Abstract

Self-organized, community-built wireless mesh networks are among the fastest-growing and most cost-effective methods for providing real-time connectivity to underserved communities. WMNs autonomously route their own traffic, compute balances, and self-heal, allowing for optimal partitioning and community mingling. We present initial results from a deployed Village network that utilizes instant neighbor discovery, crowd mesh, and partitioned delay networks. We used real smartphones and laptops, along with a volunteer-provided community mesh network, at a semi-urban test site. We measured throughput, Latency, drop rate, and network control over several busy operational days. Compared to older, rigid mesh models, our approach boosts throughput by 25 percent, reduces delay by 18 percent, and maintains packet delivery above 95 percent. We also map out interference, growth, and attack issues, then suggest tweaks to broaden the network. By publishing working code and field data, we add a practical layer to community networking research and point other teams toward future projects and deployments.