Temporal Obfuscation

Temporal obfuscation is ZEclipse’s defense against timing-based blockchain analysis, based on the design in docs/TEMPORAL_OBFUSCATION.md.

It adds controlled randomness to when hops and transfers execute, so observers cannot correlate incoming and outgoing transactions purely by time.

Components

• TemporalObfuscationManager – low-level scheduler that decides when to send transactions.
• TimeObfuscator – convenience wrapper for obfuscating single and multi-transfer flows.
• TimingEnhancedConnector – SDK connector that combines anonymous transfers with timing privacy.

These live in the TypeScript SDK (app/src/timing and connector modules).

Timing Features

Random Execution Delays

Each hop is executed with a random delay:

• Delay range depends on the hop index and selected strategy.
• Prevents easy mapping of “in” and “out” transactions across hops.

// Conceptual example from the SDK
const obfuscator = new TimeObfuscator(connection)
const executeAt = await obfuscator.obfuscateTransfer(transaction, hopIndex)

Time-Sliced Execution

For multi-recipient transfers, ZEclipse spreads transactions across time slices within a configurable window:

• Avoids bursts of related transfers.
• Makes outputs appear as independent events.

const schedule = await obfuscator.obfuscateMultiTransfer(transactions, hopIndex)
// schedule maps each transaction to a specific execution time slice

Timing Strategies

From TEMPORAL_OBFUSCATION.md, strategies roughly behave as:

Strategy	Min Delay	Max Delay	Window	Privacy Level	Typical Use Case
MINIMAL	200 ms	2 s	30 s	★★☆☆☆	Small transfers, testing
STANDARD	500 ms	8 s	60 s	★★★☆☆	General-purpose transfers
BALANCED	1 s	10 s	60 s	★★★★☆	Medium-value transfers
MAXIMUM_PRIVACY	2 s	30 s	180 s	★★★★★	High-value / sensitive txs
CUSTOM	configurable	configurable	configurable	depends on config	Advanced users only

Programmatically, you typically configure this via the connector:

connector.setTimingStrategy(TimingStrategy.MAXIMUM_PRIVACY)

Configuration

The core configuration interface is:

export interface TemporalObfuscationConfig {
  minDelay: number              // ms
  maxDelay: number              // ms
  randomBatchOrder: boolean
  timeSlicedExecution: boolean
  timeWindowSize: number        // ms
  timeSliceInterval: number     // ms
}

Strategies map to concrete configs, which you can override if you need more control.

Diagnostics & Entropy

The timing system exposes diagnostics so you can quantify privacy:

const stats = connector.getTimingStats()
/*
{
  averageHopDelay: 2543,
  timeDistribution: [25, 30, 10, 85, 40, ...],
  entropyScore: 87,
  correlationResistance: 79,
}
*/

Key ideas:

• Entropy score measures how unpredictable the timing distribution is.
• Correlation resistance approximates resistance to hop-to-hop timing correlation.

For high-value transfers, the docs recommend:

• Entropy score > 75.
• Correlation resistance > 80.

Effect on Anonymity

Base ZEclipse already achieves about 5.3M possible paths from the multi-hop split graph.

Temporal obfuscation:

• Multiplies this effective anonymity set by roughly 1–5× (strategy dependent).
• Makes it difficult to confidently rule out candidate paths using timing alone.

The tradeoff is straightforward:

• More delay ⇒ more privacy.
• Less delay ⇒ faster UX.

For most DApps, BALANCED or STANDARD strategies offer a good middle ground.