V.G. Kartashevskiy - Dr. Sc. (Eng.), Professor, Dean of Faculty of Telecommunications and Informatics, Volga State University of Telecommunications and Informatics (Samara). E-mail: kartash@psati.ru K.A. Belskiy - analitic, LLC «NetCracker» (Samara). E-mail: kirill8belsky@gmail.com O.S. Konyaeva - Senior Lecturer, Volga State University of Telecommunications and Informatics (Samara). E-mail: konyaeva2012@gmail.com A.A. Filimonov - Master, Volga State University of Telecommunications and Informatics (Samara). E-mail: Alfhero@gmail.com

In article the following problem is considered: the reception of orthogonal frequency division multiplexing (OFDM) with quadrature amplitude modulation (QAM) in channels with memory using technology Multiple Input Multiple Output (MIMO), which can significantly increase the speed of signal transmission over a communications channel, wherein the increase in speed in proportion used on the transmission antennas. It is expedient to use MIMO technology to transmit OFDM signals, since in the latter for maintaining orthogonality of subcarriers at the output of the communication in channel with memory can spend more than 20% of the channel capacity due to formation of a prefix and a postfix each OFDM symbol. Due to the specific structure of the OFDM signal with QAM the solution of demodulation problem (the problem of the decision concerning the transmission of binary characters) in the memory channel cannot be implemented by exhaustive search algorithms that implement the procedure «reception a whole» it is required, in this situation, a fantastic number sorting off hypotheses. (When using a QAM-16 the number sorting off hypotheses, at least 1616). Therefore task of estimating the samples of the envelope symbol OFDM precedes the solution of the demodulation problem; task of estimating is implemented with the using of equalizers solving the inverse problem of the recovery signals by algorithms such as the Zero Forcing (ZF) and the Minimum Mean Square Error (MMSE). In this paper to solve the problem of estimation suggested the combination MMSE algorithm with the algorithm «reception a whole with elementwise decision» (RWED) using the «feedback assessment», which leads to a significant reduction in computing costs while preserving the properties of estimates obtained algorithm MMSE. Statistical modeling of the considered algorithms in the channel memory confirmed the a priori idea of their capabilities and characte-ristics of noise immunity.

 

1. Bakulin M.G., Varukina L.A., Krejjndelin V.B. Tekhnologija MIMO: principy i algoritmy. M.: Gorjachaja linija - Telekom. 2014. 244 s.
2. Yong S.C., Jaekwon K., Won Y.Y., Chung G.K. MIMO-OFDM wireless communications with MATLAB. JohnWiley & Sons (Asia) PteLtd, 2 ClementiLoop. # 02-01. 2010. 457 p.
3. Kartashevskijj V.G., Belskijj K.A., Slipenchuk K.S. Priem signalov OFDM v kanalakh s rassejaniem // Radiotekhnika. 2015. № 2. S. 62−68.
4. Kartashevskijj V.G., Konjaeva O.S., Semenov E.S. Priem «v celom» dlja sistemy MIMO v kanale s pamjatju // Uspekhi sovremennojj radioehlektroniki. 2015. № 11. S. 49−53.
5. Volkov L.N., Nemirovskijj M.S., SHinakov JU.S. Sistemy cifrovojj radiosvjazi / M.: EHko-Trendz. 2005. 392 s.
6. Kartashevskijj V.G. Obrabotka prostranstvenno-vremennykh signalov v kanalakh s pamjatju / M.: Radio i svjaz. 2000. 272 s.
7. Ahmed Hesham Mehana, Aria Nosratinia. Diversity of MMSE MIMO Receivers // IEEE Transactions on Information theory. November 2012. V. 58. № 11. P. 6799−6805.
8. Tikhonov A.N., Arsenin V.JA. Metody reshenija nekorrektnykh zadach. M.: Nauka. 1979. 288 s.