Tree Search Based Symbol Detection for Multi-Band Petahertz Communication With Adjacent-Band Crosstalk

For Petahertz communication (PetaCom) employing multi-band transmission, the crosstalk mainly occurs between adjacent bands when a suitable optical filter is adopted at each detector. In this case, the channel matrix can be modeled as a tridiagonal matrix. Utilizing this fact, this work proposes a simplified tree search method for symbol detection in multi-band PetaCom, which does not require QL decomposition and can significantly reduce the computational complexity compared with conventional maximum-likelihood (ML) detection sphere decoding (SD). Experimental measurement verifies the model of tridiagonal channel matrix. Besides, simulation results demonstrate that the proposed method can achieve the same symbol error rate as ML detection and SD, while involving much less running time.