Exploiting Partial Reconfiguration through PCIe for a Microphone Array Network Emulator

Figure 4

The data fusion front-end is capable of simulating a sound field with multiple sound sources (green diamond) and multiple nodes (pink circles). The front-end generates PDM signals for each microphone in each node that are then sent to the FPGA back-end. The FPGA generates the corresponding polar steering response map (a) which is then fed to the data fusion algorithm to generate a probability map (b) and estimate the localization error.

(a) Fetching polar steering response map from the FPGA

(b) Fusing the data to locate the sound source