This paper proposes a novel energy-conscious access point ( $e$ AP) system with cross-layer design to increase the energy efficiency of IoT devices in IEEE 802.11 Wi-Fi networks for reliable IoT services. The proposed $e$ AP system controls the energy resources of IoT devices to extend the lifetime of the IoT device. For this purpose, we develop a new $e$ AP system that considers a cross-layer design with a prompt TCP ACK transmit function, a caching-and-retransmit IoT data function, and a multiple IoT data aggregate function to improve the energy efficiency of the IoT device. In addition, the proposed $e$ AP system has a device energy management module that precisely controls operating parameters, such as the transmission period of IoT packets, the delivery of traffic indication message (DTIM) value of IoT devices, and the transmitting power of IoT devices. These features extend the lifetime of the battery-powered IoT devices while satisfying service requirements for reliable IoT services. The long listening time of TCP ACK messages in receive mode (Rx) results in the high energy consumption of IoT devices due to the large round trip time. The proposed $e$ AP system reduces the reception time of TCP ACK messages in the IoT device, using the prompt TCP ACK transmit function in the $e$ AP. This reduces the long Rx mode time for TCP ACK reception, and increases the short sleep mode time, which results in increase of the energy efficiency of the IoT device. In the energy-saving analyses, we formulate an energy consumption model for the IoT device, and determine the energy-saving gain when the IoT device uses the $e$ AP system model, compared to a legacy AP system model. Our performance evaluation results verify that the proposed $e$ AP system achieves a maximum improvement in energy efficiency of approximately 88%, and 8.4 times improvement in the expected lifetime of the IoT device, compared to the legacy AP system model.

With the emergence of various Internet of Things (IoT) technologies, energy-saving schemes for IoT devices have been rapidly developed. To enhance the energy efficiency of IoT devices in crowded environments with multiple overlapping cells, the selection of access points (APs) for IoT devices should consider energy conservation by reducing unnecessary packet transmission activities caused by collisions. Therefore, in this paper, we present a novel energy-efficient AP selection scheme using reinforcement learning to address the problem of unbalanced load that arises from biased AP connections. Our proposed method utilizes the Energy and Latency Reinforcement Learning (EL-RL) model for energy-efficient AP selection that takes into account the average energy consumption and the average latency of IoT devices. In the EL-RL model, we analyze the collision probability in Wi-Fi networks to reduce the number of retransmissions that induces more energy consumption and higher latency. According to the simulation, the proposed method achieves a maximum improvement of $53\%$ in energy efficiency, $50\%$ in uplink latency, and a $2.1$-times longer expected lifespan of IoT devices compared to the conventional AP selection scheme.

This paper proposes an energy-efficient multi-level sleep mode control for periodic transmission (MSC-PUT) in private fifth-generation (5G) networks. In general, private 5G networks meet IIoT requirements but face rising energy consumption due to dense base station (BS) deployment, particularly impacting operating expenses (OPEX). An approach of BS sleep mode has been studied to reduce energy consumption, but there has been insufficient consideration for the periodic uplink transmission of industrial Internet of Things (IIoT) devices. Additionally, 5G New Reno’s synchronization signal interval limits the effectiveness of the deepest sleep mode in reducing BS energy consumption. By addressing this issue, the aim of this paper is to propose an energy-efficient multi-level sleep mode control for periodic uplink transmission to improve the energy efficiency of BSs. In advance, we develop an energy-efficient model that considers the trade-off between throughput impairment caused by increased latency and energy saving by sleep mode operation for IIoT’s periodic uplink transmission. Then, we propose an approach based on proximal policy optimization (PPO) to determine the deep sleep mode of BSs, considering throughput impairment and energy efficiency. Our simulation results verify the proposed MSC-PUT algorithm’s effectiveness in terms of throughput, energy saving, and energy efficiency. Specifically, we verify that our proposed MSC-PUT enhances energy efficiency by nearly $27.5\%$ when compared to conventional multi-level sleep operation and consumes less energy at $75.21\%$ of the energy consumed by the conventional method while incurring a throughput impairment of nearly $4.2\%$. Numerical results show that the proposed algorithm can significantly reduce the energy consumption of BSs accounting for periodic uplink transmission of IIoT devices.

This paper proposes a cooperative multi-agent online reinforcement learning-based (COMORL) bias offset (BO) control scheme for cell range expansion (CRE) in dense heterogeneous networks (HetNets). The proposed COMORL scheme controls BOs for CRE to maximize the number of user equipments (UEs) that satisfy their quality of service (QoS) requirements, especially in terms of delay and data rates. For this purpose, we developed a QoS satisfaction indicator that measures a violation of delay requirements by considering both QoS requirements and signal-to-interference-plus-noise ratio (SINR). In addition, we formulated a Markov decision process (MDP) model that is solved with a cooperative multi-agent online reinforcement learning algorithm. The proposed COMORL scheme maximizes the global utility for load-coupled base stations. Our simulation results verify the proposed COMORL scheme’s effectiveness in terms of throughput, delay satisfaction ratio, and fairness. Specifically, we verify that the proposed COMORL scheme achieves a maximum of approximately 27% and 30% improvement of the delay satisfaction ratio, which is how many UEs satisfy their delay requirement among all of the UEs in a serving BS under medium and full traffic loads, respectively, in a dynamic scenario in comparison to the max-SINR scheme.

The proliferation of 5G networks has revolutionized wireless communication by delivering enhanced speeds, ultra-low latency, and widespread connectivity. However, in heterogeneous cloud radio access networks (H-CRAN), efficiently managing inter-cell interference while ensuring energy conservation remains a critical challenge. This paper presents a novel energy-efficient, dynamic enhanced inter-cell interference coordination (eICIC) scheme based on deep reinforcement learning (DRL). Unlike conventional approaches that focus primarily on optimizing parameters such as almost blank subframe (ABS) ratios and bias offsets (BOs), our work introduces the transmission power during ABS subframes (TPA) and the channel quality indicator (CQI) threshold of victim user equipments (CTV) into the optimization process. Additionally, this approach uniquely integrates energy consumption into the scheme, addressing both performance and sustainability concerns. By modeling key factors such as signal-to-interference-plus-noise ratio (SINR) and service rates, we introduce the concept of energy-utility efficiency to balance energy savings with quality of service (QoS). Simulation results demonstrate that the proposed scheme achieves up to $70\%$ energy savings while enhancing QoS satisfaction, showcasing its potential to significantly improve the efficiency and sustainability of future 5G H-CRAN deployments.

Despite recent advances in automatic speech recognition (ASR), performance remains limited for Korean-accented English due to the limited availability of accent-specific speech data, including pronunciation and prosodic variations. To address this limitation, we propose a synthetic data generation framework for improving Whisper-based ASR performance. Synthetic speech is generated by converting English text into Hangul-based phonetic transcriptions using an intermediate IPA representation to reflect the phonological characteristics of Korean-accented English. The ASR model is fine-tuned using Low-Rank Adaptation with a mixture of synthetic and authentic speech data. Experimental results demonstrate relative reductions of up to 16.40% in the character error rate, 14.93% in the word error rate, and 14.81% in the phoneme error rate compared to the pretrained baseline.