31
2007.09.13 RFIC WS, A. Matsuzawa 1 Matsuzawa & Okada Lab. Matsuzawa & Okada Lab. Direction of RF-CMOS tuner technology Akira Matsuzawa Department of Physical Electronics Tokyo Institute of Technology

Direction of RF-CMOS tuner · PDF fileDirection of RF-CMOS tuner technology ... PRML Read Channel Servo DSP Analog Front End Analog ... 1/10 ~1/20 pcs 1

Embed Size (px)

Citation preview

2007.09.13 RFIC WS, A. Matsuzawa

1

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Direction of RF-CMOStuner technology

Akira Matsuzawa

Department of Physical ElectronicsTokyo Institute of Technology

2007.09.13 RFIC WS, A. Matsuzawa

2

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

SoC for DVD systems

World first fully integrated mixed signal SoC for DVD systems has developed.

Variable Gain Amp.Variable

Gain Amp.Analog

FilterAnalog

FilterA to D

ConverterA to D

ConverterDigital

FIR FilterDigital

FIR FilterViterbiError

Correction

ViterbiError

Correction

ClockRecoveryClock

RecoveryVoltage

ControlledOscillator

Voltage ControlledOscillatorData In

(Erroneous)

Data Out(No error)

Analog circuit

Digital circuit

Pickup signal

Processed Signal

Variable Gain Amp.Variable

Gain Amp.Analog

FilterAnalog

FilterA to D

ConverterA to D

ConverterDigital

FIR FilterDigital

FIR FilterViterbiError

Correction

ViterbiError

Correction

ClockRecoveryClock

RecoveryVoltage

ControlledOscillator

Voltage ControlledOscillatorData In

(Erroneous)

Data Out(No error)

Analog circuit

Digital circuit

Pickup signal

Processed Signal

PixelOperationProcessor

PixelOperationProcessor

IOProcessor

IOProcessor

AVDecode

Processor

AVDecode

Processor

Back -EndBack -End

SystemCont-roller

SystemCont-roller

CPU1CPU1CPU2CPU2

VCOVCO

ADCADC

Gm-CFilterGm-CFilter

PRMLRead

Channel

PRMLRead

ChannelServo DSPServo DSP

AnalogFront EndAnalog

Front End

Front-EndFront-EndAnalog FE+Digital R/C

PixelOperationProcessor

PixelOperationProcessor

IOProcessor

IOProcessor

AVDecode

Processor

AVDecode

Processor

Back -EndBack -End

SystemCont-roller

SystemCont-roller

CPU1CPU1CPU2CPU2

VCOVCO

ADCADC

Gm-CFilterGm-CFilter

PRMLRead

Channel

PRMLRead

ChannelServo DSPServo DSP

AnalogFront EndAnalog

Front End

Front-EndFront-EndAnalog FE+Digital R/C

Okamoto, et. al., ISSCC 2003.0.13um CMOS, 24Mtr

Digital signal processing enables perfect cure for the damaged signals.

Digital read channel technology

DVD

7b, 400MHz, 40mW ADChas been developed

My last work in Panasonic

Before talking about RF-CMOS for tuners

2007.09.13 RFIC WS, A. Matsuzawa

3

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Power of SoC

Model;2000 Model; 2003Before SoC After SoC

SoC has enabled performance increase and cost decrease.Many components and ICs have been kicked out from the PC board.

Panasonic DVD recorder

2007.09.13 RFIC WS, A. Matsuzawa

4

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Contents

• Conventional AM/FM tuner• Analog-centric CMOS tuner• Digital-centric CMOS tuner• Feature of CMOS technology• Conclusion

E-mail: [email protected]: http://www.ssc.pe.titech.ac.jp/

Courtesy Niigata-Seimitsu Co., Ltd.

2007.09.13 RFIC WS, A. Matsuzawa

5

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Current AM/ FM tuner system

Bipolar IC = 1 (RF)CMOS IC = 2 (PLL, RDS)External Components=187 12 adjustment points

AM/FM Tuner for home use

Current AM/FM tuner uses 3 ICs and large # of external components.Furthermore 12 adjustment points are needed.

Large # of products, but not expensive product.More efforts for the cost reduction are still needed.

2007.09.13 RFIC WS, A. Matsuzawa

6

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Block Diagram of Current FM/AM tuner

FMLNA

FMMIX

FM IFBPF

LIMITER FMDEMOD

STEREODECODER

LOCALOSC(AM)

LOCALOSC(FM)

AMLNA

AMMIX

AM IFBPF

AM IFA AMDEMOD

SERIALINTERFACE

FREQUENCYSYNTHESIZER

RDSDECODER

SW LEFT

RIGHT

AM Bar Antenna and Varactor

LO inductorand Varactor

AM IFT andCeramic filter

FM AntennaTunig L and varactor

FM inter-stageTunig L and varactor

FM IFT andCeramic filters

De-couplingCapacitors

Ceramic resonatorfor Stereo decoderand LPF for PLL

-VCC

De-couplingCaps for amplifier

LPF for Synthesizer

AGC smoothingCapacitor

RSSILevel

XtalElement

forSynthe.

shows alignment required

Large # of external components. They should be integrated on a chip.

2007.09.13 RFIC WS, A. Matsuzawa

7

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

External parts used in existing IC

System clock, Reference for PLL synthesizerXtal Osc. element

FM: Single conversion super heterodyne. IF=10.7MHzAM: Single or Double conversion super heterodyne IF=450KHz or

10.7MHz + 450KHz

System

RSSI level alignment, volume controlSemi-fixed and Variable resistor

RF tuning, Local oscillatorVariable capacitance

Home tuner and radio cassette tuner : around 165pcsCar tuner : 80 to 130pcs

Total number of external parts

FM and AM IF BPF for channel filter Ceramic filterNoise canceller, LPFAnalog filter

RF tuning, local oscillator, IF transformer, FM detectorInductor

AGC smoother, power-ground decouplingElectrolytic capacitor

RF bypass, coupling, de-coupling Ceramic capacitorSmall value capacitor

AGC, bias, LPF for PLLResistor

Blocks to be usedExternal Parts

Large # of external components are needed due to analog signal processing.

2007.09.13 RFIC WS, A. Matsuzawa

8

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Issues and conventional solutions of AM/FM tuner

Lower frequency AM: 522 KHz to 1710 KHzSW: 2.3MHz to 26MHzFM: 87.5 to 108 MHz

Larger Inductance and capacitance

High dynamic range

Serious 1/f noise

Sharp and fine filter

AM: 14 dBuV to 126 dBuVFM: 0 dBuV to 126 dBuV

High linearity ckt.

Bipolar

External components

External filters (Ceramic)

External varactorsBipolar

Application of CMOS technology to AM/FM tuner looks very difficult,due to lower frequency and high dynamic range.

2007.09.13 RFIC WS, A. Matsuzawa

9

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

1st trial by CMOS technology

FMLNA

FMMIX

FM IFBPF

LIMITER FMDEMOD

STEREODECODER

LOCALOSC(AM)

LOCALOSC(FM)

AMLNA

AMMIX

AM IFBPF

AM IFA AMDEMOD

SERIALINTERFACE

FREQUENCYSYNTHESIZER

RDSDECODER

SW LEFT

RIGHT

AM Bar Antenna and Varactor

FM inter-stageTunig L and varactor

-VCC

AGC smoothingCapacitor

RSSILevel

XtalElement

forSynthe.

FMDemod

FM IFT andCeramic filters

AM IFT andCeramic filter

De-couplingCapacitors

Ceramic resonatorfor Stereo decoderand LPF for PLL

FM AntennaTunig L and varactor

LO inductorand Varactor

LPF for Synthesizer

De-couplingCaps for amplifier

Can be integrated on a chip

1st trial to realize AM/FM tuner by CMOS technology,many external components should be reduced.

2007.09.13 RFIC WS, A. Matsuzawa

10

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Result of analog-centric CMOS tuner

External components 187 69

Characteristics is affected by process variation easily.Element mismatch causes DC offset, noise, distortion, and low filter performance.The reduction of # of external components is not attractive for users.

2007.09.13 RFIC WS, A. Matsuzawa

11

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Analog-centric CMOS tuner technology

Needs large capacitor for low audio frequency

Time division charge and dischargeAGC smoother

ProblemsMethods for on-chipParts

Poor THD (0.5%)Pulse count FM detectorFM Demodulator

1.poor selectivity(-45dB), 2. SCF Switch noise 3. Center frequency shift by DC offset4. Poor image rejection ratio (25 to 35dB)

1. Low IF( a few hundred KHz)2.Gm-C BPF with auto alignment, SCF

AM/FM IF BPF

High impedance required, Difficult for low frequency

Stages Direct connection, use small value coupling capacitor

Capacitors

Too much sharp C-V curve, distorted signal

MOS varactorVaractor

Can’t cover all process cornerSignal detector with DC compensation

RSSI Level adj.

Large variation of free-run frequencyStill need external LPF for PLL

Multi-vibrator VCO, SCF filter Stereo Decoder

1st trial was analog-centric CMOS tuner technology.

Circuits have been replaced by CMOS, however still use analog technology.Thus it had many issues and many external components were still needed.

2007.09.13 RFIC WS, A. Matsuzawa

12

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Issues and advanced solutions of AM/FM tuner

Lower frequency AM: 522 KHz to 1710 KHzSW: 2.3MHz to 26MHzFM: 87.5 to 108 MHz

Larger Inductance and capacitance

Larger signal dynamic range

Serious 1/f noise

Sharp and fine filter

AM: 14 dBuV to 126 dBuVFM: 0 dBuV to 126 dBuV

High linearity ckt.

PMOS

Digital filter, Mixer, PLLGHz OSC with divider

High resolution ADCSwitch mixerWatching desired and undesired signals

Digital Signal processingWith high resolution ADCIF Freq. changed from 10.7 MHz to several 100 KHz

2007.09.13 RFIC WS, A. Matsuzawa

13

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Advanced CMOS tuner

FMLNA

-

FMMIX

Anti-AliasLPF

VGA ADC DAC

LOCALOSC(FM)

FREQUENCYSYNTHESIZER

AMLNA

SERIALINTERFACE

REGISTER

Cap.Array

AM Bar Antenna(No need for Car radio)

AGC

AGC

AGC

AGC

FM Tuneor BPF

Xtal

XOSC

SYSCLKGEN

PowerDecoupling

Cap

VCC

DSP

STEREODECODER

FMIFBPF

FMDEMOD

RDSDEC

DECI.LPF

AGC GENERATOR

AMMIX

AMIFBPF

AMDEMOD

DIGITALAM LO

To/FromMPU

LEFT

RIGHTCap.Array

Digital-centric CMOS tuner has been developed.

2007.09.13 RFIC WS, A. Matsuzawa

14

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Digital-centric CMOS tuner

Full CMOS one-chip solution# of external components are 11

No adjustment points

Sensitivity: FM: 9dBuV, AM: 16dBuVSelectivity: FM/AM >65dBSNR: FM: 63dB, AM: 53dBStereo sep: 55dBImage ratio: FM: 65dB, AM: InfinityDistortion: FM: 0.09%, AM=0.25%

One-chip CMOS tuner has been successfully developed.It can attain high tuner performance andcan reduce the # of external components.Furthermore it can realize no adjustment points.

2007.09.13 RFIC WS, A. Matsuzawa

15

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Digital-centric CMOS tuner technology

ADC DSPVGA+FilterMIXERLNA

LNA

FM

AM

DSP processes

I

Q

1. AM/FM demodulations2. Stereo decoder3. AM mixer4. Channel select filter5. Support for image reject6. Watch the signal revel and control gain of each stage7. Parameter control and adjustment with MCU

Main signal processing is done by DSP.

2007.09.13 RFIC WS, A. Matsuzawa

16

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Demodulation of AM/FM signal

xReceivedsignal

[ ] ( ) ( ) )(1expexp)(1 tStjtjtS cc +=ω−×ω⋅+

Demodulatedsignal

1) AM demodulation

2) FM demodulation

( ) eredrebetosignalBasebandmiationAmplitudetR

offsetFrequency

cov:var:)(

:

τ

ωΔ

I

Q

( )∫ ττ+ωΔ dmjKtjtR d )(exp)(

∫ ττ+ωΔ=θ dmKt d )(θ

dtdθ

)(tmKdtd

d+ωΔ=θ

AM/ FM signals can be demodulated by simple arithmetic operations

)(tR

m(t) can be demodulated

2007.09.13 RFIC WS, A. Matsuzawa

17

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Stereo decoder

Level

L + R

0

L - R(lower sideband)

Sub-carrier=38KHz

L - R(Upper sideband)

Frequency Spectrum of FM Stereo Signal

Pilot tone=19KHz

15K 23KBasebandFrequency

53K

( ) ( ) tKtRLRLtS ps ω+ω−++= coscos)(

KHztonePilotKHzcarrierSub

p

s

19:38:

=ω=−ω

PLL StereoDetector

38KHz

19KHz

Left

Right

DecoderMatrix

fromDemodulator

LPF L+R

L-R

( ) ( )( ) ( ) RRLRL

LRLRL22

=−−+=−++

PLL locks the pilot tone and generates 38KHz for sub-carrier

Te stereo signal can be reconstructed by numerical PLL, mixer, and filter.

2007.09.13 RFIC WS, A. Matsuzawa

18

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Image rejection

ADCVGA+FilterMIXERLNA

FMI

Q

Deci.LPF

Vari.Delay

Vari.Gain BPF

IMO Controller

Deci.LPF

Fixed.Delay BPF

DSP

to DSP

From ADCs

Image frequency oscillator

IM detect

Image Rejection Ratio >60dB

The dummy image signal is generated by IMO and the controller controls signal delay and amplitude on Q path to minimize the I/Q imbalance.

2007.09.13 RFIC WS, A. Matsuzawa

19

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Impact of components reduction

1/6 pcs or lessPrinted board

Tuner makers are not necessary any more.Unit manufacturers fix IC directly onto unit baseTuner module

Bipolar IC exclusive for RF is not necessary any more.

About 1 billion pcs per year will be reduced to tens of millionspcs.

In AV area, about 50 mil. pcs per year will be reduced to 0.

In AV area, about 1.5 billion pcs per year will be reduced to 0.

Estimated 600 mil. pcs per year will be reduced to 0.

Components # will be less than half the # of existing pcs, but still some remain.

In AV area estimated 3 billion pcs per year will decrease to less than 500 mil. pcs. Aluminum consumption is expected to decrease by 2 thousand ton per year.

Components # will be reduced by more than 15 billion pcs per year.

Components # will be reduced by more than 7 billion pcs per year.

Impact on the Industry

Incorporated into Full CMOS

0

0

0

0

1/2 pcs or less(0~4pcs)

1/10 ~ 1/20 pcs

1/10 pcs or less

1/10 pcs or less

Reduction ratio

Bipolar IC for tuner

Intermediate-frequency transformer

PIN diode

Varactor diode

FM/AM Ceramic filter

Chip inductor

Electrolytic capacitor

Ceramic capacitor

Chip resistor

Reduced components

*Assuming that units manufactured per year are : 100 mil. units for car radios, 80 mil. units for home radios.

2007.09.13 RFIC WS, A. Matsuzawa

20

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Feature of CMOS technology

• Digital: By Scaling theory

– Cost/transistor: 0.5x – Speed/ transistor:1.4x – Power: 0.5x

• Analog:– fT increase: 1.4x– Large mismatch, large PVT fluctuations– Low gm (1/3 vs. Bipolar) – Affected by digital noise seriously

For one technology generation advance

2007.09.13 RFIC WS, A. Matsuzawa

21

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Integration and power dissipation trend

0.1 0.3 1.01

3

10

30

100

300

1000

Technology generation [μm]

Gat

e d

ensi

ty[k

G/m

m ]

2

0.1 0.3 1.01

3

10

30

100

300

1000

Technology generation [μm]

Gat

e d

ensi

ty[k

G/m

m ]

2

0.1 0.3 1.0

0.01

0.03

0.1

0.3

1

3

10

Technology generation [μm]

Pow

er

dis

sipat

ion

[μW

/M

Hz/

G]

0.1 0.3 1.0

0.01

0.03

0.1

0.3

1

3

10

Technology generation [μm]

Pow

er

dis

sipat

ion

[μW

/M

Hz/

G]

Integration level increases and power dissipation decreases with scaling

2007.09.13 RFIC WS, A. Matsuzawa

22

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Feature of Analog CMOS technology

• Pros– Can use switch and voltage controlled conductance– Smaller distortion– No carrier accumulation – Can use switched capacitor circuits– Can increase fT by scaling– Easy use of complementally circuits– Easy integration with digital circuits

• Cons– Low gm/Ids– Larger mismatch voltage and 1/f noise– Lower operating voltage with scaling– Difficult to enable impedance matching– Easily affected by substrate

Akira Matsuzawa, “Mixed Signal SoC Era,” IEICE, Trans. Electron., Vol. E87-c, No. 6, pp. 867-877, June, 2004.

2007.09.13 RFIC WS, A. Matsuzawa

23

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

1/f noise issue in CMOS

1E-19

1E-18

1E-17

1E-16

1E-15

1E-14

1E-13

1E+02 1E+03 1E+04 1E+05 1E+06 1E+07

L=0.4um

L=1.0um

Bipolar

nMOSVdd=3VId=1mA

1E-19

1E-18

1E-17

1E-16

1E-15

1E-14

1E-13

1E+02 1E+03 1E+04 1E+05 1E+06 1E+07

nMOS

pMOS

W/L=800/0.4Vdd=3VId=1mA

Bipolar

2

Inpu

t ref

erre

d no

ise

volta

ge (V

2 /Hz)

Frequency (Hz) Frequency (Hz)

Inpu

t ref

erre

d no

ise

volta

ge (V

2 /Hz)

Nch/Pch 0.4um Nch 0.4um/1.0um

1/f noise of MOS is much larger than that of bipolar.

22 , oxvfvf

nf TSff

LWSV ∝

Δ=

For the lower 1/f noise, the larger gate area is needed.

2007.09.13 RFIC WS, A. Matsuzawa

24

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Why CMOS?

• Low cost– Must be biggest motivation– CMOS is 30-40% lower than Bi-CMOS

• High level system integration– CMOS is one or two generation advanced– CMOS can realize full system integration

• Stable supplyment and multi-foundries– Fabs for SiGe-BiCMOS are very limited.

Slow price decrease and limited product capability

• Easy to use– Universities and start-up companies can use CMOS with

low usage fee, but SiGe is difficult to use such programs.

2007.09.13 RFIC WS, A. Matsuzawa

25

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Is CMOS cheaper?

0

500

1000

1500

2000

2500

3000

3500

4000

4500

0.35

_C

0.25

_C0.

25Si

Ge_

BC0.

25_S

iGe_

BC0.

18um

_C0.

15um

_C0.

13um

_C7M

6M

5M

4M

3M

Low costSiGe

45%45% 23%

30%

Dataquest March 2001+SiGe estimation

Wafer cost of SiGe BiCMOS is 30-40% higher than CMOS at the same generation,however almost same as one generation advanced CMOS.

2007.09.13 RFIC WS, A. Matsuzawa

26

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Cost up issue by analog parts

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.35um 0.25um 0.18um 0.13um0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.35um 0.25um 0.18um 0.13um

(0.35um : 1)

Chip area Chip cost

I/OAnalog

Digital

Cost of mixed A/D LSI will increase when using deep sub-micron device, due to the increase of cost of non-scalable analog parts.

Large analog may be unacceptable.Some analog circuits should be replaced by digital circuits

Wafer cost increases 1.3xfor one generation

Akira Matsuzawa, “RF-SoC- Expectations and Required Conditions,”IEEE Tran. On Microwave Theory and Techniques, Vol. 50, No. 1, pp. 245-253, Jan. 2002

2007.09.13 RFIC WS, A. Matsuzawa

27

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Technology trend in RF-CMOS LSI

Analog-centric Digital-centric

Analog circuitsAnalog processing+External component

Signal processing DSP+ADC+ Small and robust analog ckts.

Adjustment External Digital on chip, no external

External components Large # No or less

Analog-centric RF CMOS will be replaced by digital-centric RF CMOS.High performance, low cost, stable and robust circuits, no or less external components, no adjustment points, and high testability are the keys. DSP and ADC will play important role.

2007.09.13 RFIC WS, A. Matsuzawa

28

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Technology trend in RF CMOS LSI

M. Zargari (Atheros), et al., ISSCC 2004, pp.96 K. Muhammad (TI), et al., ISSCC2004, pp.268

Discrete-time Bluetooth0.13um, 1.5V, 2.4GHz

Wireless LAN, 802.11 a/b/g0.25um, 2.5V, 23mm2, 5GHz

Analog centric RF CMOS will be replaced by digital centric RF CMOS.

2007.09.13 RFIC WS, A. Matsuzawa

29

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Wide-band delta-sigma ADC

50mW (analog), 6mW (digital)Power consumption

1.2VSupply voltage

71dB / 69dBPeak SNR / SNDR*

0.2pJ/conv. (FOM=P/(2^enob*2*BW))Figure-of-merit (FOM)

77dB (97dB @ 200kHz, 115dB @ 3kHz)Dynamic range*

0.5mm2Active chip area

>55dB (for -1MHz input tone)Image rejection

1Vp (differential)Max. input voltage

20MHz @ 10.5MHz IFBandwidth

340MHzSampling frequency

CT quadrature cascaded ΣΔ modulator (2-2, 4b)Architecture

90nm CMOS, 1P6MTechnology

50mW (analog), 6mW (digital)Power consumption

1.2VSupply voltage

71dB / 69dBPeak SNR / SNDR*

0.2pJ/conv. (FOM=P/(2^enob*2*BW))Figure-of-merit (FOM)

77dB (97dB @ 200kHz, 115dB @ 3kHz)Dynamic range*

0.5mm2Active chip area

>55dB (for -1MHz input tone)Image rejection

1Vp (differential)Max. input voltage

20MHz @ 10.5MHz IFBandwidth

340MHzSampling frequency

CT quadrature cascaded ΣΔ modulator (2-2, 4b)Architecture

90nm CMOS, 1P6MTechnology

(*1MHz input signal, signal bandwidth is 20MHz)

90nm CMOS、BW=20MHz, DR(=SNR)=77dB, 50mmW, FoM=200fJ/conv.AD

C1i

ADC

2i

DAC2iDAC1i DAC3i

DAC1q

ADC

1q

DAC2q DAC3q

ADC

2q

+

+-

- +

+-

- +

+-

- +

+-

-

+

+-

-+

+-

-+

+-

-+

+-

-

R2iR1i R3i R4i

R5i R6i R7i R8i

R5q R6q R7q R8q

R2qR1q R3q R4q

C1i C2i C3i C4i C5i C6i

C1q C2q C3q C4q C5q C6q

4b 4b

4b 4b

Vpi

Vmi

Vpq

Vmq

Y1i Y2i

Y1q Y2q

ADC

1i

ADC

2i

DAC2iDAC1i DAC3i

DAC1q

ADC

1q

DAC2q DAC3q

ADC

2q

+

+-

- +

+-

- +

+-

- +

+-

-

+

+-

-+

+-

-+

+-

-+

+-

-

R2iR1i R3i R4i

R5i R6i R7i R8i

R5q R6q R7q R8q

R2qR1q R3q R4q

C1i C2i C3i C4i C5i C6i

C1q C2q C3q C4q C5q C6q

4b 4b

4b 4b

Vpi

Vmi

Vpq

Vmq

Y1i Y2i

Y1q Y2q

L. J. Breems, et., al. “A 56mW CT Quadrature Cascaded SD Modulatorwith 77dB in a Near aero-IF 20MHz Band.ISSCC 2007, pp. 238-239.

Wide band and high dynamic range delta-sigma ADC is the key for digital centric systems

2007.09.13 RFIC WS, A. Matsuzawa

30

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Trend of delta-sigma ADC

4

6

8

10

12

14

16

18

0.01 0.1 1 10 100 1000

FlashSubrangingpipelinedSARfoldingDelta-sigma

Nyquist ADC:

2sigCVSNR ∝

Delta-sigma ADC:

α⋅∝ MCVSNR sig2

The bandwidth of sigma-delta ADCs has been increased up to 20MHzwith effective resolution of about 12 bit.

Effe

ctiv

e bi

t

Bandwidth (MHz)

2007.09.13 RFIC WS, A. Matsuzawa

31

Matsuzawa& Okada Lab.Matsuzawa& Okada Lab.

Conclusion

• Radio tuners have not been replaced by CMOS technology in contrast to many other wireless systems; cellular phones and wireless LANs have been replaced by CMOS.– Low frequency:

• External large L and C, Filters• 1/f noise

– High dynamic range: • External sharp and fine tuning filters

• Analog-centric CMOS technology is not effective– No attractive performance and affected by PVT fluctuation seriously.– Cost increase for further technology scaling– Still need large # of external components and adjusting points

• Digital-centric CMOS technology must be right way– High performance and very robust against PVT fluctuations– Further performance increase and cost reduction are expected by using more

scaled technology– No or less external components and no adjustment points