• ETRI Journal
• /
• v.46 no.2
• /
• pp.205-217
• /
• 2024

We propose a bandwidth prediction approach based on deep learning. The approach is intended to accurately predict the bandwidth of various types of mobile networks. We first use a machine learning technique, namely, the gradient boosting algorithm, to recognize the connected mobile network. Second, we apply a handover detection algorithm based on network recognition to account for vertical handover that causes the bandwidth variance. Third, as the communication performance offered by 3G, 4G, and 5G networks varies, we suggest a bidirectional long short-term memory model with time distribution for bandwidth prediction per network. To increase the prediction accuracy, pretraining and fine-tuning are applied for each type of network. We use a dataset collected at University College Cork for network recognition, handover detection, and bandwidth prediction. The performance evaluation indicates that the handover detection algorithm achieves 88.5% accuracy, and the bandwidth prediction model achieves a high accuracy, with a root-mean-square error of only 2.12%.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.949-950
• /
• 2008

The intra prediction technique is the one of the key factors to the success of H.264. There are nine optional prediction modes for each $4{\times}4$ luma block and 4 modes for each $16{\times}16$ luma block. To reduce the intra mode bits efficiently, the most probable mode (MPM) is estimated by using the intra modes of the adjacent blocks, since intra modes for neighboring $4{\times}4$ luma blocks are correlated. In this paper, a new method for estimating the MPM is proposed by using the statistical distribution of intra modes of adjacent intra blocks. Experimental results show that the proposed method can achieve a coding gain of about 0.1dB.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.875-878
• /
• 2015

In this paper, we propose an intra prediction hardware architecture with less processing time, computations and reduced hardware area for a high performance HEVC encoder. The proposed intra prediction hardware architecture uses common operation units to reduce computational complexity and uses $4{\times}4$ block unit to reduce hardware area. In order to reduce operation time, common operation unit uses one operation unit to generate predicted pixels and filtered pixels in all prediction modes. Intra prediction hardware architecture introduces the $4{\times}4$ PU design processing to reduce the hardware area and uses intemal registers to support $32{\times}32$ PU processmg. The proposed hardware architecture uses ten common operation units which can reduce execution cycles of intra prediction. The proposed Intra prediction hardware architecture is designed using Verilog HDL(Hardware Description Language), and has a total of 41.5k gates in TSMC $0.13{\mu}m$ CMOS standard cell library. At 150MHz, it can support 4K UHD video encoding at 30fps in real time, and operates at a maximum of 200MHz.

• The KIPS Transactions:PartA
• /
• v.9A no.4
• /
• pp.497-502
• /
• 2002

Conditional branches can severely limit the performance of instruction level parallelism by causing branch penalties. 2-level adaptive branch predictors were developed to get accurate branch prediction in high performance superscalar processors. Although 2 level adaptive branch predictors achieve very high prediction accuracy, they tend to be very costly. In this paper, set-associative cached correlated 2-level branch predictors are proposed to overcome the cost problem in conventional 2-level adaptive branch predictors. According to simulation results, cached correlated predictors deliver higher prediction accuracy than conventional predictors at a significantly lower cost. The best misprediction rates of global and local cached correlated predictors using set-associative caches are 5.99% and 6.28% respectively. They achieve 54% and 17% improvements over those of the conventional 2-level adaptive branch predictors.

• The Journal of Industrial Distribution & Business
• /
• v.8 no.4
• /
• pp.47-52
• /
• 2017

Purpose - This current study will investigate the average financial ratio of top and failed five-star hotels in the Jeju area. A total of 14 financial ratio variables are utilized. This study aims to; first, assess financial ratio of the first-class hotels in Jeju to establishing variables, second, develop distress prediction model for the first-class hotels in Jeju district by using logit analysis and third, evaluate distress prediction capacity for the first-class hotels in Jeju district by using logit analysis. Research design, data, and methodology - The sample was collected from year 2015 and 14 financial ratios of 12 first-class hotels in Jeju district. The results from the samples were analyzed by t-test, and the independent variables were chosen. This was an empirical study where the distress prediction model was evaluated by logit analysis. This current research has focused on critically analyzing and differentiating between the top and failed hotels in the Jeju area by utilizing the 14 financial ratio variables. Results - The verification result of the accuracy estimated by logit analysis has shown to indicate that the distress prediction model's distress prediction capacity was 83.3%. In order to extract the factors that differentiated the top hotels in the Jeju area from the failed hotels among the 14 chosen, the analysis of t-black was utilized by independent variables. Logit analysis was also used in this study. As a result, it was observed that 5 variables were statistically significant and are included in the logit analysis for discernment of top and failed hotels in the Jeju area. Conclusions - The distress prediction press' prediction capability was compared in this research analysis. The distress prediction press prediction capability was shown to range from 75-85% by logit analysis from a previous study. In this current research, the study's prediction capacity was shown to be 83.33%. It was considered a high number and was found to belong to the range of the previous study's prediction capacity range. From a practical perspective, the capacity of the assessment of the distress prediction model in the top and failed hotels in the Jeju area was considered to be a prominent factor in applications of future hotel appraisal.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.23 no.4
• /
• pp.295-300
• /
• 2013

The noise map can be applied to predict the effect of noise and establish the noise reduction measure. But the predicted value in the noise map can vary depending on the input variables. Thus, we surveyed the several prediction models and analyzed the changes corresponding to the variables for obtaining the coherency and accuracy of prediction results. As a result, we know that the Schall03 and CRN model can be applied to predict the railway noise in Korea and the correction value, such as bridges correction, multiple reflection correction, curve correction must be used for reflecting the condition of the prediction site. Also, we know that the prediction guideline is an essential prerequisite in order to obtain the unified and accurate predicted value for railway noise.

• The Journal of Information Systems
• /
• v.14 no.2
• /
• pp.277-292
• /
• 2005

This paper uses a data mining approach to develop bankruptcy prediction models suitable for traditional (off-line) companies and electronic (on-line) companies. It observes the differences in the composition prediction models between these two types of companies and provides interpretation of bankruptcy classifications. The bankruptcy prediction models revealed the major variables in predicting bankruptcy to be 'cash flow to total assets' and 'gross value-added to net sales' for traditional off-line companies while 'cash flow to liabilities','gross value-added to net sales', and 'current ratio' for electronic companies. The accuracy rates of final prediction models for traditional off-line and electronic companies were found to be $84.7\%\;and\;82.4\%$, respectively. When the model for traditional off-line companies was applied for electronic companies, prediction accuracy dropped significantly in the case of bankruptcy classification (from $70.4\%\;to\;45.2\%$) at the level of a blind guess ($41.30\%$). Therefore, the need for different models for traditional off-line and electronic companies is justified.

• Journal of the Korea Concrete Institute
• /
• v.12 no.4
• /
• pp.69-78
• /
• 2000

Creep of concrete is the most dominating factor affecting time-dependent deformations of concrete structures. Especially, creep deformation for design and construction in prestressed concrete structures should be predicted accurately because of its close relation with the loss in prestree of prestressed concrete structures. Existing creep-prediction models for special applications contain several impractical factors such as the lack ok accuracy, the requirement of long-term test and the lack of versatility for change in material properties, ets., which should be improved. In order to improve those drawbacks, a methodology to modify the creep-prediction equation specified in current Korean concrete structures design standard (KCI-99), which underestimates creep of concrete and does not consider change of condition in mixture design, is proposed. In this study, short-term creep tests were carried out for early-age concrete within 28 days after loading and their test results on influencing factors in the equation are analysed. Then, the prediction equation was modified by using the early-age creep test results. The modified prediction equation was verified by comparing their results with results obtained from long-term creep test.

• Atmosphere
• /
• v.24 no.4
• /
• pp.533-540
• /
• 2014

Statistical forecast models for the prediction of the summertime Changma precipitation have been developed in this study. As effective predictors for the Changma precipitation, the springtime sea surface temperature (SST) anomalies over the North Atlantic (NA1), the North Pacific (NPC) and the tropical Pacific Ocean (CNINO) has been suggested in Lee and Seo (2013). To further improve the performance of the statistical prediction scheme, we select other potential predictors and construct 2 additional statistical models. The selected predictors are the Northern Indian Ocean (NIO) and the Bering Sea (BS) SST anomalies, and the spring Eurasian snow cover anomaly (EUSC). Then, using the total three statistical prediction models, a simple ensemble-mean prediction is performed. The resulting correlation skill score reaches as high as ~0.90 for the last 21 years, which is ~16% increase in the skill compared to the prediction model by Lee and Seo (2013). The EUSC and BS predictors are related to a strengthening of the Okhotsk high, leading to an enhancement of the Changma front. The NIO predictor induces the cyclonic anomalies to the southwest of the Korean peninsula and southeasterly flows toward the peninsula, giving rise to an increase in the Changma precipitation.

• International Journal of Highway Engineering
• /
• v.15 no.4
• /
• pp.21-29
• /
• 2013

PURPOSES : The objective of this study is to provide for the overall SPL (Sound Pressure Level) prediction model by using the NCPX (Noble Close Proximity) measurement method in terms of regression equations. METHODS: Many methods can be used to measure the traffic noise. However, NCPX measurement can powerfully measure the friction noise originated somewhere between tire and pavement by attaching the microphone at the proximity location of tire. The overall SPL(Sound Pressure Level) calculated by NCPX method depends on the vehicle speed, and the basic equation form of the prediction model for overall SPL was used, according to the previous studies (Bloemhof, 1986; Cho and Mun, 2008a; Cho and Mun, 2008b; Cho and Mun, 2008c). RESULTS : After developing the prediction model, the prediction model was verified by the correlation analysis and RMSE (Root Mean Squared Error). Furthermore, the correlation was resulted in good agreement. CONCLUSIONS: If the polynomial overall SPL prediction model can be used, the special cautions are required in terms of considering the interpolation points between vehicle speeds as well as overall SPLs.