|
Performance analysis of high rate packet capture
on multiprocessor platform
Fan YANG, Yinan DOU, Zhenming LEI, Hua YU,
Front. Electr. Electron. Eng.. 2010, 5 (1): 36-42.
https://doi.org/10.1007/s11460-009-0070-6
With the continuous improvement of transmission rate, high speed network links require good performance of packet capture. The multiprocessor platform has strong computational capabilities, and brings new chance for high rate packet capture. In this paper, we analyze the performance of common packet capture approaches that are based on general-purpose multiprocessor platform. The analysis contains two aspects: one is the maximum packet capture rate and throughput on multiprocessor platform, the other is central processing unit (CPU) load under the maximum capture rate. By comparing and analyzing the experimental result, we give the maximum packet capture rate and throughput of different capture approaches. Furthermore, we analyze the CPU load which is produced by two capture processes run on the multiprocessor platform simultaneously and make a comparison with the single capture process.
References |
Related Articles |
Metrics
|
|
Variable length dynamic addressing based on network
traffic distribution in wireless sensor networks
Jun ZHAN, Bo YANG, Aidong MEN,
Front. Electr. Electron. Eng.. 2010, 5 (1): 43-48.
https://doi.org/10.1007/s11460-009-0074-2
In this paper, a novel dynamic addressing scheme for wireless sensor networks (WSNs) is proposed by using variable length coding. A WSN is typically composed of numerous tiny energy-constrained sensor nodes with limited information processing and data storage capabilities; thus, the energy-efficient strategy is the key issue in designing protocols for WSN. Traditional addressing strategies adopt flat addressing (static and uniform addresses) for sensor nodes. However, the proposed variable length dynamic addressing (VLDA) for sensor nodes is based on the fact that different nodes in the network have uneven traffic loads. Therefore, nodes with more data to receive or send are allocated with shorter addresses. Whether a node is busy or not is determined by the network traffic distribution (NTD), which is defined as the number of data packets each node has received or sent in a period of time. Sensor nodes’ energy is saved by VLDA scheme; hence, the wireless sensor network’s lifetime is extended. In the simulation, a 20% improvement has been achieved through the addressing scheme compared to traditional flat addressing.
References |
Related Articles |
Metrics
|
|
Near optimal MIMO detection with reduced search
space
Rongrong QIAN, Tao PENG, Yuan QI, Wenbo WANG,
Front. Electr. Electron. Eng.. 2010, 5 (1): 59-64.
https://doi.org/10.1007/s11460-009-0075-1
A multi-input multi-output (MIMO) detection scheme that requires considerable low complexity but still achieves the near optimal performance is proposed. The fundamental idea of the proposed MIMO detection scheme consists of two points: 1) the computational complexity is restrained by a complexity limit in low signal-to-noise ratio (SNR) region; 2) while in high SNR region, the complexity is significantly reduced by the proposed search space method. Comparing with existing fixed-complexity techniques of MIMO detection (e.g., K-best sphere detector and reduced-search maximum-likelihood (RS ML) detection), the significant benefit of proposed detection scheme is that less computational power will be spent for the given data rate, or the throughput of detector can be increased for high SNR cases. According to the simulation results, the near optimal performance can be obtained while the detection complexity is kept considerable small.
References |
Related Articles |
Metrics
|
|
Product HMM-based training method for acoustic
model with multiple-size units
Hao WU, Xihong WU, Huisheng CHI,
Front. Electr. Electron. Eng.. 2010, 5 (1): 65-71.
https://doi.org/10.1007/s11460-009-0076-0
Multiple-size units-based acoustic modeling has been proposed for large vocabulary speech recognition system to improve the recognition accuracy with limited training data. By introducing a limited number of long-size units into unit set, this modeling scheme can make better acoustic model precision than complete short-size unit modeling without losing model trainability. However, such a multiple-size unit acoustic modeling paradigm does not always bring reliable improvement on recognition perform-ance, since when a large number of long-size units are added in, the amount of training data for short-size units will decrease and result in insufficiently trained models. In this paper, a modified Baum-Welch training method is proposed, which uses product hidden Markov models (PHMMs) to couple units with different sizes and enables them to share same portions of training data. The validity of proposed method is proved by experiment results.
References |
Related Articles |
Metrics
|
|
Size-self-adaptive recognition method of vehicle
manufacturer logos based on feature extraction and SVM classifier
Wenting LU, Honggang ZHANG, Kunyan LAN, Jun GUO,
Front. Electr. Electron. Eng.. 2010, 5 (1): 77-84.
https://doi.org/10.1007/s11460-009-0072-4
Besides their decorative purposes, vehicle manufacturer logos can provide rich information for vehicle verification and classification in many applications such as security and information retrieval. However, unlike the license plate, which is designed for identification purposes, vehicle manufacturer logos are mainly designed for decorative purposes such that they might lack discriminative features themselves. Moreover, in practical applications, the vehicle manufacturer logos captured by a fixed camera vary in size. For these reasons, detection and recognition of vehicle manufacturer logos are very challenging but crucial problems to tackle. In this paper, based on preparatory works on logo localization and image segmentation, we propose a size-self-adaptive method to recognize vehicle manufacturer logos based on feature extraction and support vector machine (SVM) classifier. The experimental results demonstrate that the proposed method is more effective and robust in dealing with the recognition problem of vehicle logos in different sizes. Moreover, it has a good performance both in preciseness and speed.
References |
Related Articles |
Metrics
|
14 articles
|