DE2250305A1 - ARRANGEMENT FOR DIGITIZING THE BEARING VOLTAGES OF A MULTI-CHANNEL BEARING DEVICE - Google Patents

Description

LICE H TI A 9 9 R ΓΠ Π R

Patent-Verwaltungs-GmbH ^{LL u} °

6000 Frankfurt (Main) 70, Theodor-Stern-Kai 1

Ulm (Danube), October 11, 1972

POMJL / Schz / sa

UL 72/133

"Arrangement for digitizing the DF voltages of a multi-channel direction finder "

The invention relates to an arrangement for digitization the DF voltages of a multi-channel DF device.

The invention can advantageously be used, for example, in a direction finder with a conventional detection system that has two direction finding channels X and Y and a third channel, e.g. B. a ¹ round channel Z, feeds for the page identifier. The • third channel Z can also be connected to a separate fret antenna. The DF information is available at the outputs of the three channels as alternating voltages of the intermediate angular frequency ω for further processing. Usually these voltages are fed to a cathode ray tube in such a way that what is shown on the display screen

409817/0549

- 2 - UL 72/133

Radius the bearing value can be read off by a radio operator. A digital display is more advantageous than this optical analog display, an dioüo ο ich. also good long-distance transmission. In addition, digital representations are better suited to modern automated systems.

For such a type of representation, however, there is a preparation the analog information that digital technology makes use of is necessary. The latter brings z. B. the advantage with that much more than it used to be possible to design and develop at the desk with analog technology and some development-related issues Difficulty getting around. But automation demands much more than just the translation of analog technology into the digital technology. Due to the extensive exclusion of the human being, a technical substitute must now be made for human abilities in terms of the senses, hands and mind.

The invention is based on the object of an arrangement of the introductory specified type, which allows a very extensive evaluation of the channel information for direction finding purposes.

To achieve this object, according to the invention, in an arrangement of the initially mentioned type suggested that at

409817/0 5Ά9

- 3 - UL 72/133

a direction finder with a round channel sample and hold circuits -provided, the voltages of the Query the direction finding channels correctly and that the scanning and holding circuits A / D converters are connected to the sampled voltages into a sequence of digital values, which uniquely "contains the azimuth information.

The following are exemplary embodiments based on the figures and further developments of the invention are described in more detail.

3s has long been known to be used for unambiguous bearings the Watson-Watt principle in addition to the two bearing channels X and T a third channel, the so-called round voltage channel Z, is required «When a wave of amplitude A, des The azimuths α and the circle frequency / o are then the channel voltages

ZA ¹ sin Wt (ta)

XsA cos cc. cos o) t Ob)

X = A sin α, eos'ut (1o)

(The amplitude A ¹ of channel Z generally does not coincide with the amplitude A in the DF channels) _Λ Dew Z-channel has a phase difference of 90 ° compared to the DF channels.

409817 / 0SA9

72/133

If the voltage in the Z-channel now goes through O, namely from - to +, and if you measure ^{the voltage values of X and Y at this point in time (ints O 0} , 36O ⁰ , ...), you get according to

x ^ β A cos.α (2a)

V _1s A sin α (2b) ^s

Channel values from which A and α "can be calculated unambiguously.

In Fig. 1 these relationships are shown graphically. One sees that x,. and y- can also be negative, even if, as always assumed here, A is positive.

The corresponding circuit is shown in FIG.

Here, the Z-channel 1 generates the zero-crossing pulse by means of a circuit 2 known per se, with which the sample-and-hold circuits 3 or 4 are queried, which at the outputs of the X-channel 5 or X-channel 6 and its A / D converter 7 and 8 follow, which the digital values x>, and y * according to the GIn. (2) make it available in a few microseconds.

BATHROOM ORIGINAL - 5 -

409817/0549

72/132

In principle, it is "irrelevant which phase is unsciiiied the bearing channels and the Z channel. It is only important that a control pulse is derived from your channel, which enables the amplitudes of x. and χ, to be determined with the correct sign. Since the above version is also technically sensible and feasible, it should be assumed in the following.

It does not matter that the timing of the control pulse from the Hull passage of the Z-channel is exactly the right one The maximum of the DF channel voltages falls. Doesn't deviate from anything 0 from, e.g. B. around the value Ψ, then the measured values, (see Fig. 1)

X _x . β A cos α. cos Ψ (3a) X ₁ = A sin α. cos Ψ (3b).

As you can see immediately, this does not change the determination of a by aretan - ~ -. As long as Ψ is within - 90 remains, is the uniqueness in the determination of α guaranteed. It is not so easy to keep the Z-channel in its phase relation to the DF channels, especially when large frequency ranges are to be achieved. The insensitivity of the method to phase shifts the bearing channels and the Z-channel are therefore very useful,

409817/0549

Ψ - UL 72/133

This is where ζ serves. 3. an arrangement according to Fig. 4-, which one at known a device 10 for checking the sign of x ". and 7 "contains and according to predetermined rules from the · ascertained The sign relationships determine those quadrants, in which the unambiguous, d. H. the correct azimuth value α is.

Then an attempt is made with the amounts of x ^. and y ,. in the 1st quadrant through

α τ = arctan - (4-a)

to form a preliminary value, which depending on the quadrant is 0 ° or 180 ° must be added or subtracted.

Often it is useful to get the singularity of the tan at 90 °

to bypass (- ^ oo), an examination regarding the order of 4-5

to execute rotated quadrant. The test specification is given in FIG. 5 for each of the rotated quadrants. Thereafter depending on the quadrant with

α «arctan ~ - (4-a)

X / l

y I

or α - arccot - ^ (4-b)

the azimuth is determined and classified in the respective quadrant,

- 8 BAD ORIGINAL

409817/0549

- UL 72/133

Pig. 6 shows an arrangement suitable for this, "in the outputs of the A / D converters 7 ″ and 8 are a switching device 11 is connected, which - controlled by a decision device 12 - according to the GIn. 4- the output voltages the A / D converter feeds a computing circuit, which in the illustrated embodiment consists of arctan and arccot formers 13 and 14- ", but also in one Formers can be summed up of the relationship

exploits.

So far you have the GIn. (4) solved via software. The normal one Polynomial approximation according to Mac Laurin

Z Z Z arctan Z = Z- -τ + -p- - -rr + ...

converges badly in the range α ^ - 4-5 ° "or IZi- 1, as one can easily convince oneself by inserting IZ / - 1 in the row. It will therefore also be common for this one Purpose, the polynomial approximation according to Chebyshev or the continued fraction method according to Gauss is used with sufficient accuracy « However, the time required is corresponding. Recently, the technology of semiconductor integrated circuits The possibility opens up of functional relationships like here the arctan function via tables practically within

- 9 - BAD ORIGINAL

409817/0549

to edit memory access times. Hence ask these integrated circuits represent inexpensive read-only memories, from which by addressing the address x-, y * the Azinuth value α can be read out (ROMs. Read Only Memory).

If x ^ and j ^ are measured with an accuracy of 1% o, alao given with o'e IO bits or an address length of 20 bits, and the azimuth wire α over 360 with 1/3 ° accuracy, i.e. 10 bits, would have to be determined If you only want to use a ROM, the ROM can be designed for 1 million memory locations of 10 bits each. In this form, the arrangement is disadvantageous in terms of price.

Fig. 7 shows such an arrangement in which to 7 and 8 a ROM 15 is connected as a storage device, in which the functional relationship between the length of the digital values and the azimuth of the incident V # lle stored and the associated unique azimuth value when the sequence of digital values is fed to its address inputs outputs (Fig. 7).

But if one creates the two-dimensionality in x,., Y ^ (A and α) away by adding x ^ and y, - as below

, 2 - and

V 2 2 V "2 2

- 10 ORIGINAL BATHROOM

409817 / 0549Ö

UL 72/133

normalized and in this way gets cos α and sin α, so you only need to provide as many addresses and thus storage spaces as α values are possible: 1000 addresses with a cc word length of 10 bits. Such memories are relatively inexpensive to obtain. A corresponding arrangement Fig. 8 shows, in which between the A / D converters 7 and 8 and the memory device 18, which corresponds to the ROM 15 of Fig. 7, per se known computing circuits 16 and 17 are inserted, the sequence of them supplied digital Normalize values to the square root of their sum of squares.

The fastest and most advantageous, however, is when, after the quadrant determination discussed above, the quotient . ™ and in a ROM the arctan function between 0 and 90 ° is saved.

For example, an arrangement according to FIG. 9 is used for this purpose, in which a decision device 22 is connected to the A / D converters 7 and 8, which determines the quotient according to predetermined rules from the signs of x. and j * determines and, depending on the decision, emits one of four predetermined values, as well as a dividing circuit 19 are connected, the latter forms the quotient from the amounts of its input signals and a memory device 20, which is preferably a ROM

- 11 '- 409817/0549

6AD ORIGINAL

UL 72/133

is executed as an address. The functional relationship between

and. the asimuth value α of the incident wave between ^(! uriM 'S) · it! j';<ui | K'iuh <: rt. We: iJ <<; rain iub o.ino (hm, f-vui'lt'Ut.Ιι_ν- ^Λ Σΐ Asimutwert α-emitting addition circuit 21 for the output values of the storage device 20 and the decision device 22 are provided.

The storage device 15 or 18 and 20 in the arrangements according to FIGS. 7 or 8 and 9 supplemented by an interpolation device.

Hit the help of suitable computing circuits can also provide a statement made over the elevation angle £ of the incident wave will.

The antenna voltages at the incidence of a wave are shown in a complex representation

0: A _o e ^{iü) t} . ''

_Λ . . ika cos S cos α i «) t

_{o,. Q} -ika cos £. cos α iwt

-r, ika cos £ sin α i Ο- ^ ₀ - e

- 12 ORIGINAL BATHROOM

409817/0549

- AS- - Uli 72/135

Assuming the same amplification and phases!. On the part of the Channels X, X and "Z is

Z = A _o e ^{ia) t}

-ν, f ika cos £ cos α -ika cost cos cu icot ~

2iA sin (ka cos £. Cos α) β ^ιω Y_.j ^ / ■ ika cos £ sin a_ -ika cos £ .sin α- »iwt == ■

2iA sin (ka cos C sin oc) e ^ia)

In the usual approximation, in which the sin is replaced by its argument, the antenna circle diameter is then d = 2a

Z = A _o e ^{iü) t}

X s. IA _Q kd cosicos α e ^{la) t} Y = iA kd cos ^ sin α e ¹

These formulas show again that as in the gin. (1) the Z-voltage is shifted by 90 ° with respect to the bearing channels, but that between the amplitudes A ¹ and A of the GIn. (1) there is a defined relationship. If you also measure the amplitude of the Z-channel

409817/0

- UL 72/133

■ |; iHto _t by forming the Hilb transformed of Z via an OO ° -rharsenGchiober, then A can be

^T ο

circuit (quotient) can be eliminated.

The quantity 2 · π χ is read out from a ROM after being called at the set frequency and also from x by forming the quotient. and jy. removed. Now only the sizes are in the two bearing channels

x. * «cos £. cos α

y y a. cos C. sin α

those about the context

cos £ «pc ^ + j ^
Allow the elevation angle £ - to be read from a RON.

To digitize the round voltage in the sense explained above, in the embodiment of the invention according to FIG. 10 a 90 ° phase shifter 23 that transforms the Hilbert Forms round voltage, and a sample and hold circuit is provided, the latter queries itself according to the round voltage, and an A / D converter is connected to the sample and hold circuit (FIG. 10).

409817/0549

UL 72/133

The embodiment of the invention according to FIG. 11 is characterized in that a multiplication stage 26 for the product formation from the digitized round voltage A and the value kd = ~ - (with d - antenna base of the direction finder and λ = wavelength) it is provided that a computing circuit 27 is provided which is the square root of the sum of squares the preferably two bearing voltages forms that a division circuit 28 is provided, which through the root value divides the product formed and the resulting quotient equals the cos of the elevation angle (£.) of the incident Shaft is (Fig. 11).

A further embodiment of the invention, which is shown in FIG. 12, is characterized in that a multiplication stage 30 for the product formation from the digitized round voltage A and the value kd * = 5- (with d == ■ antenna base of the direction finder and λ - Wavelength) is provided that a computation circuit 31 »32 is provided, which in each DF channel the DF voltage through the formed product A _Q. kd divides that an arithmetic circuit 33 is provided which forms the square root of the sum of squares of the quotients, and that the root thus formed is equal to the cos of the elevation angle of the incident wave (FIG. 12).

- 15 -

409817/0549

Claims (12)

- UL 72/133 Claims ί 1. ^ arrangement for digitizing the DF voltages of a multi-channel DF device, characterized in that in a DF device with a round channel (1) sampling. and hold circuits (3 »^ O are provided, which in time ranges (Pig. 1) from the voltage (z) in the round channel are derived (z), the voltages (x, y) of the direction finding channels (5, 6) interrogate correctly, and that A / D converters (7, 8) are connected to the sample and hold circuits which convert the sampled voltages into a sequence of digital values which uniquely contains the azimuth information (FIG. 2). 2. Arrangement according to claim 1, characterized in that with two bearing channels X and Y, whose output voltages x ", or χ, Let a computing circuit be connected to the outputs of the A / D converters (7, 8) (9) which is the sequence of digital values 7 _Λ by arctan formation according to the equation α = arctan —— in the ambiguous Azimuth value α of the incident wave transformed (Fig. 3) 3. Arrangement according to claim 2, characterized in that one device (10) known per se is provided for checking the sign of x ^ and X ₁ , which according to predetermined rules from the determined sign ratios determines those quadrants - 16 - 409817/0549 true, in which the unambiguous, i.e. the correct azimuth value α lies. 4. Arrangement according to. Claim 1, characterized in that at two bearing channels X and T, whose output voltages x "and χ- (7,8) ^{Λ Λ} to the outputs of the A / D converter / a switching device (11)
management / is connected, which - controlled by a decision (12)
establishment / - according to > o π ⁺ Zf. -N 5 <■ α <+ 2Ξ ar et at ? ctan arccot Ύ _Λ < <ο the output voltages of the A / D converter of a computing circuit (13,14-) feeds which, depending on the switch position of the reversing device, the Sequence of digital values according to the equation ⁷ 1
α ^ arctan - or according to the equation ^X 1 α - arccot - reshaped. (Fig. 6) -17- 409817/0549 -Vr- UL 72/133 5. Arrangement according to claim 1, characterized in that on the A / D converter includes a storage device, e.g. B. a ROK (15) (Head Only Memory) is connected, in which the functional Relationship between the sequence of digital values and the Azimuth of the incident wave is stored and the "when the sequence of digital values is supplied to their address input- ■ gange outputs the associated unique azimuth value (FIG. 7). 6. Arrangement according to claim 5 »characterized in that between the A / D converters (7, 8) and the memory device (18) a computing circuit (16, 17) known per se is inserted, which normalizes the sequence of digital values to the root of its square root (Fig. 8). 7. The arrangement according to claim 1, characterized in that with two bearing channels X and Y, the output voltages from x ^. "or y ^ are both a decision device (22) to the A / D converter, which determines the quadrant according to predetermined rules from the signs of x- and y ,," and, depending on the decision, one of four predetermined values (e.g. B. O, 5, π or 22. ) emits »and a dividing circuit (19) are connected, the latter from the Amounts | xJ and \ jA form the quotient, - and a memory ^{i ι l x} i 1 device (20) as the address in which the functional supply l ^ il relationship between .—, and the azimuth value α of the incident wave l ^x il between 0 and + Sj is stored, and that a den - 18 BATH ORIGINAL 4098 17/0549 - - TJL 72/133 unambiguous azimuth value α emitting. Addition circuit (.21) is provided for the output values of the storage device and the decision device (J ¹ Ig. 9). 8. Arrangement according to "one of claims 5" to. 7 »characterized-, that the storage device by an interpolation device ". ■ - ■" - "■ '-;'" - "''. - '■ is added 9. Arrangement, 'according to claim 1y characterized in that - that for. Digitization of the round clamping a 90 ° phase shift "b.er '(23), which forms the HilTDert transform of the voltage and a additional sample and hold circuit (24) are provided, the latter is queried according to the voltage and voltage, and that an A / D converter is connected to the additional sample and hold circuit (24) "(25) is connected (Fig. 10). ■ ■■ '■■' ■" ■ ■ "-. 10. The arrangement according to claim 9, characterized in that one Multiplication stage (26) for the product image from the digitized E and voltage A and the value kd - ^ - - (with d = -Antenna- "basis of the direction finder and λ «■ wavelength) is provided that ' a computing circuit (27) is provided which forms the square root of the sum of the squares of the preferably two bearing voltages "that a division circuit (28) is provided which divides the root value by the product formed and the product thus formed Quotient equal to the cos of the elevation angle (£.) Of the incident Shaft is (Fig. 11). - 19 40 9 817 / Ö 5 4 9 © AD ORIGINAL 11. Arrangement according to claim 9, characterized in that a multiplication stage (30) for product formation the digitized round voltage A and the value kd -π ~ (with d = Anbenncnbaais dos roilßcrätea and λ, * Wollcnlün ^ o) provided is that a computation circuit (31, 32) is provided, which in each i'uilkannl the filing tension by the formed Product A_. kd divides that a computing circuit (33) is provided is the root of the sum of squares of the quotients forms, and that the root thus formed is equal to the cos of the angle of elevation of the incident wave (Fig. 12). 12. The arrangement according to claim 9 »characterized in that with two bearing channels X and T, whose output voltages x ^, or Y _x . are, a phase comparison circuit (34) is provided which only .Phasengleichheit of X _x. and V _x . releases the time ranges which are derived from the voltage (z) in the round channel, that on the one hand the direction finding channels (7, 8) each have an addition (35 j 36) and a subtraction stage (37 _? 38) 'and on the other hand a subtraction stage (39) is connected to the sum channel (25), which continuously process only two immediately successive values of the digital sequence, and that an inversion stage (40, 40, 41) are connected for the sign which, in the case of negative output values of the differential stage (39) of the round duct, reverse the sign of the differential values supplied to them (FIG. 13). BATH ORIGINAL 409817/0549 Blank page