ZEKASTRA / FORENSIC / SOURCE-LINKED ANALYSIS
Communication Record Workbench
Go beyond a single timestamp. Compare two or more CSV sources, retain original file and row identities, normalize time offsets and explore communication pairs.
Files are processed in browser memory only; no upload, account or local storage. This is a defined training CSV schema, not a native operator HTS/CDR decoder. Use synthetic or explicitly authorized material. Exported files retain any identifiers entered.
01 / Load sources
Two complete fictional source datasets
Start with 18 invented records in two CSV files, including an intentionally repeated charging-style row. Or open up to four local CSV files (2 MiB each, up to 10,000 combined records).
Exact columns: event_id, timestamp_iso, originating_number, terminating_number, service, duration_seconds, cell_id. All timestamps need Z or an explicit offset; numbers are strings, not verified subscriber identities.
Advanced: manually map different CSV column names
This bridge does not guess operator semantics or time units. The original input must contain absolute ISO timestamps with explicit UTC offsets, E.164-shaped phone numbers and whole-second durations. Service labels are mapped only using the exact text you specify.
No sources loaded.
02 / Source chronology and relationships
Source-bound records, not inferred identities
Source manifest and exact file hashes
Filter records without altering originals
Number-pair summary (direction retained separately)
| Number pair | Rows | VOICE / SMS | Voice s | Direction counts | Source references |
|---|
Potential matching rows — never auto-deleted
Cross-source time-window intersections
Shows different CSV sources with at least one matching number string in the selected time window. This is temporal proximity only, not proof of a shared location or action. Processing is capped, and an incomplete result is explicitly marked.
| Source references | Shared numbers | Gap | Earlier UTC | Later UTC |
|---|
Normalized UTC timeline with original values
| Source row | UTC | Source clock | Originating | Terminating | Service | Seconds | Cell ID |
|---|
Cell ID is kept as an opaque source value. It is not a mapped tower position, a location fix or proof of device presence. Similar rows may represent multiple CDRs for one chargeable event; no automatic deduplication is performed.
03 / Source-linked output
Export exactly what was analyzed
Downloads contain the currently filtered entries; every record retains its original source row and raw timestamp. CSV neutralizes spreadsheet formula prefixes; JSON keeps exact identifiers.
The evidence workspace package includes the currently filtered records, source metadata, explicit-offset UTC normalization and method references. Original CSV bytes are not included. The pseudonymization option above is applied to this download.
Pseudonymized files still include exact dates and patterns, which may be identifying. Use only in a controlled sharing context.
Method and primary references
ITU-T E.164 describes number structure, not subscriber attribution. 3GPP TS 32.297 covers CDR format and transfer; these sample CSV columns are educational, not an operator export. Source documentation and case acquisition must be evaluated independently.