We analyze the performance of trellis-coded pulse-position modulation with
block decision-feedback equalization (BDFE) and parallel decision-feedback
decoding (PDFD) on indoor, wireless infrared channels, We show that the red
uced complexities of BDFE and PDFD as compared to maximum-likelihood sequen
ce detection allow for better codes whose increased coding gain more than c
ompensates for the penalty due to suboptimal detection. We quantify these n
et gains in performance over a range of dispersive channels, indicating whe
re BDFE and PDFD provide the best performance, Finally, we present Monte Ca
rlo simulation results to verify our analysis.