Zuo Zhang

An applicable short-term traffic flow forecasting method based on chaotic theory

Short-term traffic flow forecasting plays a very important role in urban traffic management and control. In this paper, According to the chaotic property of urban traffic flow, we compute the parameters of phrase space reconstruction for traffic flow system. Meanwhile, a local-forecasting method is introduced to predict urban road short-term traffic flow based on the theory of phrase space reconstruction. Self-organizing Map (SOM) network is introduced to seek the near neighbor. Case study using real traffic flow data from UTC-SCOOT system proves the validity of the method. The research in this paper is a significant attempt to forecast traffic flow from the viewpoint of non-linear time series.

Urban traffic network modeling and short-term traffic flow forecasting based on GSTARIMA model

This paper introduces a novel model—Generalized Space-Time Autoregressive Integrated Moving Average (GSTARIMA) methodology—into the field of short-term traffic flow forecasting in urban network. Compared to traditional STARIMA, GSTARIMA is a more flexible model class where parameters are designed to vary per location. Having proposed the model, a forecasting experiment based on actual traffic flow data in urban network in Beijing, China is constructed to verify the practicability of GSTARIMA model. After analysis and comparison with the traditional STARIMA model, the prediction results prove meritorious and the application of GSTARIMA improves the performance of urban network modeling.

Using LSTM and GRU neural network methods for traffic flow prediction

Accurate and real-time traffic flow prediction is important in Intelligent Transportation System (ITS), especially for traffic control. Existing models such as ARMA, ARIMA are mainly linear models and cannot describe the stochastic and nonlinear nature of traffic flow. In recent years, deep-learning-based methods have been applied as novel alternatives for traffic flow prediction. However, which kind of deep neural networks is the most appropriate model for traffic flow prediction remains unsolved. In this paper, we use Long Short Term Memory (LSTM) and Gated Recurrent Units (GRU) neural network (NN) methods to predict short-term traffic flow, and experiments demonstrate that Recurrent Neural Network (RNN) based deep learning methods such as LSTM and GRU perform better than auto regressive integrated moving average (ARIMA) model. To the best of our knowledge, this is the first time that GRU is applied to traffic flow prediction.

A buffer-driven approach to adaptively stream stored video over Internet

Since Internet conditions change rapidly with time, it is very important for a streaming media server to scale the quality of transmission depending on prevailing network conditions. We propose a novel buffer-driven adaptive streaming scheme for stored video. Instead of performing quality adaptation based on bandwidth estimation, this scheme scales quality of transmission based on both receiver buffer occupancy and sender buffer occupancy. Simulation results demonstrate that the proposed algorithm can achieve very smooth playback quality while preventing client buffer underflow or overflow. Besides, unlike existing quality adaptation scheme, our scheme can effectively utilize the available bandwidth without bandwidth estimation.

Short-Term Traffic Flow Forecasting Based on MARS

A promising traffic flow forecasting model based on multivariate adaptive regression splines (MARS) is developed in this paper. First, the historical traffic flow data is obtained from the loop detectors installed on the road network of Beijing. Then, part of the data is selected for training the MARS model while the rest is used to test the method. The results based on MARS method are compared with those of other methods such as the neural networks. The proposed MARS method is proved to have a considerable accuracy. Moreover, the model constructed with MARS can be described with analytical functions, which helps a lot in the further research on traffic flow forecasting.

Pattern-Based Study on Urban Transportation System State Classification and Properties

Abstract In this paper, the authors propose a novel pattern-based approach to model the classification and transition properties of traffic flow. First, fuzzy-set classification method is utilized to divide the traffic states, where the states are partitioned into a number of patterns. Then, fuzzy qualitative reasoning is applied to analyze the transitions between these states. Based on the probability of transition, stability of the traffic states is further investigated. Finally, a case study on urban transportation system is performed to demonstrate the usage of the proposed approach.

A Method for Urban Traffic Data Compression Based on Wavelet-PCA

Due to limitation of storage space and cost, the massive amount of urban detected traffic data becomes a great burden. How to efficiently reduce these data and store them becomes more and more urgent. In this paper, an effective method for urban traffic data compression based on Wavelet-PCA is proposed. After preprocessing, the dataset is decomposed using wavelet and then multi-scale PCA is applied to reduce them to different dimensions. Simulation results prove that this method can greatly compress original data at the cost of acceptable recovery error and outperforms conventional PCA. Finally, we develop a prototype system specifically for urban traffic data compression using Visual C#.NET and Matlab.

Analysis on Urban Traffic Network States Evolution Based on Grid Clustering and Wavelet De-noising

Traffic state and evolution are important for better knowledge of urban traffic properties, as well as for the better traffic control and management. Therefore, it has attracted much attention recently. Model-based and data-driven are two kinds of methods in handling with such issues. With the wide deployment of ITS, large volume traffic data are available and data-driven methods such as clustering analysis have found their applications in ITS. According to physical characteristics of urban traffic flow, the paper follows the data-driven analysis and develops a grid-based clustering method for traffic state extraction and state evolution analysis. It also designs a wavelet transformation as a filter to decrease the noise in raw traffic data. Results on de-noised signals show more definite trends for traffic state evolutions.

Sensor network nodes deployment based on Artificial Potential Functions

With the rapid development of detecting technology, sensors have been more and more applied in this area. In this paper an algorithm is proposed for sensor network nodes deployment based on artificial potential functions. First, sensor network nodes are initialized to fulfill the completeness with no regard for optimization. Then, according to a series of principles, they are adjusted to achieve the aims of guaranteeing the completeness of information for each road, the complete communication coverage and the lowest costs on the sensor devices deployment as well. In addition, with some experiments, the proposed algorithm is proved to be more suitable for a dense road network than for a sparse one. In order to improve the algorithm and solve some possible problems, some long-distance sensor nodes have been deployed in certain places and division of sensor nodes have been done to those heavy-load nodes.

A research on bus information service system using DSRC

Advanced Public Transportation Systems (APTS) are required to collect traffic information dynamically and provide real-time information for travelers and traffic management departments. DSRC is a short-range communication system that is applied specifically in the field of transportation. It can be used for identification and location of vehicles. The concept of the Bus Information Service (BIS) system using DSRC is proposed in this paper, and the system architecture, patterns of operation, and information service available are discussed. The establishment of this system will provide an open platform for public transportation management, and offer a practical information service for both travelers and bus dispatching companies.