Device Identification via Multi-Cookie Hashing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods, such as cookies, are inadequate for reliably and persistently identifying computing devices, especially when users attempt to avoid recognition or when identification is needed across multiple sites, as they can be easily deleted or do not provide sufficient assurance of device consistency.
Innovation Solution
A method using multiple cookies to form a device identifier, where individual cookie parameters are represented as irreversible hashes, allowing for anonymous comparison and correlation to determine device recognition, ensuring that changes in cookies do not prevent accurate identification by quantifying the stability and uniqueness of cookie attributes and comparing them against a predetermined threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cookies are used to identify client devices, then device recognition is enabled, but reliability is reduced when users delete cookies or use multiple sites
Solution Approach 1:
The device identifier is segmented into multiple independent cookies stored across different hosts. Instead of relying on a single cookie, the system divides the identification function across multiple cookies (e.g., cookie1 on host1, cookie2 on host2), where each cookie contains a portion of the overall device identifier. This segmentation ensures that deletion or modification of individual cookies does not completely defeat device recognition.
Solution Approach 2:
The cookie-based identification system is designed to serve multiple functions: it works for single-site recognition, cross-site identification, and maintains functionality even when some cookies are deleted. The device identifier formed from multiple cookies can be used universally across different websites and contexts, providing adaptable identification regardless of the specific browsing scenario.
2Stability of the object's composition
If individual cookies are used for identification, then simplicity is maintained, but stability is reduced when cookies change over time
Solution Approach 1:
Multiple cookies are merged to form a single device identifier. The system combines information from multiple cookies (stored on different hosts or with different attributes) into one comprehensive device identifier. This merging process stabilizes the identification by ensuring that the device can be recognized even if individual cookies change or are deleted, as long as the overall pattern across multiple cookies remains consistent.
Solution Approach 2:
The system creates multiple copies of identification data across different cookies rather than relying on a single copy. Each cookie contains a portion of the device identifier, creating redundant copies that can be used for recognition. This copying strategy ensures that if one cookie changes or is deleted, other copies remain available to maintain device identification stability.
3Adaptability or versatility
If cookies are made persistent across sites, then cross-site identification is enabled, but privacy concerns increase
Solution Approach 1:
Each cookie is stored with specific local characteristics (different hosts, different paths, different attributes) rather than being a uniform universal identifier. This local quality allows the system to gather identification information across sites while maintaining some degree of locality and control at each host level, reducing the privacy intrusion compared to a single universal tracking cookie.
Data Source
AI summary
Multiple cookies of a client device are used to form an identifier of the client device such that a change in one or even several browser cookies does not defeat proper device recognition. The cookies are included in the identifier such that individual cookies can be parsed for separate comparison with corresponding cookies of known devices. However, to protect privacy of all devices, individual parameters of the constituent cookies are represented with irreversible hashes of the respective parameters. Recognition involves quantification of a degree of correlation between the cookies and corresponding cookies of each of the known devices. To quantify the degree of correlation, the observed stability and uniqueness of each cookie, and each cookie attribute, is considered.