Node Selection-Based Anonymous Network Performance Optimization Method

As the most renowned anonymous network, Tor has a broad user base. However, with the increase in users, it faces some performance issues. For instance, relays may become overloaded as more users connect to Tor, leading to network con-gestion. Additionally, changes in network conditions may increase circuit construction latency. These issues severely impact user experience. Studies indicate that the latency for Tor clients is five times higher than for direct connections. This dissatisfaction leads to decreased active users, reducing the network's anonymity. User experience in network access is mainly concerned with network response and data download time. We represent response time by Time to First Byte (TTFB) and Time to Last Byte (TTLB) as the download time. This paper proposes a Multi-Metrics Node Selection (MMNS) method, considering node bandwidth, uptime, and circuit build time. Experimental results show that our proposed method significantly improves network response and data download time, especially for low-latency response applications. Furthermore, we analyzed anonymity and security, confirming that this method ensures security and anonymity for an anonymous communication network.