19 October 2021 in IEEE Transactions on Cognitive Communications and Networking
Dynamic spectrum access (DSA) is an effective solution for efficiently utilizing the radio spectrum by sharing it among various networks. Two primary tasks of a DSA controller are: 1) maximizing the quality of service of users in the licensee’s network and 2) avoiding interference in communications towards the incumbent network. These two tasks become quite challenging in a distributed DSA network due to the lack of a centralized controller to regulate the sharing of the radio spectrum between incumbents and licensees. Hence, optimization-driven techniques to design power allocation schemes in such a network often become intractable. Accordingly, in this paper, we present a distributed DSA based communication framework based on multi-agent reinforcement learning (RL), where the multiple cells in the multi-user multiple-input multiple-output (MU-MIMO) licensee network act as agents, and the average signal-to-noise ratio value is the reward. In particular, by considering the physical layer parameters of the DSA network, we analyze various RL algorithms, namely Q-learning, deep Q-network (DQN), deep deterministic policy gradient (DDPG), and twin delayed deep deterministic (TD3), whereby the licensee network learns to obtain the optimal power allocation policies for accessing the spectrum in a distributed fashion without the need for a central DSA controller to manage the interference towards the incumbent. Trade-offs are identified for the considered algorithms with respect to performance, time complexity and scalability of the DSA network.
11 August 2020 in IEEE Communications Letters
Generalized Spatial Modulation (GSM) is being considered for high capacity and energy-efficient networks of the future. However, signal detection due to inter channel interference among the active antennas is a challenge in GSM systems and is the focus of this paper. Specifically, we explore the feasibility of using deep neural networks (DNN) for signal detection in GSM. In particular, we propose a block DNN (BDNN) based architecture, where the active antennas and their transmitted constellation symbols are detected by smaller sub- DNNs. After N-ordinary DNN detection, the Euclidean distancebased soft constellation algorithm is implemented. The proposed B-DNN detector achieves a BER performance that is superior to traditional block zero-forcing (B-ZF) and block minimum mean-squared error (B-MMSE) detection schemes and similar to that of classical maximum likelihood (ML) detector. Further, the proposed method requires less computation time and is more accurate than alternative conventional numerical methods.
9 April 2021 in IEEE Signal Processing Letters
Intelligent reflecting surface (IRS) is being considered as a prospective candidate for next generation wireless communication due to its ability to significantly improve coverage and spectral efficiency by controlling the propagation environment. One of the ways IRS increases spectral efficiency is by adjusting phase shifts to perform passive beamforming. In this letter, we integrate the concept of IRS aided communication to the domain of multi-direction beamforming, whereby multiple receive antennas are selected to convey more information bits than existing spatial modulation (SM) techniques at any specific time. To complement this system, we also propose a successive signal detection (SSD) technique at the receiver. Numerical results show that the proposed design is able to improve the average successful bits transmitted (ASBT) by the system, which outperforms other state-of-the-art methods proposed in literature.
23 September 2022 in IEEE Transactions on Vehicular Technology
Reconfigurable Intelligent Surfaces (RISs) have recently gained significant attention as they enable controlling the wireless propagation environment to improve information transmission with the help of reflective elements. This paper studies a RIS-assisted generalized spatial modulation (GSM) multiple input multiple output (MIMO) system to improve the signal quality and spectral efficiency. Due to inter-channel interference, signal detection in GSM becomes a critical issue. Therefore, we propose a low complexity block deep neural network (DNN) detector for the RIS-GSM MIMO system to detect the active antennas and the transmitted symbols at the receiver. In order to optimize the phase shifts at the RIS elements, we propose cosine similarity-based and semidefinite relaxation (SDR)-based algorithms. Further, conventional signal detectors such as maximum likelihood (ML), block zero-forcing (B-ZF), and block minimum mean square error (B-MMSE) are applied. In addition, we explore and analyze the performance of the proposed framework in terms of bit error rate (BER) and time complexity. Numerical results reveal that both optimize-phase-shifts algorithms assisted DNN signal detector has almost the same BER performance as ML, but it is four times faster than the linear detector (B-ZF and B-MMSE), and seven times faster than ML. Clearly, this scheme outperforms the alternative traditional signal detectors with the least time complexity.
20 October 2025 in IEEE Internet of Things Journal
Precise and energy-efficient localization is a critical requirement in many Internet of Things (IoT) applications, particularly in large-scale deployments such as asset tagging, agriculture, and smart cities, where long battery life and cost-effectiveness are crucial. The Semtech SX1280 longRange (LoRa) transceiver presents a promising solution for IoT localization. It combines low cost, low power, and precise ranging capability over distances of up to 1 km. However, the ranging process requires two devices to be simultaneously active, one initiating the ranging request and the other responding to it, which can lead to significant energy expenditure if not properly managed. Despite the transceiver’s excellent performance, no existing system-level framework effectively manages sleep–wake coordination and role assignment needed for energy-efficient operation. This article presents a coordination framework that significantly reduces power consumption while maintaining the inherent precise ranging capability of the chip. The framework schedules short, synchronized wake-up windows between the initiator and the responder, allowing devices to remain in deep sleep for most of their duty cycle. This scheduling strategy minimizes reliance on precise continuous timing and mitigates drift in low-cost oscillators. To validate the framework, we designed and developed custom nodes that are compliant with the framework’s protocol. Experimental results show that the proposed approach allows a node to stay in ultralow-power mode and wake periodically to check for instructions. The node can remain in standby mode for up to nine months on a single coin cell battery and can perform ranging operations on demand in near real time, all while maintaining a localization accuracy within 5 m.