Non-Orthogonal Multiple Access Uplink System Design and Analysis

碩士 === 國立清華大學 === 通訊工程研究所 === 104 === With the rapid growth of number of mobile devices, to achieve the huge throughputs becomes an important issue in the future. In the current mobile communication systems (such as Long-Term Evolution(LTE)), Orthogonal Multiple Access (OMA) techniques has been adop...

Full description

Bibliographic Details
Main Authors: Lin, Jing Yan, 林敬彥
Other Authors: Wu, Jen Ming
Format: Others
Language:en_US
Published: 2015
Online Access:http://ndltd.ncl.edu.tw/handle/j34wt7
Description
Summary:碩士 === 國立清華大學 === 通訊工程研究所 === 104 === With the rapid growth of number of mobile devices, to achieve the huge throughputs becomes an important issue in the future. In the current mobile communication systems (such as Long-Term Evolution(LTE)), Orthogonal Multiple Access (OMA) techniques has been adopted. However, it does not fully utilize the spectrum resource. Therefore, Non-Orthogonal Multiple Access (NOMA) was proposed to enhance the performance of throughputs. Different from OMA system, each subband can be allocated to more than one users in NOMA system, which achieves better spectrum efficiency. However, the users cause interference to other users in the same subband, making detection process more complex. In the current NOMA researches, it mainly focus on downlink system, the Successive Interference Cancellation (SIC) was used for detection. However, the performance of Bit Error Rate (BER) is not good enough. In this thesis, we focus on uplink NOMA system. We propose an algorithm to allocate the subbands to users, and we try to find the best combination of L(number of users on each subband) and Ns(number of subbands that each user can use) to achieve the maximum sum rates for fixed K(the number of users). At the base station, we perform detection subband by subband, since subbands will not cause interference to each other. In uplink system, multiple antennas can be used for detection. Several low complexity multiple input multiple output (MIMO) detection algorithms are applied. The performance of BER will be better than downlink system.