Tree structure-based bit-to-symbol mapping for multidimensional modulation format

Bit-to-symbol mapping is one of the key issues in multidimensional modulation. In an effort to resolve this issue, a tree structure based bit-to-symbol mapping scheme is proposed. By constructing a tree structure of constellation points, any neighboring constellation points become nearest-neighbor constellation points with minimum Euclidean distance, which in turn, changes the bit-to-symbol mapping problem in multidimensional signal modulation from random to orderly. Then, through the orderly distribution of labels, the minimum Hamming distance between the nearest neighboring constellation points is ensured, eventually achieving bit-to-symbol mapping optimization for multidimensional signals. Simulation analysis indicates that, compared with random search mapping, tree mapping can effectively improve the bit error rate performance of multidimensional signal modulation without multiple searching, reducing the computational cost.