Defeating the Ghost: A Geometric Defense Against Sybil Attacks in Mobile Social Networks

TL;DR

Mobile Social Networks (MSNs) are vulnerable to "Sybil attacks," where one malicious device assumes multiple identities to hijack routing. This paper proposes a dual-layer defense: clients measure physical proximity via RSSI, while servers use Trilateration to verify if those nodes actually "fit" into a 2D physical space. If a node's geometry doesn't add up, it's kicked out of the routing table.

Background & Positioning

As a subset of Delay Tolerant Networks (DTNs), MSNs rely on the "Store-Carry-and-Forward" (SCF) mechanism. Efficiency depends on social properties (friendship, encounter frequency). However, this social-awareness is a double-edged sword. The paper sits at the intersection of Social-Aware Routing and Network Security, specifically addressing the gaps left by traditional identity management systems in highly dynamic environments.

The Problem: The "Reasonable Forge" Dilemma

In MSNs, nodes report their locations to a server to facilitate routing. A malicious node can easily create "Sybil nodes" (virtual identities) with fake locations. If the attacker knows where its neighbors are, it can forge a virtual location that looks perfectly plausible to the routing algorithm.

The Core Insight: Even if an attacker can lie about a coordinate, it is mathematically difficult to lie consistently about relative distances to multiple neighbors simultaneously.

Methodology: Geometry as a Truth Machine

The proposed solution splits the workload between the Mobile Client and the Location Server.

1. Client-Side: Distance Measurement

Instead of just trusting GPS, nodes use Received Signal Strength Indicator (RSSI).

• Node $v$ sends a signal.
• Neighbors $u$ report the received power $p_{uv}$.
• Node $v$ calculates the distance $d_{ij}$ using free-space wave propagation.

2. Server-Side: The Consistency Check

When the server receives these distances, it runs Algorithm 1: Eliminate_Sybil_Node.

• Trilateration: The server picks three nodes to define a local coordinate system.
• Physical Mapping: It calculates the expected position of a fourth node $k$ based on its reported distances.
• The "Truth" Test: If the calculated distance between two nodes $|p_i-p_j|$ significantly differs from the measured distance $d_{ij}$ beyond a threshold $ϵ$, the system identifies a conflict. Since real physical objects cannot exist in two places at once (or fail Euclidean geometry), the "impossible" nodes are flagged as Sybil identities.

Experiments & Analysis

The authors provide a theoretical proof of the algorithm's effectiveness:

• Theorem 2: If the reference nodes are real, no Sybil node can pass the consistency check.
• Theorem 3: If a Sybil node is accidentally used as a reference, the set of "valid" nodes will shrink, but the Sybil node will still likely be isolated because it cannot maintain consistency with the wider network of honest nodes.

Cost Considerations

• Energy: For a neighborhood of $n$ nodes, the process requires $(2n+1)n$ packet switches. This is the primary trade-off—security comes at the cost of communication overhead.
• Computational: The server handles the heavy lifting of Trilateration, which is efficient for modern location servers.

Critical Insight & Conclusion

Takeaway

The genius of this approach is moving away from cryptographic trust (which can be stolen) to physical trust (which is bounded by the laws of geometry). By forcing nodes to prove their "physicality" through distance consistency, the bar for a successful Sybil attack is raised from simple software spoofing to complex signal manipulation.

Limitations

The paper assumes a "Free-Space" propagation model for RSSI. In real-world urban environments, multipath fading and obstacles (buildings, moving vehicles) can cause RSSI values to fluctuate wildly, potentially leading to "False Positives" where honest nodes are flagged as Sybil nodes.

Future Outlook

To make this production-ready, Integrating Machine Learning to filter RSSI noise or using Time-of-Flight (ToF) for more accurate distance measuring would be the logical next step for securing the next generation of 6G/IoT social networks.