Clearly, at this or at lower frequencies the problem of determining ADA to any desired degree of resolution may require more elaborate antenna systems but which are known per se and require no further description here.

The apparatus may also be adapted by suitable forms of receiver means to operate in other parts of the electromagnetic spectrum, for example, at optical wavelengths, or with different forms of radiation, for example, acoustic energy.

On the basis that most emitting sources of interest optical wavelengths (which includes the far infrared) are based upon laser emitters and operate in a few well defined ranges of wavelength/frequency one or more optical receivers and appropriate filters may be employed as the receiving means.

Similar consideration may be given to acoustic energy and the frequency ranges in which it may be expected by the provision of suitable microphone sensing means and filters.

Once reception is effected all records are treated identically and distinguished only on the basis of set values with which tagged.

Clearly the ranges of frequencies associated with all types of radiation are discontinuous within the overall possible frequency spectrum but as indicated above with reference to Figure 5 the extent of the range of frequencies each set value is associated with is immaterial provided it offers a range which effects a balance between the number of sources which may AIR 85/Jll be expected to fall within it and the total number of set values for which storage h-q- to bp providpd and analysis Pffected.

The me t hod outlined above and the implementing apparatus of Figure 3 may be subject to modifications in respect of features thereof without departing from the scope of the invention.

For instance, the record storage means may define a three - dimensional storage matrix corresponding to the notional one described with reference to Figure 7 in which storage areas are reserved corresponding to the matrix locations and in which the records are stored. This enables tle storage densities to be determined more readily and also possibly other functions which involve the variation of parameters of records in surrounding locations.

The receiver means may also take a form in which pulses are received in channels corresponding to the ranges of frequency and/or angle of arrival defined for the set whereby there is a direct correspondence between each received pulse record and the set values with which it is tagged for storage or the real matrix location in which it is stored.

Similarly where the apparatus is moveable, instead of determining angle of arrival with respect to the apparatus and transposing this to an external reference direction, the sensing means, for exapmle, the antenna array may be steered physically or electronically to maintain its relationship with the external reference direction.

- 21 AIR 85/JH

Claims (7)

Claims

1. - p11 1 sp 9 i ú?ng 1 S A method c, f deinterleavinp rad i at Pd received by receiving apparatus comprising defining at least one time frame for reception of radiated signals, defining a set of ranges of possible angle of arrival of a received signal with respect to a predetermined reference direction related to the receiver, defining a set of ranges of possible radiation frequency of a received signal and relating values of said sets to define by each pair of set values the coordinate values of at least a notional matrix of storage locations, determining, for each signal pulse received in the time 'frame, in which range value of each set its parameters lie, storing a record of the reception in the matrix storage location corresponding to its associated set values, analysing at the end of the time frame the storage density of records in the matrix and identifying therefrom the existance of separate radiating sources distinguished by the range of frequencies and angles of arrival of signals.

2. A method as claimed in claim 1 in which the storage density of records in the matrix is analysed by examining the number of records in the set values defining each storage location and deriving therefrom any record density peaks with respect to neighbouring set values, being two dimensional peaks with respect to the matrix, and associating each two dimensional peak with a separate radiation emitting source distinguished by the range of frequencies and angles of arrival associated with the storage location in which it occurs.

3. A method as claimed in claim 2 in which each two dimensional peak is determined by examining the storage locations of the matrix defined by the set values for each of successive values of one set and, if any of the locations associated therewith have a record storage density in excess of a predetermined minimum for each value of the other set with said value of said one set, comparing the storage density of the location with that of the next to determine any one-dimensional peak density values and thereafter examining the locations of said onp dimensional density peaks with respect tc surrounding 1 r r, AIR 85/JH locations to determine storage locations having storage density peaks in both coordinates of the matrix.

4. A method as claimed in claim 2 or claim 3 comprising determining each two dimensional peak as a candidate peak and setting threshold conditions between storage locations of neighbouring candidate peaks, the failure of which candidate peaks to satisfy the conditions being indicative of the candidate peaks relating to a single pulse radiation emitting source.

5. A method as claimed in claim 4 in which the threshold conditions include ascertaining that the number of records associated with a location forming a candidate peak is greater than the number associated with each adjacent location and is separated within the matrix from any other candidate peak by at least two locations.

6. A method as claimed in claim 4 or claim 5 in which the threshold conditions include ascertaining that the number of records in at least one matrix location falling between the locations of two candidate peaks is less than a predetermined fraction of the number of records of at least one of the peaks.

A method as claimed in any one of claims 1 to 5 including tagging each record of a received pulse with parameters relating to the pulse and/or its arrival.

8. A method as claimed in claim 7 in which said parameters include at least the time of arrival of the pulse.

A method as claimed in any one of the preceding claims including tagging the record with the radiation frequency and angle of arrival of the receive pulse.

10. A method as claimed in any one of the preceeding claims including storing records by tagging each record with the set values representing the storage matrix locations in which its parameters lie and defining a notional matrix the storage locations of which are identified by the tagged set values.

11. A method as claimed in Claim 10 in which the set value tags replace the tags of radiation frequency and angle of arrival.

S AIR 85/J1i 12. A method as claimed in any one of the preceding claims including defining different ranges of possible frequencies tor different values of the set.

13. A method as claimed in any one of the preceding claims including defining different ranges of possible angle of arrival for different values of the set.

14. A method as claimed in claim 12 or claim 13 in which the ranges of possible values covered by each set are not all contiguous.

15. A method as claimed in any one of claims 1 to 13 comprising defining the set of ranges of angle of arrival to give a total angle coverage of 360 with respect to the predetermined reference direction.

16.

1 A method as claimed in any one of the preceding claims for receiving apparatus movable in operation including defining the set of possible angles of reception of the storage matrix with respect to a predetermined reference direction, determining the azimuth angle of said reference direction in the receiving apparatus with respect to an external reference direction and referring the angle of reception of each pulse, determined with respect to the receiving apparatus, to the external reference direction for defining the storage location of the pulse record.

17. A method as claimed in any one of the preceding claims including deinterleaving for at least two successive contiguous time frames by identifying the existance of separate radiation sources within each time frame from records stored in the preceeding time frame.

18. A method as claimed in any one of the preceding claims in which the, or each, time frame extends for a fraction of a second.

19. A method as claimed in any one of the preceding claims in which the receiving apparatus is arranged to receive signals over the total range of possible frequencies and possible angles of arrival simultaneously.

20. A method as claimed in any one of the preceding claims including redefining the set values of records associated with a ocat-or, -'n which a peak record a a L 1 .. ntrix ad a - '14 - e- AIR 85/JH density has been determined as the set values of the peak density location for subsequent use.

21. A method as claimed in any one of the preceding claims including the further steps of storing a library of radiation parameters of known pulse emitting sources and for each source distinguished by radiation frequency examining the appropriate library entry with respect to other data with the received pulses of that source are tagged to identify the source.

22. A method of deinterleaving radiated pulse signals substantially as herein described with reference to, and as shown in, the accompanying drawings.

23. Apparatus for deinterleaving radiated pulse signals comprising receiver means operable to receive pulses in a set of frequency ranges and in a set of angle of arrival ranges and to determine for each pulse received at least in which range of values of frequency and angle of arrival its parameters lie, and processing means including storage means operable to store a record of each pulse received in a time interval in accordance with the set values of the pulse frequency and angle of arrival, and record analysis means operable after the time interval to determine numbers of stored records associated with different values of the sets and to identify set values for which record storage density peaks occur for both set values and relate these each to a radiation pulse emitter distinguished by the range of frequencies and angles of arrival of the set values.

24. Apparatus as claimed in claim 23 in which the receiver means is arranged to receive pulses at all possible frequencies and angles of arrival of the ranges simultaneously.

25. Apparatus as claimed in claim 23 or claim 24 in which the receiver means includes an instantaneous frequency measurement receiver.

26. Apparatus as claimed in claim 27 in which the receiver means is operative in at least the microwave radar part of the electromagnetic spectrum.

27. Apparatus as claimed in claim 26 in which the receiver is a contiguous filtering receiver.

19 0 j AIR 85/JH 28. Apparatus as claimed in claim 26 or claim 27 in which the receiver means include,; radiation sensing means in the form of antenna means comprising an array of antenna elements and means to determine from signals received thereby and the array configuration the angle of arrival with respect to the array in which a received signal originates.

29. Apparatus as claimed in claim 28 in which the receiving elements of the array are radially directed from a reference axis with reception fields of at least adjacent elements overlapping to define by the relative amplitudes of signal received thereby the angle of arrival range with respect to the array.

30. Apparatus as claimed in claim 29 in which the array comprises four elements symmetrically disposed with directional reception axes orthogonally to each other in substantially the same plane.

31. Apparatus as claimed in claim 23 in which the receiver means is operative at optical wavelengths of the electromagnetic spectrum and includes radiation sensing means operable to distinguish emissions on the basis of frequency and wavelengths and angle of reception with respect to the sensing means.

32. Apparatus as claimed in claim 23 in which the receiver means is responsive to acoustic energy and includes acoustic radiation sensing means operable to distinguish emissions on the basis of angle of reception with respect to the sensing means.

33. Apparatus as claimed in any one of Claims 23 to 32 in which the receiver means is operable to produce a record of the pulse parameters including the radiation frequency and angle of arrival within the measurement resolution of the receiver means.

34. Apparatus as claimed in of claim 33 in which the processing means is operable to define the ranges constituting each value of each set and the storage means is operable to tag each record of pulse arrival with the set values of frequency range and angle of arrival range and store it in a storage location defined by a store address, said record analysis means being operable to identify those records associated with any vatue of the frequency or angle of arrival sets by addressing R AIR 85/JH the store locations appropriately taRged.

35. Apparatus as claimed in claim 34 in which the storage means is operable to replace the tagged values of radiation frequency and/or angle of arrival of each record from the receiver means with the set values of the ranges in which the values of frequency and angle of arrival lie.

36. Apparatus as claimed in any one of claims 23 to 35 in which each record of pulse arrival is tagged with the time of arrival.

37. Apparatus as claimed in any one of claims 23 to 36 in which each record of pulse arrival is tagged with the amplitude of the pulse.

38. Apparatus as claimed in any one of claims 23 to 37 in which the record analysis means is operable to determine for each of successive values of one of the sets for which the number of records is in excess of a predetermined threshold number each value of said other set for which a one dimensional density peak, associated with said one set value, is defined by the pair of set values, and operable to examine the density of the stored records associated with said pairs of set values representing one dimensional density peaks in relation to the density of surrounding set values to determine set values defining two dimensional storage density peaks by both set values of the pair.

39. Apparatus as claimed in claim 38 in which the record processing means is operable to define each two dimensional to identify thnse in c! ud ine, records density peak as being a candidate peak potentially distinguishing a pulse radiation emitting source and responsive the failure of two neighbouring candidate peaks to satisfy a threshold relationship condition between them to relate both peaks to a single pulse radiation emitting source.

40. Apparatus as claimed in claim 39 in which the processing means is response to both set values of a candidate two dimensional peak being separated from those of another candidate two- dimensional peak by less than two increments of R 1? - 27 AIR 85/JH the set values to relate both peaks to a single pulse radiation emitting source.

41. Apparatus as claimed in claim 39 or claim 40 in which the processing means is responsive to the record density of each pair of set values separating two candidate peaks exceeding a predetermined fraction of the record density of at least one of the peaks to relate both peaks to a single pulse radiation emitting source.

42.. Apparatus as claimed in claim 41 in which the 10 predetermined fraction is one half.

43. Apparatus as claimed in any one bf Claims 23 to 42 in which the storage means is responsive to determination of a density peak by the record analysis means to replace the set values of other records having neighbouring set values and associated with the density peak by the set values at which the density peak occurs.

44. Apparatus as claimed in any one of claims 23 to 43 capable of rotation in the azimuth plane in operation and in which the radiation sensing means is fixed in relation to the apparatus and including heading means operable to determine the azimuth heading of the apparatus with respect to an external reference direction and transpose the angle of arrival of each pulse determined with respect to a predetermined reference direction in the apparatus to an angle of arrival with respect to said external reference direction.

45. Apparatus as claimed in claim 44 in which the external reference direction is the nominal terrestrial or inertial direction to which the heading means refers.

46. Apparatus as claimed in any one of claims 23 to 45 in which the ranges forming the the set values are not all of equal extent.

47. Apparatus as claimed in any one of claims 23 to 46 in which the ranges forming the set value do not form a continuous total range.

48. Apparatus as claimed in any one of claims 23 to 46 in which the sum of the ranges the ranges of angle of arrival with - '18 - 5 AIR 85/JH respect to the apparatus extends 360'with respect t o' Lhe predetermined reference direction.

49. Apparatus as claimed in any one of claims 23 to 48 in which the record processing means is operable to distinguish each radiation source within the duration of the storage interval.

50. Apparatus as claimed in any one of Claims 23 to 49 including timing means operable to define the storage time interval as a time frame of predetermined duration.

51. Apparatus as claimed in Claim 50 in which the timing means is operable to define each time frame for a fraction of less than one second.

52. Apparatus as claimed in Claim 50 or Claim 51 when dependent on Claim 49 in which the timing means is operable to define a succession of contiguous time frames.

53. Apparatus as claimed in Claim 52 in which the procesing means is operable to store and analyse records for a number of successive time frames.

54. Apparatus as claimed in claim 53 in which the storage means contains the capacity to store the records of pulses received in at least two successive time frames.

55. Apparatus as claimed in Claim 53 operable to store at least some of the records of each time frame for fewer time frames than said number of time frames, the storage locations of discarded records being available for storage of records derived in a subsequent time frame.

56. Apparatus as claimed in any one of claim 23 to 55 in which the storage means and record analysis means comprises a programmed digital computer.

7. Apparatus for deinterleaving radiated pulse signals substantially as herein described with reference to, and as shown by, the accompanying drawings.

Published 1989 at The Patent Office, State House. 6671 High Holborn. London WC1R 4TP. Further copies maybe obtamed from The Patent officeSales Branch, St Mary Cray. Orpmgton, Kent BR5 3RD. Printed bY Multiplex techniques ltd, St Mary Cray, Kent, Con, 1187