Adaptive Linear Minimum BER Reduced-Rank Interference Suppression Algorithms Based on Joint and Iterative Optimization of Filters

Abstract

In this letter, we propose a novel adaptive reducedrank strategy based on joint iterative optimization (JIO) of filters according to the minimization of the bit error rate (BER) cost function. The proposed optimization technique adjusts the weights of a subspace projection matrix and a reduced-rank filter jointly. We develop stochastic gradient (SG) algorithms for their adaptive implementation and introduce a novel automatic rank selection method based on the BER criterion. Simulation results for direct-sequence code-division-multiple-access (DS-CDMA) systems show that the proposed adaptive algorithms significantly outperform the existing schemes.

Adaptive Reduced-Rank Receive Processing Based on Minimum Symbol-Error-Rate Criterion for Large-Scale Multiple-Antenna Systems

“…After the optimum filter lengths are determined for all the branches, we select the optimum branch of the JPDF scheme based on the following criterion Full-Rank-LMS 2LP + 1 2LP Full-Rank-MSER [35] 3LP + 1 2LP MSER-JIO [40] 8LP D + 7D + 2LP + 9 7LP D + 2LP − 1 EIG [13] O((LP ) 3 ) O((LP ) 3 ) MSER-MSWF [38] D Full-Rank-LMS 2LP + 1 2LP Full-Rank-MSER [7] 6LP + 5 5LP MSER-JIO [40] 10LP D + 7D + 4LP + 17 9LP D + 4LP + 3 EIG [13] O((LP ) 3 ) O((LP ) 3 ) MSER-MSWF [38] D(LP ) 2 + 5LP D + 5D + 2LP + 11…”

“…In Table III, we focus on the case of BPSK and show the number of additions and multiplications per symbol of the proposed adaptive reduced-rank algorithm, the existing adaptive MSER-based reduced-rank algorithms [38], [40], the conventional adaptive LMS full-rank algorithm [11] and the adaptive full-rank algorithm based on the SER criterion [35]. Table IV shows the corresponding figures for the case of QAM symbols.…”

“…2 and Fig. 3, we show the computational complexity against the number of received antenna elements for the recently reported MSER-based reduced-rank algorithms [38], [40] and the proposed MSER-JPDF reduced-rank algorithms. In particular, we consider a commonly used configuration with P = 3, I = 12 and D = 10.…”

Decoupled signal detection for the uplink of massive MIMO in 5G heterogeneous networks

“…The interference between antennas and users, propagation effects such as correlation, path loss and shadowing, thermal noise and signal degradation due to the hardware imperfections need to be suppressed. Linear detection techniques [11], [12], [13], [14], [15], [16], [17], [18], [19] such as maximum radio combining (MRC) and zero forcing (ZF) are a good option in terms of computational complexity, however, their performance is not compatible with the growing demand for high data rates. The performance of linear detectors can be improved using some nonlinear suboptimal detector based on successive interference cancellation (SIC) [20], [21], [22], [23], e.g., multibranch SIC (MB-SIC) [24], [25], [26] and multifeedback SIC (MF-SIC) [27], [28], [29], [30].…”

Reduced-rank interference suppression algorithm based on generalized MBER criterion for large-scale multiuser MIMO systems

“…These algorithms have better performance than those based on other criteria. An iterative optimization of the subspace projection matrix and RR filter [10] according to the MBER criterion has been presented using the normal kernel density estimation method. However, the normal kernel density method employed is not appropriate for heavier and lighter tailed distributions [11].…”