1

2
Please read this disclaimer before
proceeding:
This document is confidential and intended solely for the educational
purpose of RMK Group of Educational Institutions. If you have received this
document through email in error, please notify the system manager. This
document contains proprietary information and is intended only to the
respective group / learning community as intended. If you are not the
addressee you should not disseminate, distribute or copy through e-mail.
Please notify the sender immediately by e-mail if you have received this
document by mistake and delete this document from your system. If you
are not the intended recipient you are notified that disclosing, copying,
distributing or taking any action in reliance on the contents of this
information is strictly prohibited.

3
 R.M.D ENGINEERING
COLLEGE

DEPARTMENT OF ELECTRONICS AND

COMMUNICATION ENGINEERING

21EC948 RFIC DESIGN

Department : ECE
Batch/Year : 2021-2025/III
Created by : R.M .SENTHIL PRIYA, A.P,ECE

Date : 28.09.2023

4
 Table of Contents
S.N Contents Page
o Numbe
r
1 Course Objectives 7

2 Pre Requisites 8

3 Syllabus 9

4 Course outcomes 11

5 CO- PO/PSO Mapping 12

6 UNIT 2 – HIGH FREQUENCY AMPLIFIER

DESIGN
7 13
6.1 Lecture Plan
8 14
6.2 Activity based learning
9 15
6.3 Lecture Notes
10  15
3.1 Mixers
11  16
3.2 Fundamentals of mixers
12  21
3.3 Multiplier-based mixers
13  29
3.4 Sub sampling mixers
14  31
3.5 Diode-ring mixers

5
 Table of Contents
S.N Contents Page
o Numbe
r
15. 6.4 Assignments 33

16. 6.5 Part A Question & Answer 34

17. 6.6 Part B Questions 42

18. 6.7 Supportive online Certification courses 43

19. 6.8 Real time Applications in day to day life and 44

to
Industry
20. 6.9 Content beyond the Syllabus 45

21. ASSESMENT SCHEDULE 47

22. PRESCRIBED TEXT BOOKS AND REFERENCE 48

BOOKS
23. MINI PROJECT SUGGESTION 49

6
1. Course Objectives

● To study the various impedance matching techniques

used

in RF circuit design

● To study amplifier design

● To analyse oscillators performance

● To understand the functional design aspects of LNAs,

Mixers, PLLs and VCOs

● To understand frequency synthesis

7
2. Pre Requisites:

• 21EC301-ELECTRONIC CIRCUITS

• 21EC303-DIGITAL ELECTRONICS

• 21EC202-ELECTRONIC DEVICES

8
3. SYLLABUS
21EC948 RFIC DESIGN LTPC 3
0 03

UNIT I IMPEDANCE MATCHING IN AMPLIFIERS

7

Characteristics of passive IC components at RF frequencies –

Definition of ‘Q’, Series Parallel Transformations of Lossy
Circuits, Impedance Matching Using ‘L’, ‘Pi’ and T Networks,
Integrated Inductors, Resistors, Capacitors, Tunable Inductors,
Transformers

UNIT II HIGH FREQUENCY AMPLIFIER DESIGN

10

High frequency amplifier design – zeros as bandwidth

enhancers, shunt-series amplifier, fTdoublers, Low noise
amplifier design – LNA topologies, impedance matching,
power constrained noise optimization, linearity and large
signal performance

UNIT III ACTIVE AND PASSIVE MIXERS

10

Mixers – fundamentals of mixers, multiplier-based mixers, sub

sampling mixers, diode-ring mixers.

9
21EC948 RFIC DESIGN LTPC
3 0 03

UNIT IV OSCILLATORS
9

Oscillators– Feedback View of Oscillators, Colpitts oscillator,

Hartley oscillator, describing functions, tuned oscillators, negative
resistance oscillators.

UNIT V PLL AND FREQUENCY SYNTHESIZERS

9

Phase Detector/Charge Pump, Analog Phase Detectors, Digital

Phase Detectors, Frequency Dividers, Loop Filter Design, Phase
Locked Loops, Phase Noise in PLL, Loop Bandwidth, Basic Integer-N
Frequency Synthesizer, Basic Fractional-N Frequency Synthesizer.

10
 4. COURSE OUTCOMES
After Successful Completion of the course, the students should be able to

Course Statement Highest

Outcome Cognitiv
s e Level
CO1 To understand the principles of operation of an RF receiver
K2
front end
CO2 To design and apply constraints for LNAs, Mixers and
K3
frequency synthesizers
CO3
To analyze and design mixer K3

CO4 To design different types of oscillators and perform noise

K3
analysis
CO5
To design PLL and frequency synthesizer K3

CO6 To understand passive components at RF frequencies and

K2
required circuit theory

11
 Program Outcomes PSO
C Level of PO PO PO PO PO5 PO PO PO PO PO PO PO PS PS PS
Outcome 1 2 3 4 6 7 8 9 10 11 12 O O O
1 2 3
K3 K4 K4 K5 K3/ A3 A2 A3 A3 A3 A3 A2 K3 K3 K3
K5/
K6

K2 3 2 2 2 3 1 - - - - 1 1 3 3 3

K3 3 3 3 2 3 1 - - - - 1 1 3 3 3

K3 3 2 2 1 3 1 - - - - 1 1 3 3 3

K3 3 3 3 2 3 1 - - - - 1 1 3 3 3

K3 3 2 2 1 3 1 - - - - 1 1 3 3 3

K2 3 2 2 1 3 1 - - - - 1 1 3 3 3
5.CO-PO/PSO mapping

3 2 2 2 3 1 - - - - 1 1 3 3 3

12
 6.1 LECTURE PLAN
UNIT III – ACTIVE AND PASSIVE MIXERS

Mode of Delivery
Taxonomy level
Proposed Date
No. of Periods

Pertaining CO
Actual Date

Reason for
Deviation
Topic
S.No

1 Mixers 2 CO3 K2 Black -

Understand Board

2 Fundamentals of 2 CO3 K3 PowerPoin -

mixers Apply t Black
Board

3 Multiplier-based 2 CO3 K3 PowerPoin -

mixers Apply t Black
Board

4 Sub sampling 2 CO3 K3 PowerPoin -

mixers Apply t Black
Board

5 Diode-ring mixers 2 CO3 K3 PowerPoin -

Apply t Black
Board

Total No. of Periods : 10 13

 6.2.Activity based learning – UNIT III
Activity 1- TECHNICAL QUIZ

1. An active device is one which _____

A. Mechanically controls electron flow
B. Electrically controls electron flow
C. Pneumatically controls electron flow
D. Automatically controls electron flow
2.Which of the following elements comprise of the passive devices?
A. Resistors, Capacitors and SCRs
B. Vacuum Tubes, SCRs and Diodes
C. Transformers, Inductors and Diodes
D. Transformers,
. TRIACs and DIACs
3.How do amplifiers work without violating Law of Conservation of Energy?:
A. They amplify one factor of the input and reduce others
B. They work on the law of conservation of mass
C. They violate the Law of Conservation of Energy
D. They amplify the signal by taking an input from an external
source
4.To overcome the problem of representation of large values of gain, which
of the following units was introduced?
A. Decibels
B. Joules
C. Pascals
D. Farads
5.Active devices can also be used as_________
A. Amplifiers
B. Choppers
C. Converters
D. Inverters

14
 6.3 LECTURE NOTES
UNIT 3 - ACTIVE AND PASSIVE MIXERS
3.1 INTRODUCTION

Most circuit analysis proceeds with the assumptions of linearity and time
invariance. Violations of those assumptions, if considered at all, are usually
treated as undesirable. However, the high performance of modern
communications equipment actually depends critically on the presence of at
least one element that fails to satisfy linear time invariance: the mixer. We
will see shortly that mixers are still ideally linear but depend fundamentally
on a purposeful violation of time invariance. The superheterodyne1 receiver
uses a mixer to perform an important frequency translation of signals.
Armstrong’s invention has been the dominant architecture for 70 years
because this frequency translation solves many problems in one fell swoop
(see Figure 1)In this architecture, the mixer translates an incoming RF signal
to a lower frequency,3 known as the intermediate frequency (IF). Although
Armstrong originally sought this frequency lowering simply to make it easier
to obtain the requisite gain ,other significant advantages accrue as well. As
one example, tuning is now accomplished by varying the frequency of a local
oscillator, rather than by varying the center frequency of a multipole band
pass filter. Thus, instead of adjusting several LC networks in tandem to tune
to a desired signal, one simply varies a single LC combination to change the
frequency of a local oscillator (LO). The intermediate frequency stages can
then use fixed band pass filters. Selectivity is therefore determined.

FIGURE 1. Super heterodyne receiver block diagram.

by these fixed-frequency IF filters, which are much easier to realize than

variable- frequency filters. Additionally, the overall gain of the system is
distributed over a number of different frequency bands (RF, IF, and
baseband), so that the required re- ceiver gain (typically
15 120–140 dB on a
power basis) can be obtained without much worry about potential oscillations
arising from parasitic feedback loops. These im- portant attributes explain
 3.2 MIXER FUNDAMENTALS

Linear, time-invariant systems cannot produce outputs with spectral components not
present at the input, mixers must be either nonlinear or time-varying elements in order
to provide frequency translation. Historically, many devices (e.g., electrolytic cells,
magnetic ribbons, brain tissue, and rusty scissors – in addition to more traditional
devices such as vacuum tubes and transistors) operating on a host of diverse principles
have been used, demonstrating that virtually any nonlinear element can be used as a
mixer.4 At the core of all mixers presently in use is a multiplication of two signals in the
time domain. The fundamental usefulness of multiplication may be understood from
examination of the following trigonometric identity:

Multiplication thus results in output signals at the sum and difference frequencies of the
input, signals whose amplitudes are proportional to the product of the RF and LO
amplitudes. Hence, if the LO amplitude is constant (as it usually is), any amplitude
modulation in the RF signal is transferred to the IF signal. By a similar mechanism, an
undesired transfer of modulation from one signal to another can also occur through
nonlinear interaction in both mixers and amplifiers. In that context the result is called
cross-modulation, and its suppression through improved linearity is an important design
consideration . Having recognized the fundamental role of multiplication, we now
enumerate and define the most significant characteristics of mixers

3.2.1 CONVERSION GAIN

One important mixer characteristic is conversion gain (or loss), which is defined as the
ratio of the desired IF output to the value of the RF input. For the multiplier described
by Eqn. 1, the conversion gain is therefore the IF output, AB/2, divided by A (if that is
the amplitude of the RF input). Hence, the conversion gain in this example is B/2, or
half the LO amplitude. Conversion gain, if expressed as a power ratio, can be greater
than unity in active mixers; passive mixers are generally capable only of voltage or
current gain at best.5 Conversion gain in excess of unity is often convenient since the
mixer then provides amplification along with the frequency translation. However, it
does not necessarily follow that sensitivity improves, since noise figure must also be
considered. For this reason, passive mixers may offer superior performance in some
cases despite their conversion loss

16
 3.2.2 NOISE FIGURE: SSB VERSUS DSB

Noise figure is defined as one might expect: it’s the signal-to-noise ratio (SNR) at the input

(RF) port divided by the SNR at the output (IF) port. There’s an important subtlety, however,

that often trips up both the uninitiated and a substantial fraction of practicing engineers. To

appreciate this difficulty, we first need to make an important observation: In a typical

mixer, there are actually two input frequencies that will generate a given intermediate

frequency. One is the desired RF signal, and the other is called the image signal. In the

context of mixers, these two signals are frequently referred to collectively as sidebands.

The reason that two such frequencies exist is that the IF is simply the magnitude of the

difference between the RF and LO frequencies. Hence, signals both above and below ωLO

by an amount equal to the IF will produce IF outputs of the same frequency. The two input

frequencies are therefore separated by 2ωIF. As a specific numerical example, suppose that

our system’s IF is 100 MHz and we wish to tune to a signal at 900 MHz by selecting an LO

frequency of 1 GHz. Aside from the desired 900-MHz RF input, a 1.1-GHz image signal will

also produce a difference frequency component at the IF of 100 MHz's.

The existence of an image frequency complicates noise figure computations because noise

originating in both the desired and image frequencies therefore becomes IF noise, yet there

is generally no desired signal at the image frequency. In the usual case where the desired

signal exists at only one frequency, the noise figure that one measures is called the single-

sideband noise figure (SSB NF); the rarer case, where both the “main” RF and image signals

contain useful information, leads to a double sideband (DSB) noise figure. Clearly, the SSB

noise figure will be greater than for the DSB case, since both have the same IF noise but

the former has signal power in only a single sideband. Hence, the SSB NF will normally be 3

dB higher than the DSB NF.6 Unfortunately, DSB NF is reported much more often because it

is numerically smaller and thus falsely conveys the impression of better performance, even

though there are few communications systems for which DSB NF is an appropriate figure of

merit. Frequently, a noise figure is stated without any indication as to whether it is a DSB or

SSB value. In such cases, one may usually assume that a DSB figure is being quoted. Noise

figures for mixers tend to be considerably higher than those for amplifiers because noise

from frequencies other than at the desired RF can mix down to the IF. Representative

values for SSB noise figures range from 10 dB to 15 dB or more. It is mainly because of this

larger mixer noise that one uses LNAs in a receiver. If the LNA has sufficient gain then the

signal will be amplified to levels well above the noise of the mixer and subsequent stages,

so the overall receiver NF will be dominated by the LNA instead of the mixer. If mixers were

not as noisy as they are, then the need for LNAs would diminish considerably

17
 3.2.3 LINEARITY AND ISOLATION

Dynamic range requirements in modern, high-performance telecommunications

systems are quite severe, frequently exceeding 80 dB and approaching 100 dB in
many instances. As discussed in the previous chapter, the floor is established by the
noise figure, which conveys something about how small a signal may be processed,
whereas the ceiling is set by the onset of severe nonlinearities that accompany large
input signals. As with amplifiers, the compression point is one measure of this
dynamic range ceiling and is defined the same way. Ideally, we would like the IF
output to be proportional to the RF input signal amplitude; this is the sense in which
we interpret the term “linearity” in the context of mixers. However, as with amplifiers
(and virtually any other physical system), real mixers have some limit beyond which
the output has a sub linear dependence on the input. The compression point is the
value of RF signal8 at which a calibrated departure from the ideal linear curve occurs.
Usually, a 1-dB (or, more rarely, a 3-dB) compression value is specified. One may
specify either the input or output signal strength at which this compression occurs,
together with the conversion gain, to allow fair comparisons among different mixers.
The two-tone third-order intercept is also used to characterize mixer linearity. A two-
tone intermodulation test is a relevant way to evaluate mixer performance because it
mimics the real-world scenario in which both a desired signal and a potential
interferer (perhaps at a frequency just one channel away) feed a mixer input. Ideally,
each of two superposed RF inputs would be translated in frequency without
interacting with each other. Of course, practical mixers will always exhibit some
intermodulation effects, and the output of the mixer will thus contain frequency-
translated versions of third-order IM components whose frequencies are 2ωRF1 ±
ωRF2 and 2ωRF2 ± ωRF1. The difference frequency terms may heterodyne into
components that lie within the IF passband and are therefore generally the
troublesome ones, while the sum frequency signals can usually be filtered out. As a
measure of the degree of departure from linear mixing behavior, one can plot the
desired output and the third-order IM output as a function of input RF level. The third-
order intercept is the extrapolated intersection of these two curves. In general, the
higher the intercept, the more linear the mixer. Again, one ought to specify whether
the intercept is input- or output-referred, as well as the conversion gain, to permit fair
comparisons among mixers. Additionally, it is customary to abbreviate the intercept
as IP3, or perhaps IIP3 or OIP3 (for input and output third-order intercept point,
respectively). These definitions are summarized in Figure 2. Cubic nonlinearity can
also cause trouble with a single RF input. As a specific example, consider building a
18 455
low-cost AM radio. The standard IF for AM radios happens, unfortunately, to be
kHz (mainly for historical reasons). Tuning in a station at 910 kHz (a legitimate AM
 FIGURE 2. Definition of mixer linearity parameters.

Another parameter of great practical importance is isolation. It is

generally desirable to minimize interaction among the RF, IF, and LO
ports. For instance, since the LO signal power is generally quite large
compared with that of the RF signal, any LO feed through to the IF output
might cause problems at subsequent stages in the signal processing
chain. This problem is exacerbated if the IF and LO frequencies are
similar, so that filtering is ineffective. Even reverse isolation is important
in many instances, since poor reverse isolation might permit the strong
LO signal (or its harmonics) to work its way back to the antenna, where it
can radiate and cause interference to other receivers.

3.2.4 SPURS
Mixers, by their nature, may heterodyne a variety of frequency
components that you never intended to mix. For example, harmonics of
some signal (desired or not) could lie (or be generated) within the
passband of the mixer system and subsequently beat against the local
oscillator signal (and its harmonics). Some of the resulting components
may end up within the IF passband. The undesired signals that do
ultimately emerge from the output of the mixer are known as spurious
responses, or just spurs. Evaluation of mixer spurs is straightforward in
principle but highly tedious in practice (so much so that a hazing ritual for
newly minted RF engineers in days past often included evaluation of
mixer spurs).10 The availability of software tools to take care of this task
has eliminated the tedium, but it’s instructive to describe the process,
just the same.Let m and n be the harmonic numbers of the RF input and
19
LO frequencies, respectively. Then the spur products present at the
output of the mixer (prior to any filtering) are given by
The apparent simplicity of this equation is misleading: The calculation must be
repeated for all combinations and signs of m and n, ranging up to the maximum
harmonic order you care to consider. To make a laborious procedure even more so,
one must actually consider RF signals of frequencies below the nominal input
passband – at least down to the lower passband edge frequency divided by the
maximum value of m. One may also have to consider input frequencies somewhat
above the nominal upper passband frequency of the RF filter. Since no filter is
perfect and since no LO is completely free of distortion, harmonics of the LO can still
heterodyne with nominally out-of-band RF signals that leak through the filter. The
resulting interaction can produce spurs at the mixer output that happen to lie within
the IF passband. If the out-of-band interferer is strong enough, the spurious IF signal
can severely degrade receiver performance. For each (m, n) pair, examine the spur
frequency and then determine whether it lies within the IF passband, or sufficiently
close to it, to merit further consideration. For each spur that does, work backward to
the implied RF input frequency and evaluate the likelihood that there will be a signal
of sufficient strength at that frequency to be a source of trouble. Then make
appropriate modifications either to the input filtering or to the choice (or quality) of
LO or IF, if necessary, to avoid those troubles. This exercise is sometimes performed
with the worst-case assumption that there is no filtering of any kind at the RF input
port. In that case, the number of calculations grows very large quite quickly. If one is
patient enough, however, the information generated can be used to guide the
design of the input filter (or the frequency plan and other architectural details of the
receiver).

As a specific example, suppose we wish to design a mixer for an FM receiver, whose

nominal input passband is to accommodate signals spanning 88.1 MHz to 108.1
MHz. With a 10.7-MHz IF (conventional for consumer FM receivers), the LO needs to
tune from 77.4 MHz to 97.4 MHz (assuming low-side injection). To keep the numbers
easy, assume a bit unrealistically that the IF system possesses a nominal bandwidth
of approximately 200 kHz. Further assume that the LO is pure enough – and the
input RF filtered enough – that we need not consider harmonic orders higher than
3.12 With these assumptions, we can construct Table below

20
 3.3 MULTIPLIER-BASED MIXERS
We have seen that nonlinearities produce mixing incidentally through the
multiplications they provide. Precisely because the multiplication is only incidental,
these nonlinearities usually generate a host of undesired spectral components.
Additionally, since two-port mixers have only one input port, the RF and LO signals are
generally not well isolated from each other. This lack of isolation can cause the
problems mentioned earlier, such as overloading of IF amplifiers, as well as radiation of
the LO signal (or its harmonics) back out through the antenna. Mixers based directly on
multiplication generally exhibit superior performance because they ideally generate
only the desired intermodulation product. Furthermore, because the inputs to a
multiplier enter at separate ports, there can be a high degree of isolation among all
three signals (RF, LO, and IF). Finally, CMOS technology provides excellent switches,
and one can implement outstanding multipliers with switches.

3.3.1 SINGLE-BALANCED MIXER

FIGURE 2. Single-balanced mixer.

One extremely common family of multipliers first converts the incoming RF voltage into
a current and then performs a multiplication in the current domain. The simplest
multiplier cell of this type is sketched in Figure 2 In this mixer, vLO is chosen large
enough so that the transistors alternately switch (commutate) all of the tail current
from one side to the other at the LO frequency.23 The tail current is therefore
effectively multiplied by a square wave whose frequency is that of the local oscillator:
iout(t) = sgn[cos ωLO t ]{IBIAS + IRF cos ωRF t }. (2)

Because a square wave consists of odd harmonics of the fundamental, multiplication of

the tail current by the square wave results in an output spectrum that appears as
shown in Figure 3 (ωRF is here chosen atypically low compared with ωLO to reduce
clutter in the graph). The output thus consists of sum and difference components, each
the result of an odd harmonic of the LO mixing with the RF signal. In addition, odd
harmonics of the LO appear directly in the output as a consequence of the DC bias
current multiplying with the LO signal. Because of the presence of the LO in the output
spectrum, this type of mixer is known as a single-balanced mixer. Double-balanced
mixers, which we’ll study shortly, exploit symmetry to remove the undesired output LO
component through cancellation.
21
 FIGURE 3. Representative output spectrum of single-balanced mixer

FIGURE 4. RF transconductors for mixers .

Although the current source of Figure 3 includes a component that is perfectly

proportional to the RF input signal, V–I converters of all real mixers are
imperfect. Hence, an important design challenge is to maximize the linearity of
the RF transconductance. Linearity is most commonly enhanced through some
type of source degeneration, in both common-gate and common-source
transconductors; see Figure 4. The common-gate circuit uses the source
resistance Rs to linearize the transfer characteristic. This linearization is most
effective if the admittance looking into the source terminal of the transistor is
much larger than the conductance of Rs. In that case, the transconductance of
the stage approaches 1/Rs. Inductive degeneration is usually preferred over
resistive degeneration for several reasons. An inductance has neither thermal
noise to degrade noise figure nor DC voltage drop to diminish supply headroom.
This last consideration is particularly relevant for low-voltage–low-power
applications. Finally, the increasing reactance of an inductor with increasing
frequency helps to attenuate high frequency harmonic and intermodulation
components.

22
 FIGURE 5 Single-balanced mixer with linearized transconductance.

A more complete single-balanced mixer that incorporates a linearized

transconductance is shown in Figure 5. The value of VBIAS establishes
the bias current of the cell, while RB is chosen large enough not to load
down the gate circuit (and also to reduce its noise contribution). The RF
signal is applied to the gate through a DC blocking capacitor CB. In
practice, a filter would be used to remove the LO and other undesired
spectral components from the output. The conversion transconductance
of this mixer can be estimated by assuming that the LO-driven
transistors behave as perfect switches. Then the differential output
current may be regarded as the result of multiplying the drain current of
M1 by a unit-amplitude square wave. Since the amplitude of the
fundamental component of a square wave is 4/π times the amplitude of
the square wave, we may write:

where gm is the transconductance of the V–I converter and Gc is itself

a transconductance. The coefficient is 2/π rather than 4/π because the
IF signal is divided evenly between sum and difference components.

3.2.2 ACTIVE DOUBLE-BALANCED MIXER

To prevent the LO products from getting to the output in the first place,
two single balanced circuits may be combined to produce a double-
balanced mixer; see Figure 6. We assume once again that the LO drive
is large enough to make the differential pairs act like current-steering
switches. Note that the two single-balanced mixers are connected in
antiparallel as far as the LO is concerned but in parallel for the RF
signal. Therefore, the LO terms sum to zero in the output, whereas the
converted RF signal is doubled in the output. This mixer thus provides a
high degree LO–IF isolation, easing filtering requirements at the
23
output. If care is taken in layout, IC realizations of this circuit routinely
provide 40 dB of LO–IF isolation, with values in excess of 60 dB possible
 FIGURE 6 Active double-balanced mixer.

FIGURE 7. Linearized differential RF trans conductor for double-balanced

mixer
LO–IF .
isolation, easing filtering requirements at the output. If care is taken in layout, IC

realizations of this circuit routinely provide 40 dB of LO–IF isolation, with values in excess

of 60 dB possible. As in the single-balanced active mixer, the dynamic range is limited in

part by the linearity of the V–I converter in the RF port of the mixer. So, most of the

design effort is spent attempting to find better ways of providing this V–I conversion. The

basic linearizing techniques used in the single-balanced mixer may be adapted to the

double-balanced case, as shown in Figure 6. In low-voltage applications, the DC current

source can be replaced by a parallel LC tank to create a zero-headroom AC current

source. The resonant frequency of the tank should be chosen to provide rejection of

whatever common-mode component is most objectionable. If several such components

exist, one may use series combinations of parallel LC tanks. With such a choice, a

complete double-balanced mixer appears as shown in Figure 7. The expression for the

conversion transconductance is the same as for the single-balanced case.

3.2.3 POTENTIOMETRIC MIXERS

Gilbert-type mixers first convert an incoming RF voltage into a current through a
transconductor, whose linearity and noise figure set a firm bound on the overall mixer
linearity and noise figure. An alternative to using voltage-controlled current sources in V–I
converters is to use voltage-controlled resistances. For example, consider varying the
resistance of a triode-region MOSFET in a manner inversely proportional to the incoming
RF signal. If the voltage between drain and source is maintained at a fixed value, the
current flowing through the device will be a faithful replica of the RF voltage, and if vds
varies with the LO then the current will be proportional to the product of the LO and RF
24
signals.
 FIGURE 8. Potentiometric mixer

One possible implementation of this idea is sketched in Figure 8 The four

MOSFETs perform the mixing, while the capacitors remove the sum
frequency component as well as higher-order products. The RF input
drives the gates of the transistors, while the LO drives the sources. A
simplified analysis assumes that the resistances of the transistors are
inversely proportional to the RF signal. In that case, the current through
the devices is

Because the current is then the result of a multiplication of the RF and LO

signals, there are components at the sum and difference frequencies, as
desired. This current flows through the feedback resistors so that the IF signal
is available as an output voltage. The op-amp need only have enough
bandwidth to handle the difference frequency component, since the sum
component is filtered out by the four capacitances. Note that, for good
linearity, the gate overdrive must greatly exceed vLO. Hence, vRF must
possess a sufficiently large DC component to satisfy this inequality for as large
a value of vrf that must be accommodated. Practical mixers of this type may
exhibit good linearity (e.g., 40 dBm IIP3) but high noise figures (e.g., 30 dB).
The high noise figures are the result of the resistive thermal noise of the input
FETs (which is worst when the signal levels are small) and the difficulty of
providing a good noise match with the broadband op-amp. As a consequence,
the overall dynamic range of this type of mixer is typically about the same as
conventional Gilbert-type current-mode mixers.

3.2.4 PASSIVE DOUBLE-BALANCED MIXER

FIGURE 9. Simple double-balanced passive CMOS mixer. 25

Considering that CMOS technology offers excellent switches, high-performance
multipliers based on switching are naturally realized in CMOS form. In the active
mixers considered so far, representations of the RF signal in the form of currents,
rather than the RF voltages themselves, are effectively multiplied by a square-
wave version of the local oscillator. An alternative that avoids the V–I conversion
problem is to switch the RF signal directly in the voltage domain. This option is
considerably easier to exercise in CMOS than bipolar form, which is why bipolar
mixers are almost exclusively of the active, current-mode type. The simplest
passive commutating CMOS mixer consists of four switches in a bridge
configuration (see Figure 9). The switches are driven by local oscillator signals in
antiphase, so that only one diagonal pair of transistors is conducting at any given
time. When M1 and M4 are on, vIF equals vRF, and when M2 and M3 are
conducting, vIF equals −vRF. A fully equivalent description is that this mixer
multiplies the incoming RF signal by a unit-amplitude square wave whose
frequency is that of the local oscillator. Hence, the output contains many mixing
products that result from the odd-harmonic Fourier components of the square
wave.Luckily, these are often readily filtered out, as discussed previously. The
voltage conversion gain of this basic cell is easy to compute from the foregoing
description. Assuming multiplication by a unit-amplitude square wave, we may
immediately write

GC = 2/π.

Here, the 2/π factor again results from splitting the IF energy evenly
between the sum and difference components.33 In practice, the actual
voltage conversion gain may differ somewhat from 2/π because real
transistors do not switch in zero time. Hence, the incoming RF signal is not
multiplied by a pure square-wave signal in general. Perhaps contrary to
intuition, however, the effect of this departure from ideal assumptions is
usually to increase the voltage conversion gain above 2/π. A more general
expression for the voltage conversion gain is somewhat cumbersome to
derive, so we will state only the relevant insights here.34 The output of
The function gT (t) is the time-varying Thévenin-equivalent conductance
the mixer may be treated as the product of three time-varying
as viewed from the IF port, while gT max and gT are the maximum and
components and a scaling factor:
average values, respectively, of gT (t). The mixing function, m(t), is
defined b
26
where g(t) is conductance of each switch and TLO is the period of the LO drive. The mixing
function has no DC component, is periodic in TLO, and has only odd harmonic content
because of its half-wave symmetry. The Fourier transform of the first bracketed term in
Eqn. 26 has a value of 2/π at the LO frequency for a square-wave drive (as asserted
earlier) and a value of 1/2 for a sinusoidal drive, so the effective mixing function indeed
contributes a higher conversion gain for a square-wave drive. However, the second
bracketed term is unity for a square-wave drive (because the peak and average
conductance are equal) but π/2 for a sinusoidal drive. The overall conversion gain is
greater with a sinusoidal drive because the second term more than compensates for the
smaller contribution by the (effective) mixing function. The difference is not particularly
large, however. With a sinusoidal drive, the conversion gain is π/4 (−2.1 dB), compared
with the 2/π gain (−3.92 dB) obtained with the square-wave drive. Because of the
spectrum of the (effective) mixing function, undesirable products can appear at the IF port
of this type of mixer. The subject of filtering therefore deserves careful consideration,
especially in connection with the issue of input and output terminations. In discrete
designs, the source and load impedances are usually real and well-defined (50 ', for
example), but the sources and loads for IC mixers are usually on-chip and not at all
standardized. Far from a liability, this lack of standardization is a degree of freedom that
the IC engineer can exploit to improve performance. As a specific example, reactive
source and load terminations might be preferable because they do not generate noise.
Because it is difficult to obtain broadband operation with reactance's, narrowband
operation is implied for most practical mixers with reactive terminations. Fortunately,
there are many applications for which this restriction is not a serious limitation.

In CMOS implementations, the load at the IF port of the mixer is frequently capacitive to
an excellent approximation. In such cases, the loading is easily accommodated as forming
a simple low-pass filter in conjunction with the resistance of the switches. A detailed
analysis35 reveals that the transfer function of this filter is simply

We see that the pole frequency is simply the ratio of the average conductance (again, as
viewed from the IF port, back through the switches) to the load capacitance. This inherent
filtering action may be exploited to provide a much desired attenuation of unwanted
mixer products. A somewhat more elaborate passive mixer that further exploits the
freedom to select source and load terminations appears as Figure Note that this mixer
assumes a capacitive load, represented as CL in the schematic. This assumption reflects
the typical situation in fully integrated CMOS circuits, and it stands in contrast with the
resistive terminations common in discrete designs. A capacitive load generates no
thermal noise of its own, and also helps filter out high-frequency noise and distortion
27
 FIGURE 10. Low-noise, narrowband passive mixer.

The input network consists of an L-match in cascade with a parallel tank. The L-match,
comprising L1 and part of the tank capacitance, provides an impedance transformation
that moderately boosts the RF signal voltage to help reduce the voltage conversion loss.
The parallel tank, formed by L3 and C3+CL, filters out-of-band noise and distortion
components present at the input and generated by the mixer itself. Resistor R1 sets the
common-mode potential for the input circuit. Because any nonlinearity in the tank
capacitance reduces IP3, C3 is best implemented as a metal– metal capacitor. To reduce
the area consumed, a good choice is to use a lateral flux or fractal capacitor. Because of
the small voltage boost provided by the L-match, the voltage conversion loss can be
somewhat better than the 3.92 dB that a simple switch bridge would exhibit ideally. As an
example, one implementation in a 0.35-µm technology exhibits a 3.6-dB voltage
conversion loss with a 1.6-GHz RF and 1.4-GHz LO. Both noise figure and IP3 are strong
functions of the LO drive, since the resistance of the switches in the “on” state must be
kept low and constant to optimize both parameters. The IP3 is also a function of the
amount of voltage boost provided by the L-match. This boost may be adjusted downward
to trade conversion gain for improved IP3 and, in some cases, it may be appropriate to
remove the L-match altogether. Typical SSB noise figures of 10 dB and input IP3 of 10 dBm
are readily achievable with an LO drive amplitude of 300 mV.38 As a crude estimate, the
SSB noise figure of this type of mixer is approximately equal to the power conversion loss.
The first edition of this book contained an assertion that the absence of DC bias current in
a passive mixer implied the absence of 1/f noise. That assertion is not quite correct,
because a mixer is a periodically time-varying system. As such, noise centered at integer
multiples of the local oscillator can fold down to DC, for example. Thus, 1/f noise may still
appear at the output of the mixer without requiring any DC bias in the mixer itself. In cases
where it is important to minimize 1/f noise in the mixer output, it is generally helpful to (a)
reduce the LO drive to the minimum value consistent with acceptable mixing action and
(b) design the local oscillator carefully to minimize its close-in phase noise in particular (a
topic we take up in detail in Chapter 18). These considerations are particularly important in
the design of receivers that are sensitive to 1/f noise, which include the direct-conversion
(also known as the homodyne or zero-IF) receiver and the low-IF receiver.
28
 3.4 SUB SAMPLING MIXERS
The high quality of CMOS switches has also been exploited to realize what are
sometimes called subsampling mixers. This type of mixer exploits the
observation that the information bandwidth of modulation is necessarily lower
than the carrier frequency. Hence, one may satisfy the Nyquist criterion with a
sampling rate that is also lower than the carrier frequency, effecting down
conversion in the process.

FIGURE 11. Illustration of subsampling.

FIGURE 12 Track-and-hold subsampling mixer (simplified).

As can be seen in Figure 11, the higher-frequency signal is sampled at the

instants indicated by the dots, while the down converted signal is shown
as the lowerfrequency reconstruction. The theoretical advantage of this
approach is that it may be easier to realize samplers that operate at a
frequency well below that of the incoming RF signal. From the figure, it
should be clear that a properly designed track-and-hold circuit (Figure 12)
serves as a subsampling mixer.

29
In the sample (track) mode, transistors M1 through M5 are turned on while
transistors M6 and M7 are placed in the “off” state. Devices M3, M4, and M5 put
a voltage equal to the common-mode voltage level VCM on the right-hand
terminals of the sampling capacitors, while input switches M1 and M2 connect
the capacitors to the RF input signal. Because M6 and M7 are open, the op-amp
is irrelevant in this tracking mode, and the tracking bandwidth is simply set by
the RC time constant formed by the total switch resistance and sampling (and
parasitic) capacitance. Because the system operates open-loop in this mode, it is
easy to obtain tracking bandwidths far in excess of what can be achieved with a
feedback structure. For example, it is trivial to obtain tracking bandwidths
greater than 1 GHz in a 1-µm technology. In the hold mode, all switch states are
reversed, so that the only conducting transistors are the two feedback devices
M6 and M7. In this mode, the circuit degenerates to a pair of charged capacitors
feeding back around the op-amp. The settling time of this system need only be
fast relative to the (slow) sampling period, rather than to the RF signal period.
Thus, the bandwidth penalty associated with feedback is not serious.

Although a subsample is clocked at a relatively low frequency, the sampler must

still possess good time resolution or else sampling errors result. Therefore,
beyond an adequate tracking bandwidth, one must also have low aperture jitter
(i.e., low uncertainty in the sampling instants), and this requirement places
extraordinary demands on the phase noise of the sampling clock. Hence, even
though the frequency of the sampling clock need only satisfy the Nyquist
criterion applied to the modulation bandwidth, its absolute time jitter must be a
tiny fraction of the carrier period.

Another problem is that the sampling operation converts more than just the
signal. Noise at the input to the sampler undergoes folding into the IF band,
resulting in an unfortunate noise boost roughly equal to the ratio of RF and IF
bandwidths. Because the RF bandwidth typically exceeds the IF bandwidth by
large amounts, subsampling mixers can exhibit large noise figures (e.g., 25-dB
SSB NF). The large linearity implied by the high third-order intercepts often
exhibited by these types of mixers is offset by their poor noise performance, so
that the dynamic range of the mixer is frequently no better (or even worse) than
what one may achieve with conventional architectures. In fact, the noise and IP3
30
performance of many subsampling mixers can be replicated by preceding a
conventional mixer with a resistive divider. In principle, an LNA with sufficient
 3.5 DIODE-RING MIXERS
The four-diode double-balanced mixer has particularly good
characteristics and is nearly ubiquitous in high-performance discrete
equipment.

FIGURE 13. Simple diode mixer.

The simplest and oldest passive mixer uses a single diode, as seen in Figure
13. In this circuit, the output RLC tank is tuned to the desired IF, and vIN is
the sum of RF, LO, and DC bias components. The nonlinear V–I
characteristic of the diode provides diode currents at a number of harmonic
and intermodulation frequencies, and the tank selects only those at the IF. It
is tempting to reject this circuit as hopelessly unsophisticated. It does not
provide any isolation, and it doesn’t provide any conversion gain, for
example. However, at the highest frequencies, it may be difficult to exploit
other types of nonlinearities, and such simple mixers may be suitable. In
fact, all of the detectors 40 for radar sets developed in WWII were single-
diode circuits.41 Additionally, many early UHF television tuners also used
mixers of this type. Much of the modern work in the millimeter-wave bands
simply would not be possible without such mixers. As another note on this
circuit, it can be used as a crude demodulator for AM signals if the input
signal is the AM signal (at either RF or IF). When used in this manner, the
output inductor is removed entirely, no LO is used, and a simple RC network
provides the output filtering. Millions of “crystal” radio sets used this type of
detector (known in this context as an envelope detector), and even most
AM super heterodyne radios built today use a single-diode demodulator.
3.5.1 TWO-DIODE MIXERS

FIGURE 14 Single-balanced diode mixer

There are several other ways to use diodes as mixers. As we’ll see, it will
appear that a diode bridge can be used as just about anything, depending
on which terminals are defined as input and output and which way the 31
diodes point.
With two diodes, it’s possible to construct a single-balanced mixer. In this case, one may obtain isolation
between LO and IF, but there is poor RF–IF isolation; see Figure 14. Assume that the LO drive is sufficient to
make the diodes act as switches, regardless of the magnitude of the RF input. With a positive value for vLO,
both diodes will be on (note the reference dots on the transformer windings), effectively connecting vRF to
the IF output. When vLO goes negative, the diodes open-circuit and disconnect vRF. Hence, this mixer acts
the same as the active commutating mixer studied previously. The poor RF–IF isolation should be self-
evident from the comment that the diodes connect the RF and IF ports together whenever the diodes are
on. Similarly, it should be evident that symmetry guarantees excellent RF–LO isolation. Whenever the
diodes are on, the RF voltage can only develop a common-mode voltage across the transformer windings,
so no voltage can be induced at the LO port.

3.5.2 DOUBLE-BALANCED DIODE MIXER

By adding two more diodes and one more transformer, we can construct a doublebalanced mixer to provide
isolation among all ports (see Figure 15). Once again, assume that the LO drive is sufficient to cause the
diodes to act as switches. In the circuit shown, the left pair of diodes is on whenever the LO drive is
negative, whereas the right pair of diodes is on whenever the LO drive is positive. With the LO drive
positive, the voltage at “Right Mid” must be zero by symmetry, since the center tap of the input
transformer is tied to ground. Thus, vIF equals vRF (again, note the polarity dots). With the LO drive
negative, it is “Left Mid” that has a zero potential, and vIF equals −vRF. Hence, this mixer effectively
multiplies vRF by a unit-amplitude square wave whose frequency is that of the LO.

FI
FiGURE 15. Double-balanced diode mixer.

Isolation is guaranteed by the symmetry of the circuit. The LO drive forces a zero potential at either
the top or bottom terminal of the output transformer, as noted previously. If the RF input is zero,
there will be no IF output. Hence, this configuration provides LO–IF isolation. Similarly, we can show
LO–RF isolation by considering a zero IF input. Since, again, there is a zero potential at either the top
or bottom terminal of the output transformer, there will be no primary voltage and therefore no
secondary voltage. These passive mixers are available in discrete form, and perform exceptionally
well. The upper limit on the dynamic range is typically constrained by diode breakdown, and
isolation is a function of the matching levels achieved. With a single quad of diodes, typical double-
balanced mixers routinely achieve conversion losses in the neighborhood of 6 dB and isolation of at
least 30 dB, and they can accommodate RF inputs of up to 1 dBm at the 1-dB compression point
while requiring an LO drive of 7 dBm. Higher RF levels can be accommodated if series connections
of diodes are used in place of each diode of Figure 13.26, the drawback being an increased LO drive
requirement to guarantee switching operation of the diodes. Using a total of sixteen diodes, for
example, extends the RF input range to around 9 dBm but also requires a whopping 13 dBm of LO
32
drive.
 6.4.ASSIGNMENT

ASSIGNMENT
CO’S Blooms
Level
1. Assume you are developing an SDR system that utilizes sub- CO 3 K2
sampling mixers. Explain how sub-sampling mixers can be
integrated into an SDR architecture. Discuss the advantages and
limitations of this approach for applications like spectrum sensing
and cognitive radio.

2. In a sub-sampling mixer-based receiver, designing an anti- CO 3 K2

aliasing filter is crucial. Design an anti-aliasing filter for a sub-
sampling mixer receiver operating at a specific frequency range.
Calculate the filter specifications and evaluate its performance.

3. Discuss the challenges and considerations involved in CO 3 K2

implementing sub-sampling mixers, particularly concerning
aliasing and synchronization.

4. You have access to a satellite receiver system with a multiplier- CO 3 K3

based mixer. Design an experiment to evaluate the mixer's
performance in terms of conversion loss, isolation, and linearity.
What measurements would you take, and how would you
interpret the results?

5. Consider a wireless communication system like a Wi-Fi router CO 3 K3

or a cellular base station. Explain how mixers are used in such
systems for frequency conversion and signal processing. Discuss
the key requirements for mixers in these applications.

33
6.5. PART A Q &
A (with K level &
CO)

34
 PART A CO’S Blooms
Level

1. What is the primary purpose of a mixer in an RF CO 3 K1

system?
The primary purpose of using a mixer in RF IC design is to
facilitate frequency translation. Mixers are used to convert signals
from one frequency to another, which is crucial in RF applications
for tasks like upconversion (increasing frequency) and
downconversion (decreasing frequency) of signals. This enables
RF ICs to process, modulate, demodulate, or amplify signals at
different frequency bands efficiently.
2.Write the significance of linearity in RF mixer design. CO 3 K2
Linearity is a critical aspect of RF mixer design because RF ICs
often deal with small-signal and high-frequency applications.
Linearity refers to the ability of the mixer to produce output
signals that are proportional to the input signals over a wide
range of input power levels. In RF IC design, maintaining linearity
is essential to avoid distortion and spurious products, ensuring
that the mixer accurately processes RF signals without
introducing nonlinearities, which can lead to signal degradation or
interference. Achieving good linearity often involves careful
selection of mixer topologies, biasing schemes, and the choice of
transistors or active devices used in the mixer design.
3. What is the role of the LO (Local Oscillator) signal in an CO 3 K2
RF mixer, and how does its frequency affect mixer
operation?
The LO signal in an RF mixer serves as a reference frequency that
is mixed with the RF input signal. The frequency of the LO signal
determines the frequency of the output signals produced by the
mixer. Specifically, when the LO frequency is added to or
subtracted from the RF input frequency, it generates sum and
difference frequency components. The choice of LO frequency is
critical as it dictates the frequency translation characteristics of
the mixer, allowing designers to shift RF signals to the desired
frequency range for processing or demodulation in RF ICs.
4. In RF mixer design, what is the trade-off between CO 3 K2
conversion gain and noise figure, and why is it important?
In RF mixer design, there is often a trade-off between conversion
gain and noise figure. Conversion gain refers to the ability of the
mixer to increase or decrease the signal power level, while noise
figure represents how much additional noise the mixer introduces
to the signal. These parameters are inversely related, meaning
that as you increase conversion gain, the noise figure tends to
worsen, and vice versa.
5. What are the common mixer topologies used in RF IC design? CO 3 K2
Gilbert cell mixers, Double-balanced mixers, Single-balanced
mixers

6. What is the primary advantage of a multiplier-based mixer over

other mixer topologies in RFIC design?
The primary advantage of a multiplier-based mixer in RFIC design
35
is its high linearity, which makes it suitable for applications
requiring minimal distortion and high signal fidelity.
PART A CO’S Blooms
Level

7. Why is balanced operation essential for multiplier- CO 3 K2

based mixers in RFIC design?
Balanced operation in multiplier-based mixers helps suppress
even-order harmonic distortions and common-mode noise,
improving overall performance and reducing unwanted mixing
products.
8. In RFIC design, what techniques are commonly used CO 3 K2
to enhance the linearity of multiplier-based mixers?
To enhance linearity, RFIC designers often employ techniques
such as biasing optimization, transistor selection, and the use
of linearization circuits or feedback mechanisms in multiplier-
based mixers.
9. What is the trade-off between conversion gain and CO 3 K2
power consumption in multiplier-based mixers, and why
is it important in RFIC design?
The trade-off between conversion gain and power consumption
in multiplier-based mixers is significant. Increasing conversion
gain often requires higher power consumption, which may not
be suitable for battery-powered RFICs. RFIC designers must
carefully balance these factors to meet the performance
requirements while minimizing power consumption.

10. What is the primary function of a sub-sampling mixer in CO 3 K2

RF IC design, and what advantage does it offer?
The primary function of a sub-sampling mixer in RF IC design is
to directly down-convert high-frequency signals to lower
intermediate frequencies or baseband without using traditional
RF mixers. It offers the advantage of simplifying the RF front-
end architecture by eliminating the need for multiple RF
mixers, LO (Local Oscillator) chains, and image rejection filters,
reducing the complexity and power consumption of the RFIC.
11. How does sub-sampling mixing differ from traditional RF CO 3 K2
mixing in RF IC design?
Sub-sampling mixing in RF IC design directly samples and
down-converts high-frequency signals to lower frequencies,
often bypassing the need for an intermediate frequency (IF)
stage. Traditional RF mixing involves mixing the RF signal with
a local oscillator to shift it to an IF, which is then further
processed.

36
PART A CO’S Blooms
Level
12. What are the key challenges associated with sub- CO 3 K2
sampling mixers in RF IC design?
Common LNA topologies include CE, CB, CC, Cascode, CG,
Some key challenges include clock jitter, aliasing, and
maintaining linearity at high input frequencies. Clock jitter
can introduce phase noise, and aliasing can cause distortion.
Addressing these challenges is crucial for reliable sub-
sampling mixer operation

13. What advantages does sub-sampling offer in terms of CO 3 K2

power consumption in RF IC design?
Sub-sampling mixers can often reduce power consumption
in RF ICs because they eliminate the need for high-
frequency LO chains and IF stages. This simplification leads
to lower power consumption compared to traditional RF
mixing approaches.
14. In what applications are sub-sampling mixers CO 3 K2
commonly used in RF IC design?
Sub-sampling mixers are commonly used in software-
defined radios (SDRs), cognitive radios, and wideband
receivers, where they provide the ability to capture and
process a wide range of frequencies with reduced hardware
complexity and power consumption.

15. How does the concept of aliasing benefit sub-sampling CO 3 K2

mixers in RF IC design?
Alias in sub-sampling mixers occurs when the input signal
frequency is higher than the Nyquist frequency (half of the
sampling rate). Aliasing allows sub-sampling mixers to fold
high-frequency components back into the desired frequency
range, effectively down-converting them to lower
frequencies without the need for additional mixing stages.

37
PART A CO’S Blooms
Level
16. What role does the clock signal play in sub-sampling CO 3 K2
mixers, and how does its frequency affect the down-
conversion process?
The clock signal in sub-sampling mixers serves as the
sampling reference for down-conversion. The clock
frequency determines the rate at which samples are taken.
Lower clock frequencies provide coarser time resolution but
reduce power consumption, while higher clock frequencies
offer finer time resolution but may consume more power.
The choice of clock frequency must balance these factors to
meet the RFIC's performance and power efficiency
requirements.
17. What is a diode-ring mixer, and what is its primary CO 3 K2
function in RFIC design?
A diode-ring mixer is a type of mixer used in RFIC design. Its
primary function is to mix or modulate high-frequency RF
signals with a local oscillator (LO) signal to produce desired
intermediate frequency (IF) or baseband signals.
18. What are the advantages of using diode-ring mixers in RFIC CO 3 K2
design?
Diode-ring mixers offer advantages such as simplicity, low
power consumption, and good linearity. They are suitable for
applications where minimizing complexity and power
consumption is essential.
19. How do diode-ring mixers achieve frequency CO 3 K2
conversion in RFICs?
Diode-ring mixers achieve frequency conversion by utilizing
the nonlinear properties of diodes. They mix the RF input
signal with the LO signal, resulting in the generation of sum
and difference frequency components.
20. What is the role of biasing in diode-ring mixers, CO 3 K2
and how does it affect their performance in RFICs?
Biasing in diode-ring mixers involves applying a DC voltage
to diodes to set their operating points. Proper biasing is
crucial for optimizing linearity and minimizing distortion in
the mixer's output.

38
PART A CO’S Blooms
Level
21. In RFIC design, what is the purpose of the LO (Local CO 3 K2
Oscillator) signal in a diode-ring mixer?
The LO signal in a diode-ring mixer serves as the reference signal
with a known frequency. It is mixed with the RF input signal to
produce the desired output frequency components through the
nonlinear behavior of diodes.

22. How do diode-ring mixers contribute to the CO 3 K2

miniaturization of RFICs?
Diode-ring mixers contribute to miniaturization by simplifying the
RF front-end circuitry. Their low complexity and reduced need for
additional components make them suitable for integration into
compact RFICs.

23. What is the trade-off between conversion loss and CO 3 K2

linearity in diode-ring mixers, and why is it important in
RFIC design?
Diode-ring mixers often exhibit a trade-off between conversion
loss (signal attenuation) and linearity. Optimizing this trade-off is
crucial to balance signal amplification and distortion in RFICs
24. What are some common applications of diode-ring CO 3 K2
mixers in RFIC design?
Diode-ring mixers are commonly used in RFICs for applications
such as frequency downconversion, upconversion, modulation,
and demodulation in wireless communication systems.

25. How do diode-ring mixers contribute to reducing CO 3 K2

power consumption in RFICs?
Diode-ring mixers are known for their low power consumption due
to their simplified circuitry. This makes them suitable for battery-
operated RFICs where power efficiency is essential.

39
PART A CO’S Blooms
Level
26. What challenges might RFIC designers face when CO 3 K2
using diode-ring mixers, and how can these
challenges be addressed?
Challenges may include achieving high linearity and
minimizing spurious products. These can be addressed
through careful biasing, component selection, and circuit
design techniques to meet specific RFIC performance
requirements..
27. What is the primary difference between active and CO 3 K2
passive mixers in RFIC (Radio Frequency Integrated
Circuit) design?
The primary difference between active and passive mixers is
that active mixers use active components like transistors or
operational amplifiers to perform frequency conversion,
amplification, and mixing, while passive mixers use passive
components like diodes, transformers, or baluns to perform
frequency conversion without amplification. Active mixers
are capable of providing gain, while passive mixers usually
have lower conversion loss.
28. In what situations might a passive mixer be CO 3 K2
preferred over an active mixer in RFIC design?
Passive mixers are preferred in situations where the primary
goal is to minimize power consumption and complexity.
Passive mixers are simpler in design and consume less
power compared to active mixers. They are often chosen in
battery-operated devices or when low power consumption is
critical.
29. What is the primary disadvantage of using an CO 3 K2
active mixer in RFIC design, and how can it be
mitigated?
The primary disadvantage of using an active mixer is its
higher power consumption compared to passive mixers. This
can lead to reduced battery life in portable devices. To
mitigate this, RFIC designers can implement power-saving
techniques, such as optimizing biasing conditions,
employing low-power transistors, or using power
management circuits.
30. What is the primary factor that determines the CO 3 K2
choice between an active or passive mixer in RFIC
design?
The primary factor that determines the choice between an
active or passive mixer in RFIC design is the specific
application's requirements. If signal amplification is
necessary, an active mixer is preferred. On the other hand,
if power efficiency and simplicity are more important, a
passive mixer may be the better choice. Designers consider 40
factors like power consumption, gain, noise figure, and cost
to make an informed decision.
6.6 PART B Qs
(with K level &
CO)

41
 PART B CO’S Blooms
Level

1. Explain the concept of frequency conversion in the context of RF CO 3

mixers. K2
2. Briefly explain the mixer design considerations. CO 3
K3
3. Discuss in detail about sub sampling mixers CO 3
K2
4.Explain about mixer design and sub sampling mixers. CO 3 K2

5. Explain the commonly used mixer topologies in RFIC design, such CO 3 K2

as Gilbert cell mixers, double-balanced mixers, and single-balanced
mixers..
6. How do single-balanced mixers differ from double-balanced CO 3
mixers in terms of complexity and performance trade-offs? K2

7. Explain the significance of balanced operation in double- CO 3 K2

balanced mixers, and why is it important in RFICs?

8. What role does linearity play in the selection of a mixer CO 3

topology for RFICs, and how do Gilbert cell mixers address K2
this aspect?
9. Brief overview of the importance of mixers in RFIC design CO 3
and introduce the focus of the discussion on diode-ring K2
mixers.

10. Describe the fundamental principles underlying the CO 3

operation of diode-ring mixers in RFICs. K2

11. Explain how diode-ring mixers utilize diodes arranged in a CO 3

ring configuration to achieve frequency conversion. K2
12. Examine the principles, advantages, limitations, and CO 3
design considerations of sub-sampling mixers in RFIC design. K2
Discuss their operating mechanisms, key performance
parameters, trade-offs, and applications in detail.?
13. Analyze the fundamental principles, benefits, constraints, CO 3 K3
and critical design factors associated with sub-sampling
mixers in the context of RFIC design.
14. Explore the underlying principles, advantages, challenges, CO 3 K2
and design considerations of multiplier-based mixers in RFIC
design.

15. Investigate the fundamental principles, benefits, hurdles, CO 3 K4

and factors to consider when designing multiplier-based
mixers for RFICs. Provide a comprehensive explanation of
how these mixers operate, the advantages they offer,
potential challenges, and key design considerations.

42
 6.7.Supportive online Certification courses

UNIT I :

NPTEL/SWAYAM:
1 .VLSI Interconnects

By Prof. Sarang Pendharker | IIT Kharagpur

8 Weeks

https://onlinecourses.nptel.ac.in/noc23_ee135/preview

2. Coursera: RFIC Design Specialization (University of California, San Diego)

This specialization includes multiple courses covering various aspects of RFIC

design, from fundamentals to advanced topics. It covers amplifiers, mixers,
filters, and more.

Website: Coursera - RFIC Design Specialization

NPTEL :: Electronics & Communication Engineering - RF Integrated Circuits

3. RFIC Design Courses on Udemy

Udemy offers various courses on RFIC design, ranging from introductory to

advanced levels. You can explore different courses based on your specific
interests and skill level.

Website: Udemy - RFIC Design Courses

RFIC Unit Lecture : Basic concepts in RF Design - YouTube

43
6.8 Real time Applications in day to day life and
to Industry

RF integrated circuit (RFIC) design has a wide range of real-time applications in

both everyday life and various industries. Here are some examples of how RFICs
are used:
1. Wireless Communication Devices:** RFICs are essential components in
smartphones, tablets, and other wireless devices. They enable wireless
connectivity, such as Wi-Fi, Bluetooth, cellular, and NFC, allowing people to make
calls, access the internet, and exchange data wirelessly.
2. Internet of Things (IoT):** RFICs are at the heart of IoT devices, which include
smart thermostats, wearable fitness trackers, and home automation systems.
These devices rely on RFICs for wireless communication and connectivity to the
internet.
3. Automotive Industry:** Modern vehicles incorporate RFICs for various
applications, including keyless entry systems, tire pressure monitoring, GPS
navigation, collision avoidance radar, and infotainment systems. RFICs enable
wireless communication within the vehicle and with external networks.
4. Healthcare:** Medical devices like remote patient monitoring systems, RFID-
based asset tracking in hospitals, and wireless communication in medical
implants all use RFICs to ensure seamless data transmission and connectivity.
5. Aerospace and Defense:** RFICs are critical in radar systems, satellite
communications, electronic warfare systems, and avionics. They are used for
communication, navigation, and surveillance in aircraft and spacecraft.
6. Industrial Automation:** In industrial settings, RFICs enable wireless
communication for industrial sensors, monitoring equipment, and process control
systems. They help improve automation, reduce downtime, and enhance overall
efficiency.
7. RF Test and Measurement Equipment:** RFICs are used in spectrum analyzers,
vector network analyzers, and signal generators, which are essential tools for
testing and characterizing RF circuits and devices.
8. RFID Technology:** Radio-frequency identification (RFID) systems use RFICs
for tracking and identification purposes. They are used in supply chain
management, inventory tracking, access control, and contactless payment
systems.
9. Telecommunications Infrastructure:** RFICs play a crucial role in the
infrastructure of cellular networks, base stations, and communication satellites,
ensuring reliable and high-speed data transfer.
10. Consumer Electronics:** RFICs are found in various consumer electronics,
including TV remote controls, wireless headphones, gaming consoles, and smart
home devices, enabling wireless connectivity and remote control.
11. Security Systems:** RFICs are used in security systems for wireless
surveillance cameras, motion detectors, and access control systems. They
facilitate communication between security devices and central monitoring
stations.

44
6.9. Contents beyond the Syllabus

5G and 6G Enhanced Broadband

Communications

45
Integrated RFIC on-chip and GPS antenna with human
body for wrist and wearable communication applications

A GPS and RFIC-on-chip antenna in 0.18 um CMOS 1p6M process is

presented. The HFSS 3-D EM simulator is employed for design
simulation. A printed 1.575 GHz and 2.4 GHz antenna has been
realized by using the CMOS RFIC-on-chip. The measured VSWR is
less than 2 from 1.575 GHz and 2.4- to 2.483-GHz. This propose
super quadric combo antenna in free space, electromagnetic
coupling between super quadric antenna and human body and
rectangular antenna for wrist watch type wireless communication
applications. The measured phase distribution of the input
impedance is quite linear and the H-plane patterns are almost
omnidirectional and field tried GPS integration. In addition, in order
to improve the way controlling this provide switch by software, a
novel circuit structure which will control antenna pattern switching
automatically by hardware is also developed for wireless healthcare
and mobile biomedical application. RFIC-on-chip GPS and antenna
also merger T/R-Switch design on single chip solution for 2.4GHz
CMOS transceiver RF front-end. The old man can monitor healthcare
and transfer to health center or passive devices by with software and
show GPS location for wearable ambulatory application.

46
 7. Assessment Schedule

Assessment Proposed Date Actual Date

Unit 1 Assignment
Assessment
Unit Test 1

Unit 2 Assignment
Assessment
Internal Assessment 1

Retest for IA 1

Unit 3 Assignment
Assessment
Unit Test 2

Unit 4 Assignment
Assessment
Internal Assessment 2

Retest for IA 2

Unit 5 Assignment
Assessment
Revision Test 1

Revision Test 2

Model Exam

Remodel Exam

University Exam

47
8. Prescribed Text Books & Reference Books

TEXT BOOKS:
1. 1. Thomas H. Lee, Cambridge, The Design of CMOS Radio-Frequency
Integrated Circuits, UK: Cambridge University Press, 2004

2. Phillip E. Allen and Douglas R. Holberg- CMOS Analog Circuit Design

Oxford

University Press -3rd Ed., -2011

3. REFERENCES:

4. Behzad Razavi, RF Microelectronics, Prentice Hall, 1998.

5. Ludwig, Rf Circuit Design, 2nd Ed., Pearson,2011.
6. Bosco H Leung VLSI for Wireless Communication, Pearson Education,
2002
7. Behzad Razavi, Design of Analog CMOS Integrated Circuits, Mcgraw-
Hill, 1999
8. Jia-Sheng Hong, Microstrip Filters for RF/Microwave Applications,
Wiley, 2001

48
9. Mini Project suggestions

MINI PROJECTS LIST

S.No Name of The Project

1. Stereo Audio Mixer

2. Channel Audio Mixer using a Single Op Amp

3. Channel Mixer with Tone Control

4. Single Transistor Audio Mixer

5 MIC Mixer Circuit

6. Stage DJ Mixer

7. Balanced Microphone Preamplifier

8. Channel Stereo Headphone Amplifier

9. Audio Equalizer Circuit

10. Low Power Audio Amplifier

49
 Thank you

Disclaimer:

This document is confidential and intended solely for the educational purpose of RMK Group of
Educational Institutions. If you have received this document through email in error, please
notify the system manager. This document contains proprietary information and is intended
only to the respective group / learning community as intended. If you are not the addressee
you should not disseminate, distribute or copy through e-mail. Please notify the sender
immediately by e-mail if you have received this document by mistake and delete this
document from your system. If you are not the intended recipient you are notified that
disclosing, copying, distributing or taking any action in reliance on the contents of this
information is strictly prohibited.

50