Cross-Modulation

Cross-modulation occurs when a strong modulated interferer and a weaker desired signal pass through a nonlinear device, typically a power amplifier (PA). The third-order nonlinearity causes the instantaneous amplitude variations of the aggressor's envelope to modulate the gain experienced by the victim signal. This effectively transfers the aggressor's modulation spectrum around the victim's carrier frequency, causing spectral regrowth that cannot be filtered out by conventional means.

The phenomenon is fundamentally driven by odd-order distortion products, primarily the third-order term (IM3). Unlike harmonic distortion, which falls far from the fundamental frequencies, cross-modulation products fall directly on top of or immediately adjacent to the desired signal's channel. This makes it a primary limiter of receiver sensitivity and transmitter spectral mask compliance in frequency-division duplex (FDD) systems.

While related, cross-modulation is distinct from classic intermodulation distortion (IMD). IMD generates discrete new frequency components from unmodulated or simple multi-tone inputs. Cross-modulation specifically describes the transfer of a complex, stochastic modulation envelope. The result is a noise-like spectral regrowth in the victim channel that mirrors the bandwidth and statistical properties of the aggressor signal.

In concurrent multi-band transmitters, cross-modulation creates cross-band distortion products that fall into adjacent transmit bands. Traditional single-band DPD is blind to these effects. Effective linearization requires 2D-DPD or multi-dimensional models that include cross-terms dependent on the instantaneous envelope magnitudes of all concurrent bands to synthesize the inverse distortion and cancel the transferred modulation.

Cross-modulation is not memoryless. Cross-band memory effects occur when the PA's nonlinear behavior in one band is influenced by the past envelope history of a signal in a different band. This is caused by:

• Thermal memory: Die temperature changes with aggregate dissipated power
• Electrical memory: Bias circuit impedance and trapping effects in GaN HEMTs
• Node impedance variations: Baseband impedance affecting the instantaneous nonlinear response

The dual-band Volterra series provides a rigorous analytical framework for predicting cross-modulation. By applying a two-frequency input signal to the passband Volterra model and extracting the baseband components, one can derive exact expressions for the cross-modulation kernels. These kernels reveal how the nonlinear transfer function mixes the spectral components of both signals, guiding the selection of basis functions for simplified models like the 2D Memory Polynomial.

The fundamental nonlinear phenomenon that generates unwanted frequency components when two or more signals mix in an active device. Cross-modulation is a specific case of IMD where the modulation envelope of an interfering signal is transferred to the desired carrier. IMD products are classified by their order (e.g., third-order IMD, fifth-order IMD), with odd-order products typically falling closest to the original signals and causing the most severe in-band and adjacent-channel interference.

Nonlinear interference products generated by the interaction of multiple carrier signals within a power amplifier that fall on top of or near the desired transmit bands. In a dual-band transmitter, the nonlinear mixing of Band 1 and Band 2 produces intermodulation products that can land directly in the receive band or in adjacent allocated spectrum. Cross-band distortion is the primary impairment that multi-band DPD architectures aim to cancel.

A behavioral model that extends the standard memory polynomial to two dimensions by incorporating cross-terms dependent on the envelope magnitudes of both concurrent bands. The 2D-MMP captures cross-band memory effects—where the nonlinear behavior in one band is influenced by the past envelope history of the signal in the other band. This model structure is foundational for synthesizing predistortion signals that cancel cross-modulation products in concurrent dual-band transmitters.

The active process of generating a correction signal equal in amplitude but opposite in phase to cross-band distortion products to neutralize them. This technique requires precise phase alignment and amplitude matching across wide bandwidths. Cross-band cancellation is typically implemented within a joint DPD architecture where a unified predistorter processes the composite multi-band signal, synthesizing cancellation signals that target intermodulation products falling into adjacent transmit bands.

A closed-loop DPD adaptation method where a post-distorter model is identified from the attenuated PA output and then copied to the predistorter in the forward path. In the multi-band context, the MB-ILA must estimate not only the in-band predistortion coefficients but also the cross-band coupling terms that characterize cross-modulation. The architecture operates by minimizing the error between the post-distorter output and the original multi-band input signal.

A key performance metric measuring the ratio of power leaked into adjacent channels to the power in the main channels for a multi-band transmitter. Cross-modulation directly degrades MB-ACLR by transferring modulation energy from one carrier into the adjacent channels of another carrier. Regulatory bodies such as 3GPP specify strict MB-ACLR requirements for carrier aggregation scenarios, making cross-modulation suppression a critical design objective for multi-standard radio designers.