Baseband model of the Turbo DeCodulation system.

Contexts in source publication

Context 1

... nodes and the independent short LDPC sub-codes a high degree of parallelization in processing can be achieved. Fig. 1 the baseband model of the proposed Turbo DeCodulation system is depicted. The inner iterative loop in the receiver corresponds to a BICM-ID system [5], while the outer iterative loop is similar to an ISCD system [8,9]. By connecting these two loops, e.g., by a common channel code, we obtain a Turbo DeCodulation receiver which performs ...

Context 2

... L [ext] SDSD (x) = 0 for the non-systematic bits. In Fig. 1 this is indicated by the dashed ...

Context 3

... the upper left part of Fig. 2 we have the parameter estimation nodes P, which represent (8). Note, the parameter estimation is actually part of SDSD as seen in Fig. 1, but we will consider it separately here because it is based on a different equation and accepts other messages, yielding a new type of graph node. Similarly to the SDSD nodes S the nodes P are only connected to the systematic variable nodes V and the edges are permuted by an interleaver block π out /π -1 out . Using (8) the nodes P ...

Context 4

... the proposed Turbo DeCodulation scheme exploiting the iterative possibilities of the complete Tanner graph of Fig. 2 (i.e., three iterative loops in Fig. 1) the gain with respect to the non-iterative soft decision case increases significantly further to ∆ E b /N0 = 4.2 dB. This is more than the combined gain of BICM-ID and ISCD, proving the effectiveness of the presented graph-based Turbo DeCodulation. The dash-dotted line in Fig. 3 depicts the theoretical limit for all simulated schemes ...