Prof. Shlomi Arnon

We propose an indoor wireless infrared downlink scheme for high-data-rate multiuser connectivity with diffuse channels. The scheme is based on synchronous code-division multiple access with unipolar Hadamard codes. The orthogonality of unipolar Hadamard codes enables multiuser operation with relatively short codes. Thus, practical downlink rates of tens of Mb/s for each user can be obtained. However, multipath reflections in diffuse channels cause strong multipath dispersion and, consequently, severe distortion. This distortion becomes even more severe in a multiuser environment, as the dispersed incoherent infrared radiation of all users aggregates together. To mitigate this distortion, we use a novel adaptive multilevel serial composite decision feedback and feedforward equalizer. We investigate the system's performance with the proposed equalizer, and compare it with the performance of the same system, both composite decision-feedback and feedforward equalizers, and with a conventional decision-feedback equalizer (DFE). Our results show that the proposed scheme enables a high-data-rate multiaccess link and eliminates most of the multiuser distortion. Furthermore, it improves system performance in a multiaccess environment, as compared with the other composite equalizers and DFE for the same complexity. We also compare other coding schemes, and show that Hadamard codes are on top of the other codes.