Zhigang Jin

Traffic prediction using FARIMA models

Previous traffic measurements have found the coexistence of both long-range and short-range dependence in network traffic. Therefore, models are required to predict traffic that has both long-range and short-range dependence. This paper provides a procedure to model and predict traffic using FARIMA (p,d,q) models. Our experiments illustrate that the FARIMA model is a good model and is capable of capturing the property of actual traffic. We provide guidelines to simplify the FARIMA model fitting procedure and thus to reduce the time of traffic modeling and prediction.

Prediction-based admission control using FARIMA models

The FARIMA (p,d,q) model is a good traffic model capable of capturing both the long-range and short-range behavior of a network traffic stream in time. In this paper, we propose a prediction-based admission control algorithm for an integrated service packet network. We suggest a method to simplify the FARIMA model fitting procedure and hence to reduce the time of traffic modeling and prediction. Our feasibility-study experiments showed that FARIMA models which have number of parameters could be used to model and predict actual traffic on quite a large time scale.

Architecture and fuzzy adaptive security algorithm in intelligent firewall

The architecture of an intelligent firewall is presented and then its modules are described in detail. A fuzzy controller is the core module and the characteristics of packets are fuzzified as its inputs. Based on the fuzzy adaptive security algorithm, the overall security level of each kind of packet can be figured out and adjusted according to the varying states of the packets. The intelligent firewall can respond to these changes and take different actions accordingly. Therefore, the intelligent firewall resolves the conflict between security and speed and retains both high security and high performance.

An adaptive real-time intrusion detection system using sequences of system call

Intrusion detection is the process of monitoring computer networks and systems for violations of security policy. IDS may be of the network (NIDS) or host (HIDS) type. Traditionally, NIDS is the only way of preventing intrusion before an intrusion occurs through analyzing data packages, but has higher false rates. Although with lower false rates, HIDS detects anomalous actions by auditing host system logs that means the intrusion has took place. In this paper, we present one collaborate IDS module to make a real-time detection and block intrusions before occurrences, based on HIDS using sequences of system call anomaly detection.

Prediction-based network bandwidth control

The fractional autoregressive integrated moving average, FARIMA (p,d,q) model, is capable of capturing both the long-range and short-range dependence in the video traffic. We propose a prediction-based bandwidth allocation scheme for integrated service packet networks using the FARIMA (p,d,q) model. Our simulation experiments show that this scheme can significantly reduce the requirement of buffer size and the frame loss rate.

Designing and implementing a mobile ad hoc network testbed

This paper describes our experience in designing and implementing a mobile ad hoc network testbed. The design goal is high efficiency with low cost. Each node runs both the unicast routing protocol and the multicast routing protocols on a dynamic-on-demand basis, and interfaces seamlessly with the existing Internet infrastructure. The most important feature of this testbed is that it can generate user-defined traces and replay whatever the traces obtained in the recorded packet sequences. Through this function of the testbed, we can conduct more precise and flexible experiments.

The impact of self-similar traffic on network delay

The effect of self-similar traffic on the delay of a single queue system is studied through the use of the measured traffic and models as input process. A model-driven simulation-based method is then proposed for the computation of mean line delay in a network design. Both the hybrid-FGN and the FARIMA algorithms have been used to synthesize self-similar sample paths. The comparison results with real-traffic data sets firmly establish the usefulness of the proposed model-driven simulation-based method. A practical database method is also introduced that helps the designer to determine the parameters in network design. This approach may play an important role in network design and analysis.

Architecture and development of distributed WWW caching proxy

In order to alleviate heavy network overloads and slow network response, peer group, a new kind of distributed WWW caching proxy, is presented. Each caching proxy is a peer in the group and stores local data and maintains the cache index. Required data stored in local or remote proxy is accessible through the indexes. When peer proxies exchange cache information, only indexes are included, thus alleviating traffic overload. The peer group algorithm keeps track of any changes in the group by periodically broadcasting the cache. Therefore, the WWW distributed caching proxy is able to automatically update cache information, share cache indexes with other peers and alleviate slow convergence and inconsistency among peers and traffic overloads.

Enhanced polling mechanism of Bluetooth for realtime traffic

Bluetooth which operates in the 2.4 GHz ISM band provides short range, low cost, low power pico-cellular wireless connectivity. It intends to become the wireless personal area network standard itself. The Bluetooth specification proposes a round robin scheduler as possible solution for scheduling slaves in a Bluetooth piconet. However, this basic scheme performs badly under asymmetric traffic conditions. Recently, several polling schemes have been proposed to improve performance on asymmetric transmissions and to enhance the performance of real-time traffic. In this paper, we present a new polling algorithm for Bluetooth piconet that supports both delay and bandwidth guarantees and aims to remain fair and efficient with asymmetric flow rates. We present an extensive set of simulations and provide performance comparisons with round robin. And the results show that the performance of our algorithm can be better than that of the round robin. In addition, our modification does not introduce any control overhead

Enhancement of the On-Demand Multicast Routing Protocol in the ad hoc network

This paper proposes an enhanced ad-hoc multicast routing protocol, which is based on ODMRP (On-Demand Multicast Routing Protocol). We use active probing to enhance the end-to-end routing availability and improve the utilization of link bandwidth. The probing interval is shorter than ODMRP and the reaction to path failure is more flexible. In addition, the probing information also provides the end to end status of the link. Therefore, the overall information can be used by the routing process.