This paper presents an enhanced deep learning-based Non-Orthogonal Multiple Access (NOMA) receiver that can mainly be used in low signal-to-noise channels. We show how a better dataset generation strategy for training Deep Learning (DL) could result in better generalization capabilities. Then, we apply hyperparameter tuning using exhaustive search to optimize the DL network. A Long-Short-Term-Memory (LSTM) DL architecture is used. The results show superior Symbol Error Rate vs Signal-to-Noise Ratio performance compared to the state-of-the-art methods such as Maximum Likelihood, Minimum Mean Square Error, and Successive Interface Cancellation, even though the network is only half as complex as previously proposed DL networks in the literature.
A widely deployed cellular network, supported by direct connections, can offer a promising solution that supports new services with strict requirements on access availability, reliability, and end-to-end (E2E) latency. The communications between vehicles can be made using different radio interfaces: One for cellular communication (i.e., cellular communication over the cellular network based on uplink (UL)/downlink (DL) connections) and the other for direct communication (i.e., D2D-based direct communications between vehicles which allows vehicular users (V-UEs) to communicate directly with others). Common cellular systems with licensed spectrum backed by direct communication using unlicensed spectrum can ensure high quality of service requirements for new intelligent transportation systems (ITS) services, increase network capacity and reduce overall delays. However, selecting a convenient radio interface and allocating radio resources to users according to the quality of service (QoS) requirements becomes a challenge. In this regard, let’s introduce a new radio resource allocation strategy to determine when it’s appropriate to establish the communication between the vehicles over a cellular network using licensed spectrum resources or D2D-based direct connections over unlicensed spectrum sharing with Wi-Fi. The proposed strategy aims at meeting the quality of service requirements of users, including reducing the possibility of exceeding the maximum delay restrictions and enhancing network capacity utilization in order to avoid service interruption. The proposed solution is evaluated by highlighting different conditions for the considered scenario, and it is demonstrated that the proposed strategy improves network performance in terms of transmitted data rate, packet success rate, latency, and resource usage
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.