[Legacy Report] Real-time Seismic Tomography in Sensor Networks

---

Existing seismic instrumentation systems do not yet have the capability to recover physical dynamics with sufficient resolution. At present, raw seismic data are typically logged at a few stations then manually retrieved months later for post processing and tomographic imaging at a central server. Thus neither real-time nor high-resolution tomography imaging are possible today which limited our understanding of geospatial dynamics. We are developing a VolcanoSRI (Volcano Seismic Realtime Imaging) system, a large-scale mesh network of low-cost seismic stations, that sense and analyze seismic signals, and compute real-time, three-dimensional fluid dynamics of a volcano conduit system (e.g., 4D volcano tomography) within the sensor network. Realizing such a VolcanoSRI system requires a transformative approach to tomography computation algorithm, collaborative signal processing, and the associated sensor network design. In this talk, we present our recent research on distributed tomography algorithms that process data and invert volcano tomography in the network, while avoiding costly data collections and centralized computations. The new algorithm distributes the computational burden to sensor nodes and performs realtime tomography inversion under the constraints of network resources.

---

---

---

---