Mesh Networking for Seismic Monitoring - The Sumatran cGPS Array Case Study

In this paper, the use of mesh networking for the Sumatran continuous-Global Positioning System array (SuGAr) was proposed to optimize the telemetry between the nodes in the array and a centralized remote data server. Currently, data from the array is transferred directly to the data server via dedicated satellite links from each node in the array. This requires each node to be connected to a satellite modem and is cost-ineffective as subscriptions for each link have to be maintained. In this paper, the use of long range multi-hop radios to establish mesh networks was proposed in order to reduce the number of satellite links required as well as to perform cluster-based compression to reduce the bandwidth requirements. Due to the geographical distances between the nodes and the limitations of the radio range, clusters of mesh networks will be formed and each cluster will require only a minimum of only one satellite link. Furthermore, rather than compressing the collected data separately at each GPS station, the cluster-heads can also perform compression on all the collected data within the cluster prior to transmission to the data server. Due to the inherent correlation between the data from the GPS stations, the compression of the larger dataset will reduce the total bandwidth required to transfer the observed data. To demonstrate the optimization achieved, actual GPS data collected from the SuGAr network from the last two months (61 days) of 2007 was used in the evaluation of the proposed solution and results showed that the proposed solution significant reduced the number of bytes transmitted to the data server via the satellite links.