Open RAN is transforming mobile networks into open and flexible environments through the integration of modern software and artificial intelligence. However, the existing standardization documents provide insufficient...
详细信息
ISBN:
(数字)9798350364637
ISBN:
(纸本)9798350364644
Open RAN is transforming mobile networks into open and flexible environments through the integration of modern software and artificial intelligence. However, the existing standardization documents provide insufficient guidelines on the issuance and revocation of certificates, posing potential risks for the system. As an initial effort to overcome this limitation, we study the challenges related to the certificate management framework over the Open RAN fronthaul interface, with a particular focus on the mutual authentication and the coordination of security policies across different vendors.
Forecasting fruit sales is vital for the retail sector to establish an efficient supply chain, control inventory levels effectively, and satisfy customer needs. This research employs a machine learning algorithm to pr...
详细信息
ISBN:
(数字)9798331505264
ISBN:
(纸本)9798331505271
Forecasting fruit sales is vital for the retail sector to establish an efficient supply chain, control inventory levels effectively, and satisfy customer needs. This research employs a machine learning algorithm to predict upcoming fruit sales by analyzing past data. We aim to create a dependable prediction system by examining various data elements like past month's sales, seasonal patterns, promotions, holidays, and weather conditions. Linear regression, decision trees, and neural networks are being evaluated for their ability to forecast sales volumes. The model is created and coached with sales data, then evaluated for accuracy using different performance measures. Our findings in the fruit retail industry demonstrate that employing advanced forecasting methods decreases wastage and improves decision-making (resulting in increased profits). This research demonstrates that implementing data-driven sales forecasting within the fresh produce industry may result in more effective and environmentally friendly business operations.
In today's rapidly changing digital world, security and trust remain as an essential element supporting a range of communications, transactions, and distributed systems. With the invention of blockchain technology...
详细信息
ISBN:
(数字)9798350396157
ISBN:
(纸本)9798350396164
In today's rapidly changing digital world, security and trust remain as an essential element supporting a range of communications, transactions, and distributed systems. With the invention of blockchain technology, a trust may now be provided as “Trust as a Service (TaaS),” a decentralized, verifiable, and essential asset rather than a centralized digital network. In a decentralized and distributed ledger network, the blocks are connected in chronological order, and information is saved using distributed ledger techniques. Since the start of the deployment, blockchain has kept a record of all transactions. This research study analyses the explores the blockchain technology and analyze how it functions as a trusted provider across a range of industries. Also, this study investigates the fundamental ideas behind blockchain, its development, and the cryptographic security systems. This research study illustrates how blockchain maintains confidence in data quality, security, transparency, and immutability across a range of applications, including banking, supply chain, healthcare, and governance. This study also emphasizes the difficulties and possibilities present in the rapidly developing blockchain ecosystem and further discusses about the issues and challenges and various blockchain attacks, solutions, and tools. Finally, the open issues and solutions are analysed.
Demand side management (DSM) applications rely on the exchange of load profiles to effectively manage the operation of energy systems. However, sharing detailed energy profile data raises substantial privacy concerns,...
Demand side management (DSM) applications rely on the exchange of load profiles to effectively manage the operation of energy systems. However, sharing detailed energy profile data raises substantial privacy concerns, such as potential misuse of personal information. To mitigate these concerns, we investigate the potential of time aggregation (TA), which involves merging multiple samples of a profile into a single value representing multiple intervals. TA reduces data exchange and computational requirements in energy management and helps preserve user privacy by reducing granularity of user data. We show that, for an effective implementation of TA in energy management, it is important to make the right choice of TA method. We compare and evaluate seven different TA methods. Furthermore, we perform TA across various time frames using an optimization based DSM approach. Our findings reveal that if we aggregate load profiles from 15 minutes to 4 hours, we obtain both enhanced privacy and a 21% decrease in the required number of iterations with the investigated DSM method, albeit at the cost of a 15% decrease in objective value performance. Based on this, we conclude that depending on the application needs, TA with a carefully selected aggregation method has the potential to bring value to energy management, even when aggregating to a considerable extent.
Hybrid memory systems composed of Non-volatile memory (NVM) and DRAM to exploit the high density of NVM and low access latency of DRAM. Phase Change Memory (PCM), a type of non-volatile memory, is a viable choice for ...
详细信息
ISBN:
(数字)9798350380408
ISBN:
(纸本)9798350380415
Hybrid memory systems composed of Non-volatile memory (NVM) and DRAM to exploit the high density of NVM and low access latency of DRAM. Phase Change Memory (PCM), a type of non-volatile memory, is a viable choice for main memory. High write latency and high voltage requirements for PCM lead to Biased Temperature Instability (BTI) aging and performance degradation. De-stressing the memory circuit at regular intervals controls BTI aging. Memory performance can be enhanced by migrating the highest write count memory pages across memory units. Migration and de-stress halt the service of regular requests and affect the performance of the system. Therefore, it is crucial to control migration and de-stress to enhance hybrid memory performance while mitigating BTI aging. We propose DOPMig, a de-stress-aware page migration technique. The policy migrates write-intensive pages to DRAM at regular intervals but opportunistically parallel to the de-stress operation. This method of background migration helps to reduce the migration overheads and improves performance. DOPMig achieves a performance gain of 22%, and improves memory service rate by 15%, and increases DRAM access by 24%.
This paper presents an innovative approach to real-time sign language recognition using Long Short-Term Memory networks (LSTM), aimed at enhancing communication accessibil-ity for the deaf and hard-of-hearing communit...
详细信息
ISBN:
(数字)9798350384895
ISBN:
(纸本)9798350384901
This paper presents an innovative approach to real-time sign language recognition using Long Short-Term Memory networks (LSTM), aimed at enhancing communication accessibil-ity for the deaf and hard-of-hearing community. We address the challenge of understanding and interpreting sign language, which is critical for millions worldwide, yet restricted to those proficient in it. Our research contributes to bridging this communication gap by developing a deep learning model capable of recognizing a broad spectrum of sign language gestures and sentences with high accuracy and speed. Utilizing a rich dataset comprising diverse sign language gestures, collected in collaboration with a professional video production studio and proficient sign language users, we employ LSTM networks integrated with Dense layers to effectively capture the complex spatial and temporal patterns of sign language. The architecture of our model is specifically designed to accommodate the nuanced dynamics of sign language, with an emphasis on real-time processing. Through rigorous training and validation, our model demonstrates an outstanding accuracy rate of 92 % on a comprehensive testing dataset, alongside remarkable real-time processing capabilities. The sys-tem's efficiency in recognizing a wide array of sign gestures nearly instantaneously underscores its potential applicability in various real-world scenarios, including assistive technologies and human-computer interaction. This study not only showcases the practicality and efficacy of LSTM networks in real-time sign language recognition but also marks a significant step towards more inclusive and accessible communication technologies. Our future work includes integrating this system with the Langue des Signes Quebecoise website, further advancing the goal of universal communication accessibility.
Specific Language Impairment (SLI) in children, a neurodevelopmental disorder presenting significant hurdles in language acquisition. Motivated by the prevalence and complexity of SLI diagnosis, this article introduce...
详细信息
ISBN:
(数字)9798350370249
ISBN:
(纸本)9798350370270
Specific Language Impairment (SLI) in children, a neurodevelopmental disorder presenting significant hurdles in language acquisition. Motivated by the prevalence and complexity of SLI diagnosis, this article introduces an methodology using Mel-Frequency Cepstral Coefficients (MFCC) and chroma features combined with the XGBoost algorithm for classification. By analyzing a diverse dataset encompassing both SLI-diagnosed and non SLI children, the proposed model achieves efficiency of $\mathbf{9 9. 3 \%}$ accuracy on unseen data. Notably, the model’s strategic feature selection and utilization of XGBoost demonstrate superior performance compared to existing methods, highlighting its importance in SLI diagnosis and intervention planning. This work significantly advances machine learning-based diagnostic tools and holds promise for improving the quality of life for individuals impacted by SLI.
Traffic congestion is a pervasive problem causing severe environmental and economic issues. In recent years, traffic signal control using reinforcement learning (RL) has come a long way. Most existing studies focus on...
详细信息
ISBN:
(数字)9798331507862
ISBN:
(纸本)9798331507879
Traffic congestion is a pervasive problem causing severe environmental and economic issues. In recent years, traffic signal control using reinforcement learning (RL) has come a long way. Most existing studies focus on using distributed agents with data exchange among neighbors, which, however, increases network complexity and usage and still suffers from the lack of broader coordination. Meanwhile, the attention mechanism has achieved tremendous success, and advances in vehicle-to-infrastructure (V2I) communications have enabled real-time collections of granular data. However, integrating these technologies into traffic signal control remains under-explored. Therefore, we present GreenLight, a forward-thinking and eco-friendly traffic signal control framework that can be applied to V2I-equipped fog computing environments. For a large urban area, traffic signals are divided into clusters, each coordinated by a fog node with an RL agent. Intra-cluster indexed self-attention is applied to extract context-aware features that the fog-residing RL agent utilizes to determine the proper signal control command. Results of simulation experiments using both synthetic and real-world scenarios show that the presented framework yields lower waiting time, emissions, and fuel consumption compared to baseline methods, indicating its potential for next-generation transportation systems.
Since potatoes are a staple crop in India, there has been a noticeable growth in their cultivation. Given the significant nutritional and economic value of potatoes, this study addresses the significant issue of leaf ...
Since potatoes are a staple crop in India, there has been a noticeable growth in their cultivation. Given the significant nutritional and economic value of potatoes, this study addresses the significant issue of leaf diseases. The study enhances farmer lives and supports agricultural success in India by using machine learning techniques to categorize and detect diseases in real time. The major areas of study are image processing, feature extraction, and machine learning models like Random Forest, Naive Bayes, and K-Nearest Neighbors (KNN). The paper proposes a systematic framework that comprises steps for data collecting, picture preprocessing, feature extraction, training, and classification in order to reliably diagnose diseases. It also demonstrates how these procedures can be combined. The datasets used include approximately 5,000 labeled pictures of potato leaves that show early blight, late blight, and healthy leaves. Among the preprocessing methods are data augmentation, contrast enhancement, and image normalization. When it came to classifying diseases, the trained models fared extremely well; Naive Bayes, Random Forest, and KNN were especially accurate. The study demonstrates how integrating image processing and machine learning could aid in addressing potato leaf disease issues in agriculture..
In this study, we address the challenge of predicting air quality indices (AQIs) by leveraging Long Short-Term Memory (LSTM) networks. Recognizing the dynamic nature of air quality data, we explore the efficacy of thr...
详细信息
ISBN:
(数字)9798350364866
ISBN:
(纸本)9798350364873
In this study, we address the challenge of predicting air quality indices (AQIs) by leveraging Long Short-Term Memory (LSTM) networks. Recognizing the dynamic nature of air quality data, we explore the efficacy of three LSTM variants—standard LSTM, LSTM with an attention mechanism, and Bidirectional LSTM (BiLSTM)—in capturing temporal dependencies and improving forecast accuracy. Our methodology involved preprocessing a comprehensive dataset of air quality measures, followed by an experimental design that tested each model variant under different sequence lengths (30 and 365 days). The models were evaluated based on their mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE) across three response variables: Nitrogen Oxides (NO), Particulate Matter up to 10 micrometers in size (PM10), and Particulate Matter up to 2.5 micrometers in size (PM2.5). Notably, when comparing the RMSE values across the three models, the LSTM with the attention mechanism consistently outperformed the standard LSTM and BiLSTM models, particularly with a sequence length of 30 days, indicating a superior ability to capture the intricacies of temporal dependencies in environmental data. Moreover, the shorter sequence length generally yielded better predictive performance, suggesting that longer sequences might introduce unnecessary complexity or overfitting. These indications of our study suggest a promising direction for the deployment of attention-based models in real-time AQI prediction systems, which could enhance the accuracy of environmental forecasts and inform public health decisions more effectively. Future work will explore the integration of additional environmental factors, the refinement of model frameworks, and the expansion of the dataset to include multi-city comparisons.
暂无评论