جستجوی مقالات مرتبط با کلیدواژه « خوشه بندی » در نشریات گروه « مهندسی دریا »

It is a big challenge for a ELINT system to distinguish between some same type radars pulses that received from one direction and a frequency hopping radar. The aim of this paper is to classifying same type radar with frequency hopping that work in a near frequency band in the most complicated state. In existing methods, a unit's information is collected in peacetime and distinguished from same type radars in wartime by classification. In this research, the received signals from same type radars without prior knowledge are separated. This method is based on radar signature that parameters extracted from the pulse envelope, are used for pulse separation and determining the accurate number of radars. Due to the uncertainty of the number of radars, clustering is made in some iteration from 1 to k clusters resulting from frequency separation. Validation Criteria are then used to determine the correct number of clusters. The proposed method was tested by randomly generating 100 radar pulses from 5 classmate radar samples and the result shows and its results show the higher accuracy of Davis Boldin and Silhouette criteria than Gap and Kalinsky criteria in different SNRs. Also, Skewness and kurtosis properties have been shown to be relatively more stable at low SNRs.

Acoustic source separation is considered as one of the challenging issues in the signal enhancement field and it becomes even more problematic when some acoustic sources are mixed in the presence of noise. In this paper, the weighted gaussian mean shift algorithm method for dual-channel systems is generalized to multichannel systems. The source separation mechanism is blind and it has the ability to extract several sources from various mixtures. The proposed method is compared with a number of existing algorithms for acoustic source separation under noisy conditions recorded by multiple hidrophones. The results demonstrate the effectiveness of the proposed method.

Sensor nodes have limited energy in wireless networks, so one of the main challenges in these networks is the energy consumption. Clustering-based methods are commonly used to improve energy consumption. In this way, clusters are selected from the existing nodes and have the task of aggregating the received data from the sensors and sending information to the sink. In this paper, the Leach algorithm is used in the first round and in the following rounds, the cyclone movement pattern in manta ray is used to perform clustering and load balancing in network. Using this approach causes to select the cluster-head from the high density location of each cluster, which reduces the energy consumption of the nodes during the data transfer. The proposed approach is implemented in MATLAB simulation and compared with the previously presented methods in two different scenarios. The results indicate an improvement about 27% and 14% respectively, in terms of the remaining energy of the nodes and the number of live nodes in a large number of nodes in network.

In this study, wave height for four buoyes in the Gulf of Mexico predicted using wind and wave characteristics such as mean wind speed, wind gust, wind pressure, air temperature and water temperature difference, duration and fetch length. For this purpose, the combination of kmeans algorithm and the neural network has been used. Initially, the data were sorted to eliminate the effect of the seasonal process in calculating the wave height, and then the meteorological data clustered using the kmeans algorithm from k=1 to k=10. The specified clusters were introduced as inputs of the MLP model and the mean RMSE index for k cluster was calculated. The results showed that in most buoys, the optimum number of clusters is between 8 and 10. In addition, Comparison of the results of this study with other studies showed that in previous studies, the prediction was based on the inputs of the recorded wave height of adjacent buoys or wave time delay has been applied. While in this research, only meteorological parameters have been used as inputs to increase the accuracy of prediction, which has a fairly good accuracy. Therefore, using the clustering technique increases the accuracy of the estimation at wave height.

Underwater wireless sensor networks are a type of sensor networks that are submerged and communicate with each other by sound waves. The fundamental differences between underwater sound propagation and radio waves at ground level, on the one hand, and energy resource constraints, on the other hand, require new criteria for the design of underwater routing protocols. Most routing protocols are based on the location of the sensor nodes. On the other hand, maintaining the position of nodes in underwater sensor networks is the most important issue. In this paper, we propose a position-independent and cluster-based routing protocol in underwater wireless sensor networks called EECRU. The EECRU algorithm uses the concept of special nodes that act as headers. If no header is found, then this algorithm transfers data using the neighbor node method. It also uses the concept of CDMA to obtain information from different nodes and special nodes. The simulation results using MATLAB software confirm that the proposed algorithm is effective in increasing network lifetime and reducing power consumption.