Optimal Joint Decoding of Correlated Data Over Orthogonal Multiple-Access Channels With Memory

摘要: Motivated by the structure of basic sensor networks, we study an optimal joint decoding problem in which real-valued outputs two correlated Gaussian sources are scalar quantized, bit assigned, and transmitted, without applying channel coding or interleaving, over a multiple-access that consists orthogonal point-to-point time-correlated Rayleigh fading subchannels used with soft-decision demodulation. Each subchannel is modeled nonbinary Markov noise discrete was recently shown to effectively represent it. The have memory captured time-varying correlation coefficient governed two-state first-order process. At receiver side, design sequence maximum posteriori (MAP) decoder exploit between sources, their temporal memory, redundancy left quantizers' indexes, channels' outputs, memory. Under simple practical case using two-level source quantization, propose model estimate behavior quantized sources. We then establish necessary sufficient conditions under delay-prone MAP can be reduced instantaneous symbol-by-symbol decoder. illustrate our analytical results system simulation demonstrate appropriately harness characteristics achieve improved signal-to-distortion ratio performance for wide range conditions.