RF Cascade Budget Analyzer
Compute cascaded noise figure, gain, IIP3, and P1dB for an N-stage RF signal chain using Friis formulas. Paste your stage list as JSON, set your system specs, and get a line-by-line cascade table, sensitivity analysis, and Monte Carlo yield statistics across component tolerances.
How It Works
The Friis noise figure formula reveals that the first stage dominates system NF:
F_total = F₁ + (F₂−1)/G₁ + (F₃−1)/(G₁G₂) + ···
Each subsequent stage's contribution is divided by all preceding gains. A high-gain, low-NF LNA in position 1 suppresses everything downstream — until the last high-power stage limits the IIP3.
The cascaded IIP3 works in the opposite direction:
1/IIP3_total = 1/IIP3₁ + G₁/IIP3₂ + G₁G₂/IIP3₃ + ···
Here, later stages with high accumulated gain before them dominate. The last amplifier before the ADC is typically the IIP3 bottleneck. This tension — put the gain early for NF, but spreading gain increases IIP3 burden on later stages — is the fundamental cascade design tradeoff.
The Monte Carlo sweeps stage gain (±0.5 dB σ), NF (±0.3 dB σ for active stages), and IIP3 (±2 dB σ) to compute the distribution of system metrics across a manufacturing lot.
Related Calculators
FAQ
Why does my IIP3 get much worse after adding a high-gain amplifier?+
Cascaded IIP3 is dominated by stages with large gain in front of them. Adding a 20 dB amplifier before a weak stage amplifies the signal 100×, so even a tiny nonlinearity in that weak stage produces large intermodulation products at the output. The Friis IIP3 formula (1/IIP3_total = Σ G_cumul/IIP3_i) makes this explicit: each stage's IIP3 contribution is divided by its input accumulated gain, not reduced by it.
What do the MC confidence bands on NF mean?+
The p5/p50/p95 NF bands come from running 50,000+ trials where each stage's gain is perturbed by ±0.5 dB (σ) and each active stage's NF by ±0.3 dB (σ). The p95 NF is the worst-case 95th percentile — 95% of manufactured units will have NF better than this value. Write your system specification against the p95 curve, not the nominal. The yield percentage tells you what fraction of units meet all specs simultaneously.