Adjacent Channel Leakage Ratio (ACLR)

ACLR is defined as the ratio of the filtered mean power centered on the assigned channel frequency to the filtered mean power centered on an adjacent channel frequency. It is typically expressed in dBc (decibels relative to the carrier).

• Formula: ACLR = 10 log₁₀(P_adjacent / P_carrier)
• A more negative value (e.g., -45 dBc) indicates better linearity and less interference.
• Measurements require specific measurement bandwidths and channel spacings defined by standards bodies like 3GPP.

ACLR degradation is a direct consequence of spectral regrowth, which is caused by the intermodulation distortion (IMD) products generated when a modulated signal passes through a non-linear power amplifier.

• The third-order intercept point (IP3) of an amplifier is a key predictor of its ACLR performance.
• AM-AM and AM-PM distortion create spectral components outside the intended channel bandwidth.
• Higher-order modulations (e.g., 256-QAM, OFDM) with high Peak-to-Average Power Ratio (PAPR) are particularly susceptible.

ACLR and Error Vector Magnitude (EVM) represent a fundamental trade-off in transmitter design. Both are degraded by power amplifier non-linearity, but they are often addressed by competing techniques.

• EVM measures in-band signal quality (constellation fidelity).
• ACLR measures out-of-band emissions (spectral containment).
• Digital Pre-Distortion (DPD) is the primary technique to improve both simultaneously, but aggressive Crest Factor Reduction (CFR) can improve ACLR at the direct expense of EVM.

Accurate ACLR measurement requires a controlled test environment to isolate the device under test (DUT). The standard setup includes:

• A vector signal generator (VSG) to produce the test waveform.
• A spectrum analyzer or vector signal analyzer (VSA) with ACLR measurement capability.
• The measurement must use the correct channel bandwidth, adjacent channel offset, and root-raised cosine (RRC) filter with the specified roll-off factor (α).

Digital Pre-Distortion (DPD) is the most critical technique for improving ACLR without sacrificing efficiency. DPD applies an inverse model of the power amplifier's non-linearity to the baseband signal.

• Neural Network DPD architectures can model complex memory effects that limit traditional Volterra-based methods.
• Effective DPD can improve ACLR by 10-20 dB, enabling operation closer to the amplifier's saturation point.
• This directly translates to higher Power-Added Efficiency (PAE) and lower operational expenditure (OpEx) for network operators.