PIC Receiver for MIMO MC-CDMA Systems Based on LCRLSCMA

Abstract

Parallel interference canceller (PIC) is a powerful algorithm to combat multi-user interference (MUI). In this paper, a PIC receiver for space-time block coded multiple input multiple output multi-carrier code division multiple access (STBC MIMO MC-CDMA) based on linearly constraint recursive least squares constant modulus algorithm (LCRLSCMA) is studied. Simulations indicate that the proposed algorithm has better BER performance than the traditional algorithms among different environments.Keywords-Space-time block coded multiple input multiple output multi-carrier code division multiple access; linearly constraint recursive least squares constant modulus algorithm; Parallel interference canceller; multi-user detection. I.I NTRODUCTION Space-time block coded (STBC) multiple input multiple output (MIMO) systems can increase the channel capacity and spectral efficiency without adding extra bandwidth and power consumption. Multi-carrier code-division multiple access (MC-CDMA) , which combines the advantages of orthogonal frequency division multiplexing (OFDM) with code division multiple access (CDMA), has the high frequency efficiency and interference suppression ability.In STBC MC-CDMA systems, the minimum mean square error (MMSE) receiver has low spectral efficiency for training sequences; MMSE receiver based on subspace[1, 2], which requires the estimation of the signal and noise subspace, has high complexity; In[3, 4], a robust Multi-user Detection based on the variable diagonal loading of the sample covariance is raised, which has high complexity for matrix inverse and subspace estimation.The CMA algorithm is widen used in multi-user detection for its robustness and efficiency. In [5],the author proposes a multi-user detection for MC-CDMA systems based on stochastic gradient descent constant modulus algorithm (SGD-CMA), SGD methods are quite sensitive to the selected step size. A small step size will lead to slow convergence rate, whereas a large step size will result in the adapted weight oscillation. In [6],the author proposes a joint subspace MMSE and CMA multi-user detection algorithm, which requires twice subspaces decomposition, has error spread when the imperfect estimation of the first subspace and will lead to the performance descent; Recursive least-squares CMA has rapid convergence rate, in multi-user (multi-mode) environment, CMA has local minimum, the algorithm[7] may converge to interfere user. In this paper linearly constraint recursive least squares (LCRLSCMA) combining with PIC structure is raised, which can eliminate MAI and suppress noise, by using generalized sidelobe cancellation structure, we transform the constraint cost function into a non-constraint issue, and iterative formula of the LCRLSCMA is derived. Finally, parallel interference cancellation technology after the LCRLSCMA is implemented for desired user's detection, and multi-user interference is further eliminated. Simulations indicate that the proposed algorithm has better performance than that of the traditional algorithms amon...