Modeling and Analysis of Task Scheduling Strategies for Jitter Minimization in Real-Time Systems

In real-time computing systems, maintaining a strict temporal determinism is as important as ensuring functional correctness. A primary pain-point in these architectures is timing jitter. Timing Jitter is defined as the difference in task execution intervals relative to a predefined isochronous schedule. Such latency variations can degrade the closed-loop control stability and network throughput in time critical applications. This study is done using MATLAB simulation to evaluate scheduling heuristics with focus on temporal predictability. We analyze both time-triggered deterministic models and probabilistic non-deterministic models. While deterministic scheduling enforces a rigid periodic duty cycle, the non-deterministic model introduces stochastic noise to emulate real-world interrupt latencies and resource contention. We quantify jitter through statistical signal analysis, measuring mean temporal drift, variance and worst-case timing error. Our comparative data reveals that deterministic scheduling preserves the phase alignment with minimal jitter. Whereas non- deterministic execution exhibits significant timing skew. This highlights why rigorous scheduling is essential for the operational integrity of time-sensitive systems.