1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 60 61 62 63 64 65 XXXVII SIMP ´ OSIO BRASILEIRO DE TELECOMUNICAC ¸ ˜ OES E PROCESSAMENTO DE SINAIS - SBrT2019, 29/09/2019–02/10/2019, PETR ´ OPOLIS, RJ Antenna Selection in MIMO-OFDM systems David Nu˜ nez Cuadrado, Jo˜ ao Cal-Braz and Raimundo Sampaio-Neto Abstract— MIMO-OFDM results from combination of two well-known effective strategies used in many of today’s com- munication technologies. It unites the well-known advantages of MIMO communication, such as high data rates and im- proved link reliability to the characteristics of OFDM, known as intersymbol-interference-free communication and simple frequency-domain equalization. This paper presents in detail the signal modelling of MIMO-OFDM system and the adequate processing at the receiver to perform decoupled detections per OFDM subsymbol. In addition, antenna selection strategies are proposed to increase system performance by exploiting a scenario when the transmitter is equipped with more antennas than the number of radiofrequency (RF) chains. Simulation results evidence that antenna selection strategies result in significant system performance improvement. Keywords— Multiple-Input Multiple-Output (MIMO), Orthog- onal Frequency Division Multiplexing (OFDM), antenna selection I. I NTRODUCTION Wireless communication systems have experienced an ac- celerated evolution in the last decades caused by the strin- gent requirements in terms of data rates, latency and energy efficiency. Among the recent developments achieved so far, Orthogonal Frequency Division Multiplexing (OFDM) and Multiple-Input and Multiple-Output (MIMO) figure in most of the current communication technologies, due to their indis- putable effectiveness. International mobile telephony standards organization has settled OFDM as the main waveform for the fifth generation (5G) mobile service radio access [1], endorsing the well- known advantages of OFDM communication. OFDM tech- nique is a spectrally efficient modulation scheme that trans- forms a broadband channel with frequency-selective fading into a set of parallel narrowband channels with frequency-flat fading, avoiding the occurrence of intersymbol interference and simplifying the system in terms of equalization [2]. MIMO communication is an emerging technique that of- fers various advantages through the deployment of multiple antennas at the communicating nodes. Channel capacity that increases linearly with system size is achieved, resulting in high data rates by the exploitation of the spatial domain. Secondly, the diversity gain experienced by the multiantenna systems offers improved reliability of the transmission link, mitigating the deleterious effect of fading [3]. Most MIMO literature assumes frequency-flat fading chan- nel and, when stated otherwise, assumes the use of OFDM transmission to build an equivalent system model that reverts David Nu˜ nez Cuadrado, Raimundo Sampaio Neto, CETUC, Pontifical Catholic University of Rio de Janeiro, Rio de Janeiro-RJ, Brazil. e-mails: david@cetuc.puc-rio.br, raimundo@cetuc.puc-rio.br. Jo˜ ao Cal-Braz. National Institute of Metrology, Quality and Technology INMETRO, Rio de Janeiro- RJ, Brazil. email: jabraz@inmetro.gov.br the frequency-selective environment back into frequency-flat, taking the particularities of MIMO-OFDM system for granted or leaving them unspecified [4], [5]. Thus, the combination of both techniques results in system with enhanced performance and more resilient to communi- cation errors. The main drawback of a MIMO-OFDM imple- mentation is the cost, in terms of size, power consumption and hardware complexity, which is scaled with the number of antennas since there is a radio frequency (RF) chain and an OFDM modulator associated to each antenna element [6]. In order to overcome this challenge, strategies that can lower the energy consumption and the cost of implementation and operation are required. Antenna selection strategies have been considered in several frequency-flat MIMO scenarios as viable solutions that reduce the hardware complexity through the use of a number of RF chains smaller than the number of available antennas in the system [6], [7], [8] . The underlying idea is to use a reduced number of RF chains and, based on the current channel characteristics, choose a subset of available antennas more adequate to communication to which the RF chains should be connected. Several antenna selection schemes have been developed, mainly for single-user MIMO communications. The selection criteria range from minimizing the symbol error rate [7] or the channel matrix condition number, to maximizing channel capacity [9], among others. When the multiuser scenario is considered, relevant works consider the maximization of the signal-to-leakage and noise ratio [10], [11]. The goals of this paper are twofold: develop a detailed sig- nals and system modeling in MIMO-OFDM scenario. First, a standard MIMO-OFDM is considered, followed by a precoded MIMO-OFDM, aiming at complexity reduction for detection at the receiver. Having the complexity reduction motivation at aim, two antenna selection techniques are presented, namely Mutual Information Method and γ -Parameter Method. Notation: Boldface letters will be used for matrices (capital letters) and vectors (lowercase letters); (·) ∗ , (·) T and (·) H de- note conjugate, transpose and Hermitian (conjugate transpose), respectively; [ A ] † is the pseudo-inverse of matrix A; E [ · ] is used to represent expectation and Tr{·} to represent trace. I N denotes the N × N identity matrix. Further, diag{v} stands for a diagonal matrix with the components of vector v on its main diagonal. We have used i.i.d to refer to independent, identically distributed random quantities. II. SYSTEM MODEL Before addressing the Single User MIMO-OFDM (SU- MIMO-OFDM), we briefly describe the baseband discrete model of classical Cyclic-Prefix (CP)-OFDM. SBrT 2019 1570558553 1