Time Encoding of Bandlimited Signals: Reconstruction by Pseudo-Inversion and Time-Varying Multiplierless FIR Filtering

We propose an entirely redesigned framework of bandlimited signal reconstruction for the time encoding machine (TEM) introduced by Lazar and Toth. As part TEM consists in obtaining integral values a input over known intervals, it theoretically amounts to applying linear operator on input. then approach general question pseudo-inversion this operator. perform task numerically iteratively using projections onto convex sets (POCS). The algorithm can be implemented exactly discrete with multiplications that are all reduced scaling signed powers two, thanks use relaxation coefficients. Meanwhile, achieves rate convergence similar For real-time processing, we approximate time-varying FIR implementation, which avoids splitting into blocks. finally some preliminary semi-convergence analysis under data noise.