Equalization Based Soft Output Data Detection for Massive MU-MIMO-OFDM Using Coordinate Descent

For the next generation of wireless communication networks to advance, massive multi-user multiple-input multiple-output orthogonal frequency division multiplexing (MU-MIMO-OFDM) systems are essential. Nevertheless, due to the concurrent presence of several users and the frequency-selective fading channel, identifying sent data in such systems proves to be a daunting issue. This study proposes a huge MU-MIMO-OFDM system-compatible coordinate descent-based equalization-based soft output data identification technique. This algorithm's major goals are to improve the estimation of transmitted data symbols and effectively deal with inter-user interference. By exploiting the sparse nature of the channel impulse response, the data detection problem as a joint sparse signal recovery and symbol detection task. Then, the coordinate descent algorithm, which iteratively updates the estimated symbols and exploits the sparsity structure of the channel has been implemented. These soft outputs can be utilized in subsequent stages of the communication system, such as channel decoding or interference cancellation. The simulation results clearly illustrate the superiority of the proposed method over existing detection techniques in terms of bit error rate (BER) performance. The algorithm showcases remarkable enhancements in detection accuracy, even in challenging scenarios involving a substantial number of users and severe channel conditions. When the number of base stations is increased from 32 to 128, the proposed algorithm demonstrates a substantial 76% reduction in bit error rate (BER). In contrast, conventional methods only achieve a value of approximately 60% reduction in BER under the same conditions.