FAQ

Q: What frequency ranges do you refer to microwave and RF?
A: Generally, frequency range up to 1 GHz is referred to RF and from 1 GHz to 30 GHz as microwave, although the definitions of the bands may vary somewhat. At 30 GHz, the free-air wavelength is 1 cm and since the wavelength of the higher frequencies is shorter, they are called millimeter waves (mm-waves). Most of today's wireless telecommunications take place in the 500 kHz to 12 GHz frequency range although the current trend is to increase more and more in frequency. Some of the recent applications into this area are LMDS (28GHz) and vehicle collision avoidance systems.
Q: What's the difference between Analog and Digital?
A: Some superstitions regarding Digital vs. Analog: Digital is easy and Analog is hard. If it works, it's digital. These statements oversimplify things, however they reveal a lot of truth. One important reason for digital design being superficially easy is that its building blocks are so well defined. Provided we follow certain rules (loading, speed, etc) we know that the building blocks (AND, OR, NOT, etc) will perform their specified function exactly. This makes the interconnection of digital building blocks very well defined. However, with analog circuits, we cannot hope to replicate this ease of interconnection. The reason for this is that with digital circuits what matters is only whether a signal is below or above some level. The actual value of the signal does not matter, but with analog circuits any deviation in the value of a signal from the ideal value is either an error or noise. What is more the errors and noise in an analog signal propagate, so they get worse as we progress through the system. The end result is that we can make a digital system that is essentially error-free, but an analog system always has some error and noise and all we can do is to design it so that the error and noise are below some acceptable limits. By following good engineering practices a RF/analog engineer is capable of designing systems where the error and noise are so low that for practical purposes they can be considered not present.
Q: What is the Smith Chart?
A: The Smith Chart was originally developed by Philip Smith, an AT&T Bell Lab engineer, as an RF engineering tool in 1933. Although Smith's original paper was rejected by the IRE, predecessor of the IEEE, it has become one of the most popular design aids for RF and microwave engineers and over 50 million copies have been distributed throughout the world. Initially, the chart was created to simplify the task of plotting impedance variation in the passive transmission line circuits, then later additions were developed for active circuit design. On the Software and Modeling archive, you can find plenty of good Smith Chart programs: download it and learn about RF circuit designs from it!
Q: What is SPICE?
A: The SPICE I am talking about is neither a cable TV channel nor the British pop music stars, Spice Girls. The name SPICE is an acronym for Simulation Program with Integrated Circuit Emphasis, which is certainly not so exciting as the other two. SPICE was developed by the EECS Department at the University of California, Berkeley twenty five years ago as the latest version of computer program that predicts the electrical characteristics of an integrated circuit, following in the steps of BIAS, SLIC, TIME, and CANCER, as well as numerous other lesser known programs. Because it was developed primarily as a teaching tool to provide students insight into integrated circuit performance, SPICE was rapidly embraced by many engineering colleges throughout the world. SPICE has achieved acceptance as an industry standard without the usual help of standards bodies, committees, meetings, position papers, and bureaucracy.
Q: What does it mean by AC coupling a RF signal and DC coupling? Can you please give me applications of each of them in real life?
A: AC coupled is when you want to "block" DC from one circuit getting to the next circuit. This means, in one way or another, a capacitor in series. DC coupling might be used when you need a very wide bandwidth in a system since there is no such thing as an ideal capacitor. A good example is a low frequency spectrum analyzer, which is AC coupled (goes to 1.5 GHz or so), and a wideband, which is DC coupled (goes to 26 GHz and beyond). In the wideband spectrum analyzer, there's a yellow warning label that tells you the max spec is 0VDC.
Q: What are Multiplexers? What are they used for?
A: A multiplexer combines many signals, each of which transmitted through its own channel (a wire, waveguide, coax, etc.) in some way into one signal, which may be transmitted on a single channel. The individual signals are still recoverable, but they are all available within this one large signal.
Q: What are Attenuators? What are they used for?
A: The term attenuate means to lessen the value of something. In RF and microwave, an attenuator lessens a signal. In an electronic circuit, we can attenuate a signal by pacing something that will impede the signal's progress in the path of the signal. That is fairly simple. However, a good attenuator is not only to attenuate the signal but also to maintain a good matched condition in the process.
Q: What are Filters? What are they used for?
A: The one basic function of filters, passing a particular thing and rejecting everything else, still applies to RF and microwave filters. In RF and microwave, filters reduce or eliminate spurious signals or harmonics by passing the one and only desired signal or band of signals and rejecting any other signals. There are four typical types of filters: low pass, high pass, bandpass and band reject also called notch. However, the need to use filters should be considered carefully because whenever you add a filter to a circuit, you are adding loss, VSWR from the filter, ripple, and even some delay in the circuit.
Q: What do dBm and dBc stand for? How to calculate them between mW?
A: 1) Decibel is an absolutely arbitrary way that we agreed to describe the ratio between two things. The definition of decibel is just like the definition of a dozen. In the sense that it is completely arbitrary. The origin of the dozen as a commercial unit of measure was because the number twelve was a CONVENIENT number since it could be divided by two, three and four. The same happened with decibels. We defined the decibel because it was CONVENIENT. While observing naturally occurring phenomena, we (the human beings) realized that there were certain processes such as sound levels whose linear perception was actually the result of exponential increasing absolute levels. Using the same unit to describe largely dissimilar signal levels was just NOT PRACTICAL. Using amplifiers as an example, a 60dB power Gain of a 1 mW input signal is 1,000,000mW. Since power gain in a certain stage can be actually understood as a multiplication factor then if we didn't have the decibel, we would have to use the term "amplifier multiplication factor" instead of amplifier gain. For the previously described amplifier we would need to say that it has a "one million multiplication factor" instead of a 60 dB gain. If we cascaded an additional 10dB stage we would need to multiply the million by ten and say that the amplifier is now a "ten-million multiplication factor" amplifier instead of 70dB. By using dBs we can get away by doing lazy sums of gains or losses or other ratios. Somebody though that, instead of using such large number to express the gain of an amplifier, it would be easier to use the exponent resulting in expressing the multiplication factor as powers of ten. By using this approach it's easy to understand why adding the exponents is actually the same as multiplying the factors. So we agreed to define the decibel for power ratios as: 10 log(A/B) being A/B the ratio. Going back to the 60dB power gain example the ratio between the 1mW to1000,000mW is: 10 log (1,000,000/1) = 10 log (10^6) = 10 x 6 = 60dB However, if you find it somewhat tedious to convert the exponential power ratio to dB, the JavaScript-based Online Interactive dB Calculator can make the process easy and simple. 2) Now what is a dBm? Simple, it's just a decibel that always uses 1mW as its B portion. Since the B portion is defined, then the dBm is not only a measure of a ratio only but also a known ratio to 1mW therefore dBm IS a way to express a certain POWER. The same applies to dBW where B = 1W. When using voltages the definition changes to 20 log(B/A) just because power expressed in voltages is V^2/R so the square power goes outside of the log as2x10=20. 3) Now what is dBc? dBc is still a RATIO but not a measure of power or voltage or anything. dBc is a UNITLESS measure. And it expresses the ratio between a certain carrier (c) and another level. For instance, when doing a two-tone test the IMD levels are expressed as dBc because it's the difference in dB or the ratio between the intermodulation products and the two-tone signals. You might be probably asking yourself "how come if I measure the two tones in dBm's, which is a measure of power, then I end up with dBc, which is a unitless measurement?" The answer to that question is that you actually have both. Explanation: As explained before dBm was a measurement using a mW as your "zero" 1mW=0dBm (B=1mW). So what you're actually saying indirectly is also that 40dBm is the power you would get if you added 40dB of gain to a 0dBm (1mw) signal. Now, think about the two tone test: If you measure an IMD level of 10dBm and a two tone level of40dBm although you say that IMD is (40-10=30) 30dBc, what you're actually saying indirectly is that the two tone level (40dBm) is the level you would get if you added 30dB of gain to your IMD level (10dBm). But since the part you're interested in is expressing linearity then what you're interested in expressing is the DIFFERENCE in dB or ratio between the two levels and we call that dBc for "dBs to the carrier". The phrase "absolute reference" can be used when referring to the "B" such as the 1mW when defining dBm. Also the absolute power reference could be chosen as any value, not just restricted to 1 mW, however then you probably need to accurately define your new reference since dBm (1 mW) is the most widely used and known. In the broadcast industry, it's fairly common to use dBk when working with high power transmitters, the dBk reference point or "B" being 1 kilowatt. In some disciplines (EMI/EMC testing, for instance), you may see terms like dBV and dBuV (where that u is a Greek micron for micro and the V in both cases is Volts), and hence is unrelated to power unless a load is specified.
Q: Spread Spectrum technology was initially used for military purposes. It's now being used commercially in GPS systems, wireless LANs, and cellular/PCS systems. What is it about the SS technology that applies to both military and commercial systems?
A: Secure communications and bandwidth efficiency are the two attributes of SS that attract most to military and commercial applications, respectively. Spread Spectrum exploits multipath and multiple access and has high tolerance against jammers. With Spread Spectrum, information is inherently encrypted. These are important in both commercial and military systems, e.g. an interference in a cellular system can be seen as a jammer. It took some time though before the price of SS chips became low enough to enable commercial applications.
Q: I'm confused: give me some clear differentiation for PCS and Cellular? Is it only a matter of Digital vs. Analog?
A: In the US, cellular started with analog AMPS. However, with introduction of digital systems such as GSM IS-95, and IS-54 which used almost the same frequency band, what we call cellular is not a technology but a frequency band, which is around 800 MHz. PCS is actually not a technology but a collection of services. All of the existing digital technologies (GSM, North American TDMA, and CDMA) can be, and are, implemented at a frequency band around 1800 MHz.