Orthogonal Expanded Memory Polynomial Model for Circular Complex Gaussian Processes

Publisher:
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Publication Type:
Journal Article
Citation:
IEEE Transactions on Signal Processing, 2024, 72, (99), pp. 3287-3302
Issue Date:
2024-01-01
Full metadata record
This paper proposes an orthogonal expanded memory polynomial (OEMP) model for circular complex Gaussian processes. The basis functions of the OEMP model are constructed from those of the expanded memory polynomial (EMP) model recently derived by expanding the conventional memory polynomial (MP) model for I/Q imbalanced signals, and proved orthogonal to each other thanks to the derived closed-form bivariate distribution moments. In addition to the perfect duality with its power spectral density expression, the OEMP model also demonstrates unique minimum residual power property. It is believed that the OEMP model will be highly beneficial to nonlinearity mitigation for wideband wireless communication systems. Simulation and experimental results obtained from a terahertz system with 4.25 GHz bandwidth and 252 GHz carrier frequency confirm that significant performance improvement can be achieved for digital post-cancellation with low-complexity after applying the proposed OEMP model.
Please use this identifier to cite or link to this item: