Oral-History:Paul Rosen

About Paul Rosen

Paul Rosen, a graduate of Tufts University, was head of the ground equipment satellite group at Lincoln Laboratory. While at the Air Force Cambridge Research Laboratory (AFCRL), which later became Lincoln Laboratory, he worked on the SAGE (Semi Automatic Ground Environment) program. He also designed a modem that depended on amplitude alone, which was adopted by Bell Labs.

In this interview, Rosen discussed his work with AFCRL, Lincoln Laboratory and his interest in radars and Spread-spectrum communications. He specifically mentions the collaborative work environment at Lincoln Laboratory, his interaction with MIT professors and his work with PPIs (Pulse Position Indicators) and Spread-spectrum communications.

About the Interview

PAUL ROSEN: An Interview Conducted by Michael Geselowitz, IEEE History Center, 22 April 2004

Interview #440 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, IEEE History Center, 445 Hoes Lane, Piscataway, NJ 08854 USA or ieee-history@ieee.org. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

Paul Rosen, an oral history conducted in 2004 by Michael Geselowitz, IEEE History Center, Piscataway, NJ, USA.

Interview

Interview: Paul Rosen

Interviewer: Michael Geselowitz

Date: 22 April 2004

Place: IEEE History Center, New Brunswick, New Jersey

Family background; early interest in engineering

Geselowitz:

Let's begin with your early years and how you got involved in engineering.

Rosen:

I can remember distinctly as a little kid, five years old, saying I was going to be an engineer. I did not have the foggiest idea what an engineer did really. I knew they built bridges.

Geselowitz:

What did your father do?

Rosen:

My father was an immigrant who came to the United States in 1920. I was born in 1922. He came here speaking no English and his first job was washing milk cans at a Jewish dairy manufacturing plant that made cream cheese, sour cream and so on. He had taught himself arithmetic and how to read. Over a ten-year period he learned the Jewish dairy chemistry business such that he became the dairy chemist for that company, and that is what he did for most of his life. He bought a house within ten years of having come to the States. He also learned how to do electricity, plumbing and painting. Fortunately or unfortunately, I was the eldest child and therefore he had a permanent helper. That was peachy until I got to be an adolescent and liked girls, in which case I began to resent the heck out of it. I remember taking radios apart. At that time radios were built with three TRF circuits lined with copper, so I could sell the copper. I did that to earn a few pennies, and I've got to tell you that infuriated my father. My father used to say something about me in Yiddish, translating roughly as "He likes to look to see where the legs are growing from". He was talking about a kind of innate curiosity, like sexual curiosity. As a kid I built things.

Geselowitz:

In some way you had an idea that engineers built things, that is why you wanted to be an engineer? Was this in Boston?

Rosen:

I just thought I’d like to do it, whatever the “it” was. We lived in a Boston suburb, Chelsea, which at that time was largely a Jewish ghetto. Then we moved to Revere which was a more upscale ghetto. I don't know if you are familiar with these areas.

Geselowitz:

Yes. I lived in Cambridge for eighteen years.

Rosen:

Okay. I always liked to do things with my hands. I told my grandson yesterday -- who is studying the oboe -- that I took piano lessons. I have a very nice grand piano at my house now. My mother was dying for me to take piano lessons. I really was not all that interested in taking piano lessons, but I did because it was in the middle of the Depression and people did all kinds of things to make a living. One thing that people did was teach the piano. I don't think my teacher was a very good musician, but he knew how to make radios. I had gotten a copy of F. E. Terman’s book, which was written about 1922 or 1923. Crystal radios were my thing. I would wind the coils, and at one point I made my own detectors. They were not really lead sulfide detectors as mined and packaged, but I discovered that could make lead sulfide, using solder as lead which I filed and put sulfur in it. Then heated the stuff. If you were lucky you got a detector. I just loved making my own detectors.

Geselowitz:

Many engineers of your generation that I have interviewed got their start with radio sets and crystal radios.

Rosen:

Yes.

Geselowitz:

It concerns me that the current generation, even the ones who are quote-unquote ‘tech savvy,’ are programming on computers. If they pull the backs off the computers they have printed circuits that they cannot possibly manipulate. How are they going to learn the kind of building real world systems that you learned?

Rosen:

The answer to your question is they don't and they will not. And what is going to happen as a result is that people will learn to depend more and more on mathematics and the transformation of that mathematics into computing the behavior of electronic devices. If I can do a Fourier transform and make a filter mathematically instead of winding and soldering, what the heck. I wish I could do the mathematical part, but I cannot.. My oldest son is an MIT graduate and he is something of an anomaly, since he was graduated a long time ago. His forte is POTS – plain old telephone systems -- making the stuff that goes between the guy who brings in the wide band signal and the user. He is probably going to be preserved in plastic pretty soon and stuck in a museum somewhere, because that is not the direction in which the world is going. He designs circuits, real component based circuits.

Geselowitz:

What year did he graduate from MIT? Do you remember?

Rosen:

He graduated MIT in 1967. I went to school with him. I took a couple classes with him because I was re-treaded that year. There is a picture somewhere in one of the MIT books of him and me sitting and gazing raptly at a guy talking about a transistor.

Education

Geselowitz:

That was your interest. What was your formal education? Did high school help or hinder your in your interests?

Rosen:

It was neutral. Revere High School was not Boston Latin School. I guess I did okay in algebra and geometry, but I did equally well in German. I wrote well and was very active in the Drama Club. I was a thespian. I did that a lot, which makes me different from most engineers. I don't remember what I did technically when I was in high school. I probably fixed bicycles. I wanted to become a ham radio operator but could not because it was prewar or wartime. Because I was living at home and my father had been an immigrant, I could not get a license.

Geselowitz:

From where was your father an immigrant?

Rosen:

Russia. He was naturalized by virtue of his father having come here and been naturalized. His father died here. His mother refused to come here because the heathen were here. He became a naturalized citizen, but because he was naturalized and because we (the US) were a tiny bit less civilized than we are now, I could not get a license, so I did not do it. I built a radio. Then I went to Tufts College. Tufts College cost $300 and MIT cost $600. My father was a very prudent man and we elected for me to attend Tufts. That decision was largely based on naiveté, because I did not know about scholarships. I did not know about borrowing money to go to school. It was cash on the table. Therefore I went to Tufts. It was a lousy decision. It was during the war. I started there around 1940. They had a tiny engineering school. Their professors had all been inducted into the military. The guy who was the head of the department was a good guy. He had worked at the Rad Lab and taught Maxwell's equations and the higher end stuff of the time. However at that time at Tufts, the engineering school in particular was sort of a place that did dirty-handed engineering. I learned how to survey land and make cast iron ingots and file them into a square shape. The school had a government-sponsored program at that time where a person could earn 40¢ an hour, so I worked in the machine shop making things for the school. I learned some machine work, how to be a draftsman and how to be an engineer's engineer – dirty fingernails. In my view the engineering courses themselves were sort of incoherent for the most part. First of all, there was no sense of cohesion between where to start, what it (a given subject) was for, where it was going and what would be the final result. I learned math from Professor Mergendahl the first year. I just loved the guy and I loved the kind of math he taught.

Geselowitz:

How is that spelled?

Rosen:

M-e-r-g-e-n-d-a-h-l. Nice man. He had a sense of humor. In the springtime when the windows were open and the students was undoubtedly thinking of their dates for that night. Mergendahl would dance around the room singing, "I wonder who's kissing her now?" That would get people's attention. He taught me some math. The next semester I had Dr. Otto Joseph Michael Smith, a newly minted PhD, visit me. He taught integral calculus. He was also my advisor. Mid-semester he called me in and said, "You know, you really should leave the engineering school. I see you are very good at English and some of these other things, but you will never make a real engineer." I suspect he did not even know what the heck an engineer did, because he said, "A guy like you will end up as an oiler somewhere oiling machinery. Go to the liberal arts school." I did not. And I flunked his course. This was a disaster for me. You have to understand I was the eldest child of a totally ignorant immigrant family. I went to college, and colleges were better to go to than synagogues.

Geselowitz:

They were sacred temples.

Rosen:

It was a sacred temple, and I had violated it somehow. There was a guy named Bob Isaacs in my class, a smart guy, and I asked Bob to tutor me because I could take a makeup exam. He tutored me and I got an A in the makeup exam. I got an F for the class but I did not have to repeat it. I was branded internally. That sort of set up my view of Tufts College. I hated it. I did not have the brains to go see the head of the department, who was a perfectly good guy, to whom I never told this story until I ran into him ten years ago. I made my way through school and did all right on the math. I ended up with sort of a B-minus net average. I learned Maxwell's equations and all that, though I did not know what to do with it. I also learned static mechanics and chemistry. They were okay. I was launched with a bachelor's degree from Tufts College. I really had a lousy education in general, although the head of the department thought I was a good student.

Naval service

Geselowitz:

You graduated in 1944 and the war had not yet ended.

Rosen:

I graduated in 1944. Two things happened. I signed up because I did not wish to be drafted. I wanted to finish college, so I signed up with a Navy program called the V7 program. You may have heard of it. It was a 90-day wonder program. When you got out of school you were sent to midshipmen school. Three months later you would come out as a bona fide officer. This was toward the end of the war and they decided they needed quite as many engineering officers, so they did away with the V7 program and made it into another program, V12. However instead of becoming an officer, you were transformed into Seaman First Class while you were in school. I lived my senior year at the school doing pushups, running in the morning, vomiting because I had run too and far too fast. I learned to scale walls and all of that. I was graduated one semester early, in January of 1944. I don't think I want to talk too much about my Navy career except that it was bizarre. There were so many midshipmen being produced at that point in time that we weren't sent to a midshipmen school but to a pre-midshipmen school in a hotel in Asbury Park, New Jersey. They stuffed all six or twelve floors with swabbies, ten guys to a room. If you were lucky you ate once a day. When you did eat you had “horse dung” (Spam) and beans. It was terrible. It was a storage place. I got out of there at the end of a month and was sent on to midshipmen school in Notre Dame. I went all the way through that. I had a physical just before graduation and had an 18/20 eye on my eye exam so I could not be a deck officer and they did not need engineering officers, so they asked me whether I wanted to do defusing of unspent weapons and I said no.

They said, "Okay. In that case we cannot give you a commission. We are going to make you a Seaman First Class," and they did. They did not have any uniforms there for Seaman First Class at Notre Dame, but they had uniforms left over from the First World War ROTC. I was expecting to go home and get married with my ensign's uniform and my shoulder bar and all that, but they set me up with a hat from the First World War with a brim about four inches high. They also set me up with a pair of pants whose crotch was down around my knees and the jumper did not quite cover the pants. Then they handed me my orders telling me to report to Little Creek, Virginia, for service as a swabbie. I went home, got married and went to Little Creek.

There they outfitted me with a current uniform. I also discovered that the set of people to whom I had been assigned were illiterate. Our learning task was to fill a net with rocks near a ship mockup. Then a guy on a fake ship running the winches would take the rocks and dump them into the hold. Then there would be another set of idiots like me at the bottom unloading the rocks into the hull of the ship practicing to load and unload. There was a guy named Blue, a bosun's mate who would come out every day with his hat cocked over his head and his thumbs in his belt carrying a 45 caliber pistol and say, "All right you people, line up." We'd line up and march off and go load the ships. It turned out they were training us to ride on the backs of Liberty ships. I doubt that you are familiar with a Liberty ship. These ships, eggshecks, were cranked out by the thousands. It had a 3-inch gun on the aft end for defense, and it had a Navy crew to man the guns and load and unload the ship if it ever got where it was going. That was what I was there to do. That seemed unreasonable to me. I asked the commanding ensign, an Irishman from South Boston, "Could I go see the personnel officer?" He gave permission.

The personnel officer looked at my record and said, "Well, you're here." I said, "Who do you send here?" He said, "Illiterates." Because I had never taken the general classification test and I was a Seaman First Class it was automatically recorded that I was an illiterate. That was why I was there. I told the guy that I had a degree, so he said, "I'll give you the general classification test" and he did. He looked at the results and said, "This is amazing. I have never seen a score this high." At that time they had an Eddy test – after a Captain Eddy – that tested to see whether one had an aptitude for electronics. Then I started down this other track of getting to be an electronic technician. That went through a large number of perturbations with my wife following me to Mississippi and getting pregnant and going back to Boston. Then I found myself at the electronic training school at Great Lakes Naval Training Station in Chicago. I was learning resistors and capacitors and how the capacitors charge, things like that.

Geselowitz:

Some of which you already knew.

Rosen:

Of course. Anyway, that is what I was learning. And E = IR. Then I got scarlet fever and went to the hospital. As a result I got mastoiditis in my left ear. They could not decide what to do with me. They talked about operating and not operating. Penicillin had just come out, so they injected me with penicillin. I went around and injected other people with penicillin. They did not operate, but while in the hospital I got orders to report for a physical examination because they needed officers again. I reported for a physical while in the hospital with mastoiditis (a severe infection of the bone back of the ear) and they tested my 18/20 eye. I had not been exercising it very much, so it was 20/20. Therefore I qualified as a deck officer. They also tested my hearing. I was deaf as a post because I had mastoiditis. At that time the tests were not what you might call sophisticated. A corpsman would stand at one end of the room and he'd say a number and then repeat it. He'd say "one" and I'd say, "What?" "One." "What?" "One." Loudly "Oh, one." "Good." "20/20. You're done." It is probably more than you want to know, but it is sort of an interesting story.

I went back to learning electronics, and it was not more than about three days before the Exec [executive officer] called me up and said, "I don't know what the heck you did, but you are going to get out of here today. You are going overto Northwestern to midshipmen school." "Aye-aye, sir." I packed myself up and went to Northwestern. It was a Friday. I was going to go through the midshipmen school business again, but this time I was going to be a real deck officer. It had been almost a year since I had learned how to lay out courses and look at silhouettes of ships and so on. Every month in midshipmen school they gave a test. If you passed the test you went on to the next month; if you did not pass the test, you were out and you were a Seaman First Class. The very next day (Saturday) they were going to give a test. I had wonderful shipmates. I don't know who they were, but they were terrific guys. They took a blanket and hung it up on the door so the light would not show through – because we had an ensign, Jackson, who walked up and down looking to see who was not asleep. They tutored me and I passed the test the next morning.

Once more I became an officer – a gentleman now. However, I never went into the engineering business. The reason was that in my record there was this item that I had been at Little Creek, Virginia, learning about loading ships. Therefore I was a ship loading guy, so they sent me – where else? – back to Little Creek. People there said, "Are we glad to see you. We're leaving for the west coast tomorrow. You are joined up with logistics support company number 250." These were guys who loaded and unloaded ships. There were also guys who went ashore after the Seabees and brought supplies ashore. There were three white officers, one black officer and everybody else was black. I got a physical and a dental examination and the dentist said, "You are going to go overseas. If you still have your wisdom teeth they are going to pull them out." And they did. All four of them. Then we got on a cattle car and went to the west coast. Ultimately I found my way to Guam, where in the fullness of time, since I was the junior officer initially, I became the commanding officer. I had been to Naval Justice School so I could administer Naval Justice as becomes an officer and a gentleman. I ran this company of 250 guys loading and unloading ships. The best part about it was that we had a lot of beer and my last officer colleague and I, an Italian from New York, had built a wooden box so we could go to the Officers Club and get ice shavings, put them into the box, and then put the beer in it in the morning. The officially available water, stored in a wooden cistern in the sun was like pee, warmth and all. When you drank that beer at 3 o'clock in the afternoon after working on the docks, there was nothing like it, not even sex.

Employment at the Air Force Cambridge Research Laboratory (AFCRL)

Geselowitz:

At this point the war was still going on and presumably Guam was a transit point to send supplies to the islands?

Rosen:

No, the war was already over but they were sending people through and we were supporting that. All of my cohorts who somehow got their electronics training became radar officers and came out prepared to go to work. I came out having loaded and unloaded ships for a long time and not having looked at diagrams or soldering irons or anything else for a long time. I got home and started to look for work.

One of the summer jobs I had in college was working for the National Radio Company. They are now extinct, but they built wonderful radios. I worked there one summer and had a ball.

When I went overseas I had an RME-69 receiver that I bought in California and took with me. That went all over the Pacific with me. Anyway, I started to look for a job. I was married and we were living with my mother-in-law, which wasnot the greatest thing for either of us. The Air Force had a place in Rome, New York, called the Air Force Research Laboratory. I don't know what its function was. It is still there, is it not? They had opened a branch – and I don't know why – close to MIT called the Air Force Cambridge Research Laboratory (AFCRL). I found out about it and went there and applied for a job. A guy gave me a test and asked me to solve a differential equation for a capacitor charging through a resistor. I did it and got a job. That is how I got to be bona fide engineer. I had a really good guy for a boss there for a short time. His name was Bob Wooley. This place did not have a mission so far as I could tell. The particular thing that we seemed to be doing was building a gun-directing analog computer – only we did not have the pieces from which to do it. Therefore I Analog circuits of some sort, I don't remember now. Wooley was a nice man and a very good teacher.

Geselowitz:

How is his name spelled?

Rosen:

Robert W-o-o-l-e-y. I'm not sure but I think he had been at GE during the war. He was a good circuit designer and he was my boss. He would sit behind the glass partition elevated a little bit from the rest of us peons – there must have been about four of us – and he would sit and gaze out at us. I interpreted, in my paranoid way, his gazing at me as wondering just what the heck I was doing. Finally, I could not stand it anymore and I spoke to him about it. He said, "Look, Rosen. If I have something to say to you, I'll say it. If I'm just looking, I'm looking. So knock it off. Go to work." That was very good advice. I worked there for a while. There is just one thing that really stands out in my mind.

Melvin Baller ran this thing. He had come from Rome and he was the supervisor. He was Wooley's boss as well. For reasons I don't remember, since they were building an analog computer, they needed a gear reduction box. Whether they did not have the money to take it to a machine shop or what, I undertook to make it having had all these 20 or 30 hours of experience with in a machine shop at Tufts at 40 cents an hour.

I got some gears and shafts and worked on it, but it was clumsy. It was a big box. Baller said to me, "Did you say you worked in a machine shop or a shoemaker shop?" I said, "A machine shop, sir." That sort of struck home. I decided I did not like this guy, so I was going to go somewhere else.

Employment at Raytheon Manufacturing Company; return to AFCRL

Rosen:

I had a friend named George Freedman who was a metallurgist. He had the sense to go to MIT. George was working at Raytheon Manufacturing Company. At that time Raytheon was making all sorts of vacuum tubes in large vacuum tube factory. It was a major producer in the United States on Chapel Street in Newton. They had a homebred genius there named Paul Stutsman. Stutsman's claim to fame was that he had invented the OZ4. OZ4 was a major contribution to the automotive industry because until that time they had used mechanical vibrators as the device to make and break the field in a primary of a high-voltage transformer to supply the high voltages required for the vacuum tubes. The OZ4 was the thing that had a gas that had a very low breakdown voltage. It would break down, causing a pulse in the transformer primary, recover and then the voltage across the OZ4 would rise again to repeat the breakdown cycle. The transformer secondary high voltage winding would then feed a conventional vacuum tube rectifier producing high DC voltage for the tubes in the auto radio. The OZ4 was a big moneymaker.

I went to work for this guy. As it turned out, being an electronics engineer in a vacuum tube manufacturing place does not offer a lot of challenge. What I did for him was design test equipment for new gaseous discharge tubes he was trying. The germanium transistor had just come out. It was a rectifier, and he was interested in what could possibly be done with it as something to satisfy his business. He also had some weird ideas about experiments. I don't know what motivated them, but he had me doing voltage current curves with on germanium transistors and trying things in various ways that really did not make a lot of sense to me. I decided that this was not getting me anywhere either, though I learned a lot about making racks for testing OZ4s and other things. Therefore I went to look for a job again. I went back to Air Force Cambridge Research Laboratory, which was under new management at that point. Jack [John V.] Harrington, and some other people were attempting to build processors to look at radar data and transform that analog data into digits. Simultaneously at MIT there were a bunch of guys working on the Whirlwind computer. You should excuse the expression.

Geselowitz:

Why do you say that?

Rosen:

Because it was hardly a whirlwind. You probably don't know vacuum tubes very well. The computer used vacuum tubes with high-transconductance/high-current capability. The reason they wanted high-transductance/high-current was because a capacitor could be charged and discharged quickly in a flip-flop when fed by such tubes. They had two 6AG7s per flip-flop, each gate/flip-flop mounted on a 3½" x 17” panel. One cannot make a very big computer usingsuch large, power hungry devices. They used storage tubes for memory. These were devices where the beam would essentially impinge on a capacitive surface. The impinging beam would leave a charge, and then the next beam would sweep across the array of charged dots. When it did one could sense that the charge had been stored indicating a digital “one” and an output pulse would be obtained. That was the storage system/system memory.

Semi Automatic Ground Environment (SAGE) system and Pulse Position Indicators (PPI)

Rosen:

The United States' relationship with Russia at that time was not all that wonderful, and there was some concern that there might be an attack on the U.S. with Russian bombers flying over the Pole. Therefore a study was done, and that study led to the conclusion that we ought to try to build a defense system against bombers. The study members looked around for who was going to do this.

Some of the participants in the original study in Washington had been MIT people. They decided they were going to launch an experiment where one would in fact try to process the radar data at a number of radar sites, long-range for distant detection and little radars to fill in underneath; and combine all of that data in a computer. Then that computer-processed data would be made available as vectors to dispatch interceptors. That was the beginning of the SAGE system [Semi Automatic Ground Environment].

I did not go to work on the computer end of the system, but for Jack Harrington, who working on ground environment equipment including radar data processors and communications techniques for sending the processed computer data to the distant computers.

I did a lot of things there. For instance I worked on PPIs – rotating Pulse Position Indicators. For a while all I did was work on digital circuits. I don't really remember what they were for, but I learned to design flip-flops, multivibrators, pulse generators and all that kind of stuff, and I loved it. I just had a feel for what to do. When this SAGE System program came up we were conjoined with the MIT computer people.

Cold War research at MIT's Radiation Lab

Rosen:

Then Harrington's group (including me) were moved from the Air Force Cambridge Research Laboratory to one of the old Radiation Laboratory buildings at MIT.

Geselowitz:

Was that in Building 20?

Rosen:

Yes. We were still AFCRL employees but we worked over there. George Valley, a professor of physics, became our division head. He had been one of the Rad Lab guys. There also was a man named Jay Forrester, of whom you may have heard.

Geselowitz:

Yes. He is well known to us.

Rosen:

I think he ultimately formed Forrester Associates, the consulting people, but I'm not certain. Jay Forrester was the guy who headed the MIT computer set. These two organizations, the external environment people and the computer people respected one another technically, but for cultural reasons but they really worked in two different domains. The computer people worried about things like how to build big, fast memory and how to program the computer resources needed to solve the computer environment problems. Whereas the group of people with whom I worked worried about how to determine from where and at what angle a raw radar signal was coming, the location of the target and whether or not it was a real target as opposed to noise. People – not me – had worked out some probabilistic notions about the likelihood of seeing a target at a given range given the noise environment, and then making a machine to make that determination, and then coding those coordinates onto a telephone line to the computer. I worked on a couple of aspects of that system. I worked on the components of the radar data processors. These were largely flip-flops arranged in some logical way to do these probabilistic calculations. I also worked on large-scale memory systems for these machines and on the modulation systems for transmitting the processed radar data to the computers.

Analog-to-digital conversion; Irvin Reed’s network of 1N34 diodes.

Geselowitz:

Was it analog-to-digital conversion and early digital processing?

Rosen:

Yes, exactly. One of the earliest things we did was to design an analog-to-digital converter for converting the analog angle of rotation of the antenna to a digital number. Irving Reed is well known. You must know him well.

Geselowitz:

I know the name.

Rosen:

Irving designed a network of 1N34 diodes, which took the voltage proportional to the angle of the antenna as it rotated, and made a digital number out of the analog voltage. Reed had designed a matrix on a board with 100 or 200 1N34 diodes soldered into some kind of mathematical array such that when the signal came down from the antenna servo, it impinged on these diodes and the diodes would generate a number representing the direction of the antenna. We also had a rudimentary digital processor that did some of the statistics. I don't remember what we used for memory at that time. It was my job to have all that digital stuff working.

The problem was that 1N34 diodes are notorious for their temperature sensitivity and we were stuck doing this experiment at big radar in Truro on the Cape [Cod] and we were given space next to the radar in a concrete block building. It was clear that the temperatures were too darn high. I had an Italian technician working with me named Dick Tringale. He was a gung-ho guy. I instructed him to go find a sledgehammer and he broke a large hole in the wall of the building. We had fan and stuck it front of it and cooled the diodes. Needless to say the guy who was in charge of the base at the time was very unhappy, but it worked. That was just an experiment all by itself, a processor and radar with data coming off of it being recorded in some way. I don't know what the recordings were like.

Modem design

Rosen:

Ultimately we built a machine whose input was a digital signal that said, "This is an airplane and these are its coordinates." Next the question was how you get that information from A to B. I designed a modem with which to do that. Because I had not had a very intense education in the mathematics of modulation I did not, at the time, do an optimum modem. I designed a modem that depended on amplitude alone, a technique that did not phase information as well.

However I had made lots of tests on telephone lines. These were pretty lousy lines. Some of them ran on posts set in the ground used to keep out cattle. Therefore the telephone lines had propagation characteristics that were very different from one another. There were two things one really could not predict – the bandwidth and the phase response. I discovered, simply by messing around, that if I arbitrarily picked something like 2 kilohertz as the upper end of what a reasonable telephone line would do, and put a simple low-pass filter on the modem output, the phase of the filter would roll off gently before the signal encountered the severe phase distortion caused by the telephone line itself. The result was that this line could be pinged and a fairly representative pulse would be obtained at the end. The detector I used was a very simple amplitude detector, a diode and a capacitor. Of course there was amplification in between and there was digital stuff that somehow synchronized with what was coming off the radar.

At the time Bell Labs had great men working on this, and they said if one drives a telephone line over 600 bits per second, one has arrived at the end of the flat earth, and you've had it. With my simple device I did 1800 bits per second consistently. That was a big deal, because they were going to install thousands of miles of transmission. They (the Bell Labs people) adopted my modem – initially. Jack Harrington and I got a patent on it. I did the grunt work and Jack did the analysis post facto that it would work. One of my colleagues said, "Oh, you're going to get rich on this." We had an incompetent attorney at MIT. You have been with academics. I think you may know that some of the administrative people associated with academic institutions believe that they have tenure until ten years after they die. This attorney was terrible. The problem was that he let me write a patent application that was much too constrained.

Geselowitz:

Too narrow?

Rosen:

Much too narrow.

Geselowitz:

You could have made much broader claims based on what you had done?

Rosen:

Much, much broader claims. The result was that when this system ultimately went into the field indeed my design was the one that Western Electric (under Bell Labs aegis) built for its early stuff. There was an Air Force contracts officer who thought that Harrington and I should have gotten something out of it. We got point-triple-O nothing because no one would buy the patent and it belonged to the government. The government gave it to Western Electric and that was the end of it.

In the meantime, having worked at these sites first on the components that made up these machines, and second on these machines and third on installing them at sites we began to make a system. I became the leader of the group that built and installed these pieces of equipment at the various sites.

SAGE system and radars

Geselowitz:

Is this a digital data network?

Rosen:

This is a digital data network fed by processed radar data, with the digits going to the computer on telephone lines.

Geselowitz:

Was it still strictly for the purpose of radar or was it now expanded to a more general purpose?

Rosen:No. The SAGE System was in concept a very simple system. It had radars distributed along the perimeter from which attacks were expected, basically of along the northern and eastern perimeter of the United States. There were already fairly good-sized radars but they had the defect of looking far out, but seeing sea clutter, and blind to the space under the beam through which an enemy plane could fly. There were also smaller radars that already existed called gap fillers whose function it was to fill the gaps under the larger radar beams. Thus there were two classes of digital data processors – one for the big radars and one for the small radars. The manufacturer that got the contract for building these experimental radar data processors, the smaller stuff, was a vacuum cleaner manufacturing company, Lewyt. They built them but they were still experimental devices and they were exact replicas of what we had built in the laboratory.

Magnetic drum design; application in IBM computers

Rosen:

We did something that was unique for the time I think. At that time memory for computers was punched cards and some tape. Obviously both of these were too slow for the real time needs of the radar data processors. I am not talking about memories used for computers but memories used in the kind of application we had. The guys I worked for, Harrington and Company, decided we ought to try a magnetic drum. It was my job to design that drum, and so I did. We bought it from the Burroughs Manufacturing Company. They made a drum that had forty heads on it. The heads were adjusted in proximity to the magnetic surface using an Allen wrench, which expanded the end of the head and brought it closer and closer to the frangible magnetic drum surface. It was a nerve-wracking job and it was my job. How could you tell if the recorded signal was big enough? You hung an oscilloscope at the end of the head amplifier and ooched away at the Allen wrench until the signal looked big enough to process.

The guys working for me knew how intense I was about this. The speed control for this drum was very simple. It was a Variac. One morning they put an electric drill in series with the Variac and put the electric drill in the wastebasket, so when I turned on that drum, the drill started in the tin wastebasket. You can imagine the racket. It was a good joke. I could have killed them of course. But in fact we sort of did the prototypical design for the use of rotating drums in computer systems at that time and I did most of it.

The drum had forty heads and each track was assigned to one stroke of the radar beam. The drum was used as a device to store the analog returns of forty successive sweeps of the radar that could be operated on statistically, and that is what the rest of the equipment did. I don't know what probabilistic computations were done, but it was my job to make it all play at a lot of sites. Later in the development game we also had an experimental radar operating at 400 MHz. It was a big radar with a 110-foot antenna running on trolley tracks on the top of a tower in Bath, Maine. It was my job to make that work. I did not design the tower and incurred the everlasting dislike if not hatred of the people from the mechanical engineering division because I kept telling them they could not seem to get that darned antenna to run in a plane. They sort of ultimately did. That was very interesting, because a 400-MHz radar antenna has an entirely different sea-ground reflection in the higher frequency, and it could see a lot further. In addition to the experimental site in Bath, my recollection is that we had on the order of four or five large sites and about fifteen little ones and I was in charge of seeing to it that it all played. Of course there were a lot of people engaged in installing equipment at these sites but there were times when I did troubleshooting because I had worked on most of the experimental site equipment.

In the fullness of time we then went around and interviewed companies as prospects to build this equipment in large numbers for when another manufacturer would come along. IBM was one of the outfits we visited, and both we and the computer people concluded that IBM should build the computers. That is what launched the IBM Big Blue computer business. Those computers were the first ones that IBM built.

SAGE at Lincoln Laboratory

Geselowitz:

To be clear to the listener, because I think we glided over that, when they moved the Air Force group over to MIT they affiliated with Lincoln Laboratories.

Rosen:

It became the Lincoln Laboratory.

Geselowitz:

It became the Lincoln Laboratory which became MIT's sort of military research arm which later spun off as quasi-separate with its own board and so forth. That was much later, when the students complained about military work being done there.

Rosen:

Yes. The students complained. And they particularly did that rather late in the game.

Geselowitz:

Right. I'm talking about after you were there. I'm talking about in the late '60s and in the '70s.

Rosen:

I felt like a turkey, because here I was working my butt off and I thought doing good work, and these guys would come around and accuse me of being a warmonger. I had a son who hated being an engineer because I was an engineer and so on. All this SAGE activity transpired between something like 1950 and 1958.

During that period we did all of the basic architectural design and certainly laid the foundations for a lot of the hardware that was to follow. A flip-flop is a flip-flop is a flip-flop. It may look different, but it does the same thing. And AT&T was picked as the major contractor with Western Electric as the installing arm of the production SAGE System, They went off and they did that At the point in time when it was decided that there was going to be a production system the MITRE organization was formed. Forrester and his people then moved out and became the MITRE Corporation. They did very, very good work, wonderful work on computers. Forrester and his guys devised the core memories and the guys who did that then went off and formed DEC using the PDP something-or-other that had been built at the Lincoln Laboratories. That was a spin-off. Those guys were very good. But MITRE Corporation then went off to be a non-hardware thing and they still exist.

Geselowitz:

Sort of as a think tank.

Rosen:

Yes, as a think tank ultimately in the Washington area. Lincoln Laboratory on the other hand remained a hardware outfit. In the meantime, although the initial impetus for forming it was the SAGE project, other branches of the military electronic arts moved into it – communications, solid-state physics and radar. The radar work ultimately elided into the ballistic missile area looking stuff and the solid-state impinged on many areas. I'm not really competent to comment on it except that they did a lot of stuff like solid-state lasers for instance – some very, very useful devices including components for designing low-noise receivers.

Space communications, speech processing, and spread-spectrum communications work

Rosen:

I was out of a job when SAGE moved out. I was invited to go to MITRE, but my societal leanings were not with that set of fellows. I was not a computer guy. I did not know how to add one and one. I also knew they were tending to be think-tankish, and I did not want to do that. There was then this question about what the heck we were going to do. It was decided that space communications was the next thing.

Geselowitz:

What year was this?

Rosen:

It was 1960.

Geselowitz:

It was not that far after Sputnik in other words.

Rosen:

I think it was stimulated by Sputnik. I was so low on the totem pole that I only heard whispers about Sputnik. It was decided we were going to go into the space communications business, but it was not clear who was going to do what. There were a lot of genuine communications people, among them Walter Morrow, who ultimately became the director of the laboratory.

William H. Radford, an MIT professor, who was the head of the communications division and ultimately became a director. The reason Walter sticks in my mind is that it was his initial notion on using the so-called Needles experiment that gave us the motivation for doing some real experiments on satellites and satellite communications. I did not talk enough about George Valley; an MIT professor who had worked at the Rad Lab. Valley inspired me. Working in Building 20 George would come around at 8 o'clock at night. He was then in charge of the ground environment – not the computers, but all the rest. Valley was on leave as a physics professor and was our boss. George would come around at 8 o'clock at night. He would wear a green sweater, a pair of gray pants and a jacket. I had a colleague who joked that that uniform had a zipper down the back. He would climb into it like a teddy bear thing and zip it up. That was George. He was a very inspiring man because he coupled with the troops. He stuck with the troops. And for me, professors were godlike. I liked him. He was a good man of integrity but I suspect he was not a very good politician. He ultimately became associate director of the laboratory. I don't think he ever became director. He was kissed off by being sent to be secretary of the Air Force for a while. Then he came back, did his thing at MIT but he never came back to the laboratory. I got a note from him many years later including a published review of what had been going on at the Lincoln Laboratory. His note said, "Paul, those were the wonderful days." They were. I did not get to see much of George because he was the director and I was a sort of a peon, but he was a very, very good guy. However his sense of politics and dealing with the Air Force and the MIT administration was perhaps not so sharp. I thought he got a raw deal.

Oh, and this is important. With Radford and others from the Rad Lab there came a bunch of people who had been involved in communications and they were the guys who I think invented spread-spectrum communications developed during the Berlin blockade to circumvent jamming. Bill Davenport, Paul Green and Bob Price were some of the communicators. I am not sure whether Ben Gold was with that group at that time. I think he was. When they decided to put together a ground equipment satellite group I was anointed the head. I really was not qualified technically to do that. I did not know much about communications. I knew nothing about spread-spectrum communications. And some of these guys had a university logo on their heads. Working steady hours between 9 and 5 was not what they did. They tended to work when they wanted to work, and that mode was not my thing.

I also inherited a bunch of mathematicians, some of them wonderful guys – Irving Reed, Gus Solomon of the Reed-Solomon codes, and guy named Weiss. Not Herb Weiss. Herb Weiss was a radar guy. It was another Weiss who was a mathematician. I inherited all of these guys and I really did not know what to do with them, and they did not know what the heck to do with me either. However I had enough of a sense to know that there was a lot of promise in this spread spectrum stuff. At the time I was interested in looking at the spread-spectrum systems that would be easier to synchronize on simple vehicles using simple equipment. Therefore I sort of fell in love with frequency hopping as a spectrum spreading technique. My personal technical analytic ability to demonstrate that fact was close to zero, but my hardware sense about what could be done on a moving vehicle was good. I also knew that conservation of spectrum bandwidth and power was very important, given the constraints imposed by satellites. But people wanted to talk. They did not like to use code. And speech cost bandwidth and power.

Therefore I really thought that speech processing was a thing on which we ought to work. A big deal at that time was vocoding. I'm wandering a little bit. It was all a mess of trying to decide what the heck this group was going to do. Ben Gold, whom you've interviewed, I understand said, "Paul Rosen hollered at me." I probably did, because Ben liked to walk in at 10 o'clock with his tennis racket swung over his shoulder. Ben was a very interesting guy. You must have talked to him.

Geselowitz:

I was not the one who interviewed him, but we have a long interview with him.

Rosen:

Ben was a very, very bright man. He is still working by the way.

Geselowitz:

Right.

Rosen:

Ben was a very bright guy with special ideas about what great technical people ought to be like. He has a strong artistic bent and he had just come back from a year in Italy painting. Getting a grubby-fingered person like me and Ben to work together somehow was a challenge. Anyway, I said, "Ben, we're going to work on speech." There was also a guy named Charlie Rader.

Geselowitz:

Whom we have also interviewed.

Rosen:

You have interviewed Charlie. Charlie was his confidante and protégé. Charlie was younger. Those two guys were both my charges. They worked on vocoders. My boss at the time, Gerald P. Dineen, had been in Jack Harrington's group when we were working on SAGE. Dineen was one of the guys who did some of the mathematics on the probabilities of what was required to declare something a real radar hit. He was a mathematician.

Geselowitz:

Yes.

Rosen:

I started to talk about spread-spectrum communications. Using the moon as a reflective target made it a very good satellite for communications. The problem with the Moon is that it is a rough surface. Therefore if you pulse it, you receive a time-dispersed signal. If you use a known which is sufficiently complex and then operate on that signal by certain integrative processes (sort of a definition of spread spectrum signaling) you end up with something that looks like the original pulse. I don't know who developed the actual signal design. I think Green was involved in that. They designed a modulation system to do some communications using the Moon and we got our feet wet. I did not have anything to do with that except that some of the people then assigned to my group, for whom I did not have any particular job, worked on that modulation system. Then Walter Morrow, a very bright man, thought of a satellite system that would be peculiarly resistant to jamming with a thing called Needles which comprised of millions of dipoles in orbit around the Earth. It could not be jammed because it was such a dispersed medium. The laboratory as a whole devised a way of winding wire and then coating it, casing it and cutting it in such a way that when this spinning device went into space the stuff that was coated dispersed and a bunch of dipoles formed a ring around the Earth. That was a considerable accomplishment. I was offered the job of running that. I did not do it.

Geselowitz:

You were saying that, even though you tried not to take that management position, you did get involved.

Rosen:

Yes. Some of the people who worked for me worked on the demodulation equipment and the antennas and that stuff. It was a clever and well-executed program. Do you know about Westford?

Geselowitz:

Yes.

Rosen:

The Needles project used the Westford antenna, a huge 84-foot thing that had to be used in order to see anything since the cross section of these needles was very small. Nonetheless it got our feet wet in the space business. We began to build a space laboratory and ground equipment. Now it is at about this point that Bill Ward became involved. Bill was a radar guy. He had a doctorate from the University of Texas. He was a modest, mild, very smart man. He became involved in the big antenna measurements business for communications purposes, and he was involved in the initial communications experiments with Needles. I was responsible for seeing to it that that equipment got together. However Needles was not really very practical. Its useful bandwidth was very, very small; the modulation system was very complex; and its cross section was tiny. While it would be a suitable vehicle for very important strategic communications like saying, "Go. Now is the time," it was not much good for anything else. But we got our feet wet.There then ensued the experimental design and construction of a series of satellites and terminals. The satellites were designated LES-1 (Lincoln Experimental Satellite 1), LES-2, LES-3 and so on. We also had experimental terminals, the first of which was LET-1(Lincoln Experimental Satellite 10). I had a lot to do with that, because I was interested in digital processing for communications.

Lincoln Experimental Satellite and Lincoln Explain Terminal

Rosen:

I want to mention a very important people component of my work as leader of the satellite terminal program at the Laboratory. The Lab and MIT had established a sort of exchange program where MIT professors were invited to work in some group or division at the Lab for a year or so. The professors were exposed to the problems of realizing some of their theoretical ideas and the Lab people benefited from the extraordinary exposure to leaders in various fields. I became acquainted and became friends with John M. (Jack) Wozencraft, who was a professor at the Institute and author (with Irwin Jacobs) of Principles of Communication Engineering. It is a core book. I also got to know Bob [Robert] Fano, also a professor at MIT, who was also in the communications and coding business. Both these men worked with me in the course of my work with the satellite terminal group. They were profoundly stimulating for me in setting goals and programs for our work, given my interest in the digital processing of data to be used for communications purposes. That interest drove me to push vocoding at the outset. When the possibility of fast Fourier transforms to be used as filters in speech processing showed up because Gold and some other guys thought this might be doable I thought that was a great idea and pushed hard to get money to do that.

En route, we built vocoders, and we put vocoders into one or two of the early terminals that we built. I was interested in frequency hopping – again, a digital way of communicating – so we designed antijamming frequency-hopping modulation systems. We squeezed all of this stuff into mobile terminals that we built, such as the LET-1. Ultimately the LET-1 operated at SHF because that is where the DSCS (Defense Satellite Communications System) operated. Let's look at the LES catalogs. You have it there.

Geselowitz:

Yes, I do have it. "Lincoln Laboratory's space communications program after Project Westford began in 1960 [unintelligible phrase] and demonstrated military space communications systems."

Rosen:

Westford was the Needles project.

Geselowitz:

Right. "The initial program objective was to build, launch and field a LES and a LET," meaning Lincoln Experimental Satellite and Lincoln Explain Terminal.

Rosen:

Right.

Geselowitz:

"They worked here as a system and demonstrated practical military satellite communications, using Project Westford's advanced super high frequency technology or SHF."

Rosen:

Right.

Geselowitz:

And it says here in this note in the 7 to 8 GHz range.

Rosen:

Right.

Geselowitz:

"Contributed to the decision to design LES and LET using that band."

Rosen:

Right. That was because the dipoles were that long.

Geselowitz:

LES-1 was actually launched from Cape Canaveral on 11 February 1965.

Rosen:

But notice that it operated at a fairly high frequency. That is why LET-1 was built at that frequency. Parenthetically, that was not a good frequency for platforms like airplanes which at that time could use only simple antennas like a dipole, implying a much lower frequency. It was an interesting terminal in that it had frequency hopping anti-jamming, vocoding and it was built air-transportable. That brings up a rather funny story. We could not just go out and buy an antenna and a mount, so we used a 5-inch gun mount as a pedestal. We put the gun mount on a tripod, and had an antenna built. I kept asking the people in the mechanical engineering division whether this thing was going to meet the weight standards for their part of the project. They said, "Absolutely." It's sort interesting. Because I own a summer place in Bath, Maine.

Geselowitz:

You must have become attracted to Maine when you were in Bath at the test facility.

Rosen:

No, as a matter of fact that is when I got distinctly un-attracted to Maine. When it was my job to look at that site I went there by flying to Maine from Boston or from Hanscom Field, getting out at the Naval Air Station, riding in a station wagon up to the site, spending a rather miserable day there – because nothing worked – and then going back to the Naval Air Station. My view of Maine was restricted to driving up and down Route 1.

Now particularly at that time Route 1 was not a beautiful place. For a number of years a guy who worked for me importuned me to come visit him there. He had a place there. I kept saying, "I don't like the sand," "I've been to Maine, I don't like it." I finally gave in. The fellow was working for me and I really could not be impolite anymore. It went well. About a mile and a half from the radar site this guy has a cottage on the New Meadows River. The sun was setting through the pine trees while I was sitting there with an ice cold martini in my hand, with a guy tooling a little sailboat back and forth about 50 yards out, and I just flipped.

The next day we bought land and we have been there ever since. That is how we got to Maine.

Now the big UHF radar antenna has long since gone off, but the tower is festooned in cellular phone towers and stuff like that. Anyhow, the place that was going to build this air transportable mount was a company called Bath Shipbuilding. Bath people are a wonderful people. I really truly mean that. I love going there. I feel at home there. But they have an ethos that is their own and it tends to be kind of an independent ethos. If it is time to go hunting deer they go hunting deer. And if that meathead from MIT wants this thing because there is a schedule – eh [dismissively].I had a hard time extracting this thing from them. The point that is germane however is this. Remember when I said I wanted to be engineer when I was little and I wanted to build bridges? These guys working on the mount for this terminal would know how to build bridges. It turned out that the tripod as designed was too heavy. What did they do? They burned holes in the legs of the tripod, but they were concerned that the tripod legs would be too weak. So they welded rings around the lightening holes. That made it heavier than before. I had one heck of a time getting an air-transportable terminal built. But I don’t think it was ever air transported.

The Army ultimately decided it would accept it – though rather ungraciously I thought. It was a very nice terminal and it would do good things for people who wanted to experiment with it, but the Army people at that time – I don't know about them now – were very conservative people. They did not really know much about satellite communications. Look at these guys running around in Iraq now. They know some satellite communications. They did not really care a lot. They just wanted to make sure I didn't mess with their turf too much. But we used it and all the techniques that we talked about worked. The vocoders worked, the jamming stuff worked, it met specs and so on.

Despite the fact that it was not used a lot, we demonstrated certain fundamental ideas. In particular, we advanced the antijam modulation business for mobile terminals because the modulation system was simple. And we pulled it off in part because of the help and encouragement I got from the MIT people. Jack Wozencraft was a friend and colleague. First he was my associate division head and then, when I left for a while, he became the division head. Bob Fano and I were also friends. Lovely, wonderful people. I have a photograph here. Maybe you would like to copy it. This photograph is of Professor J. M. Wozencraft, Paul Rosen and Claude Shannon standing in front of a sequential decoder. I am sure you are familiar with Shannon.

Geselowitz:

Of course.

Rosen:

The big deal at that time when I was the group leader was: "Great. Shannon has this terrific proof. How the heck do you do it?" Reed did it one way, Solomon did it another way, Reed-Solomon did it another way and there have been a thousand codes written. Wozencraft was one of the early inventors and he wrote a code called sequential decoding. We wanted to see whether this stuff worked, so we built a sequential decoder. And that's it in the photograph. We said you could probably build that…and we did. I was pleased and honored to see Shannon come out and talk and watch it perform. And it worked. We used a telephone line as the channel. Very proud of it.

In summary, I pushed. I cannot say it much better. I pushed the application of digital circuitry into the communications field to enhance the ability to communicate under difficult conditions – by coding and by any other kind of data processing including voice processing. And out of that ultimately the fast Fourier transform was designed and built as a real machine. Again, these things were not enthusiastically received by some of the laboratory management, and it cost me professionally in some ways. However the fact is that these things have endured. I feel that my people and I contributed very substantially to this whole area of the application of flip-flops, digits, to processing data to use it more efficiently and squeeze it into smaller and smaller spaces.

Leadership positions; reflections on achievements

Rosen:

I ultimately became the head of the whole communications division. I made decisions about satellites and terminals and who was going to do what. You'll find that in the records there. For a while, because the mechanical engineering division was rather inept, I took on the job of being its associate division head. There was a division head, but I ran the division. During that period I got some people from MIT to come into the division. One of the problems was that the guys working in this place were worried about the possibility of bringing in people who really knew how to use computers to do indeterminate structures for example. I transformed that division into a high quality structural engineering/mechanical engineering division and it built wonderful satellites. Part of that was done under my supervision – not under my technical direction, but on the basis of my guessing who was likely to do a good job and then watching very carefully to see what was happening and when. And I knew when I was being told news that was overly optimistic. We did some stuff which has not been publicized except tangentially in the history.

Geselowitz:

Is there anything in particular that you would like to mention for the record of which you are particularly proud?

Rosen:

Yes. There was a fellow named Leon Ricardi. You will find papers by him. If he has not been honored by the IEEE, he should have been. He was the foremost antenna designer in the world as far as I am concerned. With my help – not with my technical direction but with my help and my encouragement – he designed the multibeam antennas that ultimately ended up on the defense satellite communications systems. Those were a large advance because antennas could be selectively pointed in different directions without moving anything. Everything was saved electrically. Leon did that. Leon did the antennas on LES-6. LES-6 is a monument to really good high-quality engineering. It lasted for thirty years. I think it just died recently. It provided a foundation for military satellite communications that I believe endures to this day for mobile terminals. It was a UHF satellite. FLEETSAT, the first really useful satellite for multiple communications, built by the Navy, had technology that was based largely on that LES-6 satellite, built while I was the Division Head. The data processing used in it and in the terminals with its frequency, their frequency hopping, that was our frequency hopping propelled by me as an enthusiast.

Geselowitz:

Is there a laser technology in GPS right now?

Rosen:

No.

Geselowitz:

That is a different channel.

Rosen:

No, that is a different guy. All of our institutions tended to vie with one another. GPS was done by—

Geselowitz:

It came out of NavStar. Brad Parkinson was involved in it.

Rosen:

NavStar. I don't know.

Geselowitz:

The earlier Naval satellite.

Rosen:

Different. No, GPS is an Air Force thing. Ivan Getting, who just died, was the guy who pushed that as head of the Aerospace Corporation. I didn't have a low opinion of him but I was not very happy with him because he had a low opinion of the Lincoln Laboratory. At some study or other he and I got into a pissing contest about who was ahead of whom. Anyway, Getting is the guy who pushed this and he deserves a whole lot of credit. That GPS thing is another major milestone in man's use of space. We had nothing to do with that, unfortunately.

Geselowitz:

This is strictly military communication?

Rosen:

Yes, strictly military, but not a communications system but a navigation system. We did not do anything with that. I had a lot to do with Aerospace Corporation of which Getting was the head, because Aerospace Corporation was the outfit that helped the Air Force to propel Lincoln Laboratories' demonstration devices into production. Aerospace oversaw the production programs but what they oversaw was the elaboration of the model or design that we did.

Is there anything else I should say? This whole business of multibeam antennas and their use came from my division – encouraged by me and propelled by me but designed and executed by Ricardi. He died recently. I'm sorry. I loved that guy. I didn't mention EHF communications. One reason I did not mention it is that I do not know where EHF communications is in the security business these days. EHF was a very desirable communication with a very desirable frequency at which to work because one could get a very high density in a beam using a small aperture. It could be easily gotten on an airplane. A couple or even three of them could be installed on the deck of a ship. For all I know they are mounted on tanks now – I really don't know – but we pushed EHF communications in our satellites. The end of the paper in your hand overlaps my stay in Washington as the Defense Department satellite systems architect.

Geselowitz:

They say in their conclusion, "Concerning the large bandwidths that will be required, new systems will most likely use EHF at least for uplinks and downlinks."

Rosen:

Yes. Right. We demonstrated that. We demonstrated it on the satellites, on the ground and on vehicles. Beyond that I don't think I should say anymore.

IRE and IEEE

Geselowitz:

Okay. You have been very thorough. I have one final question that I need to ask because of my own professional affiliation. You mentioned the IEEE a couple of times tangentially. Were you a member of the IRE and did you have any involvements? A lot of people you mentioned did.

Rosen:

I never became very involved in the IRE or IEEE though I was a member. I never did. Green was, I know.

Geselowitz:

Why was that? I'm curious. I am guessing that when you were a manager at various levels in Lincoln Labs that people came to you all the time saying, "I'm not going to be at that meeting next week because I'm going to this IEEE or IRE conference" or meeting.

Rosen:

People achieve their sense of self by producing these papers. That is not how I produce my sense of self. I produced my sense of self by seeing the object, the thing, the person. I was deeply interested in people. There are people who will attest to that. There are other people who will say I was a bastard, and that is because I called a spade a spade. If you look at my jacket you will see I am wearing an IEEE Fellow's pin. I am a Fellow. How did I get to be a Fellow? General Gordon Gould was an Air Force General who was an MIT graduate. He was a very unusual Air Force General in that for practical purposes he wore a purple suit. That was very unusual for the Air Force. When I was at the Military Satellite Office (MSO) in the D.C. area my deputy was a guy who had worked for Gould when Gould ran Hanscom field. I had sort of a connection to Gould through this guy. Gould did not know me very well, and what he did stirred me very deeply. Gould decided that he wanted to nominate me as a Fellow because of what I had done for the military satellite communications business. He did nominate me, and I became a Fellow. I am very proud of that. That is my greater connection with the IEEE.

Geselowitz:

Great. And we're proud to have honored you. With that I am going to thank you very much for your time. I appreciate it greatly.