Construct the compiled causal SOS plus rolling-RMS operation

The operation uses direct-form-II-transposed second-order sections followed by a per-channel rolling root-mean-square. It is a descriptor for onChunk() and performs no processing until registered and stepped.

Usage

bandpassRmsOp(
  sos,
  window_samples,
  warmup = c("partial", "complete"),
  name = "bandpass_rms"
)

Arguments

sos: Finite second-order-section matrix with columns b0,b1,b2,a0,a1,a2.
window_samples: Exact positive RMS window length.
warmup: Whether to emit partial-window RMS values or only rows after a complete window exists.
name: Default operation name.

Value

A PipelineOperation descriptor.

Examples

# A pass-through SOS keeps the example self-contained; use
# PhysioPreprocess::sosDesign() for a real band definition.
sos <- matrix(c(1, 0, 0, 1, 0, 0), 1L, 6L,
              dimnames = list(NULL, c("b0", "b1", "b2", "a0", "a1", "a2")))
op <- bandpassRmsOp(sos, window_samples = 4L)
pipeline <- streamPipeline(chunk_size = 8L)
onChunk(pipeline, op)
x <- matrix(sin(seq_len(16)), 8, 2, dimnames = list(NULL, c("C3", "C4")))
pipelineEnqueue(pipeline, x, ingest_time_ns = 0)
pipelineStep(pipeline)$results[[1]]$output$samples
#>         C3_rms    C4_rms
#> [1,] 0.8414710 0.4121185
#> [2,] 0.8760409 0.4825975
#> [3,] 0.7199097 0.6989948
#> [4,] 0.7293079 0.6621351
#> [5,] 0.7646931 0.6633985
#> [6,] 0.6305303 0.7819422
#> [7,] 0.7074585 0.6834889
#> [8,] 0.7758979 0.6449042