Tx and Rx Architectures

Purpose
◼ Review of Digital Modulation
◼ Transceiver Architectures
Review of Digital Modulation
Base band Pass band

2-level 4-level

▪ Detectability: the quality of the demodulated signal under channel

attenuation + receiver noise
▪ Bandwidth Efficiency: the bandwidth (e.g. 25 MHz for GSM) occupied
by the modulated carrier for a given information rate in the baseband
signal
▪ Power Efficiency: the type of PA that can be used in the transmitter
ASK,PSK, FSK

ASK

PSK
PSK
 Pulse Shaping
▪ Pulse is designed to occupy a smaller bandwidth.

Raised Cosine
:0.3 to 0.5
 Signal Constellation
“Signal constellation” allows us to visualize the modulation.

PSK
Ideal Noisy

Noise corrupts the amplitude for both ZEROs and ONEs.

ASK
 FSK

Ideal
Noisy

EVM

The constellation is used to measure the deviation of

the constellation points from their ideal positions due to
the Noise Impairment, which is called “Error Vector
Magnitude” (EVM)”
Quadrature Modulation (QPSK)

Constellation Diagram

▪ Occupies half
of the
bandwidth
▪ Pulses at the I
and Q are
called symbols
(2Tb)
▪ QPSK suffers
from large
phase change
(900/1800)
Offset QPSK (OQPSK)

▪ Offset quadrature
phase-shift keying
(OQPSK) is a
variant of phase-
shift keying
modulation using
four different
values of the phase
to transmit.
▪  900 degree
change
▪ Not pass through
origin
/4 QPSK
▪ This variant of
QPSK uses
two identical
constellations
which are
rotated by
45° with
respect to one
another.
 Time Waveforms (/4-QPSK)

▪ Maximum phase step is 1350 compared with 1800 in QPSK

▪ QPSK and its variants provide high spectral efficiency but they
are required to be linear PA
 GMSK & GFGSK

▪ Gaussian minimum shift keying (GMSK), modulation index m = 0.5

▪ Gaussian frequency shift keying (GFSK), modulation index m = 0.3
 QAM
▪ QAM allows four possible amplitudes for sine and cosine, ±1, ±2
 Constellation Diagram

▪ Saves bandwidth
▪ Denser constellation: making detection more sensitive to noise
▪ Large envelope variation: need highly linear PA
 Impact on RF Circuit
◼ Conventional QPSK has transitions
through zero (i.e. 1800 phase change).
Highly linear amplifiers required.
◼ In Offset QPSK, the phase transitions
are limited to 900.
◼ In /4 QPSK the set of constellation
points are toggled at each symbol, so
transitions through zero cannot occur.
◼ All QPSK schemes require linear power
amplifiers, which are not power
efficient
Transceiver Architectures
General Conditions

▪ The BPF must provide a very high Q

 Tx-Rx Feedthrough
Heterodyne Receiver

▪ In full-duplex standards, the Tx and the Rx operate concurrently.

▪ With a 1-W Tx power, the leakage sensed by LNA can reach -20dBm,
dictating a substantially higher Rx compression point.
 Receiver Architectures
Heterodyne Receiver

▪ “Heterodyne” receivers employ an LO frequency unequal to ωin and

hence a nonzero IF
▪ Mixer performs down conversion.
 Heterodyne Receiver

▪ Constant LO: each RF channel is downconverted to a different IF channel

▪ Constant IF: LO frequency is variable, all RF channels within the band of

interest translated to a single value of IF.
 Image Frequency
Problem!

If LO<IN --low-side injection

If LO>IN --high-side injection

▪ Two spectra located symmetrically around ωLO are downconverted to the IF

 Solution !
▪ The most common approach is to precede the mixer with an “Image-
reject Filter”

▪ A high IF allows substantial rejection of the image.

Image Rejection vs Channel Selection

 Cont…

High IF

▪ A high IF allows substantial rejection of the image.

Low IF

▪ A low IF helps with the suppression of in-band interferers.

 Solution !
▪ The most common approach is to precede the mixer with an “Image-
reject Filter”

▪ A high IF allows substantial rejection of the image.

 Dual Down-conversion (I)

▪ The front-end filter selects the band while providing some image rejection
as well (Point B)
Dual Down-conversion (II)

▪ After amplification and image-reject filtering, spectrum of C obtained

▪ Sufficiently linear mixer translates desired channel and adjacent
interferers to first IF (Point D)

▪ Partial channel selection BPF3 permits the use of a second mixer with
reasonable linearity. (Point E)
▪ Spectrum is translated to second IF. (Point F)

▪ BPF4 suppresses the interferers to acceptably low levels (Point G)

▪ An optimum design scales both the noise figure and the IP3 of each
stage according to the total gain preceding that stage.
 Mixing Spurs
RF Mixer IF Mixer

▪ RF input multiplied by LO harmonics, produces unwanted

interference. If an interferer is downconverted to the same IF,
it corrupts the signal
▪ Generally, more than two down conversion steps are not used, so
as not to complete the design.
▪ Mixing spur from additional down conversion mixers become
difficult to manage.
 Modern Heterodyne Rx
Zero Second IF
▪ To avoid secondary image, most modern heterodyne receivers
employ a zero second IF.
▪ In this case, the image is the signal itself. No interferer at
other frequencies can be downconverted as an image to a zero
center frequency if ωLO2=ωIF1
Symmetrically/Asymmetrically
-Modulated Signal

AM Signal FM Signal

▪ AM signals are symmetric, FM signals are non-symmetric.

▪ Most of today’s modulation schemes, e.g., FSK, QPSK, GMSK,
and QAM, exhibit non symmetric spectra around carrier
frequency.
 Self Corruption !
▪ With zero second IF superimposes
two copies of the signal
▪ If the signal spectrum is
non-symmetric, the original signal
spectrum is corrupted

Solution 1 !
Minimum IF required to
avoids self-corruption of
non-symmetric signals?
 Solution 2!
▪ By creating two versions of the downconverted signal that have
a phase difference of 900

Quadrature downconversion
Sliding-IF Receivers (SIR)
▪ Modern heterodyne receivers employ only one oscillator
▪ The second LO frequency is therefore derived from the first by
“frequency division”
 SIR:  2 Circuit
▪ Such divide-by-2 topology can produce quadrature output
▪ The second LO waveforms at a frequency of fLO1/2
 SIR: Properties

Fractional Bandwidth:

RF
IF
Direct-Conversion Rx(DCR)
Zero IF/Homodyne ▪ However,
for narrow
channel BW,
DCR is not
employed
due to
flicker noise
problem.

▪ Low IF
offer better
solution e.g.
GSM
▪ Absence of an image greatly simplifies the design of such Rx
▪ Channel selection is performed by on-chip low-pass filter
▪ Mixing spurs are considerably reduced in number
 LO Leakage

▪ LO couples to the antenna through:

✓ Device capacitances: LO and RF ports of mixer + output and input
ports of the LNA
✓ The substrate to the input pad, especially because the LO employs
large on-chip spiral inductors
 LO Leakage Cancellation

▪ LO leakage can be minimized through symmetric layout of the

oscillator and the RF signal path
▪ LO leakage arises primarily from random or deterministic
asymmetries in the circuits and the LO waveform
 DC Offset

▪ A finite amount of in-band LO leakage appears at the LNA

input. Along with the desired signal, this component is amplified
and mixed with LO.
▪ DC offset might saturates baseband circuits, thus prohibiting
signal detection.
DC Offsets--Cancellation
Returned current by -GmF to node X
=–GmF Vout=(R1C1s + 1) and the current
due to Gm1 =Gm1Vin . The sum the two
at node X, multiply the sum by RD and
A1, and equate to Vout
DC Offsets--Cancellation

▪ Offset cancellation: High-pass Filter; But removes the signal’s

near zero frequency, thus introducing Inter Symbol Interference

Small Modulation Large Modulation

▪ Modulation schemes that contain little energy at fc could be “ac”

coupled but it slow down the transient response.
DC Offsets--Cancellation
▪ The Negative-Feedback loop realized using ADC & DAC
converges such that Vout is minimized closed to zero. But high
resolution ADC & DAC are required to achieve more accurate
results.
 Even-Order Distortion
▪ Direct-conversion receivers are additionally sensitive to even-
order nonlinearity in the RF path, and so are heterodyne
architectures having a second zero IF.

▪ RF input of the mixer may appear at the output without

frequency translation, corrupting the downconverted signal. The
beat generated by can be removed by ac coupling
Second Order Intercept IP2
If Vin(t) = A cos ω1t + A cos ω2t, then the LNA output is given by

Beat amplitude grows with the square of the amplitude of the input tones.
 Feedthrough—Solu.1
▪ If the output is sensed differentially, the RF feedthroughs in
Vout1(t) and Vout2(t) are cancelled while the signal components
add. However, this cancellation is sensitive to asymmetries.
 I/Q for IF=0
▪ Separation into quadrature phases can be accomplished by
shifting
✓ RF/LO signal waveform by 90°.

More preferred Option!

 Mismatch in I/Q
▪ Error in quadrature mixers result in imbalances in the
amplitudes and phases of the baseband I and Q outputs.
▪ Quadrature error tend to be larger in direct-conversion
receiver than in hetero-dyne topologies
Mismatch in I/Q (Cont.)
▪ RF & LO tend to suffer higher mismatches at high frequencies

▪ Delay =10ps RF & LO tend to suffer higher mismatches at high

frequencies
Mismatch in I/Q (Cont.)

Where a and b are either -1 or +1.

Mismatch in I/Q (Cont.)
Let us examine the results for two cases:
(i) ε ≠ 0, θ = 0:The quadrature baseband symbols are scaled differently
in amplitude.

(ii) ε = 0, θ ≠ 0 : The constellation is compressed along one diagonal and

stretch along other.
Mismatch in I/Q (Cont.)
The phase and gain
are corrected (by
applying RF tone)
and observing the
baseband sinusoids
in the analog or
digital domain.

With the mismatches

known, the received
signal constellation is
corrected (by circuit
arrangement shown in
the figure) before
detection.
Image-Reject Receiver--(IRR)
▪ Heterodyne and direct-conversion receivers have revealed
various pros and cons such as:
✓ Heterodyne must deals with the image and mixing spurs
✓ Direct conversion, with the even-order distortion and flicker
noise
Mixing Spur
Hetero-dyne Receiver

Direct-conversion Receiver

▪ Suppress the image without using the filter

 Shift-by-90 0
▪ Impulse at +ωc (Carrier) is rotated
clockwise and that at –ωc
counterclockwise.
▪ Multiplication by j rotates the
impulses by 90° counterclockwise.

▪ Frequency domain:

▪ Shift-by-90 ° operation is also

called the “Hilbert transform”.
 Shift-by-90 0 (Cont.)

The HPF and LPF transfer functions are given by:

HHPF-HLPF =/2 at all

frequencies and any choice of
R1 and C1.
 Shift-by-90 0 (Cont.)
Negative of Hilbert Transform
High Injection
fc<fLO

Sine term

fc>fLO Low Injection

Cos term

Sine term

High Injection Hilbert Transform

▪ The RF input is mixed with the quadrature phases of the LO so

as to translate the spectrum to a non-zero IF.
Spectrum of I+Q

Signal Component

Image Component
 Hartley Architecture
1888-1991

▪ If we shift I(t) or Q(t) by another 90 °before adding them, the

image can be removed.

▪ The low-pass filters are inserted to remove the unwanted

high-frequency components generated by the mixers.

▪ The Hilbert Transform of the Hilbert transform negates the signal, if I

or Q is shifted by another 90 °before adding them, the image is removed.
Hartley’s Architecture (Cont.)
B C

x(t) = Asig cos(ωct + Φsig) + Aim cos(ωimt + Φim)

▪ Signals at “A+C” yields doubling the signal amplitude and

cancellation of image signals
Shift-by-90 ..(Hartley)
0
▪ The 90 ° phase shift depicted in Hartley is typically realized as
a +45 ° shift in one path and -45 ° shift in the other.

▪ This is because it is difficult to shift a single signal by 90 °while

circuit components vary with process and temperature.

IF-to-Zero
 Drawbacks..(Hartley)
▪ The principal drawback of the Hartley architecture stems from its
sensitivity to mismatches i.e. amplitude error, ε, and phase error,
Δθ etc.

▪ Image power divided by signal power called “Image Rejection Ratio

(IRR) ”
▪ Average Power of the signal “a cos(t+)+b cos(t)”= a2+2ab cos
+b2)/2.
 Weaver Architecture
1894-1978
▪ The Weaver receiver, derived from its transmitter counterpart,
avoids those issues in Hartley architecture.

▪ Mixing a signal with quadrature phases of an LO takes the Hilbert

transform. As shown above, the Weaver architecture replaces the
90°phase shift network with quadrature mixing.
Weaver Architecture..(Analysis)

Assume Low-side Injection for both mixing stages (RF & IF Spectra)

(Image Removed)
 Secondary Image ..(Weaver)
▪ The Weaver architecture must deal with a secondary image if
the 2nd IF0.
▪ Suppose if 2ω2 - ωin + 2ω1 accompanies the RF signal, down
conversion of this signal to the first IF translates this to image
of the signal with respect to ω2 i.e. mixing with ω2 brings it to
the same IF at which the signal appears.
Double Quadrature ..(Weaver)
▪ The second downconversion preferably produces a zero IF, in
which case it must perform quadrature separation as well.
▪ Hartley/Weaver must include calibration (i.e. gain/phase error
cancellation) to ensure image to remain down by more than
40dB.
 Low-IF Rx
▪ It is undesirable to place the image within the signal band. If
LO is kept at the edge of the desired (200-kHz) channel, then
IF=100KHz
▪ The 1/f noise penalty is much less severe. Also, on-chip high-
pass filtering of the signal becomes feasible for a narrow-
channel standard e.g. GSM

Example
 Quadrature Phase @RF
▪ To obtain high IRR, one possible solution is to move the 90 °phase
shift in the Hartley architecture from the IF path to the RF path.

▪ The RC-CR network is centered at a high frequency and can

maintain a reasonable IRR across the band.
Quadrature Phase@Digital domain
▪ Another possibility of the low-IF architecture is shown below,
the downconverted signals are applied to channel-select filters
and amplifiers as in a direct-conversion receiver.
▪ The results are then digitized and subjected to a Hilbert
transform in the digital domain before summation.

Digital domain
Polyphase Filters/Networks

▪ Vout can be viewed as a weighted sum of V1 and V2

Special Case:

▪ The circuit computes the vector sum of V1 and V2 (= -jV1)

 Special Case (Cont.)

▪ C1 rotates V2 by another 90 °so that the result cancels the

effect of V1 at the output node

Differential Form

R1 = R2 = R C1 = C2 = C
Differential Form (Cont.)
Use the concept developed earlier to construct Image Reject Receiver

For the circuit shown in Figure, show that Vout is free from Image
Differential Quadrature Phases

▪ The TOP and BOTTOM produces quadrature outputs

▪ The LEFT (45 °more negative than +jV1) and the RIGHT
(45 °more negative than -jV1).
Differential Quadrature Phases (Cont.)
▪ Vsig+Vim , -Vsig-Vim , +jVsig - jVim and -jVsig + jVim

Signal Image
Added Removed
 Cascaded Polyphase
The output signal and image components exhibit opposite sequences.
Therefore, if the polyphase filter is followed by another, then the
image is further suppressed.

(i) How the loading of the second stage on the first is accounted?
(ii) How are the RC values chosen in the two stages?
Ref: Razavi Book; Page No. 222-226
Transmitter Architectures
General Conditions
Emphasis is on MOS circuits for
transceiver (radio).
⧫ Transmitter
✓ PA
✓ Up-Mixer
✓ Modulator etc.
⧫ Receiver
✓ LNA
✓ Down Mixer
✓ De-modulator etc.

LNA and PA are called key FRONT BLOCKS and

operate >1GHz.
Transmitter Architectures
General Conditions
Transmitter
▪ An RF transmitter performs modulation, up conversion, and power
amplification.
▪ Voice message from microphone is quantized, digitized and
modulated…,THEN frequency is raised through mixing and
composite signal power level is INCREASED through power
amplifier (PA) and finally broadcast through ANTENNA
Transmitter Architectures
▪ Each incoming pulse is mapped to the desired shape (raised
Cosine in case of GMSK) by a combination of digital and analog
techniques:

▪ Modulator used is a Quadrature as shown in the figure above

 Direct Conversion Tx
▪ This topology directly translates the baseband spectrum to the
RF carrier by means of a “quadrature upconverter”.
▪ Matching Network filter out of band components that results
from the PA non-linearity
▪ Signal levels are generally large (several hundred millivolts),
therefore noise is less critical in transmitter than receiver
 I/Q Mismatch
The I/Q mismatch in DCRs results in “cross-talk” between the
quadrature baseband outputs or, equivalently, distortion in the
constellation.

For the four points in the constellation:

 I/Q Mismatch (Cont.)
Apply two tones V0 cosωint and V0 sinωint to the I and Q inputs and
examining the output spectrum.

The power of the unwanted sideband at ωc-ωin divided by that of

the wanted sideband at ωc + ωin is given by
I/Q Mismatch (Cont.)
Case-I: When single tone V0 cosωint is applied
 I/Q Mismatch (Cont.)
Case-II: Applying a sinusoid to one baseband input while the other
is set to zero.
 I/Q Mismatch (Cont.)
Case-II: applying a cosinusoid to one baseband input while the
other is set to zero.

Gain mismatch can be adjusted so as to drive this difference to zero.

 Carrier Leakage
▪ The analog baseband circuitry producing the I/Q signals in the
transmitter also exhibits dc offsets.

▪ Carrier Leakage will lead to adverse effects like distorting the

signal constellation and power control.
 Carrier Leakage (Effect)
▪ It distorts the signal constellation, raising the error vector
magnitude (EVM) at the TX output.

QPSK Signal

▪ The baseband quadrature outputs suffer from dc offsets, i.e.,

horizontal and vertical shifts in the constellation.
Carrier Leakage (Effect)
▪ The output power of the transmitter is varied across a wide
range by varying the amplitude of the baseband signals.

▪ With a short distance (between mobile & base), due to carrier

leakage, its power dominates, making difficult to measure the
signal power.
 Carrier Leakage (Reduction)
▪ The loop consisting of the TX, the detector, and the DACs
drives the leakage toward zero, with the final settings of the
DACs stored in the register.
 Tx Linearity
▪ The distortion of a variable-envelope
signal is typically characterized by the
compression that it experiences.

▪ Maximize the gain of the PA

helps to minimize the output
swing.
 Oscillator Pulling

▪ The PA output exhibits very large swings (20Vpp for 1W

delivered to a 50- load), which couple to various parts of the
system through the silicon substrate.
▪ Therefore, an appreciable fraction of the PA output couples to
the LO.
 Direct Conversion Tx

▪ Most of today’s direct-conversion transmitters avoid an

oscillator frequency equal to the PA output frequency.
▪ This architecture is popular as it reduces injection pulling to a
great extend
▪ Proper balanced layout can suppress this effect
 Heterodyne Tx
▪ Perform the signal upconversion in two steps so that the LO
frequency remains far from the PA output spectrum

▪ One advantage of this architecture is that the I/Q upconversion

occurs at a significantly lower frequency than the carrier,
exhibiting smaller gain and phase mismatches.
 Sliding-IF Tx
▪ In analogy with the sliding-IF receiver architecture, we
eliminate the first oscillator in the above TX and derive the
required from the second oscillator
 A Single Chip Radio

Message Input
Antenna

Message Output

Example
Front-end block LMX3162 Single Chip
although shown Radio Transceiver is
separately but a monolithic, integrated
very much part of radio transceiver
SINGLE CHIP optimized for use in ISM
2.4 GHz wireless
RADIO
system fabricated using
BiCMOS
Concluding Remarks!
CMOS is used in developing
Personal Wireless System (1-10
GHz)

What makes a transceiver?

Various aspects have been
covered.
 Status Report!
Coverage>99% Finished PART

Understand key
concepts related to RF
1 Understand the
trade-off
among different
2
transceiver
architecture
COs used in RF
Understand the systems
design of key
blocks of a
transceiver keeping
3 Analyze and design
in view the system 4 RF circuits from a
requirements using given System Level
CAD Tool Unfinished AGENDA Specifications