9 April 2019 in IEEE Wireless Communications Letters
This letter presents the first attempt of exploiting deep learning (DL) in the signal detection of orthogonal frequency division multiplexing with index modulation (OFDM-IM) systems. Particularly, we propose a novel DL-based detector termed as DeepIM, which employs a deep neural network with fully connected layers to recover data bits in an OFDM-IM system. To enhance the performance of DeepIM, the received signal and channel vectors are pre-processed based on the domain knowledge before entering the network. Using datasets collected by simulations, DeepIM is first trained offline to minimize the bit error rate (BER) and then the trained model is deployed for the online signal detection of OFDM-IM. Simulation results show that DeepIM can achieve a near-optimal BER with a lower runtime than existing hand-crafted detectors.
22 July 2022 in IEEE Transactions on Vehicular Technology
Non-orthogonal communications are expected to play a key role in future wireless systems. In downlink transmissions, the data symbols are broadcast from a base station to different users, which are superimposed with different power to facilitate high-integrity detection using successive interference cancellation (SIC). However, SIC requires accurate knowledge of both the channel model and channel state information (CSI), which may be difficult to acquire. We propose a deep learning-aided SIC detector termed SICNet, which replaces the interference cancellation blocks of SIC by deep neural networks (DNNs). Explicitly, SICNet jointly trains its internal DNN-aided blocks for inferring the soft information representing the interfering symbols in a data-driven fashion, rather than using hard-decision decoders as in classical SIC. As a result, SICNet reliably detects the superimposed symbols in the downlink of non-orthogonal systems without requiring any prior knowledge of the channel model, while being less sensitive to CSI uncertainty than its model-based counterpart. SICNet is also robust to changes in the number of users and to their power allocation. Furthermore, SICNet learns to produce accurate soft outputs, which facilitates improved soft-input error correction decoding compared to model-based SIC. Finally, we propose an online training method for SICNet under block fading, which exploits the channel decoding for accurately recovering online data labels for retraining, hence, allowing it to smoothly track the fading envelope without requiring dedicated pilots. Our numerical results show that SICNet approaches the performance of classical SIC under perfect CSI, while outperforming it under realistic CSI uncertainty.
5 July 2023 in Cluster Computing
18 September 2017 in IEEE Transactions on Vehicular Technology
This paper proposes a novel framework to analyze the symbol error probability (SEP) for multicarrier index keying - orthogonal frequency division multiplexing (MCIK-OFDM) systems. Considering two different types of detections such as the maximum likelihood (ML) and low-complexity greedy detectors (GD), we derive tight, closed-form expressions for the average SEPs of MCIK-OFDM in presence of channel state information (CSI) uncertainty. We undertake an asymptotic performance analysis with respect to three CSI conditions, which ensures to provide a comprehensive insight into the achievable diversity and coding gains as well as the impact of various CSI uncertainties on the SEP performance. The SEP performance comparison between the ML and GD, is obtained under different CSI uncertainties. This interestingly reveals that the GD can achieve nearly optimal error performance as the $M$-ary modulation size is large, or even outperforms the ML under certain CSI conditions. Finally, the theoretical and asymptotic analysis are verified via simulation results, obtaining the high accuracy of the derived SEP.
13 March 2020 in IEEE Transactions on Wireless Communications
We propose a novel deep energy autoencoder (EA) for noncoherent multicarrier multiuser single-input multiple-output (MU-SIMO) systems under fading channels. In particular, a single-user noncoherent EA-based (NC-EA) system, based on the multicarrier SIMO framework, is first proposed, where both the transmitter and receiver are represented by deep neural networks (DNNs), known as the encoder and decoder of an EA. Unlike existing systems, the decoder of the NC-EA is fed only with the energy combined from all receive antennas, while its encoder outputs a real-valued vector whose elements stand for the sub-carrier power levels. Using the NC-EA, we then develop two novel DNN structures for both uplink and downlink NC-EA multiple access (NC-EAMA) schemes, based on the multicarrier MU-SIMO framework. Note that NC-EAMA allows multiple users to share the same sub-carriers, thus enables to achieve higher performance gains than noncoherent orthogonal counterparts. By properly training, the proposed NC-EA and NC-EAMA can efficiently recover the transmitted data without any channel state information estimation. Simulation results clearly show the superiority of our schemes in terms of reliability, flexibility and complexity over baseline schemes.
1 October 2018 in IEEE Transactions on Vehicular Technology
We propose a new spread orthogonal frequency division multiplexing with index modulation (S-OFDM-IM), which employs precoding matrices such as Walsh-Hadamard (WH) and Zadoff-Chu (ZC) to spread both non-zero data symbols of active sub-carriers and their indices, and then compress them into all available sub-carriers. This aims to increase the transmit diversity, exploiting both multipath and index diversities. As for the performance analysis, we derive the bit error probability (BEP) to provide an insight into the diversity and coding gains, and especially impacts of selecting various spreading matrices on these gains. This interestingly reveals an opportunity of using rotated versions of original WH and ZC matrices to further improve the BEP performance. More specifically, rotated matrices can enable S-OFDM-IM to harvest the maximum diversity gain, which is the number of sub-carriers, while benchmark schemes have diversity gains limited by two. Moreover, we propose three low-complexity detectors, namely minimum mean square error log-likelihood ratio, index pattern MMSE (IP-MMSE), and enhanced IP-MMSE, which achieve different levels of complexity and reliability. Simulation results are presented to prove the superiority of S-OFDM-IM over the benchmarks.