When processing Multi-Fold Coincidence (MFCO) data, the hardware returns a 256-element array (pattern_bins) where each index represents an 8-bit channel bitmask. For example, index 5 (binary 00000101) corresponds to a simultaneous coincidence on Channel 0 and Channel 2.
To avoid repeatedly implementing bitwise masking logic in Python, the SDK provides the small nexatomtt.analysis module. It operates on the eight-bit, 256-bin MFCO representation; this does not imply that every product has only eight physical TDC resources.
Usage:
from nexatomtt.analysis import top_patterns, exact_pattern_count
| Function | Parameters | Returns | Description |
|---|---|---|---|
pattern_channels |
pattern_index: int |
list[int] |
Decodes an 8-bit pattern index (0-255) into a list of active hardware channels (e.g., 5 $\rightarrow$ [0, 2]). |
pattern_mask |
channels: list[int] |
int |
Encodes a list of channels into the corresponding 8-bit pattern index bitmask. |
exact_pattern_count |
pattern_bins: list[int], channels: list[int] |
int |
Returns the count of coincidences where only the specified channels fired (strict exact match). |
contains_channels_count |
pattern_bins: list[int], required_channels: list[int] |
int |
Returns the aggregated sum of all coincidence patterns that include the specified channels, regardless of what other channels also fired (superset match). |
order_counts |
pattern_bins: list[int] |
dict[str, int] |
Aggregates the entire histogram and groups the counts by order: singles, doubles, triples, and higher. |
top_patterns |
pattern_bins: list[int], limit: int = 10 |
list[dict] |
Returns nonzero-count records, sorted by descending count then ascending pattern for ties. |
from nexatomtt.analysis import exact_pattern_count, contains_channels_count
# result is one MFCO callback record whose status/quality you have checked.
bins = [int(value) for value in result.pattern_bins]
only_0_and_1 = exact_pattern_count(bins, [0, 1])
includes_0_and_1 = contains_channels_count(bins, [0, 1])
# The inclusive result also counts patterns with channel 2, 3, etc. present.
Both helpers require exactly 256 nonnegative integer counts. Channel IDs must be integers in 0–7. order_counts excludes the empty pattern and groups the remaining bins by the number of set bits. These helpers neither validate completion/quality nor divide by acquisition live time. Inspect those result fields before interpreting a rate, and do not add successive WHOLE_RUN results together.