What Is ‘Data Sanitization’ in the Context of Oracle Feeds?

Data sanitization is the process of cleaning and validating raw price data received from external sources before it is used by the smart contract. This involves filtering out extreme outliers, checking for sudden, unrealistic price jumps, and ensuring the data format is correct.

The goal is to remove bad or malicious data points that could compromise the contract's financial logic. A decentralized oracle network typically performs this sanitization across its many nodes to achieve a consensus on a clean price.

Why Is the Median Calculation (MWAP) Often Considered More Robust against Outliers than the Mean (TWAP)?
How Do DONs Determine Which Data Points Are “Outliers” during Aggregation?
How Does the Number of Inputs and Outputs Affect the Size of a Transaction?
Why Is It Dangerous to Sign the Raw Transaction Data Instead of Its Hash?
How Do “Data Providers” and “Node Operators” Differ in a DON?
What Is a “Data Provider” in the Oracle Ecosystem, and How Does It Interact with the Oracle Network?
What Is “Data Aggregation” in the Context of DONs?
How Do Exchanges Ensure the Integrity of the Data Used for the Reference Rate Calculation?