Cheng Shiduan

Effective fairness queuing algorithms

In this paper we present two effective fair queuing algorithms (EFQs), which achieves fairness properties and delay properties similar to WFQ (which is currently the better scheduling algorithm among others), and uses a system potential function of O(l) complexity. In EFQs, computation of system potential function does not require any division or multiplication operations such as in minimum delay-self clocked fair queuing (MD-SCFQ). Compared to MD-SCFQ, EFQs have lower complexity and can be easily implemented on chips. We verify the effectiveness of the proposed EFQs through theoretical analysis and give some simulation results.

A PHB reservation mechanism in DiffServ network

A new DiffServ per-hop behavior (PHB) reservation protocol (DPRP) is proposed to support the end-to-end QoS guarantee in a DiffServ network through several domains. It is a reservation mechanism that inherits the advantages of RSVP but is a lightweight version of RSVP. DPRP transports and negotiates the QoS requirement between sources and destinations as well as reserves PHB for service levels in each DS domain. It negotiates a dynamic service level agreement (SLA) between various domains efficiently and guarantees the end-to-end QoS of the DiffServ network. Essential principles and advantages of DPRP are presented.

PDB-based SLS decomposition in heterogeneous IP network

In order to provide QoS-assured service for end users, service providers (SP) must obtain sufficient network resource from network providers (NP). However, current network infrastructures belong to different NPs and are divided into a number of domains. SP may need to rely on more than one NP to get the connectivity. Thus the SLA between SP and NPs may span several domains with heterogeneous QoS technologies. Without knowing the network details, such as topology, infrastructure and QoS technology, SP is difficult to sign a proper SLA with a tradeoff between end-to-end QoS and his budget. A SLS (service level specification) decomposition service is proposed here to solve the problem. The service implementation uses distributed PDB information to conceal the network heterogeneity from SP, which acts as an indication of network performance for a certain QoS level packets. The service enables SP to get an optimized SLA and achieve inter-domain QoS without knowing the network details. A SLS Decomposition Service prototype is implemented and its feasibility is validated.

Protocol conformance test suite generation

The strategy of generating test suite from protocols modelled EFSM is presented. The concept of transition/program dependency graph (TPDG) constructed from data flow of EFSM is presented. From the TPDG and the control flow graph (CFG) of EFSM, test sequences that can test both the control part and the data part can be generated. Test data (for parameterizing test sequences) selection method is discussed too. The generated test cases (parameterized test sequences) are described in TTCN according to ISO/IEC 9646. A software-test suite generation system (TSGS) based on the above strategy is introduced. The software is implemented in a Sun sparc station.

Analyzing the delay performance of server-based and receiver-based local recovery approaches for reliable multicast

Local recovery approaches for reliable multicast can be classified into a server-based and receiver-based approaches. This paper presents a delay analysis of server-based (L1) and receiver-based (L2) local recovery approaches for reliable multicast. We focus on the average delay between the initial arrival of a packet at the sender and its correct reception at a randomly selected receiver. Our analysis shows the following characteristics. (1) Except for extremely low loss, L2 usually becomes saturated under lower arrival rates than L1. (2) When the tail links per repair server increases, the saturated value of L1 decreases. However, L1 is still better than L2 even if the tail links per repair server comes to 200. (3) The delay performance of L1 will hardly change if we increase the number of tail links and keep k constant. The delay performance of L1 is more scalable than that of L2.

Differentiated service fast-TCP policy for flow control and resource management

This paper presents an enhancement of the flow control mechanism of traditional TCP, which is called a differentiated service fast-TCP policy. The basic idea of this policy is delaying ACK of differentiated services from the destination towards the source to different extents. It can not only shape the output rate of the sources and palliate congestion in the routers, but can also manage the resource being competed for among all the services, such as link bandwidth and router buffer. Simulation and results comparison show the effect of this proposed policy in a simple network.

Border collision bifurcation and chaotic control of discrete feedback transmission control protocol-random early detection system

Various types of border collision bifurcations occurring in a discrete feedback transmission control protocol-random early detection （TCP-RED） system are analyzed based on normal form method. As the system parameters change, the system experiences the border collision and a sudden jump from a periodic attractor to a chaotic attractor leading to the intense vibration of the system and serious degradation of its performance. According to the linear stability condition of fixed point, a method of chaotic control is presented. With this method, the system can be stabilized to the fixed point by perturbing the state variable in the neighborhood of the fixed point. This method can significantly improve the performance and the resource utilization of TCP-RED system, and simulation results show that it is effective.

A new synchronous digital hierarchical Bi-directional ring capacity design algorithm based on the shortest-route traffic matrix

In this paper, the concept of the Shortest-Route Traffic Matrix(SRTM) was first presented, and the generalized formula for computing ring capacity requirement in use of SRTM is given. Then, a new capacity design algorithm which is based on SRTM was presented for Synchronous Digital Hierarchical(SDH) Bi-directional Self-Healing Ring (BSHR). The algorithm simulation results demonstrate that this algorithm is very efficient for SDH BSHR capacity design and can make less project investment and make high utilization of lines and equipment. By means of the algorithm in this paper, capacity optimization assignment for SDH Hierarchical Self-Healing Ring (HSHR) and for ATM Virtual Path (VP)-based Self-Healing Ring (SHR) is also discussed.

Implementation of an ATM network and PSTN/N-ISDN interworking gateway

Broadband networks based on asynchronous transfer mode (ATM) have already been in commercial operation in some countries. The ATM has the capability of integrating various types of services into a single network. However, the traditional narrowband services still accounts for the major portion of the total telecommunication services. So it is necessary to implement the interworking function between the ATM network and PSTN/N-ISDN. We present a narrowband-broadband network interworking gateway (NB-IWG), in which we exploited the dynamic composite cell mechanism to adapt the voice traffic into ATM cells and implemented the signalling interworking between the narrowband and broadband network.

Design and implementation of an expert system on man-machine interface for the SPC switching system

It is very important to improve man-machine interface of computer system for the people. We know that the daily maintenance or management of SPC switching system requires many man-machine communications, but now many kinds of man-machine communications of SPC switching system are neither very easy nor very perceptive directly for the operators. This paper presents a user-friendly intelligent man-machine communication tool for SPC switching systems.<<ETX>>