Amplifiers for ZigBee applications

[drabos]

zigbee qpsk

I have to design an LNA and a PA for ZigBee application, before/after a ZigBee chip.

Can I use nonlinear PA for it (without linearisation)? Any experience?

 

[khouly]

what is the spec's of each ?

about the linearization , if the modulation is constant envelop , maybe , but i think it is better to use linearization , or use linear PA

khouly

 

[drabos]

ZigBee uses O-QPSK modulation.

 

[khouly]

O-QPSK , like the CDMA , i think it will need linearizer , or just use linear amp


what about the spec's of ur LNA ?
khouly

 

[drabos]

frequeny range: 2.405-2.485 GHz
gain: 20 dB
IP3: the bigger the better
NF: as least can , max. 2dB
Maximum power level at the output of the LNA: +10 dBm

 

[szekit]

QPSK is a linear modulation and would need a linear PA i guess?!

 

[melc]

frequeny range: 2.405-2.485 GHz
gain: 20 dB
IP3: the bigger the better
NF: as least can , max. 2dB
Maximum power level at the output of the LNA: +10 dBm

Any ordinary WLAN (WIFI) amplifier will work.

 

[drabos]

The modulation isn't QPSK, rather O-QPSK.

 

[khouly]

hi darbos , did u begin the design phase , and do u want to implement the LNA as CMOS "integrated or discreete?

khouly

 

[drabos]

discrete

LNA with pHEMT, PA with BJT

 

[khouly]

good
check the pHEMT od avagotech , they have a good HEMTs u can use , and work in this frequ with very good NF

and the BJTs check CEL , and avago u will find a good selction

which RF simulator will u use in ur design process ?

khouly

 

[PaulHolland]

Hi Simply use a 802.11g amplifier and your OK. It needs a linear amplifier but not that good as 802.11g or b in that matter. If you use 802.11b amp you will be able to get the full power output with no problem.

Paul.

 

[drabos]

RF simulator: ADS 2005a

 

[khouly]

good ,
ADS is great in LNA , and PA design , HB simulator and Sparamter , and design of matching circuits

about the LNA the gain is high , do u want the LNA to be two stage or single stage , and do u want any gain control in it or not


khouly

 

[szekit]

The modulation isn't QPSK, rather O-QPSK.

O-QPSK is a subset of QPSK where the constellation map of QPSk is overlaid with another constellation of QPSK which is offset by pi/4. It is still a linear phase modulation.

 

[drabos]

LNA is a single stage, there's no gain control yet.

Added after 1 minutes:

and the main datas for the LNA:
Noise Figure: 0.4 dB
IIP3: 12 dB
OIP3: 20 dB
gain: 24 dB

 

[khouly]

how is ur design going ?
to get this very hgigh gain , u shoudl use tarnsistor like


khouly

 

[drabos]

I use this pHEMT, Agilent ATF36163

Added after 1 minutes:

Avago

 

[khouly]

good ,
this transistor is very low noise , and with ordinary package with good gain at 2.4 GHz

khouly