TLCTC Blog - 2025/10/01
Mapping CWE-514 (Covert Channel) to TLCTC: A Cause-Oriented Analysis
CWE-514 (Covert Channel) maps primarily to TLCTC #8 Physical Attack, because the root cause it describes is the abuse of physical properties (timing, power, EM, cache, etc.) to leak information — i.e., a physical/side-channel vulnerability, not a misuse of a legitimate software function.
The Primary Mapping: CWE-514 to TLCTC #8
The TLCTC mapping table shows CWE-514 → #8 Physical Attack. Now the reasoning in TLCTC terms (concise, cause-oriented):
- TLCTC is cause-oriented (Axiom I & III).
- TLCTC maps each generic vulnerability to one top-level threat cluster (the cause). CWE-514 describes a weakness in the physical channel (timing/power/etc.) used to exfiltrate data — that vulnerability’s cause is physical, so it belongs to #8.
Why Not #1 (Abuse of Functions) as the Primary Cause?
#1 Abuse of Functions covers abuse of legitimate software functions or logic (business-logic abuse, improperly exposed APIs, CSRF, parameter tampering, etc.). Covert channels aren’t primarily about abusing an application’s intended logic or API semantics — they exploit side-effects of hardware/software execution that create an unintended physical signal or state. That difference (logic vs physical side-channel) is exactly why TLCTC places covert channels under #8.
Modeling Complex Chains: When to Use #1 → #8
If the initial exploit is an abuse of a legitimate function and that abuse then creates or amplifies a physical covert channel used for leakage, the attack can be modelled as a sequence: #1 → #8.
Example of a Multi-Stage Attack
An attacker abuses a legitimate API that allows fine-grained control of CPU workload (legitimate function) to create specific timing/power patterns that a co-located observer decodes. Here the first causal step is function abuse (#1); the second is information leakage via a physical channel (#8). That sequence is legitimate TLCTC notation for a multi-stage attack.
Which mapping is the “correct” one depends on the root cause you’re classifying.
Practical Guidance for Classification
- Read the CWE/CVE context: does it call out a side-channel (timing, EM, covert storage/timing)? If yes, map to #8.
- Identify the enabling cause: Does the exploit rely on misusing a legitimate function to generate or amplify that side-effect? If yes, model as #1 → #8 to show both cause steps. TLCTC encourages this attack-path notation when multiple causes are involved.
Quick Concrete Example Pair
The Problem with Broad CWEs like CWE-514
CWE-514 (“Covert Channel”) is a bit too broad on its own. It’s useful as an umbrella label, but for engineering, detection, and control mapping you almost always want the more specific subtypes. Here’s why in TLCTC terms:
- Cause clarity matters (TLCTC axioms). TLCTC is cause-oriented: each generic vulnerability should map cleanly to one threat cluster and drive precise controls. Broad CWE “buckets” blur the cause and weaken control selection.
- CWE-514 mixes mechanisms. “Covert channel” can mean timing-based leakage, storage-based leakage, cache micro-architectural effects, etc. Those sub-mechanisms have different mitigations (cache partitioning vs. rate limiting vs. constant-time ops). Using the umbrella makes it harder to pick the right #8 Physical-layer controls.
- CWE-385 Covert Timing Channel → #8
- CWE-515 Covert Storage Channel → #8
Practical Recommendations
- Use CWE-514 as a high-level tag to indicate a physical side-channel risk (#8 Physical Attack).
- For remediation and detection, relabel to the precise subtype (e.g., CWE-385 for timing vs. CWE-515 for storage) and then pick controls accordingly.
- If you need operational tracking, use TLCTC’s dual-layer notation to preserve precision:
- Strategic: #8 (Physical Attack)
- Operational ID: e.g., TLCTC-08.01 (Timing), TLCTC-08.02 (Storage)
CWE-514 is fine as an umbrella, but it’s too broad for precise defense planning. Prefer more specific variants and map them to TLCTC #8 with operational refinement so your controls and telemetry line up with the actual covert-channel mechanism.