Der-Jiunn Deng

Mobility management associating all user equipments (UEs) to proper base stations (BSs) [also known as handover (HO)] to achieve the designated performance optimization is one of the most crucial functions in mobile networks. Although effective mobility management has received considerable research attentions, existing schemes follow event-driving operations, in which HO decisions are made based on events of performance degradation that BSs/UEs passively suffer from or proactively foresee. However, the performance and decisions of these schemes are highly subject to the identified events, to lose generalization and better performance under unidentified events. To address this issue, we propose a foundation model (FM)-based mobility management for the sixth-generation (6G) mobile networks inherently supporting artificial intelligence (AI) computing, in which the FM generates the HO decisions for all UEs to maximize the overall throughput under the constraints of ping-pong rate and HO failure (HOF) rate by implicitly taking moving trajectories, traffic demands, channel conditions of all UEs, and available resources of BSs into account. To this end, a hierarchical model structure composed of long short-term memory (LSTM) networks with multihead attention (MHA) is adopted, which is trained by emulated datasets with augmentation. The performance evaluation results show that the proposed scheme outperforms the state-of-the-art schemes in terms of the average throughput over all UEs while satisfying the required ping-pong rate and HOF rate, and justify the robustness of the proposed FM under different network deployment scenarios.

With the rapid advancement of Internet of Energy (IoE), peer-to-peer (P2P) energy trading has become an important mechanism for decentralized energy balancing and distribution-network relief. However, prior works have focused mainly on single platforms or isolated communities, offering limited insight into coordinated operation across multiple communities within a unified IoE ecosystem. This work proposes an Internet of Things (IoT)-enabled hierarchical trading framework that manages both intra- and inter-community energy transactions through decentralized area platforms and a central coordinating platform. A mixed-integer programming (MIP) model is formulated to maximize the daily profit of energy-trading operators by optimizing household charging and discharging decisions, supported by distributed renewable generation and battery storage. The framework incorporates a dynamic IoT-driven pricing mechanism that adapts buying and selling prices using real-time measurements from smart meters and local IoE communication networks, ensuring responsive and data-driven market operation under varying conditions. Because the optimization problem is NP-complete, a simplified harmony search (SHS) algorithm is adopted to efficiently generate high-quality solutions. Experiments across 20 system configurations show that SHS outperforms several widely used metaheuristic baselines in both solution quality and run-to-run stability. Results further indicate that larger communities yield higher profits, with the maximum observed when each community contains 25 households. Eliminating the central trading platform causes a drastic reduction in operator profit, confirming its essential role in enabling scalable coordination and equitable energy allocation across heterogeneous communities. These findings highlight the effectiveness of the proposed hierarchical IoE architecture and underscore the importance of centralized coordination in next-generation smart energy systems.

This work explores the integration into a semiactive RFID tag of a custom, ISO-compatible LF wake-up ASIC operating at 134.2 kHz with a Bluetooth Low Energy (BLE) advertisement response strategy. This integration aims to extend the reading range, enhancing bulk RFID reading for highly efficient traceability and monitoring applications in the cattle industry and dairy farms. A holistic analysis was carried out to validate the design of the LF tag, and a measured reading distance of 14 × that of the state-of-the-art passive LF ISO11784/85 tags was achieved. The proposed tag is powered by a small coin cell battery and achieves a lifespan that exceeds that of dairy cattle, remaining low-power, robust and adhering to current standards.

The 5G technology in smart agricultural machinery, UAVs, and agriculture-related sensors is paramount for outdoor agricultural IoT operations. However, conventional cloud computing is not optimal for processing real-time data from numerous IoT devices. In addition, it is challenging to deploy these communications devices outdoors while ensuring hardware safety and cybersecurity. Consequently, this work firstly creates an integer programming model for the optimal deployment of a 5G and multi-access edge computing (MEC) system with hardware maintenance safety and cybersecurity in a smart agricultural park, enhancing system stability, scalability, and reducing response delay. Then, this work hybridizes simplified harmony search (SHS) and variable neighborhood search (VNS) to solve this problem. SHS can explore a large area of the solution space through its stochasticity and global search capability, while VNS explores the solution space by systematically changing the neighborhood structure, focusing on local search and fine tuning. By combining the two, a good balance between global exploration and local exploitation can be achieved. Simulation experiments demonstrate the proposed algorithm advantages in computational efficiency, energy consumption, and cost reduction in real deployment scenarios. This algorithm can effectively minimize the costs of installation, security maintenance, and data security, under various constraints, compared with past approaches.

The reconfigurable intelligent surface (RIS) steering reflective beam directions toward a target mobile user equipment (UE) has been a promising technology for coverage enhancement and physical-layer (PHY) security to avoid signal eavesdroppers and jammers. However, tracking and predicting the moving trajectory of UE demand significant computing burdens unaffordable to the RIS. Performing beam steering computations on the radio access network (RAN) intelligent controller in the open RAN (O-RAN) architecture therefore emerges as a revolutionary paradigm. In this paper, we propose a PHY-secured beam steering design based on a pre-processing-based Kalman filter algorithm. Through a pre-processing phase to derive the radiation feature, the proposed design accurately tracks the position of the mobile UE and quickly recovers the position search while position tracking fails. Through practical field-try experiments, we justify the capability of O-RAN RIC-empowered RIS for PHY security.

With promising benefits of splitting deep neural network (DNN) computation loads to the edge server, split computing has been a novel paradigm achieving high-quality artificial intelligence (AI) services for the energy-constrained user equipments (UEs). To satisfy the service demands of a large number of UEs, traditional edge-UE split computing evolves toward multitier split computing involving the edge and cloud servers with different capabilities, leading to a “complex” optimization involving communication and computation. To tackle this challenge, in this article, we propose a multitier deep reinforcement learning (DRL) decision-making scheme for distributed splitting point selection and computing resource allocation in the three-tier UE-edge-cloud split computing systems. With the proposed scheme, the high-dimensional optimization can be tackled by the UEs and an edge server with different control cycles through performing local decision-making tasks in a sequential manner. Based on the policies updated by the UEs and the edge server in successive stages, the overall performance of split computing can be continuously improved, which is justified through a theoretical convergence performance analysis. Comprehensive simulation studies show that the proposed multitier DRL decision-making scheme outperforms the conventional split computing schemes in terms of the overall latency, inference accuracy, and energy efficiency to practice multitier split computing.