criterion456
Member level 2
How to calculate the necessary isolation of duplexer ??
For FDD communication system, there is Tx leakage
If we don't want desense issue, we need enough isolation of duplexer.
In general, e.g. CDMA or WCDMA, the isolation should be at least 50 dB.
My question is, how is the "50 dB" derived ?
Account for LNA P1dB ?
If we assume the PA out = 27 dBm, isolation = 50 dB
That is, the Tx leakage@LNA input = 27 - 50 = -23 dBm
Nevertheless, take the LNA, SKY74092-11 of SKYWORKS, for example
its P1dB is -7 dBm in high gain mode
In other words, if just account for LNA P1dB,
it is not necessary for isolation to be at least 50 dB, even 40 dB is already enough
Besides, if the Rx band noise floor raises due to Tx max power on
No doubt, the sensitivity degrades.
Thus, in terms of Rx band noise floor
my question can be also expressed as :
How can I derive the necessary isolation that ensures the Rx band noise floor will NOT raise with Tx max power on ???
Thanks a lot~!!
For FDD communication system, there is Tx leakage
If we don't want desense issue, we need enough isolation of duplexer.
In general, e.g. CDMA or WCDMA, the isolation should be at least 50 dB.
My question is, how is the "50 dB" derived ?
Account for LNA P1dB ?
If we assume the PA out = 27 dBm, isolation = 50 dB
That is, the Tx leakage@LNA input = 27 - 50 = -23 dBm
Nevertheless, take the LNA, SKY74092-11 of SKYWORKS, for example
its P1dB is -7 dBm in high gain mode
In other words, if just account for LNA P1dB,
it is not necessary for isolation to be at least 50 dB, even 40 dB is already enough
Besides, if the Rx band noise floor raises due to Tx max power on
No doubt, the sensitivity degrades.
Thus, in terms of Rx band noise floor
my question can be also expressed as :
How can I derive the necessary isolation that ensures the Rx band noise floor will NOT raise with Tx max power on ???
Thanks a lot~!!