Asking the fleet what it is doing…
monad-knowledge Wi-Fi sensing lab · FIIT STU
PRB504 active 1 / 12 weeks

Probabilistic & Statistical Methods for Sensing

A twelve-week track on the probabilistic and statistical backbone every sensing result rests on — estimation, Bayesian inference, experimental design, and honest reporting. The throughline: an estimate without a defensible uncertainty is not a result. This is the methods course the other four lean on.

The 12-week arc

1 of 12 lectures published — each a self-contained read with a slide deck and runnable workbook. Start at the first published week and work forward.

Wk Theme & focus Read
01 Estimation theory & the CRLB start here Bias, variance, and the best any estimator can do read → slides · workbook
02 Bayesian inference for sensing Priors that earn their keep; posteriors over hidden states planned
03 Gaussian processes for spatial fields Modelling a radio/sensor field with calibrated uncertainty planned
04 Sensor fusion as inference Combining heterogeneous sources as one probabilistic model planned
05 Hypothesis testing & reproducibility p-values, power, and the replication crisis in practice. Use the lab's own canon cases: TOST semantics (pass = equivalence within Δ; fail ≠ difference — the drift-trigger powered run), the flat-day KS ladder where 4× power SHRANK the effect 0.24→0.135 while p crossed 0.05 (significance-by-sample-size), and the Zaman AUC = 1.000 session-leakage audit planned
06 Experimental design Factors, blocking, randomisation for field experiments planned
07 Bootstrap & resampling Confidence intervals when the textbook assumptions fail planned
08 Time-series & non-stationarity Drift and autocorrelation (the statistical side of WS501 wk 9) planned
09 Calibration & uncertainty quantification Are your error bars honest? Reliability diagrams and coverage planned
10 Causal thinking in field experiments Confounders, and what you can and cannot claim planned
11 Honest statistical reporting CDFs, CIs, and effect sizes over point estimates planned
12 Capstone A pre-registered analysis plan for a sensing experiment planned

Capstone

Week 12: take a real or planned sensing experiment and write a pre-registration — hypotheses, design, the exact analysis, and the uncertainty reporting — that a reviewer could hold you to.