MMIC – A Revolution in Microwave Engineering

One of the most revolutionary inventions in microwave engineering was the MMIC (Monolithic microwave integrated circuits) for high frequency applications. The major advantage of the MMIC was integrating previously bulky components into non-discrete tiny components of a chip. The subsequent image shows the integrated components of the MMIC – spiral inductors (red), FETs (blue) for example.

It is apparent that smaller transistors are present towards the input of the MMIC. This is because less power is required to amplify the weak input signals. As the signals become stronger, higher power (and hence a larger FET) is required. The input terminal (given by the arrow) is the gate and the output the drain. Like almost all RF devices, MMIC’s output and input are usually matched to 50 ohms, making them easy to cascade.

Originally, MMICs found their place within DoD for usage in phased array systems in fighter jets. Today, they are present in cellular phones, which operate in the GHz range much like military RADARs. MMICs have switched from MESFET configurations to HEMTs, which utilize compound semiconductors to create heterostructures. MMICs can be fabricated using Silicon (low cost) or III-V semiconductors which offer higher speed. Additionally, MOSFET transistors are becoming increasingly common due to improved performance over the years. The gate of the MOSFET has been shortened from several microns to several nanometers, allowing better performance at higher frequencies.

RF Spectrum Analyzers

A Spectrum analyzer (whether in the RF Domain or optical) is a tool that is dual of the oscilloscope. An oscilloscope displays a waveform in time domain. When this is represented as a function, a Fourier transform can be used on it to obtain its spectrum. A spectrum analyzer displays this content.

Spectrum analyzers are very similar to radio receivers. A radio receiver could be classified into many types: (Super)heterodyne, crystal video, etc. Similar to a heterodyne receiver, which features a bandpass filter, mixer and low pass filter, a spectrum analyzer must tune over a specific range. This range must be very narrow, which requires a high Quality factor bandpass filter to operate. This is where the YIG (Yttrium Iron Garnet) filter comes into play. YIG has a very high quality factor and resonates when exposed to a DC magnetic field. This is what determines the spectrum analyzers “resolution bandwidth”. Of course, a narrow RBW means a less noisy display and better resolution. The tradeoff for this is increased sweep time. The sweep time is inversely proportional to the RBW squared.

A sweep generator is used to repetitively scan over the frequency band. The oscillator sweeps and repetitively mixes/multiples with the input signal and is filtered with a low pass filter. The low pass filter determines the spectrum analyzer’s “video bandwidth”.

An important concept with regards to bandwidth is thermal noise. Thermal noise is the single greatest source of noise in systems under 100 GHz. Past 100 GHz and into optics, shot noise becomes more apparent. However, bandwidth is the greatest contributor to thermal noise, as noise power is given as kTB. Since k is a constant and T has a relatively negligible effect on thermal noise (the main thing is that T is nonzero. At absolute zero, you have no thermal noise. Anything above that, you have thermal noise. The difference between a pretty cold device and a scorching hot one is only maybe 10 dBm or so. Just ballparking), this means that bandwidth has a huge effect on noise. A higher RBW increases the spectrum’s noise floor and makes it harder for closely spaced frequency components to be seen, as more frequency components are passed through the envelope detector.

Video bandwidth, on the other hand, typically determines resolution between power levels and smooths the display. It is important to note that the VBW contribution happens after data has been collected and does not affect the measurement results, whereas the RBW dictates the minimum measurable bandwidth.

Phase noise is also present in a spectrum analyzer and can affect measurements near the center frequency and results from phase jitter. Since this is pretty much a phase modulation, sidebands are produced near the center frequency which can interfere with measurement. Jitter refers to deviation from periodicity of a signal.

Noise Figure

Electrical noise is unwanted alterations to a signal of random amplitude, frequency and phase. Since RADAR is typically done at microwaves frequencies, the noise contribution of most RADAR receivers is highest at the first stages. This is mostly thermal noise (Johnson noise). Each component of a receiver has its own Noise Figure (dB) which is typically kept low through the use of a LNA (Low Noise amplifier). It is important to know that all conductors generate thermal noise when above absolute zero (0K).

Noise Power

Noise Power is the product of Boltzman’s constant, temperature in Kelvin and receiver bandwidth (k*t0*B). This is typically also expressed in dBm. This value is -174 dBm at room temperature  for a 1 Hz bandwidth. If a different receiver bandwidth is present, you can simply add the decibel equivalent of the bandwidth to this value. For example, at a 1MHz bandwidth, the bandwidth ratio is 60 dB (10*log(10^6) = 60). This value can be added to the standard 1Hz bandwidth to arrive at -114 dBm. For a real receiver, this number is scaled by the Noise Figure.

 

The Noise Figure is defined as 10*log(Na/Ni) where Na is the noise output of an actual receiver and Ni is the noise output of an ideal receiver. Alternatively these can be converted to dB and subtracted. It can also be defined as the rate at which SNR degrades. For systems on earth, Noise Figure is quite useful as temperature tends to stay around 290K (room temperature). However, for satellite communication, the antenna temperature tends to be colder than 290K and therefore effective noise temperature would be used instead.

Noise Factor is the linear equivalent of Noise Figure. For cascaded systems, the noise factor gradually decreases and decreases as shown. This explains why in a receiver chain, the initial components have a much higher effect on the Noise Figure.

noisefactor

Noise Figure is a very important Figure of Merit for detection systems where the input signal strength is unknown. For example, it is necessary to decrease the Noise Figure in the electromagnetic components of a submarine in order to detect communication and RADAR signals.

Dispersion in Optical Fibers

Dispersion is defined as the spreading of a pulse as it propagates through a medium. It essentially causes different components of the light to propagate at different speeds, leading to distortion. The most commonly discussed dispersion in optical fibers is modal dispersion, which is the result of different modes propagating within a MMF (multimode fiber). The fiber optic cable supports many modes because the core is of a larger diameter than SMF (single mode fibers). Single mode fibers tend to be used more commonly now due to decreased attenuation and dispersion over long distances, although MMF fibers can be cheaper over short distances.

Let’s analyze modal dispersion. When the core is sufficiently large (generally the core of a SMF is around 8.5 microns or so), light enters are different angles creating different modes. Because these modes experience total internal reflection at different angles, their speeds differ and over long distances, this can have a huge effect. In many cases, the signal which was sent is completely unrecognizable. This type of dispersion limits the bandwidth of the signal. Often GRIN (graded index) fibers are employed to reduce this type of dispersion by gradually varying the refractive index of the fiber within the core so that it decreases as you move further out. As we have learned, the refractive index directly influences the propagation velocity of light. The refractive index is defined as the ratio of the speed of light to the speed of the medium. In other words, it is inversely proportional to the speed of the medium (in this case silica glass).

modal

In order to mitigate the effects of intermodal distortion in multimode fibers, pulses are lengthened to overlap components of different modes, or even better to switch to Single mode fibers when it is available.

The next type of dispersion is chromatic dispersion. All lasers suffer from this effect because no laser is comprised of a single frequency. Therefore, different wavelengths will propagate at different speeds. Sometimes chirped Bragg gratings are employed to compensate for this effect. Doped fiber lasers and solid state lasers tend to have much thinner linewidths than semiconductor PIN lasers and therefore tend to have less chromatic dispersion, although the semiconductor lasers has several advantages such as lesser cost and smaller size.

Another dispersion type is PMD (Polarization mode dispersion) which is caused by different polarizations travelling at different speeds within a fiber. Generally, these travel at the same speed however spreading of pulses can be caused by imperfections in the material.

For SMF fibers, it is important to cover waveguide dispersion. It is important to note that since the cladding of the fiber is doped differently than the core, the core has a higher refractive index than the cladding (doping with fluorine lowers refractive index and doping with germanium increases it). As we know, a lower refractive index indicates faster speed of propagation. Although most of the light stays within the core, some is absorbed by the cladding. Over long distances this can lead to greater dispersion as the light travels faster in the core leading to different propagation velocities.

RF Over Fiber Links

The basic principle of an RF over Fiber link is to convey a radio frequency electrical signal optically through modulation and demodulation techniques. This has many advantages including reduced attenuation over long distances, increased bandwidth capability, and immunity to electromagnetic interference. In fact, Rf over fiber links are essentially limitless in terms of distance of propagation, whereas coaxial cable transmission lines tend to be limited to 300 ft due to higher attenuation over distance.

The simple RFoF link comprises of an optical source, optical modulator, fiber optic cable and a receiver.

rfof

The RF signal modulates the optical signal at its frequency (f_opt) with sidebands at the sum and difference of the RF frequency and optical signal frequency. These beat against the carrier in the photodetector to reproduce and electrical RF signal. The above picture shows amplitude modulation and direct detection method. Also, impedance matching circuitry is generally included to match the ports of the modulator to the demodulator as well as amplifiers.

Before designing an RFoF link, it must be essential to bypass a transmission line in the first place. Will the system benefit from having a lower size and weight or immunity to electromagnetic interference? Is a wide bandwidth required? If not, this sort of link may not be necessary. It also must be determined the maximum SWaP of all the hardware at the two ends of the link. Another important consideration is the temperature that the link will be exposed to (or even pressure, humidity or vibration levels) that the link will be exposed to. The bandwidth of the RF and distance of propagation must be considered, finally.

The Following Figures of Merit can be used to quantify the RFoF link:

Gain

In dB, this is defined as the Signal out (in dBm) – Signal in (dBm) or 10log(g) where g is the small signal gain (gain for which the amplitude is small enough that there is no amplitude compression)

Noise Figure

For RADAR and detection systems where the input signal strength is unknown, Noise Figure is more important than SNR. NF is the rate at which SNR degrades from input to output and is given as N_out – kTB – Gain (all in dB scale).

Dynamic Range

It is known that the Noise Floor defines the lower end of dynamic range. The higher end is limited by spurious frequencies or amplitude compression. The difference between the highest acceptable and lowest acceptable input power is the dynamic range.

For example, if defined in terms of full compression, the dynamic range would be (in dB scale) : S_in.max – MDS. where MDS is the minimum detectable signal strength power.

Scattering Parameters

Scattering parameters are frequency dependent parameters that define the loss or gain at various ports. For two port systems, this forms a 2×2 matrix. In most Fiber Optic links, the backwards isolation S_12 is equal to zero due to the functionality of the detectors and modulators (they cannot perform each other’s functions). Generally the return losses at port 2 and 1 are what are specified to meet the system requirements.

 

 

Erbium Doped Fiber Amplifiers (EDFA)

EDFA

The above figure demonstrates the attenuation of optical fibers relative to wavelength. It can be seen that Rayleigh Scattering is more prevalent at higher frequencies. Rayleigh scattering occurs when minute changes in density or refractive index of optical fibers is present due to manufacturing processes. This tends to scatter either in the direction of propagation within the core or not. If it is not, this leads to increased attenuation. This accounts for 96% of attenuation in optical fibers. It can also be noted that lattice absorption varies wildly with the wavelength of light. From the graph, it is apparent that 1550 nm wavelength this value (and also Rayleigh Scattering) is quite low. It is for this reason that 1550 nm is a common wavelength of propagation with silica glass optical fibers. Although this wavelength allows for greater options in design, shorter wavelengths (such as 850 nm) are also used when distance of propagation is short. However, 1550 is the common wavelength due to the development of dispersion shifted fibers as well as something called the EDFA (Erbium doped fiber amplifier).

EDFAs operate around the 1550 nm region (1530 to 1610 nm) and work based on the principle of stimulated emission, in which a photon is emitted within a optical device when another photon causes electrons and holes to recombine. The stimulated emission creates a photon of the same size and in the same direction (coherent light). The EDFA acts as an amplifier, boosting the intensity of light with a heavily doped core (erbium doped). As discussed earlier, the lowest power loss for silica fibers tends to occur at 1550 nm, which is the wavelength that this stimulated emission occurs. The excitation, however, occurs at 980 or 1480 nm, which is shown to have high loss.

The advantages of the EDFA is high gain and availability to operate in the C and L bands of light, It is also not polarization dependent and has low distortion at high frequencies. The major disadvantage is the requirement of optical pumping.

EDFA

Receiver Dynamic Range

Dynamic range is pretty general term for a ratio (sometimes called DNR ratio) of a highest acceptable value to lowest acceptable value that some quantity can be. It can be applied to a variety of fields, most notably electronics and RF/Microwave applications. It is typically expressed in a logarithmic scale. Dynamic range is an important figure of merit because often weak signals will need to be received as well as stronger ones all while not receiving unwanted signals.

Due to spherical spreading of waves and the two-way nature of RADAR, losses experienced by the transmitted signal are proportional to 1/(R^4). This leads to a great variance over the dynamic range of the system in terms of return. For RADAR receivers, mixers and amplifiers contribute the most to the system’s dynamic range and Noise Figure (also in dB). The lower end of the dynamic range is limited by the noise floor, which accounts for the accumulation of unwanted environmental and internal noise without the presence of a signal. The total noise floor of a receiver can be determined by adding the noise figure dB levels of each component. Applying a signal will increase the level of noise past the noise floor, and this is limited by the saturation of the amplifier or mixer. For a linear amplifier, the upper end is the 1dB compression point. This point describes the range at which the amplifier amplifies linearly with a constant increase in dB for a given dB increase at the input. Past the 1dB compression point, the amplifier deviates from this pattern.

123

The other points in the figure are the third and second order intercept points. Generally, the third intercept point is the most quoted on data sheets, as third order distortions are most common. Assuming the device is perfectly linear, this is the point where the third order distortion line intersects that line of constant slope. These intermodulation distortion generate the terms 2f_2 – f_1 and 2f_1 – f_2. So in a sense the third order intercept point is a measure of linearity. As shown in the figure, the third order distortion has a linear slope of 3:1. The point that the line intercepts the linear output is (IIP3, OIP3). This intercept point tends to be used as more of a rule of thumb, as the system is assumed to be “weakly linear” which does not necessarily hold up in practice.

Often manual gain control or automatic gain control can be employed to achieve the desired receiver dynamic range. This is necessary because there are such a wide variety of signals being received. Often the dynamic range can be around 120 dB or higher, for instance.

Another term used is spurious free dynamic range. Spurs are unwanted frequency components of the receiver which are generated by the mixer, ADC or any nonlinear component. The quantity represents the distance between the largest spur and fundamental tone.

Discrete Time Filters: FIR and IIR

There are two basic types of digital filters: FIR and IIR. FIR stands for Finite Impulse Response and IIR stands for infinite impulse response. The outputs of any discrete time filter can be described by a “difference equation”, similar to a differential equation but does not contain derivatives. The FIR is described by a moving average, or weighted sum of past inputs. IIR filter difference equations are recursive in the sense that they include both a sum of weighted values of past inputs as well as a weighted average of past outputs.

fuck

As shown, this specific IIR filter difference equation contains an output term (first time on the right hand side).

The FIR has a finite impulse response because it decays to zero in a finite length of time. In the discrete time case, this means the output response of a system to a Kronecker delta input or impulse. In the IIR case, the impulse response decays, but never reaches zero. The FIR filter has zeros with only poles at  z = 0 for H(z) (system function). The IIR filter is more flexible and can contain zeroes at any location on a pole zero plot.

The following is a block diagram of a two stage FIR filter. As shown, there is no recursion but simply a weighted sum. The triangles represent the values of the impulse response at a particular time. These sort of diagrams represent the difference equations and can be expressed as the output as a function of weighted sum of the inputs. These z inverse blocks could be thought of as memory storage blocks in a computer.

800px-FIR_Filter_(Moving_Average).svg

In contrast, the IIR filter contains recursion or feedback, as the past inputs are added back to the input. This feedback leads to a nontrivial term in the denominator of the transfer function of the filter. This transfer function can be tested for stability of the filter by observing the pole zero plot in the z-domain.

IIR

Overall, IIR filters have several advantages over FIR filters in terms of efficiency in terms of implementation which means that lower order filters can be used to achieve the same result of an FIR filter. A lower order filter is less computationally expensive and hence more preferable. A higher order filter requires more operations. However, FIR filters have a distinct advantage in terms of ease of design. This mainly comes into play when trying to design filters with linear phase (constant group delay with frequency) which is very hard to do with an IIR filter.

The Acoustic Guitar – Intro

We will consider our study of sound by briefly analyzing the acoustic guitar: an instrument that uses certain physical properties to “amplify” (not really true as no energy is technically added) sound acoustically rather than through electromagnetic induction or piezoelectric means (piezoelectric pickups are common on acoustic-electric guitars however). A guitar can be tuned many ways but standard (E standard) tuning is E-A-D-G-B-E across the six strings from top to bottom, or thickest string to thinnest. The tuning is something that can be changed on the fly, which differentiates the guitar from something like a harp which the tension of the string cannot be adjusted.

Just like the tuning pegs on a guitar can be loosened or tighten to change the tension, the fretting hand can also be used to change the length of the string. Both of these affect the frequency or perceived pitch. In fact, two other qualities of the string (density and thickness) also effect the frequency. These can be related through Mersenne’s rule:

unnamed

As shown, the length and density of the string are inversely proportional to the pitch. The tension is proportional, so tightening the string will tune the string up.  The frequency is inversely proportional to string diameter.

The basic operation of the guitar is that plucking or strumming strings will cause a disturbance in the air, displacing air particles and causing buildups of pressure “nodes” and “antinodes”. This leads to the creation of a longitudinal pressure wave which is perceived by the human ear as sound. However, a string on its own does not displace much air, so the rest of the guitar is needed. The soundboard (top) of the guitar acts as an impedance matching network between the string and air by increasing the surface area of contact with the air. Although this does not amplify the sound since no external energy is applied, it does increase the sound intensity greatly. So in a sense the soundboard (typically made of spruce or a good transmitter of sound) can be thought of as something like an electrical impedance matching transformer. The acoustic guitar also employs acoustic resonance in the soundhole. As with the soundboard, the soundhole also vibrates and tends to resonate at lower frequencies. When the air in the soundhole moves in phase with the strings, sound intensity increases by about 3 dB. So basically, the sound is being coupled from the string to the soundboard, from the soundboard to the soundhole and from both the soundhole and soundboard to the external air. The bridge is the part of the guitar that couples the string vibration to the soundboard. This creates a reasonably loud pressure wave.

In terms of wood, the typical wood used for guitar making has a high stiffness to weight ratio. Spruce has an excellent stiffness to weight ratio, as it has a high modulus of elasticity and moderately low density. Rosewood tends to be used for the back and sides of a guitar. The main thing to note hear is the guitar is made of wood.. because wood does not carry vibrations well. As a result the air echos within the guitar instead, creating a sound that is pleasant to the ear. Another factor, of course is cost.

Strings are comprised of a fundamental frequency as well as harmonics and overtones, which lead to a distinct sound. If you fret a string at the twelfth fret, this is the halfway part of the string. This would be the first overtone with double the frequency. It is important to note that the frets of a guitar taper off as you go towards the bridge. This distance can be calculated since c = fλ is a constant. Each successive note is 1.0595 higher in pitch so the first fret is placed 1.0595 from the bridge. This continues on and on with 1.0595 being raised to a higher and higher power based on what fret is being observed.

Microstrip Antenna – Cavity Model

The following is an alternative modelling technique for the microstrip antenna, which is also somewhat similar to the analysis of acoustic cavities. Like all cavities, boundary conditions are important. For the microstrip antenna, this is used to calculated radiated fields of the antenna.

Two boundary conditions will be imposed: PEC and PMC. For the PEC the orthogonal component of the E field is zero and the transverse magnetic component is zero. For the PMC, the opposite is true.

cavity

This supports the TM (transverse magnetic) mode of propagation, which means the magnetic field is orthogonal to the propagation direction. In order to use this model, a time independent wave equation (Helmholtz equation) must be solved.

helmholtz

The solution to any wave equation will have wavelike properties, which means it will be sinusoidal. The solution looks like:

1234

Integer multiples of π  solve the boundary conditions because the vector potential must be maximum at the boundaries of x, y and z. These cannot simultaneously be zero. The resonant frequency can be solved as shown:

res

The units work out, as the square root of the product of the permeability and permittivity in the denominator correspond to the velocity of propagation (m/s), the units of the 2π term are radians and the rest of the expression is the magnitude of the k vector or wave number (rad/m). This corresponds to units of inverse seconds or Hz. Different modes can be solved by plugging in various integers and solving for the frequency in Hz. The lowest resonant mode is found to be f_010 which is intuitively true because the longest dimension is L (which is in the denominator). The f_000 mode cannot exist because that would yield a trivial solution of 0 Hz frequency. The field components for the dominant (lowest frequency) mode are given.

1x

 

 

Microstrip Patch Antennas Introduction – Transmission Line Model

Microstrip antennas (or patch antennas) are extremely important in modern electrical engineering for the simple fact that they can directly be printed to a circuit board. This makes them necessary for things like cellular antennas for GPS, communication with cell towers and bluetooth/WiFi. Patch antennas are notoriously narrowband, especially those with a rectangular shape (patch antennas can have a wide variety of shapes). Patch antennas can be configured as single antennas or in an array. The excitation is usually fed by a microstrip line which usually has a characteristic impedance of 50 ohms.

One of the most common analysis methods for analyzing microstrip antennas is the transmission line model. It is important to note that the microstrip transmission line does not support TEM mode, unlike the coaxial cable which has radial symmetry. For the microstrip line, quasi-TEM is supported. For this mode, there is a field component along the direction of propagation, although it is small. For the purposes of the model, this can be ignored and the TEM mode which has no field component in the direction of propagation can be used. This reduces the model to:

microstrip

Where the effective dielectric constant can be approximated as:

eff

The width of the strip must be greater than the height of the substrate. It is important to note that the dielectric constant is not constant for frequency. As a consequence, the above approximation is only valid for low frequencies of microwave.

Another note for the transmission line model is that the effective length differs from the physical length of the patch. The effective length is longer by 2ΔL due to fringing effects. ΔL can be expressed as a function of the effective dielectric constant.

123

 

 

 

The Helical Antenna

The helical antenna is a frequently overlooked antenna type commonly used for VHF and UHF applications and provides high directivity, wide bandwidth and interestingly, circular polarization. Circular polarization provides a huge advantage in that if two antennas are circularly polarized, the will not suffer polarization loss due to polarization mismatch. It is known that circular polarization is a special case of elliptical polarization. Circular polarization occurs when the Electric field vector (which defines the polarization of any antenna) has two components which are in quadrature with equal amplitudes. In this case, the electric field vector rotates in a circular pattern when observed at the target, whether it be RHP or LHP (right hand or left hand polarized).

Generally, the axial mode of the helix antenna is used but normal mode may also be used. Usually the helix is mounted on a ground plane which is connected to a coaxial cable using a N type or SMA connector.

The helix antenna can be broken down into triangles, shown below.

traignel

The circumference of each loop is given by πD. S represents the spacing between loops. When this is zero (and hence the angle of the triangle is zero), the helix antenna reduces to a flat loop. When the angle becomes a 90 degree angle, the helix reduces to a monopole linear wire antenna. L0 represents the length of one loop and L is the length of the entire antenna. The total height L is given as NS, where N is the number of loops. The actual length can be calculated by multiplying the number of loops with the length of one loop L0.

An important thing to note is that the helix antenna is elliptically polarized by default and must be manually designed to achieve circular polarization for a specific bandwidth. Another note is that the input impedance of the antenna depends greatly on the pitch angle (alpha).

The axial (endfire) mode, which is more common occurs when the circumference of the antenna is roughly the size of the wavelength. This mode is easier to achieve circular polarization. The normal mode features a much smaller circumference and is more omnidirectional in terms of radiation pattern.

The Axial ratio is the numerical quantity that governs the polarization. When AR = 1, the antenna is circularly polarized. When AR = ∞ or 0, the antenna is linearly polarized. Any other quantity means elliptical polarization.

itsover

The axial ratio can also be approximated by:

AR

For axial mode, the radiation pattern is much more directional, as the axis of the antenna contains the bulk of the radiation. For this mode, the following conditions must be met to achieve circular polarization.

Axial

These are less stringent than the normal mode conditions.

It is also important to consider that the input impedance of these antennas tends to be higher than the standard impedance of a coaxial line (100-200 ohms compared to 50). Flattening the feed wire of the antenna and covering the ground plane with dielectric material helps achieve a better SWR.

h

This equation can be used to calculated the height of the dielectric used for the ground plane. It is dependent on the transmission line characteristic impedance, strip width and the dielectric constant of the material used.

The Superheterodyne Receiver

“Heterodyning” is a commonly used term in the design of RF wireless communication systems. It the process of using a local oscillator of a frequency close to an input signal in order to produce a lower frequency signal on the output which is the difference in the two frequencies. It is contrasted with “homodyning” which uses the same frequency for the local oscillator and the input. In a superhet receiver, the RF input and the local oscillator are easily tunable whereas the ouput IF (intermediate frequency) is fixed.

1

After the antenna, the front end of the receiver comprises of a band select filter and a LNA (low noise amplifier). This is needed because the electrical output of the antenna is often as small as a few microvolts and needs to be amplified, but not in a way that leads to a higher Noise Figure. The typical superhet NF should be around 8-10 dB. Then the signal is frequency multiplied or heterodyned with the local oscillator. In the frequency domain, this corresponds to a shift in frequency. The next filter is the channel select filter which has a higher Quality factor than the band select filter for enhanced selectivity.

For the filtering, the local oscillator can either be fixed or variable for downconversion to the baseband IF. If it is variable, a variable capacitor or a tuning diode is used. The local oscillator can be higher or lower in frequency than the desired frequency resulting from the heterodyning (high side or low side injection).

A common issue in the superhet receiver is image frequency, which needs to be suppressed by the initial filter to prevent interference. Often multiple mixer stages are used (called multiple conversion) to overcome the image issue. The image frequencies are given below.

image

Higher IF frequencies tend to be better at suppressing image as demonstrated in the term 2f_IF. The level of attenuation (in dB) of a receiver to image is given in the Image Rejection Ratio (the ratio of the output of the receiver from a signal at the received frequency, to its output for an equal strength signal at the image frequency.

RADAR Range Resolution

Before delving into the topic of pulse compression, it is necessary to briefly discuss the advantages of pulse RADAR over CW RADAR. The main difference between the two is with duty cycle (time high vs total time). For CW RADARs this is 100% and pulse RADARs are typically much lower. The efficiency of this comes with the fact that the scattered signal can be observed when the signal is low, making it much more clear. With CW RADARs (which are much less common then pulse RADARs), since the transmitter is constantly transmitting, the return signal must be read over the transmitted signal. In all cases, the return signal is weaker than the transmitter signals due to absorption by the target. This leads to difficulties with continuous wave RADAR.  Pulse RADARs can also provide high peak power without increasing average power, leading to greater efficiency.

“Pulse Compression” is a signal processing technique that tries to take the advantages of pulse RADAR and mitigate its disadvantages. The major dilemma is that accuracy of RADAR is dependent on pulse width. For instance, if you send out a short pulse you can illuminate the target with a small amount of energy. However the range resolution is increased. The digital processing of pulse compression grants the best of both worlds: having a high range resolution and also illuminate the target with greater energy. This is done using Linear Frequency Modulation or “Chirp modulation”, illustrated below.

290px-Linear-chirp.svg

As shown above, the frequency gradually increases with time (x axis).

A “matched filter” is a processing technique to optimize the SNR, which outputs a compressed pulse.

Range resolution can be calculated as follows:

Resolution = (C*T)/2

Where T is the pulse time or width.

With greater range resolution, a RADAR can detect two objects that are very close. As shown this is easier to do with a longer pulse, unless pulse compression is achieved.

It can also be demonstrated that range resolution is proportional to bandwidth:

Resolution = c/2B

So this means that RADARs with higher frequencies (which tend to have higher bandwidth), greater resolution can also be achieved.

 

 

Mathematical Formulation for Antennas: Radiation Integrals and Auxiliary Potentials

This short paper will attempt to clarify some useful mathematical tools for antenna analysis that seem overly “mathematical” but can aid in understanding antenna theory. A solid background in Maxwell’s equations and vector calculus would be helpful.

Two sources will be introduced: The Electric and Magnetic sources (E and M respectively). These will be integrated to obtain either an electric and magnetic field directly or integrated to obtain a Vector potential, which is then differentiated to obtain the E and H fields. We will use A for magnetic vector potential and F for electric vector potential.

Using Gauss’ laws (first two equations) for a source free region:

cfr

And also the identity:

1

It can be shown that:

2

In the case of the magnetic field in response to the magnetic vector potential (A). This is done by equating the divergence of B with the divergence of the curl of A, which both equal zero. The same can be done from Gauss Law of electricity (1st equation) and the divergence of the curl of F.

Using Maxwell’s equations (not necessary to know how) the following can be derived:

3

For total fields, the two auxiliary potentials can be summed. In the case of the Electric field this leads to:

4

The following integrals can be used to solve for the vector potentials, if the current densities are known:

5

For some cases, the volume integral is reduced to a surface or line integral.

An important note: most antenna calculations and also the above integrals are independent of distance, and therefore are done in the far field (region greater than 2D^2/λ, where D is the largest dimension of the antenna).

The familiar duality theorem from Fourier Transform properties can be applied in a similar way to Maxwell’s equations, as shown.

mxw

In the chart, Faraday’s Law, Ampere’s Law, Helmholtz equations and the above mentioned integrals are shown. To be perfectly honest, I think the top right equation is wrong. I believe is should have permittivity rather than permeability.

Another important antenna property is reciprocity… that is the receive and transmit radiation patterns are the same , given that the medium of propagation is linear and isotropic. This can be compared to the reciprocity theorem of circuits, meaning that a volt meter and source can be interchanged if a constant current or voltage source is used and the circuit components are linear, bilateral and discrete elements.