27 TerraSAR-X (TS-X) time series images with 11-day interval over the west suburb of Tianjin, China were processed using the Stanford Method for PS (STMPS). Deformation over 681,364 PS pixels was retrieved. The accura...
详细信息
The load forecasting method usually starts from a single method,we usually improved prediction methods to get the better forecasting accuracy,but this often confined to the application of the method,combination foreca...
详细信息
The load forecasting method usually starts from a single method,we usually improved prediction methods to get the better forecasting accuracy,but this often confined to the application of the method,combination forecasting method can achieve superiority of various methods,the forecast accuracy is higher than single forecasting *** this paper,we used RBF neural network prediction method and support vector machine forecasting *** neural network prediction method is the more popular method in recent years,it has the better generalization ability to the traditional neural network prediction method,It can effectively avoid local minima value and has a very good learning ability;S VM prediction method is transformed into one-dimensional nonlinear prediction of linear space,it has very precise calculation process and can meet the high forecast *** on the combination of the two methods,not only from the Angle of artificial memory model prediction,and using the tight nonlinear model,ultimately meet the purpose of combined *** main innovation in this paper is that assess the result of every kind of prediction method by making the standards of error qualified,using the error rate to determine the weight of combination,finally,we can get the satisfactory results through an empirical analysis.
Classification following a non-linearly separable boundary is a cognitive task that is hard to complete for humans and animals. Feedforward neural networks are able to perform the task efficiently but they present lit...
详细信息
ISBN:
(纸本)9781467361279
Classification following a non-linearly separable boundary is a cognitive task that is hard to complete for humans and animals. Feedforward neural networks are able to perform the task efficiently but they present little correspondence with human cognition. We present a neural network model of human categorization of the n-bit parity problem, using a modification of the pi-sigma network incorporating bidirectional associative memories and self-organizing maps. The model has good cognitive validity with cognitive human processes and is efficient.
The importance of Internet of Things (IoT) lies in its far-reaching impact on our daily lives, so it is prone to risks. Therefore, security measures must be considered before deploying IoT networks to avoid intrusions...
The importance of Internet of Things (IoT) lies in its far-reaching impact on our daily lives, so it is prone to risks. Therefore, security measures must be considered before deploying IoT networks to avoid intrusions. Most of the relevant work in the literature have been dedicated to improve the intrusion detection system and security systems. However, most current Intrusion Detection System (IDS) approaches suffer from the lack of a proper testing procedures before an actual attack. In this paper, we propose a low-cost solution to detect malicious nodes by ensuring that witness nodes work properly. We propose an effective automatic detection strategy for the test sequence and selection of witness nodes. The goal is to test the automatic correct actions of a witness node when an attacker launches a replication or cloning attack. We present a case study of replication attacks in IoT and use the CupCarbon simulator to evaluate our approach and demonstrate how a compromised node could be detected among witness nodes.
暂无评论