RT Dissertation/Thesis
T1 Channel estimation techniques for next generation mobile communication systems
A1 Chen Hu, Kun
AB We are witnessing a revolution in wireless technology, where the society is demanding newservices, such as smart cities, autonomous vehicles, augmented reality, etc. These challengingservices not only are demanding an enormous increase of data rates in the range of 1000 timeshigher, but also they are real-time applications with an important delay constraint. Furthermore,an unprecedented number of different machine-type devices will be also connected to the network,known as Internet of Things (IoT), where they will be transmitting real-time measurements fromdifferent sensors. In this context, the Third Generation Partnership Project (3GPP) has alreadydeveloped the new Fifth Generation (5G) of mobile communication systems, which should becapable of satisfying all the requirements. Hence, 5G will provide three key aspects, such as:enhanced mobile broad-band (eMBB) services, massive machine type communications (mMTC)and ultra reliable low latency communications (URLLC).In order to accomplish all the mentioned requirements, it is important to develop new keyradio technologies capable of exploiting the wireless environment with a higher efficiency. Orthogonalfrequency division multiplexing (OFDM) is the most widely used waveform by the industry,however, it also exhibits high side lobes reducing considerably the spectral efficiency. Therefore,filter-bank multi-carrier combined with offset quadrature amplitude modulation (FBMC-OQAM)is a waveform candidate to replace OFDM due to the fact that it provides extremely low out-ofbandemissions (OBE). The traditional spectrum frequencies range is close to saturation, thus,there is a need to exploit higher bands, such as millimeter waves (mm-Wave), making possible thedeployment of ultra broad-band services. However, the high path loss in these bands increases theblockage probability of the radio-link, forcing us to use massive multiple-input multiple-output(MIMO) systems in order to increase either the diversity or capacity of the overall link.All these emergent radio technologies can make 5G a reality. However, all their benefits can beonly exploited under the knowledge and availability of the channel state information (CSI) in orderto compensate the effects produced by the channel. The channel estimation process is a well knownprocedure in the area of signal processing for communications, where it is a challenging task due to the fact that we have to obtain a good estimator, maintaining at the same time the efficiency andreduced complexity of the system and obtaining the results as fast as possible. In FBMC-OQAM,there are several proposed channel estimation techniques, however, all of them required a highnumber of operations in order to deal with the self-interference produced by the prototype filter,hence, increasing the complexity. The existing channel estimation and equalization techniques formassive MIMO are in general too complex due to the large number of antennas, where we mustestimate the channel response of each antenna of the array and perform some prohibitive matrixinversions to obtain the equalizers. Besides, for the particular case of mm-Wave, the existingtechniques either do not adapt well to the dynamic ranges of signal-to-noise ratio (SNR) scenariosor they assume some approximations which reduce the quality of the estimator.In this thesis, we focus on the channel estimation for different emerging techniques that arecapable of obtaining a better performance with a lower number of operations, suitable for low complexitydevices and for URLLC. Firstly, we proposed new pilot sequences for FBMC-OQAMenabling the use of a simple averaging process in order to obtain the CSI. We show that ourtechnique outperforms the existing ones in terms of complexity and performance. Secondly, wepropose an alternative low-complexity way of computing the precoding/postcoding equalizer underthe scenario of massive MIMO, keeping the quality of the estimator. Finally, we propose a newchannel estimation technique for massive MIMO for mm-Wave, capable of adapting to very variablescenarios in terms of SNR and outperforming the existing techniques. We provide some analysisof the mean squared error (MSE) and complexity of each proposed technique. Furthermore,some numerical results are given in order to provide a better understanding of the problem andsolutions.
YR 2019
FD 2019-07
LK https://hdl.handle.net/10016/29609
UL https://hdl.handle.net/10016/29609
LA eng
NO Mención Internacional en el título de doctor
DS e-Archivo
RD 19 may. 2024