CASPIAN JOURNAL

MANAGEMENT AND HIGH TECHNOLOGIES

A two-step mimo detection based on Markov chain Monte Carlo techniques and sphere detector

Read Zemtsov Dmitriy V., Savelev Boris A. A two-step mimo detection based on Markov chain Monte Carlo techniques and sphere detector // Caspian journal : management and high technologies. — 2014. — №3. — pp. 99-108.

Zemtsov Dmitriy V. - postgraduate student, Penza State University, 40 Krasnaya St., Penza, 440026, Russian Federation, mesc@live.ru

Savelev Boris A. - D.Sc. (Engineering), Professor, Penza State University, 40 Krasnaya St., Penza, 440026, Russian Federation, sba@pnzgu.ru

The article deals with the problem of ensuring high reliability of data transmission over wireless channels in multi-antenna (MIMO) systems and reducing the computational complexity of the task being solved by using a two-step MIMO detector based on Markov Chain Monte Carlo (MCMC) techniques and sphere detection. The methods and principles of the proposed algorithm are described. The authors suggest methods used for reducing computational complexity of the algorithm by using the hard decision sphere detector. Also the article demonstrates the possibility of avoiding the stalling problem encountered in MCMC algorithm at high SNRs, through the initiation of a Markov chain by a sphere detection results. Developed software models show that the BER performance of a two-step detector is better as compared with the soft decision sphere detector. Calculation of number of operations spent by algorithms indicates a lower computational complexity of a two-step detector compared to soft sphere detector.

Key words: многоантенная (MIMO) система, детектирование, методы Монте-Карло с цепями Маркова, сэмплирование по Гиббсу, сферическое детектирование, вероятность ошибки на бит (BER), отношение сигнал - шум (SNR), вычислительная сложность алгоритма, multi-antenna (MIMO)