1-D digital waveguide modeling for improved sound synthesis

Digital waveguide modeling of one-dimensional vibrating structures in musical instruments is developed further in this paper based on finite difference time domain (FDTD) formulations. It is shown that such realizations can be developed, efficient enough to run in real time on modem processors and more flexible than the computationally less expensive commuted synthesis. FDTD waveguide structures are analyzed with lossless and lossy propagation, input and output ports, terminations and scattering junctions. The formation of static initial state of displacement wave from pointwise excitation in plucked string models is demonstrated, not possible with simple waveguides. Related sound synthesis examples can be found at www.acoustics.hut.fi/demos/fd1.