Adjacent Channel Power Ratio (ACPR)

ACPR is mathematically defined as the ratio of the total power in a specified adjacent frequency channel to the total power in the main transmission channel. It is typically expressed in decibels relative to the carrier (dBc).

• Formula: ACPR = 10 × log₁₀(P_adjacent / P_main)
• Measurement bandwidth: Adjacent channel power is integrated over the channel bandwidth of the victim receiver
• Offset frequency: Specified as the distance from the carrier center frequency (e.g., ±5 MHz, ±10 MHz)
• Typical targets: -45 dBc for 3GPP LTE, -50 dBc or better for 5G NR with higher-order modulation
• Spectrum analyzer integration: Requires root-raised-cosine filtering and gated power measurements per regulatory test specifications

ACPR limits are mandated by spectrum regulatory bodies to prevent adjacent channel interference and ensure coexistence between different wireless operators and services.

• 3GPP TS 38.104: Defines ACPR requirements for 5G NR base stations, typically -45 dBc for wide area BS
• FCC Part 24/27: Specifies out-of-band emission limits for US cellular and PCS bands
• ETSI EN 301 908: European harmonized standard for IMT-2000 and beyond
• ACLR vs ACPR: Adjacent Channel Leakage Ratio (ACLR) is the 3GPP-preferred term, functionally equivalent to ACPR
• Spectrum mask compliance: ACPR is measured alongside spectrum emission masks (SEM) for full regulatory certification

ACPR degradation is a direct consequence of amplitude-to-amplitude (AM-AM) and amplitude-to-phase (AM-PM) distortion in power amplifiers. Spectral regrowth occurs because nonlinear transfer functions generate intermodulation products that spread beyond the intended bandwidth.

• Third-order intercept point (IP3): Higher IP3 correlates with better ACPR performance
• 1 dB compression point (P1dB): Operating near P1dB increases spectral regrowth exponentially
• Memory effects: Thermal and electrical memory effects cause asymmetric spectral regrowth, making upper and lower ACPR values unequal
• Crest factor impact: High peak-to-average power ratio (PAPR) signals like OFDM drive the PA into nonlinear regions more frequently, worsening ACPR
• Back-off tradeoff: Increasing PA back-off improves ACPR but reduces power efficiency

Digital predistortion (DPD) is the primary technique for improving ACPR without sacrificing PA efficiency. A well-tuned DPD system can achieve 15-25 dB of ACPR improvement.

• Linearization gain: ACPR improvement directly attributable to DPD, typically measured as ΔACPR = ACPR_without_DPD - ACPR_with_DPD
• Convergence monitoring: ACPR is used as a real-time metric to verify DPD coefficient adaptation quality
• Wideband DPD: For 5G signals with 100+ MHz bandwidth, DPD must linearize across multiple adjacent channels simultaneously
• Multi-band ACPR: Concurrent multi-band DPD must suppress cross-modulation products that appear as ACPR degradation in non-adjacent bands
• Residual ACPR: The remaining spectral regrowth after DPD correction, used to quantify linearization effectiveness

Accurate ACPR measurement requires a calibrated vector signal analyzer (VSA) or spectrum analyzer with specific test configurations to isolate PA-generated distortion from instrument noise.

• Required equipment: Vector signal generator, PA under test, directional coupler, attenuator, and VSA
• Averaging: Multiple trace averaging (typically 10-100 sweeps) reduces measurement variance
• Noise floor considerations: Instrument noise floor must be at least 10 dB below the expected ACPR level
• Crest factor reduction (CFR): Often tested in combination with DPD to evaluate end-to-end ACPR improvement
• Temperature cycling: ACPR must be verified across the full operating temperature range to ensure DPD robustness

Optimizing ACPR involves balancing multiple competing design parameters. Aggressive ACPR targets can negatively impact other system metrics.

• Efficiency vs linearity: Every 1 dB improvement in ACPR through back-off typically costs 2-3% in PA drain efficiency
• DPD complexity: Achieving -55 dBc ACPR may require higher-order memory polynomial models with 2-3× more coefficients
• Bandwidth expansion: DPD correction signals require 3-5× the original signal bandwidth, increasing DAC and feedback path requirements
• EVM correlation: ACPR and EVM are correlated but not interchangeable; excellent ACPR does not guarantee good EVM
• Cost implications: Higher ACPR requirements drive more expensive PAs, wider-bandwidth feedback ADCs, and more powerful DPD processors

A measure of in-band distortion quality that quantifies the deviation of the actual transmitted symbols from their ideal reference constellation points.

• Expressed as a percentage or in dB relative to the reference signal power
• Directly impacts bit error rate (BER) and modulation accuracy
• 3GPP specifies EVM limits per modulation scheme: 17.5% for QPSK, 3.5% for 256QAM
• While ACPR measures out-of-band emissions, EVM captures the in-band signal fidelity degradation caused by the same PA nonlinearity

A regulatory emission limit defined by standards bodies (FCC, ETSI) as a frequency-dependent power envelope that transmitted signals must not exceed.

• Specifies absolute power limits in adjacent and alternate channels, not just ratios
• ACPR measurements are used to verify mask compliance at specific frequency offsets
• Failure to meet spectral mask requirements results in type approval rejection
• Masks become more stringent for wideband signals in shared spectrum environments

A time-domain modeling accuracy metric that quantifies the normalized difference between the ideal linear output and the actual PA output after linearization.

• Calculated as the mean squared error normalized by the input signal power
• Values below -40 dB typically indicate excellent DPD performance
• Used extensively during DPD coefficient training as the primary cost function
• NMSE and ACPR are strongly correlated: improving NMSE through DPD directly reduces ACPR

A signal conditioning technique applied before the power amplifier to reduce the peak-to-average power ratio (PAPR) of the transmitted waveform.

• Clips or shapes signal peaks to allow higher average PA output power without saturation
• Works synergistically with DPD: CFR handles peak excursions while DPD corrects nonlinearity
• Excessive CFR introduces its own EVM degradation and spectral regrowth
• The CFR-DPD cascade is standard in modern base station transmitters to jointly optimize efficiency and ACPR

A figure of merit for amplifier linearity derived from two-tone testing, representing the theoretical output power level where third-order intermodulation products would equal the fundamental tones.

• Higher IP3 values indicate better inherent linearity before DPD is applied
• Output IP3 (OIP3) is typically 10-15 dB above the 1 dB compression point
• While ACPR measures spectral regrowth with modulated signals, IP3 characterizes the underlying device physics
• DPD can effectively improve the cascaded IP3 of the entire transmitter chain by 15-25 dB

The ratio of the desired signal power to the total interference power within the receiver bandwidth, directly impacting link budget and network capacity.

• ACPR from adjacent transmitters becomes the interference term in C/I calculations for neighboring cells
• Poor ACPR in one base station degrades the C/I and throughput of adjacent sectors
• Network planning tools use ACPR specifications to calculate frequency reuse distances
• 3GPP defines minimum C/I requirements per modulation and coding scheme for reliable demodulation