D3 Mobile Forensics
Evidence quality and confidence
Confidence belongs to a specific finding, not to a product name or file format.
Active database records
Structured content with identifiers and relationships can support context, but parsing and field meanings still need validation. An active database can itself be incomplete, synchronised or manipulated.
WAL, journals and freelists
These structures may retain changes or residual records. Establish the relation to the main database, transaction state, schema and acquisition time. Do not silently treat a fragment as a complete committed message.
Carved fragments
A fragment may lack sender, timestamp or conversation context. Validate its structure, avoid joining unrelated bytes and distinguish a recovered string from a reliably attributed message. A tool's deleted label is not proof of deliberate user deletion.
Caches, thumbnails and notifications
These can corroborate activity but may be stale, truncated, duplicated or detached from the original. A thumbnail is not the original image; a notification may not show the complete communication.
Screenshots and exports
Assess provenance, editability, surrounding conversation and correspondence with native records. A plausible screenshot or readable export is not self-authenticating.
Provider returns
Providers can transform media or omit fields. Preserve the returned representation and collection context. A changed hash compared with an original upload may reflect transformation rather than tampering.
Reporting confidence
For each material conclusion, identify the source, validation performed, gaps, corroboration and credible alternatives. State fact, inference and opinion separately. Do not assign unvalidated numerical confidence scores or infer human authorship solely from an account name.
Integrity is not the whole authenticity question
A matching cryptographic hash supports byte-for-byte consistency with the item hashed at that point in its history. It does not establish who created the original, whether its content is true or whether earlier material was omitted. Assess provenance, completeness, context and authorship separately.
A changed media file may reflect resizing, transcoding or metadata handling rather than fabrication. Record the supplied representation and investigate the transformation before drawing a conclusion.
Media metadata and provenance · Worked example: observation, inference and limitations
Five questions, five different tests
Integrity asks whether the item has changed since a documented comparison point. Authenticity asks whether it is what it purports to be. Completeness asks what relevant material is included or missing within a defined scope. Accuracy asks whether a recorded value correctly represents the event being assessed. Attribution asks which account, device or person can properly be associated with the activity.
A favourable answer to one question does not settle the others. An unchanged file can contain an incomplete export. A genuine application record can contain an incorrect device-clock value. A message associated with an account does not, without further evidence, identify the person who composed it.
A defensible conclusion states which question was tested, the material used, the result and the remaining boundary.
Sources and interpretation
Each entry identifies its limited supporting proposition and documentation-check date. General methodology is not a current device-support list. These checks are not laboratory validation; availability and method suitability require assessment for the exact source and configuration.
- NIST FIPS 186-5: Digital Signature Standard (2023)
Signature generation uses a private key; verification uses the corresponding public key. This reference is not a claim of laboratory certification.
Technical standard · Documentation checked
- NIST mobile device forensics guidance
Methodological background (2014), not a current device-support list.
General methodology · Documentation checked
- NIST IR 8354: Digital Investigation Techniques, final scientific foundation review (2022)
Methodological background: discovery is not exhaustive and changing software can change artefact meaning. Not a device-support list.
General methodology · Documentation checked
- SWGDE 18-Q-001-2.1: Minimum Requirements for Testing Tools Used in Digital and Multimedia Forensics, version 2.1, 7 March 2024
Testing guidance, not certification or endorsement of the practice.
General methodology · Documentation checked
Confidential case review

