Improving NOMA Multi-Carrier Systems With Intentional Frequency Offsets

Abstract

In this letter, we investigate the possible benefits of asynchrony in the frequency domain for the non-orthogonal multiple access (NOMA) schemes. Despite the common perspective that asynchrony in transmission or reception of multistream signals is harmful, we demonstrate the advantages of adding intentional frequency offset to the conventional power domain-NOMA (P-NOMA). We introduce two methods which add artificial frequency offsets between different sets of subcarriers destined for different users. The first one uses the same successive interference cancellation (SIC) method as the conventional P-NOMA except that it enjoys reduced inter-user interference (IUI) between interfering sub-carriers. The second scheme adopts a precoding at the base station and a linear preprocessing scheme at the receiving user. It decomposes the broadcast channel into parallel channels circumventing the need for SIC. As a result, it fully exploits the advantages provided by the frequency asynchrony and enables the interference-free transmission to the users. The numerical results show that both methods can outperform the conventional P-NOMA.

Deep neural network based downlink power domain multi-user NOMA-OFDM signal detection

“…(3) A deep neural network architecture has been created by initializing its weights at random with the values W l i , 1 , W i j , 2 , W j k , 3 , and W k d , 4 . (4) The difference between the accurate downlink power domain multi-user U1 and U2 NOMA-OFDM symbols X k (t) and the DNN-detected output symbols ˆ( ) X t k is determined, and the estimation error is calculated in accordance with equation ( 14).…”

“…As a result, non-idealities such as the fading channel effect and carrier frequency offset (CFO) caused by oscillator frequency mismatch reduce the symbol error rate (SER) performance for signal detection in downlink power domain multi-user NOMA-OFDM [2,4,9,10,14,15]. As a multi-carrier multiple-access technique, it has a high peak-to-average power ratio (PAPR).…”

Downlink Asynchronous Non-Orthogonal Multiple Access Systems with Imperfect Channel Information

“…If τ = 0, ANOMA degrades to NOMA and the received samples at iT and (i + τ )T are identical. The number of received samples is doubled in ANOMA (τ = 0) compared with NOMA, which then results in sampling diversity [2][3][4].…”

“…Recently, a novel scheme called asynchronous NOMA (ANOMA) has been proposed to further improve the performance of NOMA, for example, in the uplink system [2] and the cooperative network [3]. In ANOMA, the intentional introduction of timing mismatch at the transmitter and the oversampling technique at the receiver result in the sampling diversity [2][3][4]. It has been demonstrated that ANOMA outperforms NOMA in terms of the throughput performance, the power consumption, etc.…”

Joint Beamwidth and Power Optimization in MmWave Hybrid Beamforming-NOMA Systems

“…Non-orthogonal multiple access (NOMA) aims to improve the spectral efficiency and simultaneously serve more than one user at the same frequency/time/code in single-carrier and multicarrier systems [10], [11]. Especially, NOMA transmits the users' signal at the same time slot and frequency band by using superposition coding (SC) and decodes the desired signal by exploiting successive interference cancellation (SIC) at the receiver [12].…”