received by the International Bureau on 22 April 2014 (22.04.14)

1. A communication device (44, 64, 134) comprising:

a transmitter (46, 66, 136) configured to receive a digital input signal and output a radio frequency transmit signal that is transmitted by the communication device (44, 64, 134);

a main receiver (48, 68, 138) configured to receive a radio frequency receive signal and output a digital output signal, the digital output signal comprising passive intermodulation distortion that is a non-linear function of the radio frequency transmit signal output by the transmitter (46, 66, 136);

a combined digital predistortion and passive intermodulation estimate generation subsystem (74, 144) configured to:

receive the digital input signal of the transmitter (46, 66, 136);

predistort the digital input signal of the transmitter (66, 136) according to a behavioral model of an inverse of a nonlinearity of the transmitter (66,

136);

generate a digital passive intermodulation estimate signal that is an estimate of the passive intermodulation distortion in the digital output signal of the main receiver (48, 68, 138) based on the digital input signal of the transmitter (46, 66, 136) according to a behavioral model of the transmitter (46, 66, 136), a source of the passive intermodulation distortion, and the main receiver (48, 68, 138); and

subtract the digital passive intermodulation estimate signal from the digital output signal of the main receiver (48, 68, 138) to thereby provide a compensated digital output signal.

2. The communication device (44, 64, 134) of claim 1 wherein the behavioral model of the transmitter (46, 66, 136), the source of the passive intermodulation distortion, and the main receiver (48, 68, 138) is a simplified Volterra model.

3. The communication device (44, 64, 134) of claim 1 wherein the behavioral model of the transmitter (46, 66, 136), the source of the passive intermodulation distortion, and the main receiver (48, 68, 138) is a generalized memory polynomial model.

4. The communication device (64, 134) of claim 1 wherein the combined digital predistortion and passive intermodulation estimate generation subsystem (74₍ 144) comprises one or more components that are utilized for both

predistortion of the digital input signal and generation of the digital passive intermodulation estimate signal.

5. The communication device (64, 34) of claim 4 wherein the behavioral model of the inverse of the nonlinearity of the transmitter (66, 136) and the behavioral model of the transmitter (66, 136), the source of the passive intermodulation distortion, and the main receiver (68, 138) are simplified Volterra models, and the one or more components are one or more shared regressors of the simplified Volterra models.

6. The communication device (64, 134) of claim 5 wherein the simplified Volterra models are generalized memory polynomial models.

7. The communication device (64, 134) of claim 4 wherein the behavioral model of the inverse of the nonlinearity of the transmitter (66, 136) and the behavioral model of the transmitter (66, 136), the source of the passive intermodulation distortion, and the main receiver (68, 138) are trained using the same resources.

8. The communication device (64, 34) of claim 1 wherein the radio frequency transmit signal is a concurrent multi-band signal, and the digital input signal of the transmitter (66, 136) is a digital representation of the concurrent multi-band signal.

9. A communication device (44, 64, 134) comprising:

a transmitter (46, 66, 136) configured to receive a plurality of digital input signals and process the plurality of digital input signals to provide a radio frequency transmit signal as a concurrent multi-band signal, where each digital input signal of the plurality of digital input signals corresponds to a different frequency band of a plurality of frequency bands of the concurrent multi-band signal and output the radio frequency transmit signal that is transmitted by the communication device (44, 64, 134);

a main receiver (48, 68, 138) configured to receive a radio frequency receive signal and output a digital output signal, the digital output signal comprising passive intermodulation distortion that is a non-linear function of the radio frequency transmit signal output by the transmitter (46, 66, 136);

a combined digital predistortion and passive intermodulation estimate generation subsystem (74, 144) configured to:

receive the plurality of digital input signals of the transmitter (46, 66, 36);

predistort the plurality of digital input signals of the transmitter (66, 136) according to a behavioral model of an inverse of a nonlinearity of the transmitter (66, 136);

generate a digital passive intermodulation estimate signal that is an estimate of the passive intermodulation distortion in the digital output signal of the main receiver (48, 68, 138) based on the plurality of digital input signals of the transmitter (46, 66, 136) according to a behavioral model of the transmitter (46, 66, 136), a source of the passive

intermodulation distortion, and the main receiver (48, 68, 138); and subtract the digital passive intermodulation estimate signal from the digital output signal of the main receiver (48, 68, 138) to thereby provide a compensated digital output signal.

10. The communication device (64, 134) of claim 9 wherein the combined digital predistortion and passive intermodulation estimate generation subsystem (74. 1 4) comprises one or more components that are utilized for both

predistortion of the plurality of digital input signals and generation of the digital passive intermodulation estimate signal.

11. The communication device (64, 134) of claim 10 wherein the behavioral model of the inverse of the nonlinearity of the transmitter (66, 136) and the behavioral model of the transmitter (66, 36), the source of the passive intermodulation distortion, and the main receiver (68, 138) are simplified Volterra models, and the one or more shared components are one or more shared regressors of the simplified Volterra models. 2. The communication device (64, 34) of claim 11 wherein the simplified Volterra models are generalized memory polynomial models.

13. The communication device (64, 134) of claim 9 wherein the behavioral model of the inverse of the nonlinearity of the transmitter (66, 136) and the behavioral model of the transmitter (66, 136), the source of the passive

intermodulation distortion, and the main receiver (68, 138) are trained using the same resources.

14. The communication device (44, 64, 134) of claim 1 wherein the transmitter (46, 66, 136) and the main receiver (48, 68, 38) are coupled to an antenna (50, 70, 140) of the communication device (44, 64, 134) via a duplexer (52, 72, 142).

15. The communication device (44, 64, 134) of claim 1 wherein the communication device (44, 64, 134) is a base station in a cellular

communications network.

16. A method of operation of a communication device (44, 64, 34)

comprising a transmitter (46, 66, 136) configured to receive a digital input signal and output a radio frequency transmit signal that is transmitted by the

communication device (44, 64, 134) and a main receiver (48, 68, 138) configured to receive a radio frequency receive signal and output a digital output signal, the digital output signal comprising passive intermodulation distortion that is a nonlinear function of the radio frequency transmit signal output by the transmitter (46, 66, 136), the method comprising:

predistorting the digital input signal according to a behavioral model of an inverse of a nonlinearity of the transmitter (66, 136);

generating a digital passive intermodulation estimate signal that is an estimate of the passive intermodulation distortion in the digital output signal of the main receiver (48, 68, 138) based on the digital input signal of the transmitter (46, 66, 136) according to a behavioral model of the transmitter (46, 66, 136), a source of the passive intermodulation distortion, and the main receiver (48, 68, 138); and

subtracting the digital passive intermodulation estimate signal from the digital output signal of the main receiver (48, 68, 138) to thereby provide a compensated digital output signal;

wherein generating the digital passive intermodulation estimate signal and predistorting the digital input signal utilize one or more shared components of the behavioral model of the inverse of the nonlinearity of the transmitter (66, 36) and the behavioral model of the transmitter (66, 136), the source of the passive intermodulation distortion, and the main receiver (68, 138).

17. The method of claim 16 wherein the behavioral model of the inverse of the nonlinearity of the transmitter (66, 136) and the behavioral model of the

transmitter (66, 136), the source of the passive intermodulation distortion, and the main receiver (68, 138) are simplified Volterra models, and the one or more shared components are one or more shared regressors of the simplified Volterra models. 8. The method of claim 17 further comprising training the behavioral model of the inverse of the nonlinearity of the transmitter (66, 136) and the behavioral model of the transmitter (66, 36), the source of the passive intermodulation distortion, and the main receiver (68, 138) using the same resources.