When a valid user requests content, Watermarking distributes segments of content based on a pattern that’s unique to that user. If your content is pirated or redistributed, you can analyze the content and extract the user’s watermarking pattern to identify the user that originally leaked the content.
Watermarking responds to segment requests from an end user with either an "A" or "B" encoded variant and the resulting sequence of segments forms a unique pattern. This pattern is determined by a watermarking token (WMT) that's included in each request.
If your stream is shared with unauthorized users, you or your watermarking vendor can forensically analyze the segments to extract the unique pattern. You or your watermarking vendor can then identify the source of the leak by searching a database that maps patterns to authorized users.
Below is a simplified example of how the Akamai platform uses a variant path based on the watermarking pattern in the WMT. The end user doesn't know that variant segments have been retrieved from different sources and his viewing experience is seamless.
- You need to work with a third-party watermarking vendor. The watermarking vendor’s role is to generate and store watermarking patterns for you in a WMT, as well as preprocess your content into "A" and "B" variants and place it onto the origin using a strict naming convention.
- What does Akamai do? Akamai parses, verifies, and acts upon the WMT supplied by the client during segment requests. The Akamai edge verifies the token and then maps requests for an "A" or "B" variant based on the sequence number of the segment, and the user’s specific watermarking pattern. We also offer various, optional settings you can apply to further protect the WMT during transport.
- The client application has a role, too. Your client application ("player") needs to get the WMT from the watermarking vendor and then pass it along with each request to Akamai.
See the Adaptive Media Delivery documentation for complete usage instructions.
We offer several services to protect your content, including geographic restrictions such as Content Targeting and Enhanced Proxy Detection, access restrictions like Token Authorization and transport security with Media Encryption, as well as secure (HTTPS) delivery via Enhanced or Standard TLS. However, these can't help if your content has already been compromised.
Watermarking isn't used to protect your content—it's used to identify the source that compromised it.
Updated 8 months ago