(Example format)
Modern distributed systems face a fundamental trade-off: moving all data to the cloud preserves analytic depth but incurs high latency and bandwidth costs, while processing entirely on the edge reduces communication overhead but often sacrifices accuracy and coordination. Baymac (Bayesian Media Access & Coordination) proposes a hybrid approach. By embedding lightweight probabilistic models into edge nodes and coordinating their transmissions through a modified MAC protocol, Baymac enables real-time adaptation to network conditions and data priorities. baymac
The rapid expansion of Internet of Things (IoT) devices and edge computing has introduced new challenges in data throughput, latency reduction, and energy efficiency. This paper introduces Baymac , a novel middleware architecture designed to bridge the gap between centralized cloud analytics and resource-constrained edge nodes. Baymac leverages adaptive Bayesian inference and media access control (MAC)-layer optimization to dynamically allocate bandwidth and computational tasks. Our simulations show that Baymac improves data processing efficiency by up to 34% compared to traditional round-robin edge schedulers, while reducing average packet latency by 18%. The rapid expansion of Internet of Things (IoT)
Baymac reduced latency by 18% compared to the best baseline (threshold-based) and improved energy efficiency by 28%. Our simulations show that Baymac improves data processing
