The Rohde Ahead
Ulrich L. Rohde studied electrical engineering and radio communications at the universities ofMunichand Darmstadt, Germany. He holds a Ph.D. in electrical engineering (1978) and a Sc.D. (hon.,1979) in radio communications, a Dr.-Ing (2004), and several honorary doctorates.He is President of Communications Consulting Corporation; Chairman of Synergy MicrowaveCorp., Paterson, New Jersey; and a partner of Rohde & Schwarz, Munich, Germany. Previously, he wasthe President of Compact Software, Inc., Paterson, New Jersey; and Business Area Director for RadioSystems of RCA, Government Systems Division, Camden, New Jersey. He is a Professor of microwavecircuit design and has held Visiting Professorships at several universities in the United States and Europe.Dr. Rohde holds several patents and has published more than 150 scientific papers in professionaljournals, contributed a chapter entitled “Oscillators and Frequency Synthesizers” to the Handbook ofMicrowave and Optical Components, 2nd Edition; contributed a chapter entitled “Frequency Synthesizers” to The Wiley Encyclopedia of Telecommunications, as well as six other books.
Dr. Rohde is a member of the following: Fellow Member of the IEEE, Invited Panel Member forthe FCC’s Spectrum Policy Task Force on Issues Related to the Commission’s Spectrum Policies, ETAKAPPA NU Honor Society, Executive Association of the Graduate School of Business-ColumbiaUniversity, New York, fellow of the Radio Club of America, and former Chairman of the Electrical and Computer Engineering Advisory Board at New Jersey Institute of Technology. Honorary Senator of theArmed Forces University of Germany in Munich.
MWJ: You have a rather famous last name in the RF/microwave and test equipment worlds. I understand you grew up in a technically inclined family. What was that like? Did you know you would be involved in technology from a very early age?
UR: I grew up in the middle of World War II in Munich and circumstances surrounding my first five years were terrible. While my father was working on radar monitoring receivers I wasn’t really getting too much insight in these technical things until I was about ten years old. At this age I bought an AM receiver based on a crystal with a point contact whisker and was fascinated with its reception capability and was listening to the various broadcast stations including American Forces Network which made it interesting for me to learn English. As I understood what products my father’s company Rohde & Schwarz made, I was particularly interested in monitoring receivers and spent endless time listening to the various broadcasts worldwide. And as soon as it became possible, my father and I got a ham radio license (mine Dj2LR and N1UL). Needless to say, I soon became curious about what was inside these radios and there was very little literature available. It was quite difficult to find a useful book on this topic. The books, LEHRBUCH DER HOCHFREQUENZTECHNIK (Textbook of RF Design), by Dr.-Ing. habil. Fritz Vilbig (1940), Volume 1 and 2 were totally frustrating because their mathematics exceeded my possibilities but got me even more interested. Both volumes are still in by possession. I had good knowledge in the radio engineering, from this book, with bad grades in school in Latin. Reviewing my grades today, I can see that my real interest was in radio engineering and not history or Latin or related topics. Later I built a super-regenerative receiver to listen to police which was both fascinating and illegal. I guess these things affected me for the rest of my life and I stayed with radio engineering at the universities and at the companies I worked for. The recent picture on the cover page of Microwave Journal (vol. 54, No.11, November 2011) that highlighted the 100 “Divine Innovators” in the advancement of engineering, science, and microwave technology, justifies my interest in RF/Microwave field from childhood.
MWJ: Do you think there is a new generation of young microwave enthusiasts that will carry on the work of older engineers and scientists?
UR: The new generation of microwave engineers is using different techniques than when I started. I came from the tube age. The modern hybrid IC technology and discrete semiconductor technology is much more flexible and dynamic. However, the transition from discrete circuits to integrated circuits is non-trivial. In this sense the new engineers will not be educated on some of the past technology but those driven by curiosity can certainly take a look back. There are very few cases left where electronic tubes have an advantage in modern technology. Base stations and cell phones are RF based, but they are partly implemented to work RF and digital. The problem today is that too many people learn about the digital part and don’t comprehend the pitfalls of a poorly designed front end. Here a well balanced education is needed.
MWJ: You recently received your Dr. Ing habil and became an honorary member of the Board of Regions at Cottbus University. I understand that Habilitation is the highest academic qualification a scholar can achieve in several European and Asian countries, requiring a professorial thesis based on independent scholarship, reviewed by and defended before an academic committee similar to a doctoral dissertation. What was your thesis about?
UR: My Dr.-Ing. Habil Thesis is a spin-off of trying to minimize the phase noise and real estate size of microwave oscillators by avoiding discrete inductors and partially based on my previous PhD. I proposed a gyrator based novel circuit which, together with some RF feedback circuits, provides low noise oscillators from fairly low frequencies up to the millimeter wave frequencies. The design requires a rigorous understanding of DC biasing, impedance matching and noise feedback to provide the best results. This topic has been brought up in literature for a long time and so far nobody was able to find the best low noise condition. These innovative ideas are referred to by engineers worldwide and definitely recognized as outstanding contribution by the microwave community.
MWJ: What role do you see for yourself in training tomorrow’s engineers?
UR: Together with some colleagues, I have published a variety of RF`& Microwave books including a recently published textbook in mathematics (two volumes: Introduction to Differential Calculus: Systematic Studies with Engineering Applications for Beginners, John Wiley & Sons Inc. ISBN:978-1-1181-1775-0, (784 pages), Jan 2012; Introduction to Integral Calculus: Systematic Studies with Engineering Applications for Beginners, John Wiley & Sons Inc. ISBN:978-1-1181-1776-7, (432 pages), Dec 2011). Both manuscripts are unique in nature and offer fundamental basis of the Differential and Integral Calculus for the conceptualization of engineering problems, generating intrinsic desire among students for learning calculus in simplified ways so that one can find unified approaches for the solution of radio and microwave engineering problems. I am still working with a few of co- authors on some microwave related books to show a good mix between analytical, mathematical results and measurement which gives the engineer a highly practical guide. I typically write books in such a way that I explain the various decisions which are taken along the way. In addition to this I have been publishing more than 100 on related articles in reputed international journals and conferences, serving as reviewer to more than dozen journals and supervising a variety of Ph.D. dissertations with challenging topics and very promising results.
MWJ: You have been involved in the study of phase noise - its characterization in measurements and simulations. Why so much interest in this particular area?
UR: The ability to receive small signals in the presence of large interference signals depends on the gain distribution of the front end, the properties of the mixer, and the local oscillators. The local oscillator is one of the key elements which people have been designing for years. It covers very low frequencies up to very high frequencies and of particular interest are crystal oscillators, SAW oscillators and resonator circuits with electromagnetic coupling. The better the local oscillator, the higher is the dynamic range of the receiver section as can be. In a radar system the lower the phase noise the further higher the range the system has. Our high performance signal processing components are close to the limits set by physics. In RADAR systems, low noise signal source is key technology to reduce influence by near-carrier echoes by clutters. Also in high-speed radio communication systems, highly accurate QAM/OFDM modulation can be achieved only with low noise local signal sources (oscillators). For low phase noise oscillator applications, noise dynamics place stringent conditions owing to inherent high noise figure and low dynamic range caused by the uncontrolled nonlinearity at large-signal drive-level conditions. In addition, these problems become critical at high frequencies when active devices (Bipolars/FETs) are technologically scaled to obtain higher operating frequency. In general, most device parameters are extracted from linear 50 WS-parameters and DC I/V (static and pulsed) data. The first constraint which is the ability to predict behavior under extreme nonlinear conditions or non-50 Wterminations, especially for autonomous circuits, is questionable.
The second constraint for low noise oscillator design is the quality factor (Q) of the oscillators, which is the most important design parameter that determines the phase noise performance as defined by Leeson phase noise model. It is important to understand that the Leeson model is the best estimate case since it assumes the tuned circuit filters out of all the harmonics. In all practical cases, it is hard to predict the operating Q and the noise figure and the rf output power. When comparing the measured results of oscillators with the assumptions made in Leeson’s equation, one frequently obtains a de facto noise figure in the vicinity of 10 to 20 dB and an operating Q that is different than the assumed loaded Q. Attempting to match the Leeson calculated curve, the measured curve requires totally different values than those assumed. To keep my interest burning in the low noise signal source, I proposed closed form mathematical definition of the phase noise in terms of known circuit parameters that describe the effectiveness of the noise model in the field of oscillator designs. The third constraint is market demand and affordable signal source solutions. Currently signal sources revenue is being driven by the integrated Silicon (Si) technology due to the expansion of consumer electronics and computer applications. Silicon technology is moving forward to handle some of the timing tasks which were traditionally given to high non-Silicon resonator such as crystal and ceramic resonator based frequency generating timing devices. Our aim is to incorporate pros and cons for both technologies and come up with a hybrid technology using two yard sticks: first is cost and second is performance.
MWJ: You were the owner of Compact Software during the early days of nonlinear RF circuit simulation. What was it like to bring a technology like Harmonic Balance to the market? We’re designers immediately accepting or skeptical?
UR: Owning Compact Software was fascinating. At the time, I already owned a company called Communications Consulting Corporation which distributed a powerful general purpose microwave and RF program, named CADEC, and a variety of Synthesis Programs running on HP computers (HP Basic) and once Compact Software became available I was lucky to acquire it and also transported the lower cost programs to the PC. The missing parts of all software at the time was the arbitrary noise analysis, small signal and large signal, for amplifiers, oscillators, mixers, and frequency multipliers, and in a team effort with Prof. Vittorio Rizzoli, we were the first company in the world to introduce an accurate general purpose solution. The trouble with non-linear analysis is that you need, not only accurate models for the semiconductor but also need the correct parameter extraction. The data sheets of most semiconductor manufacturers give very limited information about the large signal operation and the users were rightfully skeptical about this. The harmonic balance method, a method in which the linear portion of the circuit was calculated in the frequency domain and the non-linear portion in the time domain, has a real fast computation compared to SPICE. This HB simulator also had all the necessary lumped and distributed elements. Many dialects of this were developed. The real test was whether linear and non-linear models gave the same answer for noise and gain, and I introduced a physics based noise model for semiconductors, as in those days for many transistors the noise figure was only available at some limited bias points. Later there was fierce competition amongst all the CAD companies and there was always an issue about computation time. I remember saying that some of these competitive programs calculated faster but gave the wrong answer faster. Not everyone liked this true statement. The dynamic range of harmonic balance engine is more; about 180dB by virtue of its multi-dimensional-fast-Fourier transform as compared periodic-discrete Fourier transform to reduce simulation times. The outcome of the algorithm is reduced dynamic range (74-80dB) in comparison to fast simulation speed. My advice for new designers is to consider the issue of dynamic range and possible convergence problems in advance to meet the system performance.
MWJ: Which came first, Compact Software or Synergy Microwave? Did one company spawn the other?
UR: Synergy Microwave Corporation was founded based on a contract with NSA, and I assembled a team of engineers to design a high performance receiver, one of my pet projects. The company that was actually asked to build the receiver had a sudden suspicious fire and several bulldozers were required to flatten the company including my test equipment and all the drawings. There was no sprinkler system in place and the adjacent buildings were at risk I lost the contract to another company and for days I was in tears and felt helpless and at odds with the insurance company. We then turned around and concentrated on signal processing devices which Synergy Microwave has been producing since the 1980’s. Compact Software was acquired a few years later and many of the designs that we started at Synergy were only possible because of the CAD capability of Compact Software., after the merger from Ansoft and now Ansys. These and ADS from Agilent are the core software pieces of Synergy Microwave and have helped us tremendously.
MWJ: In what area do Synergy’s products dominate in performance?
UR: The major constraint is market demand and affordable solutions. Today, Synergy Microwave Corporation is a world class leader in high performance products such as oscillators, mixers, and synthesizers, not in the high volume low cost area, but in very high performance applications. As an example, we have a mixer in the VHF range with an intercept point of 53 dBm and we have an oscillator at 1 GHz, the phase noise is better than 150 dB at 10 kHz off the carrier. Using our various patents, we built many crystal oscillators we need as reference sources, and we are just analyzing these crystal oscillators and comparing the results with those advertised to make sure we are really better.
MWJ: Can you give some examples of where the company is pushing the state-of-the-art?
UR: Our Engineers at Synergy Microwave have done major research in new noise cancelling techniques and coupled resonators. A good example is our 4 GHz planar resonator based oscillator which matches the phase noise of the better 4 GHz DROs. We have a project underway to break the record of the 1GHz-10 GHz sources as there are many applications for this. The design team under my supervision and the leadership of Dr. Ajay Poddar developed unified analytical technique for designing low cost and low noise signal sources and mixer circuits, disclosed in recently granted Patents (US Patent Nos.: 7088189, 7180381, 7265642, 7262670, 7292113, 7196591, 7365612, 7545229, 7580693, 7586381, 7605670, 7612296) enabling technical leadership in the development of RF/Microwave components and modules for the applications in modern communication systems. I contributed in the correct behavioral prediction and modeling of noise and intermodulation in a variety of novel circuitry for which over dozen patents are granted. The introduction of the large signal noise behavior for bipolar transistors and FET’s in CAD and introduction of the slow wave evanescent mode coupled resonators concept using injection-mode dynamics allowed for the first time the accurate prediction and design of microwave devices (100MHz-18GHz); a YIG like broadband low noise performance, reduced the phase noise 10-15dB below the previously reported state-of-the-art. Some of these patented techniques are also applicable for low-cost mixers, which now achieve IP3 levels above 50dBm. These novel circuits are primarily responsible for obtaining a much higher receiver dynamic range, which depends on the reciprocal mixing (Phase noise of VCO’s and synthesizers) and the IIP3 of mixers and hence also require much lower LO drive-level now than previously reported.
I have developed many variations of the low noise oscillator techniques that include: technological scaling and minimization of l/f-noise in SiGe HBTs coupled mode N-Push oscillator/VCO; multi-mode multi-band VCOs, reconfigurable-concurrent oscillators for multi-standards wireless communication systems. My research work has resulted in cost-effective, compact and power-efficient microwave source solutions, making it possible to improve oscillator figure-of-merit (FOM) by 10-20dB and produced in quantities of well over hundreds of thousands.
In the area of low micro-phonic and high efficiency signal sources, I investigated the possibility of designing microphonics-insensitive and high efficiency oscillator circuits by using adaptive mode-coupling techniques for the applications where DC-to-RF conversion efficiency is critical. A unified mathematical formula was introduced by me, wherein analysis of complex jitter and noise of autonomous system can be easily carried out, performance can be accurately predicted without using expensive CAD tools, and circuits can be reliably fabricated, thereby cutting down design cycles and manufacturing costs.
MWJ: Which markets are particularly well suited to Synergy’s family of products offering?
UR: The developments in the area of modern receiver design include
Digital signal processing RF techniques in low cost, power-efficient, and reconfigurable passive FET mixers; A unified method of designing ultra-wideband, power-efficient and high IIP3 reconfigurable passive FET mixers, and A/D based receivers with down converter.
Under my guidance, Synergy Microwave, under contract with the DARPA and US army developed world-class performing oscillators and mixers which hold the record in overall performance, including several top product awards by trade journals. To secure the business in global market, my company is leading in the technical publications in reputed international conference and journal, including several dozens of patents applications.
As we are trying to avoid low cost, high volume products and instead develop high end products, most of our customers are either in the test equipment or in the military electronics arena for two-way radio applications where performance such as adjacent channel low noise side band emissions is mandatory. Most of the products are optimized for the customer applications.