PTLM: A Passive TCP Latency Monitor for Enterprise and Living Lab Networks

Latency monitoring is a fundamental requirement for TCP-based services in scenarios where responsiveness must meet strict deadlines, typically below 100 ms. This is particularly relevant for emerging applications such as remote driving in smart mobility testbeds and remote operation of industrial machinery in Industry 5.0 environments. Traditional active monitoring techniques, which rely on probe traffic (e.g., ICMP ping), introduce additional load and may fail to capture the true latency experienced by the target service.In this paper, we present PTLM (Passive TCP Latency Monitor), a system that enables non-intrusive and passive monitoring of TCP latency by directly observing application flows at intermediate nodes along the client–server path. PTLM leverages the principles of passive latency estimation introduced in previous work (e.g., pping) but extends them with a practical and deployable solution specifically designed for enterprise networks and experimental living labs.We report the results of an extensive experimental campaign including feasibility, performance, and reliability tests. In particular, experiments conducted in the MASA (Modena Automotive Smart Area) living lab demonstrate that PTLM can successfully detect latency variations that traditional ping-based methods fail to observe. The results validate PTLM as an effective monitoring solution for latency-critical applications, with direct relevance to enterprise deployments (IGNITE 5.0 project) and experimental testbeds supporting autonomous driving and other time-sensitive services.