Current Issue : October-December Volume : 2025 Issue Number : 4 Articles : 5 Articles
In this article, we present a triple-band Doherty power amplifier (DPA) with a Schiffman phase shifter, which achieved a 90-degree phase shift to facilitate broad frequency range operations. As the cornerstone of the triple-band DPA, the Schiffman phase shifter enabled simultaneous triple-band operations. Furthermore, the entire triple-band Doherty amplifier was designed and fabricated using GaN on SiC HEMT devices, confirming its practical applicability and robust performance. It achieved an output power of 34 dBm at the low-band (LB) frequency of 0.8 GHz, accompanied by a peak drain efficiency (DE) of 53%. Similarly, at the mid-band (MB) frequency of 1.6 GHz, the amplifier maintained an output power of 32 dBm with an identical peak DE of 45%. At the high-band (HB) frequency of 2.2 GHz, the DPA continued to deliver an output power of 33 dBm, again with a peak DE of 50%....
This article explores the transformative impact of artificial intelligence and machine learning technologies on telecommunications service assurance. The industry is experiencing a paradigm shift from reactive to predictive and prescriptive approaches to network management, enabled by three key technological breakthroughs: generative AI, causal inference, and federated learning. Major telecommunications providers are implementing Large Language Models to automate incident resolution processes, reducing resolution times and improving remediation quality. Simultaneously, causal AI is advancing proactive service assurance by establishing cause-and-effect relationships between network events, enabling operators to prevent service disruptions before they occur. Federated learning implementations are solving multi-domain assurance challenges by enabling cross-operator insights while maintaining data sovereignty. Together, these technologies are not merely enhancing existing processes but fundamentally reimagining telecommunications service assurance. The convergence of these approaches promises to deliver self-diagnosing, self-optimizing networks that can anticipate and address potential issues before they impact customer experience, representing a revolutionary advancement in how service quality and reliability are managed in increasingly complex network environments....
Telecommunications networks form critical infrastructure requiring exceptional reliability amidst growing complexity. Traditional monitoring approaches based on static thresholds increasingly fall short as 5G deployments, software-defined networking, and network function virtualization create dynamic environments generating massive operational data volumes. Machine learning offers transformative capabilities for anomaly detection in these networks, enabling proactive identification of potential failures before service disruption occurs. This article explores how artificial intelligence techniques, including supervised learning, unsupervised learning, and time series analysis, can be applied to telecom network management, highlighting architectural frameworks and real-world applications such as performance monitoring, predictive maintenance, security threat detection, and root cause analysis. While implementation challenges persist around data quality, model explainability, legacy system integration, and ethical considerations, emerging technologies like federated learning, reinforcement learning, and digital twins promise to further enhance network intelligence while addressing current limitations....
In the realm of natural language processing (NLP), text classification constitutes a task of paramount significance for large language models (LLMs). Nevertheless, extant methodologies predominantly depend on the output generated by the final layer of LLMs, thereby neglecting the wealth of information encapsulated within neurons residing in intermediate layers. To surmount this shortcoming, we introduce LENS (Linear Exploration and Neuron Selection), an innovative technique designed to identify and sparsely integrate salient neurons from intermediate layers via a process of linear exploration. Subsequently, these neurons are transmitted to downstream modules dedicated to text classification. This strategy effectively mitigates noise originating from non-pertinent neurons, thereby enhancing both the accuracy and computational efficiency of the model. The detection of telecommunication fraud text represents a formidable challenge within NLP, primarily attributed to its increasingly covert nature and the inherent limitations of current detection algorithms. In an effort to tackle the challenges of data scarcity and suboptimal classification accuracy, we have developed the LENS-RMHR (Linear Exploration and Neuron Selection with RoBERTa, Multi-head Mechanism, and Residual Connections) model, which extends the LENS framework. By incorporating RoBERTa, a multi-head attention mechanism, and residual connections, the LENS-RMHR model augments the feature representation capabilities and improves training efficiency. Utilizing the CCL2023 telecommunications fraud dataset as a foundation, we have constructed an expanded dataset encompassing eight distinct categories that encapsulate a diverse array of fraud types. Furthermore, a dual-loss function has been employed to bolster the model’s performance in multiclass classification scenarios. Experimental results reveal that LENS-RMHR demonstrates superior performance across multiple benchmark datasets, underscoring its extensive potential for application in the domains of text classification and telecommunications fraud detection....
Loading....