Performance Analysis of Multiple Access Transmissions for Wireless Networks

  • Lina Bariah

Student thesis: Doctoral Thesis

Abstract

The increased proliferation of connected user devices, o_ering a variety of services at different levels of performance, represents a major challenge for broadband wireless networks. An efficient solution for such challenge would require a paradigm shift towards the development of key enabling technologies for next generation wireless networks and beyond. One of the critical challenges towards realizing the next generation networks, however, is the scarcity of spectrum, owing to the unprecedented broadband penetration rate and the growing number of broadband users in recent years. Multiple access technologies play a pivotal role in determining the feasibility of satisfying the newly imposed requirements of the future wireless networks, such as, higher data rate, enhanced spectrum efficiency, increased connectivity and reduced energy consumption. Consequently, a growing research interest is devoted towards the development of new technologies, such as, orthogonal and non-orthogonal multiple access, massive multiple-input multiple output (MIMO), and cognitive radio (CR), to address some of the aforementioned challenges that face next generation networks. In this respect, this dissertation focus on the analysis of orthogonal and non-orthogonal multiple access techniques for wireless networks. In the first part of this thesis, we investigate a spectrally and computationally efficient blind channel estimation technique for orthogonal frequency division multiplexing (OFDM) systems over frequency selective fading channels, where a hybrid frame structure is utilized to enable blind detection of data symbols. Amplitude coherent detection (ACD) is considered for the hybrid frame structure, where several modulation schemes are used to modulate particular subcarriers, such as amplitude shift keying (ASK), phase shift keying (PSK) and quadrature amplitude modulation (QAM). To evaluate the performance of the considered estimator, a novel accurate closed-form symbol error rate (SER) expression is derived for the ASK symbols. The obtained SER analytical expression is then utilized to derive an accurate formula for the mean squared error (MSE) of the initial channel estimates. The obtained analytical results, veri_ed by Monte Carlo simulations, show that the SER and MSE are highly dependent on the channel correlation factor. Yet, accurate detection and estimation can be obtained for most practical channel models. The second part of the thesis is devoted to investigate the error rate performance of non-orthogonal multiple access (NOMA) systems over generalized Nakagami-m fading channels, under realistic assumptions, such as, imperfect successive interference cancellation (SIC). In particular, we derive novel pairwise error probability (PEP) expressions to characterize the performance of all users under different fading scenarios. The obtained PEP expressions are then used to obtain a union bound on the bit error rate (BER) and to investigate the achieved diversity gain of all users. Motivated by the fact that power allocation have a critical impact on the reliability of all users, and given that optimum power allocation should be applied to realize users' fairness in NOMA systems, we exploit the derived BER union bound to formulate and evaluate an optimization problem that aims at finding the optimum power coefficients that minimize the average BER union bound, under individual BER union bounds and average power constraints. The obtained analytical and simulation results demonstrate that the error rate performance of all users and the achieved diversity order is highly susceptible to users order and fading scenarios, represented by the fading parameter m. The third part of the thesis provides a comprehensive analytical framework devoted to evaluate the error rate performance of NOMA-based underlay CR networks. In specific, we consider a secondary NOMA-based relay network, which shares the full spectrum with a primary network, under interference power constraints. Given that secondary users, who adopt NOMA, operate under interference power constraints, we investigate the PEP performance of all secondary users while considering imperfect SIC. Particularly, we derive an accurate PEP expression which is then used to evaluate the BER union bound. The derived BER union bound is then used to examine the achieved diversity order of all users and optimize the power coefficients that allow exploiting the full potential of NOMA systems. Finally, the fourth part of the thesis provides a mathematical framework to analyze the error rate performance of NOMA-based relay networks with simultaneous wireless information and power transfer (SWIPT). An exact closed form expression is derived for the PEP of each user while considering an imperfect SIC process. Then the derived PEP is used to derive a tight union bound on the BER. The derived analytical results demonstrate that the error rate of all users is highly affected by the power splitting ratio, energy conversion efficiency and power allocation scheme.
Date of AwardOct 2018
Original languageAmerican English

Keywords

  • Channel estimation
  • cognitive radio
  • NOMA
  • OFDM
  • performance analysis
  • SWIPT.

Cite this

'