A Novel Method of Signal Fusion Based on Dimension Expansion.

A novel method of signal fusion, namely multi-dimensional unified signal (MDUS) fusion algorithm, is proposed based on dimensionality expansion of the cognitive radio (CR). The paper focuses on the issue of under-utilized and overcrowded spectrum bands in CR, through multi-component signal fusion. The proposed MDUS fusion algorithm can meet the basic requirement of signal fusion, that is, transmitting multi-component signal in one signal simultaneously, while basic signal elements, such as amplitude and bit rate, are not constrained by the fusion. The one-dimensional imaginary domain of the signal can be expanded to two, three or higher-dimensional imaginary space by exploiting the property of logical recursion. The original data flows can be extended to three, four or higher dimensions according to functional demands. Thus, the complementary aims of mutually independent orthogonal space and multivariate signal transmission without interference, are achieved by fusing multiple uni-dimensional signals into a multi-dimensional signal. Hence, based on the electromagnetic environment, a more flexible solution can be provided for spectrum occupancy of CR users. Specifically, the number of Orthogonal Frequency Division Multiplexing (OFDM) sub-bands and MDUS dimension can be dynamically converted, based on the MDUS-OFDM structure. The proposed approach can be utilized to provide more accurate estimation and prediction of transmission quality for CR Networks. Simulation results show the MDUS offers both BER and dimensionality advantage compared to the OFDM. Additionally, single-dimensional spectrum patterns can be expanded to alterable multi-dimensional spectrum patterns owing to the flexibility of MDUS-OFDM structure. [ABSTRACT FROM AUTHOR]