Aviz Networks details mobile-network session data masking
Companies mentioned
Aviz Networks describes deterministic masking for mobile-network session records that preserves joins and correlations while removing raw subscriber and device identifiers. The post matters to enterprise network, security, and analytics teams that need to export data without exposing identity fields.
Research Overview
The blog centers on session records generated as users move across mobile networks and explains why those records often contain sensitive details such as subscriber IDs, device IDs, IP addresses, tower information, and session timing. It says exporting those records to analytics, SIEM, data lake, firewall, or partner systems can expose personal data if the identifiers remain in place.
It also lays out three example environments, labeled ASN-1, ASN-2, and ASN-3, to show different trust boundaries for exporting records to a threat analytics tool, a big data analytics platform, and a firewall tool. In each case, the article says the masking approach is applied before data leaves the boundary.
Technical Breakdown
ASN uses deterministic tokenization to replace sensitive values with fixed-length strings at the session-record level. The post says this preserves non-sensitive fields and allows the same original value to map to the same token each time, which supports joins, grouping, counting, and trend analysis.
The article contrasts this with packet-level masking, which it says overwrites detected patterns in raw payloads on a byte-by-byte basis. It states that packet-level masking can weaken subscriber correlation across sessions, protocols, and network elements, while ASN masks aggregated records and can cover fields such as MSISDN, IMSI, IMEI, and subscriber IP.
Operational Impact
The blog lists third-party analytics sharing, regulatory and audit reporting, cross-team troubleshooting, and long-term data lakes as use cases. It says masked records can support per-subscriber analysis across gateways and towers while keeping identity fields out of downstream systems.
It also says deterministic masking helps maintain analytical usefulness after export and reduces exposure if a stored data set is compromised, since no raw identities are recorded. The post adds that the transformation is irreversible and does not rely on reverse lookup tables.
Key Findings
The central claim is that mobile-network session data can be masked in a way that prevents reversal while still allowing correlation across downstream tools. The article says this approach is intended to avoid the tradeoff between privacy and analytics that comes with deleting fields or exporting raw identifiers.
The post concludes that the design should operate at system boundaries and avoid added overhead where computation is most important. It says the result is a system that follows policy while still supporting subscriber-level analysis.
This Blog Signals brief is a fact-based summary of the vendor blog.
Blog post, originally published by Guruprasad Hegde at aviznetworks.com.