Ordered Successive Interference Cancellation Algorithm OSIC in MIMO Signal Detection

In MIMO (Multiple-Input Multiple-Output) communication systems, signal detection is a critical component, particularly when multiple antennas transmit simultaneously, requiring the receiver to accurately separate and decode different data streams. The Ordered Successive Interference Cancellation (OSIC) algorithm is a common MIMO detection technique whose core principle involves progressively eliminating strong interference signals to enhance detection performance. The OSIC algorithm workflow primarily consists of three steps: signal ordering, interference cancellation, and iterative detection. First, the algorithm orders received signals typically based on Signal-to-Noise Ratio (SNR) or Minimum Mean Square Error (MMSE) criteria to determine the decoding sequence. Then, it sequentially decodes the strongest signal and reconstructs interference components using decoded signals, subtracting interference from the original received signal to improve detection accuracy for subsequent signals. In code implementation, this requires maintaining a dynamic list of undetected streams and updating the channel matrix after each cancellation step. In QAM16 modulated systems, the abundance of constellation points results in higher detection complexity, making sorting strategies crucial for OSIC performance. Three common ordering approaches include: ZF (Zero-Forcing) criterion-based ordering, MMSE criterion-based ordering, and optimal ordering (such as in V-BLAST). ZF ordering offers computational simplicity but weaker noise resistance, MMSE ordering performs better under balanced SNR conditions, while optimal ordering achieves best performance at the cost of higher computational complexity. From an implementation perspective, ZF ordering can be implemented using pseudo-inverse calculations, while MMSE requires additional noise variance estimation. Under AWGN (Additive White Gaussian Noise) channels, OSIC algorithm performance is significantly influenced by ordering strategies. Experimental results demonstrate that MMSE ordering typically approaches optimal ordering performance at high SNR, while ZF ordering may perform poorly at low SNR. Furthermore, since symbol detection errors in QAM16 modulation propagate to subsequent interference cancellation steps, selecting appropriate ordering strategies becomes essential for reducing bit error rates. Code implementations often include error propagation mitigation techniques like soft decision feedback or iterative refinement. In conclusion, the ordering strategy for OSIC algorithms in MIMO signal detection requires balancing computational complexity and detection performance, with QAM16 modulation under AWGN channel conditions further highlighting optimization opportunities for ordering methods. Practical implementations should consider adaptive ordering schemes that dynamically switch between strategies based on channel conditions.