This paper addresses an intelligent reflecting surface (IRS) to the uplink nonorthogonal multiple access (NOMA) served by a multiantenna receiver for effective data collection from massive devices. We aim to achieve max-min fairness of the network by optimizing receive beamforming, IRS reflection, and transmit power allocation (PA) of the devices. For this purpose, first, we design a block coordinate descent (BCD) algorithm that reduces the complexity of a conventional IRS reflection optimization. Next, we design a nonlinear optimization (NLO) problem solvable with the limited-memory Broyden-Fletcher-Goldfarb-Shanno bounded (L-BFGS-B) algorithm, which is renowned for handling large-scale problems, to cope with large IRS elements and devices. The problem is formed with a smooth but complex objective function that depends on the IRS phase shift and PA vectors for which the gradient is derived in a computationally efficient form. The results reveal that the proposed BCD and proposed NLO with the L-BFGS-B outperform the conventional BCD in performance and complexity, where the NLO approach offers a substantial complexity reduction.
28 June 2023 in IEEE Wireless Communications Letters
In intelligent reflecting surface (IRS) assisted communication, reliable channel estimation with reduced training overhead is crucial for the timely reconfiguration of wireless environments through IRS reflection control. We consider IRS cascaded channel estimation with element-grouping, which reduces the training overhead by sharing a reflection coefficient among a subgroup of elements at the cost of interference. Based on the maximum a posteriori estimation, we design a deep denoising network that supports various noise levels for noise-dominated regions and a deep unfolding network that optimizes extrapolation and denoising alternately for interference-dominated regions. The proposed networks provide a more reliable estimation through balanced noise and interference mitigation than the conventional networks proposed recently for element-grouping and super-resolution.
Sixth generation communication systems are expected to utilize extremely large antenna arrays (ELAAs) to significantly enhance the spectral efficiency by compensating for path loss at high operating frequencies. However, in ELAA systems, the required number of pilot signals can become excessive, especially in multi-user scenarios, making channel estimation for all antennas particularly challenging. In addition, ELAAs present challenges as the near-field region becomes more prominent, rendering traditional channel estimation methods inefficient. To address this, we propose to utilize a channel knowledge map for channel estimation in ELAA systems, enabling seamless near-field and far-field communication while reducing the pilot overhead. To enhance the practicality of the channel knowledge map in ELAA systems, we propose a channel angle distance map (CADM) database that provides the angles and distances of multipaths from the polar-domain for a given location. The estimated channel parameters are adopted as initial values and refined through the limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) quasi-Newton algorithm to minimize the channel estimation error. Extensive simulation results demonstrate that the CADM-based method outperforms benchmark schemes with reduced pilot overhead.
In intelligent reflecting surface (IRS) assisted communication systems, channel estimation is of great importance to realize the performance gains promised by an IRS that adapts to the channel environments through reflection pattern optimization. Recently, deep learning networks have been applied to estimate the IRS cascaded channel to find an estimator close to the optimal one which is intractable due to the non-Gaussian nature of the cascaded channel. This paper applies a deep residual denoising network to estimate the cascaded channel more robustly in the presence of a residual carrier frequency offset (CFO). Simulation results show that the learning-based channel estimation provides a significant improvement in mean square error compared with the linear estimators.
7 April 2026 in IEEE Transactions on Communications
We propose multi-mode spectrally efficient frequency division multiplexing with index modulation (MM-SEFDM), an enhanced non-orthogonal waveform that embeds distinct constellations on every subcarrier and conveys additional bits through subcarrier-index patterns. By compressing the subcarrier spacing of orthogonal frequency division multiplexing (OFDM), MM-SEFDM significantly improves spectral efficiency at the expense of controlled inter-carrier and inter-symbol interference. This joint non-orthogonal and multi-mode index modulation structure introduces significant detection complexity, necessitating a low-complexity, yet effective receiver capable of handling interference and permutation-based index-symbol decoding. To this end, we propose a two-phase Transformer-based decoder: in Phase 1, the network learns joint index–symbol constellations over AWGN channels, while in Phase 2, it continues training over Rayleigh fading channels to enhance robustness under realistic propagation conditions. Furthermore, to enhance model generalization, the detector is trained under domain randomization across different non-orthogonality levels and channel estimation errors (CSE), ensuring robust and SNR-agnostic performance under diverse interference and operating conditions. Compared to brute-force maximum-likelihood detection, the proposed learning-based decoder reduces inference complexity from exponential to polynomial time, while achieving near-optimal bit-error rate performance.