THE UNIVERSITY OF MICHIGAN RESEARCH INSTITUTE ANN ARBOR TWO INTERFEROMETER-TYPE DIRECTION-FINDING SYSTEMS Technical Report No. 89 Electronic Defense Group Department of Electrical Engineering By: S. I. Soclof Approved by:' H. W. Farris Project 2899 TASK ORDER NO. EDG-10 CONTRACT NO. DA-36-039 sc-78283 SIGNAL CORPS, DEPARTMENT OF THE ARMY DEPARTMENT OF ARMY PROJECT NO. 3-99-04-106 May 1959

TABLE OF CONTENTS ABSTRACT iii 1. INTRODUCTION 1 2. THE "OAK LEAF" DIRECTION FINDER 1 2.1 Principle of Operation 1 2.2 System Analysis 3 3. THE "SPHERICAL WAVE-FRONT" SYSTEM 10 3.1 Principle of Operation 10 3.2 System Analysis 12 3.5 Possible Method of Instrumentation 17 4. PERFORMANCE OF INTERFEROMETER DF SYSTEMS IN TERMS OF ACTUAL PROPAGATION CONDITIONS 22 5. CONCLUSIONS 28 ii

ABSTRACT The analyses of two types of radio direction-finder are presented. These systems were investigated primarily with the VHF frequency range in mind. The analysis of the systems is presented for the purpose of indicating one way of instrumenting an interferometer system and to indicate a way of eliminating the ambiguities involved. Some discussion is given on the limitations of such systems. As with most direction finders in this frequency range and lower, the limitation on accuracy is a function of the environment and not solely dependent upon the instrumental accuracy of the equipment. iti

TWO INTERFEROMETER-TYPE DIRECTION-FINDING SYSTEMS 1. INTRODUCTION This report will describe two proposed radio direction-finding systems based on the interferometer principle. From a theoretical standpoint the systems show certain advantages over other types of systems. However, it must be kept in mind that in a practical environment these advantages may not be realized because of the nature of operation of the systems. Both systems suffer in that they take samples of the wave front at only a few widely separated points and thus do not average over the wave front as would be desired. 2. THE "OAK-LEAF" DIRECTION FINDER 2.1 Principle of Operation The first of the two systems investigated is called the "Oak-Leaf" DF because of the characteristic pattern produced by its antenna system. The system utilizes five antennas, as shown in Fig. 1. Antennas 1 and 2 form a wide-spaced pair in terms of the wave length, X. If the phase of the signal at antenna 2 is continually advanced with respect to that at antenna 1, a multilobed interference pattern is produced. A similar process obtains from the antenna pair 1 and 3 except that, since the spacing is slightly different, the resultant pattern will be rotated slightly in space, although similar in shape.

s __: 4:-r / L w V _1 3 2 — I 1- s s<\ _ _ _ _ _ _ _ p 4 -s sL>X \i —-— L L m=L/S FIG.I "OAK-LEAF" DF ANTENNA POSITIONING e, — o J 1,o +o, _ _ (o-WI j FILTER v |I —- MIXER -|e, W, (<O+Wl) Wo +Wl LOCAL OSCILLATOR FIG. 2 2

Each pattern alone contains information as to the bearing of the transmitter but with ambiguities as represented by the multiplicity of lobes in the multilobed pattern. These ambiguities are resolved by a continuous rotation of one pattern with respect to the other so as to produce one major lobe which will indicate the direction of the target. There remains, however, the reciprocal bearing ambiguity, which is resolved by using another antenna array, composed of antennas 1, 4, and 5, at right angles to the first. 2.2 System Analysis The initial analysis concerns the antenna system comprised of antennas 1, 2, and 3; it is subsequently applied to the system of 1, 4, and 5. Establish antenna 1 as the reference antenna and let the voltage at this antenna be represented by: e1 = A cos ot, where A involves system parameters involved with antenna 1. Similarly, e3 = B cos (o0t + a), and e2 = C cos (Wot + ), where B and C represent system parameters peculiar to antennas 3 and 2, respectively, and 2n a = t (L-S) cos i, and = (L) cos

Now, with reference to Fig. 2, if a locally generated signal at a frequency of W1/2A is mixed with el, the following is obtained: e x e = A cos 0t x D cos w t O 1 AD = - cos ( + W )t + COS (U) - )l)twhere e = D cos e lt is the oscillator voltage. After filtering to retain only the 0o + el term one obtains, AD el -= cos (Do + ol)t Now, with reference to Fig. 3, a similar process is gone through, this time with e' and e2. e1 e LD c +2 iW- cos ( + B cos t + ) ADB t B r [ cos (2o t + l a) + a cos (a - t)] After filtering to retain only the e1 term one obtains e2 = - cos (a - olt) Going through a similar process for e3, one obtains,, ADC 3 = - cos (5 - lt) Now introduce another locally generated signal given by e = E cos e2t 4

I Wo+Ji!''eo- a | WI FILTER W,L. e2 MIXER () wo 2 FIG. 3 eI/ W2 "U+WI,2''La ----- —' -- ('d2 -WI FILTER aI +(2Ja __ MIXER ---- W — 2 -- (Wa +@ W2) e " W2 e2I LOCAL OSCILLATOR FIG. 4 5

With reference to Fig. 4 it can be seen that one obtains e2 x e A O ( wt) x cos (a - t) x E cos (t) = -AD [cos (c + w2t - wlt) + cos (a - alt - w2t)] After filtering to retain only the w2 + 1l term one obtains,,, ADBE e2' = -- cos (a - a)t - w2t) Now after the processing shown in Fig. 5, we find 2 fABDE ADC ~e ~48 - (tI W2)t x = cos (+ - Clt) 2 W I + 2 la MIXER'- 2 -2 L 2cw, + e2 a, L e" l 3 FIG. 5 6

A2D2BCE F 1 ef = - -A E [cos (a + p - lt - 2a 2t) + cos (a - + O2t) By definition, m = L/S. Now, since a = 2 (L-S) cos 9, and 2t 0f = -A (L) cos 9, we find that 0 2n 2c 2tS a + = (2L-S) A cos = = (2Sm-S) 2 oA cos = (2m-l) 2- cos, and 0'0 0 a - - _ 2s cos 9. 0 0 Now choose o1 and.2 such that = m - 1. a z= (m-l) 2 so: 2aM + 02 = 22'- a2u + o2 = 2m)2 - 2 = O2 (2m-1) Substituting all of this into the expression for ef one obtains: ef = AD BCE. {cos [(2m - (2m-l) w2t] + cos [2t --- cos @] },or ef= A2DBC [(2m-l)(au2t 2 Cos 9)+ cos (w2t - cos ) Normalize ef to prevent the resultant expression from ever becoming negative and call this F(@): F(@) = 2 + cos [ 2t - CO + cs [(m)(t - 2 ) 7

Let y = 2. Now if F(@) is applied on a circular sweep of an oscilloscope being swept at a rate of 7 = - revolutions per second one obtains a pattern given by: F(@) - 2 + cos (r - A cos 9) + cos (2m-l)(7 - cos ). 0 0 Figure 6 illustrates the particular case wherein @ = 90~, but the same pattern will appear for any value of 9 except for a rotation around the origin. Now, F(@) will have an absolute maximum when 2nS S cos 9 = 7 * 0 From this one obtains cos 9 as: cos = EW 2nS- Since S < 2 the only ambiguity involved in determining 9 will be in determining the correct algebraic sign to use. This is resolved by antennas 1, 4, and 5. Going through the same signal processing as before, using antennas 1, 4, and 5 one obtains another value for 7, to be called 7NS, from which one obtains sin 9 by the relation: / o\ sin 0 = 7NS 2AS Having obtained 7NS and YEW, one can now determine the angle of arrival, 9, by constructing the triangle as shown in Fig. 7. 8

FIG.6 "OAK-LEAF"D F F(0)=2+COSy +COS(15y) L/d = m = 8 =90" 9

ff2 Y NS= 27 S SIN & Xo = = —Yr s cos& -- - FEW YE W'X o FIG.7 DETERMINATION OF 8 Note that the value of 7 is independent of all the arbitrary amplitude constants, A, B, C, etc., so that the system will be relatively immune to amplitude unbalances. Figure 8 shows a complete block diagram for the "Oak Leaf" DF. 3. THE "SPHERICAL WAVE-FRONT" SYSTEM 3.1 Principle of Operation The "Spherical Wave-Front" system samples the incoming wave front at five points, one point serving as a reference point for the rest of the phases. Under ideal circumstances the wave propagating from a point source expands spherically, so that, by determining the shape of the wave front, one can determine not only the bearing of the target but also the range, thus determining the location or fix of the target. 10

MASTER W, OSC' (m-l) CWI I * —-----------------— MIX wo w2 2'- MIX MIX j- I!,wL1 (~o a e 3 _- MIX MIX wo e, FILTER oJ~ 2oj + w z e-~~~~~~~~~~~~~~~~3"~;a 73 SWEEP GEN. CRO AMP II IFIG.8 BLOCK DIAGRAM FOR "OAK-LEAF" D F 11

The determination of the shape of the wave front is done by comparing the relative phases of the incoming signal at each of the sampling points. Using these relative phases and certain approximations, one being that the array length, L, be much less than the distance to the transmitter, a fairly compact expression relating the bearing and range of the transmitter to the relative phases at each of the sampling points can be obtained. In the proposed system, the signals from the sampling points are processed in such a way as to give a direct indication of the bearing and range. 5.2 System Analysis First consider the signal processing necessary for the operation of the system and then briefly the circuits to perform these operations. The antenna positioning is as shown in Fig. 9. The target transmitter is at point P. Now consider only antennas 0 and E as shown in Fig. 10. One finds that if L <<R: L2 cossin2 L3 EO = R - RpE = cos - sin2 2 cos sin 4 L (6 cos2 - 1) +... 128R3 Similarly: L2 L32 WO = R - Rp = cos - sin2 + cos sin L4 2 nL- (6 cos2 - 1) +... 128R3 12

I P I N 0 XMTR I R W E L ^________0\ \________ \t;S \\_ I L>>X r ~ L **4 *44~ ~ ~~~~~~4 FIG. 9. ANTENNA FPOSITIONING FOR ItSPHERICAL WAVE-FRONT" SYSTEM S — ~~~ ~~~LFIG. 9. ANTENNA POSITIONING FOR''SPHERICAL WAVE-FRONT" SYSTEM

01'91- \ / \ 3 1% 0 33d/ / | r i dl

In terms of relative phase angles referred to antenna 0 as the reference point and dropping the higher order terms: 2i nL cos 9 L2 sin2 9 PEO = T- EO = 4R + 2i nL cos @ sL2 sin2 @ PWO = wo O x 4R -+ and similarly: rL sin Q xAL 2 ~NO k - R_ cos e + *;PSO =.. + Now add pEO and NpO to obtain pEW:.L2 2 PEW = EO + PWO = 2RA sin2 +. Similarly for cpn. 2 NS TPNO +OPSO C Cos2e + * Next take the sum of P'EW and (pNS to be called (EWNS: EWNSL= P = sin2 9 + cos2 +.. TEWNS ='PEW + PNS =' 1 L PEWNS = 2R + *** Solving this for R one obtains: * L2 R = 2EWN Now take the quotent of pEW and (EWNS: 15

nL2 sin2 Q CEW. 2R. PEWNS iL2 2Rk Solving this for sin2 9 one obtains: ~;iz2 Aid SPEW sin2 9 E PEWNS Similarly: 2 PNS cos 9.'EWNS tS 11 FIG.II DETERMINATION OF RANGE AND BEARING 16

Since sin 0' E sn /( )2 + ( N2 and v cos 9 (OW) + (0/NS) we can construct the diagram shown in Fig. 11. 3.3 Possible Method of Instrumentation'PEO is the phase angle by which the signal at antenna E leads that at antenna 0. cpWO is likewise the phase angle by which the signal at antenna W leads that at reference antenna 0. Let the voltage eE, ew, and eo be respectively: eE = E sin (aot + PE0), = W sin (c)t + cpWO) and e0 = A sin (ot)} where E, W, and A are arbitrary constants involving the system parameters. First, by the processing shown in Fig. 12, one takes the product eE and eW to be called eEW: e x eW E sin (uot + pEO) x W sin (tot + pWO) EW [cos E) - cos (a co t + PE0 + pW)] 17

2wo _ d.c. FILTER 2wo MIXER *eEW eE — ------ ~~~~~20 o eEw ewp * O2 I FIG. 12 After filtering to retain only the 2uo term one has: EW eEW = " " cos (2ot + PEO + PWO) EW - 2 cos (ao~t + PEW) Next one takes a locally generated signal at a radian frequency of ow + B to be called eB: eB = B sin (wD + ))t By processing shown in Fig. 13 one obtains the following: Wo 2woI -- eo- I 1 FILTER 2 o + MIXER (2 ) Co +'3 LOCAL OSCILLATOR FK. 13 18

e0 x eB = A s txBin ot( + s)t AB 2 cos 1t - cos (2o + P)t Now, by filtering, one takes only the oo + p component of eOB: AB OB - cos (2uo + )t Next, as shown in Fig. 14, one takes the product of eEW and eOB to be called eEWO: EW x e0B = E cos (ot + (PEW) x (ot + t) = A [cos (Pt - W) + cos (4t + t + PEW)] 2wo e~EW* C0 40o + | FILTER | MIXER,~ eEWO 2w. +~. FIG. 14 19

After retaining only the f term: EWAB eEWO = - cos (Bt - WEW) In a similar manner, the signals from antennas N and S together with that from antenna 0 could be processed to give: NSAB eNSO = 8 cos ( Pt - PNS) The next step is the derivation of a voltage proportional to /(EW and /NS.N Consider Fig. 15. SAWTOOTH Vot Vo - GENERATOR - ADD AMP Vx VW |-3 — f(A) _-Vx2 AMP -- SQUARE FIG. 15 20

The output voltage, Vx, is given by: VX/tv -... 1 7 V~2x ( 12' t For values of A, 7, and t, such that V/Vt >> -l l and 4A27V t >> 1, ~ 2A.f Vx can be represented approximately by: /Vot Vx Ay - Now one has a voltage, Vx, which is proportional to the square root of time, t. Now send eEW0 and eOB through the wave shaping processing shown in Fig. 16. e e' EW-O - AMP -- CLIP D —----- DIFFERENTIATE - EWO AMP CLIP DIFFERENTIATE 80 FIG. 16 The output elo is a series of sharp pulses with a recurrence 2A'PEW interval of time, To = -, but lags the pulses of eBO by a time, T = —, as shown in Fig. 17. 21

Now if one takes the voltage Vx, which also has a fundamental period of To = -, and considers its value at t = T, one obtains a voltage VEW proportional to / EW Similarly, VNS/ * Now, if VEW is applied to the y-axis of an oscilloscope, and VNS is applied to the x-axis, a triangle can be constructed as shown in Fig. 18. However, there still remains some quadrantal ambiguity. This ambiguity can be resolved by some additional instrumentation since this information is present in the unprocessed signals. A complete block diagram of the system is shown in Fig. 19. 4. PERFORMANCE OF INTRFEROMETER DF SYSTEMS IN TERMS OF ACTUAL PROPAGATION CONDITIONS Each of the systems discussed determines the bearing of a transmitter by sampling the wave front at a few widely separated points. In the "Oak-Leaf" DF, the bearing is determined from these samples by comparing the relative phases at each of the sampling points; from these relative phases an imaginary phase front can be drawn which under ideal propagation conditions, will coincide with the actual phase front. The bearing is determined from the phase front by a perpendicular to the phase front at the DF site. The "Spherical Wave-Front" system proceeds one step further than the previous system: not only are the relative phases compared, but the differences in relative phases are compared in order to obtain as an end result an imaginary phase front as before represented, this time, by the arc of a circle. Again, under ideal propagation conditions, the imaginary 22

~Z 083 NO NOIlVVAOJNI 9N18V38 ONV 39NVt JO- NOIIVlN3S3d'81'91J | _ — SNA M3A g.o — S- I, —-1 ------ M3A M.k/'I >4: XA

0W PREAMP WO 2wo0 + E o - MIXER CONV. ~COW. ~ FILTER IF AMP co ~~I ~~ ~OSC. Wo +,B,/1' ~ 1 -- -- MIXER F 9 B PREAMP GA FILTER W S CONV. I o IF AMP MIXER 2wo MIXER P FILTER FILTER cS PREAMP CONV. IF AMP AMP CLIP DIFF AMP - CLIP DIFF KEYED GATE SAWTOOTH ADD - AMP PEAK GEN. ---- ---- DETECTOR SQUARE AMP CRO'FIG.19 BLOCK DIAGRAM OF SPHERICAL WAVE-FRONT SYSTEM 24

phase front will coincide with the actual phase front. In this system the bearing and also the range are determined by locating the center of the circular arc. In either system the fundamental difficulty is the same. Each system samples the wave front at only a few widely separated points and in no way attempts to determine the average wave front between these points. The "Spherical Wave-Front" system is a bit worse in this respect. Not only does it compare the relative phases at the several points and then take the differences of these phases as does the "Oak-Leaf" DF, but it goes one step further and takes differences of the differences of the relative phases, thus leaving the system open to the possibility of very serious errors. Perhaps this may be seen more clearly by referring to Fig. 20. In the "Oak-Leaf" DF system the bearing of the transmitter is determined by essentially comparing length P1 with length PO to derive a difference length AL = PO-P1, which is determined in terms of phase angles. Since ALT will be a very small quantity as compared to PO or P1 it can be seen that a very small difference in two large quantities is being taken, which will lead to serious errors when actual propagation conditions cause the average velocity of wave propagation over path P1 to differ from that over path P0. Looking at Fig. 21, which represents the conditions involved in the "Spherical Wave-Front" system, it is seen that in this system that distance PO is compared to PE and also to distance PW to obtain difference lengths LE0 = PO-PE and AWO = PO-PW, respectively. Then AEO is compared to AWO to obtain the second difference, DEW = aE0 - WO0, which is expressed in terms of phase angles, all with respect to the phase of the signal at antenna 0. 25

TRANSMITTER AL= PO- PI/ AL = PO- PI/ FIG. 20 26

Lz IZ'911 11 /M ~~~d~~~~~~~~~~ / d ~I~~~i1111NSNV~I~~~~~

Since the array length L is very small compared to the distance R, it can be seen that only a very small difference in the average velocity of wave propagation over the three paths, PE, PO, and PW, can cause extremely large errors in the bearing and range. Also, since the resultant phase angle to be measured is likely to be extremely small except in the special case of the transmitter being very close to the direction-finding site, the instrumental errors are likely to be very large, if not the prime determining factor in the practicality of the system. 5. CONCLUSIONS The analysis section of this report presents two methods of instrumenting two interferometer DF systems. In order to determine the usability of these systems in a real situation, one should make an error analysis using perturbation data for the particular case involved. It appears that as one increases the aperture in the manner of an interferometer system without increasing the number of sampling points that the system may be more susceptible to environmental errors than would ordinarily be the case, and, as is pointed out in the report, more sample points in the space are needed. 28

DISTRIBalTION LIST Copy No. C 1-2 Commanding Officer, U. S. Army Signal 26 Commander, Rome Air Development Center, Research and Development Laboratory, Griffiss Air Force Base, New York, ATTN: Fort Monmouth, New Jersey, ATTN: Senior RCSSLD Scientist, Countermeasures Division 27 Commander, Air Proving Ground Center, ATTN: 3 Commanding General, U. S. Army Electronic Adj/Technical Report Branch, Eglin Air Proving Ground, Fort Huachuca, Arizona, Force Base, Florida ATTN: Director, Electronic Warfare Department 28 Commander, Special Weapons Center, Kirtland Air Force Base, Albuquerque, New U Chief, Research and Development Division, Mexico Office of the Chief Signal Officer, Department of the Army, Washington 25, 29 Chief, Bureau of Ordnance, Code ReO-l, D.C. ATTN: SIGEB Department of the Navy, Washington 25, D.C. 5 Chief, Plans and Operations Division, 30 Chief of Naval Operations EW Systems Branch, Office of the Chief Signal Officer, OP-347, Department of the Navy, Washington Washington 25, D.C., ATTN: LGEW 25, D.C. 6 Commanding Officer, Signal Corps Elec- 31 Chief, Bureau of Ships, Code 840, Departtronics Research Unit, 9560th USASRU, ment of the Navy, Washington 25, D.C. P.O. Box 205, Mountain View, California 32 Chief, Bureau of Ships, Code 843, Depart7 U.S. Atomic Energy Commission, 1901 ment of the Navy, Washington 25, D.C. Constitution Avenue, N.W., Washington 25, D.C., ATTN: Chief Librarian 33 Chief, Bureau of Aeronautics, Code EL-8, Department of the Navy, Washington 25, D.C. 8 Director, Central Intelligency Agency, 2430 E. Street, N.W., Washington 25, 34 Commander, Naval Ordnance Test Station, D.C., ATTN: OCD Inyokern China Lake, California, ATTN: Test Director-Code 30 9 Signal Corps Liaison Officer, Lincoln Laboratory, Box 73, Lexington 73, 35 Commander, Naval Air Missile Test Center, Massachusetts, ATTN: Col. Clinton W. Point Mugu, California, ATTN; Code 366 Janes 36 Director, Naval Research Laboratory, 10-19 Commander, Armed Services Technical Countermeasures Branch, Code 5430, WashingInformation Agency, Arlington Hall ton 25, D.C. Station, Arlington 12, Virginia 37 Director, Naval Research Laboratory, 20 Commander, Air Research & Development Washington 25, D.C.,ATTN: Code 2021 Command, Andrews Air Force Base, Washington 25, D.C., ATTN: RDTC 38 Director, Air University Library, Maxwell Air Force Base, Alabama, ATTN: 21 Directorate of Research & Development, CR-4987 USAF, Washington 25, D.C., ATTN: Chief, Electronic Division 39 Commanding Officer-Director, U.S. Naval Electronic Laboratory, San Diego 52, 22-23 Commander, Wright Air Development Center, California Wright-Patterson Air Force Base, Ohio, ATTN: WCOSI-3 40 Office of the Chief of Ordnance, Department of the Army, Washington 25, D.C., 24 Commander, Wright Air Development Center, ATTN: ORDTU Wright-Patterson Air Force Base, Chio, ATTN: WCIGL-7 41 Chief, West Coast Office, U.S. Army Signal Research and Development Laboratory, Bldg. 25 Commander, Air Force Cambridge Research 6, 75 S. Grand Avenue, Pasadena 2, Center, L.G. Hanscom Field, Bedford, California Massachusetts, ATTN: CROTLR-2 29

UNIVERSITYOF MICHIGAN 3 9015 03525 2017 DISTRIBUTION LIST (Cont'd) Copy No. Copy No. L2 Commanding Officer, U.S. Naval Ordnance 59-66 Commanding Officer, U.S. Army Signal Laboratory, Silver Spring 19, Maryland Research & Development Laboratory, Fort Monmouth, New Jersey. 43-44 Chief, U.S. Army Security Agency, ATTN: 1 Copy - Director of Research Arlington Hall Station, Arlington 12, 1 Copy - Technical Documents Virginia, ATTN: GAS-24L Center - ADT/E 1 Copy - Chief, Systems Branch, 45 President, U.S. Army Defense Board, Countermeasures Division Headquarters, Fort Bliss, Texas 1 Copy - Chief, Detection & Location Branch, Countermeasures 46 President, U.S. Army Airborne and Elec- Division trcnics Board, Fort Bragg, North Carolina 1 Copy - Chief, Jamming & Deception Branch, Countermeasures 47 U.S. Army Antiaircraft Artillery and Guided Division Missile School, Fort Bliss, Texas, ATTN: 1 Copy - File Unit No. 4, Mail & E&E Dept. Records, Countermeasures Division 48 Commander, USAF Security Service, San 1 Copy - Chief, Vulnerability Antonio, Texas, ATTN: CLR Branch, Signal Facilities Division h9 Chief of Naval Research, Department of 1 Copy - Reports Distribution Unit, the Navy, Washington 25, D.C., ATTN: Countermeasures Div. -File Code 931 67-74 Commanding Officer, U.S. Army Signal 50 Commanding Officer, U.S. Army Security Research & Development Laboratory, Fort Agency, Operations Center, Fort Huachuca, Monmouth, New Jersey, ATTN: Reports Arizona Distribution Unit (for retransmittal) 51 President, U.S. Army Security Agency 75-76 Commanding Officer, U.S. Army Signal Board, Arlington Hall Station, Arlington Missile Support Agency, White Sands 12, Virginia Missile Range, New Mexico, ATTN: SIGWS-EW and SIGWS-FC 52 Operations Research Office, Johns Hopkins University, 6935 Arlington Road, 77 Commanding Officer, U.S. Naval Air DevelopBethesda 14, Maryland, ATTN: U.S. Army ment Center, Johnsville, Pennsylvania, Liaison Officer ATTN: Naval Air Development Center Library 53 Commanding Officer, U.S. Army Signal 78 Dr. H. W. Farris, Director, Electronic Research & Development Laboratory, Defense Group, University of Michigan Fort Monmouth, New Jersey, ATTN: U.S. Research Institute, Ann Arbor, Michigan Marine Corps Liaison Office, Code AO-4C. 79-99 Electronic Defense Group Project File, 54 Commanding Officer, U.S. Army Signal University of Michigan Research Institue, Research & Development Laboratory, Ann Arbor, Michigan Fort Monmouth, New Jersey, ATTN: ARDC Liaison Office 100 Project File, University of Michigan Research Institute, Ann Arbor, Michigan 55 President, U.S. Army Signal Board, Fort Monmouth, New Jersey 56-58 Commanding Officer, U.S. Army Signal Research & Development Laboratory, Fort Monmouth, New Jersey, ATTN: Chief, Security Division (for retransmittal to BJSM) Above distribution is effected by Countermeasures Division, Surveillance Dept., USASRDL, Evans Area, Belmar, New Jersey. For further information contact ir.. O. Myers, Senior Scientist, telephone PRospect 5-3000, Ext. 61252. 30