Yue-Ru Chuang

A novel optical IP router architecture for WDM networks

Asynchronous transfer mode (ATM) is employed as the backbone for Internet to connect LAN or WAN. However, the IP packets need to be segmented and packaged due to the differences of format and length between IP packet and ATM cell. Therefore, it will increase the processing time and the complexity of the edge router. In this paper, we propose a novel optical IP router (OIR) architecture and switching concept to fast transmit IP packets in an optical network. The non-TDM based OIR employed genetic algorithms (GAs) with time stamps to support the switching of variable-length IP packets. By using the algorithm, we provide an efficient scheduling technique to cope the variable-length packets in non-TDM based environment.

An optimization solution for packet scheduling: a pipeline-based genetic algorithm accelerator

The dense wavelength division multiplexing (DWDM) technique has been developed to provide a tremendous number of wavelengths/ channels in an optical fiber. In the multi-channel networks, it has been a challenge to effectively schedule a given number of wavelengths and variable-length packets into different wavelengths in order to achieve a maximal network throughput. This optimization process has been considered as difficult as the job scheduling in multiprocessor scenario, which is well known as a NP-hard problem. In current research, a heuristic method, genetic algorithms (GAs), is often employed to obtain the near-optimal solution because of its convergent property. Unfortunately, the convergent speed of conventional GAs cannot meet the speed requirement in high-speed networks. In this paper, we propose a novel hyper-generation GAs (HG-GA) concept to approach the fast convergence. By the HG-GA, a pipelined mechanism can be adopted to speed up the chromosome generating process. Due to the fast convergent property of HG-GA, which becomes possible to provide an efficient scheduler for switching variable-length packets in high-speed and multi-channel optical networks.

A Pipeline-Based Genetic Algorithm Accelerator for Time-Critical Processes in Real-Time Systems

The meta-heuristic methods, genetic algorithms (GAs), are frequently used to obtain optimal solutions for some complicated problems. However, due to the characteristic of natural evolution, the methods slowly converge the derived solutions to an optimal solution and are usually used to solve complicated and offline problems. While, in a real-world scenario, there are some complicated but real-time problems that require being solved within a short response time and have to obtain an optimal or near optimal solution due to performance considerations. Thus, the convergence speed of GAs becomes an important issue when it is applied to solve time-critical optimization problems. To address this, this paper presents a novel method, named hyper-generation GA (HG-GA), to improve the convergence speed of GAs. The proposed HG-GA breaks the general rule of generation-based evolution and uses a pipeline operation to accelerate the convergence speed of obtaining an optimal solution. Based on an example of a time-critical scheduling process in an optical network, both analysis and simulation results show that the HG-GA can generate more and better chromosomes than general GAs within the same evolutionary period. The rapid convergence property of the HG-GA increases its potential to solve many complicated problems in real-time systems

An ACK-based polling strategy for the IEEE 802.11 wireless networks

In IEEE 802.11 infrastructure wireless local area networks (IWLANs), most of data frames are transmitted between access point (AP) and wireless stations (STAs). The data transmission in IWLAN is inefficiency since the excessive overhead generated from frequent handshakes between AP and STAs and contentions among STAs. For minimizing the bandwidth wastage, in this paper, we propose a new ACK-based polling strategy to efficiently coordinate and arbitrate the sequence of transmissions among STAs. The proposed strategy alters the consecutive handshake sequence between sender and receiver defined in IEEE 802.11 medium access control (MAC) protocol to minimize the handshaking overhead and to reduce or even eliminate the potential collisions in transmissions. Different from standard handshakes, the recipient of data frame postpones sending acknowledgement (ACK) control frame to schedule the timing of the next transmission from sender. The ACK frame is sent from receiver via standard contention protocol and the sender waits the ACK frame for the authorization to send the next data frame. The simulations and analytical results demonstrate that the proposed ACK-based contention strategy is able to improve the transmission efficiency by simply reducing the overheads of handshakes and collisions in the wireless networks.

An M-time inheriting transmission strategy for interactive multimedia applications in wireless networks

With the development of IEEE 802.11 WLAN techniques, the wireless network bandwidth has gradually become sufficient for many multimedia applications. However, the standard MAC protocol is originally designed for traditional data transmission fashions, which are asymmetric transmissions, and seems to be not an efficient transmission fashion for recently oncoming multimedia applications. A sort of interactive multimedia applications, such as video conference, interactive Internet game, and network phone (VoIP), etc., is more and more popular in the world. The features of these interactive applications are frequent and bi-directional data exchange, which will cause large and unnecessary transmission overheads for channel contention under the standard MAC protocol. To reduce these overheads and improve system transmission efficiency, an M-time inheriting transmission strategy (MITS) is proposed in this paper. The MITS is a dynamically bi-directional transmission strategy, which uses the token-like fashion to allow the receiver to send its data frame immediately without participating in following channel contention. Simulation results show that the MITS can indeed improve system goodput of the standard MAC protocol in the environments with interactive applications.

An ACK-based polling strategy for supporting high performance and QoS in IEEE 802.11 wireless networks

The infrastructure architecture of wireless local area networks (WLANs) has been widely established in many environments to provide convenient multimedia services. However, in this standard operation, arbitrary channel contention and frequent handshaking significantly affect data transmission efficiency between AP and stations (STAs). This paper proposes an ACK-based polling strategy (APS) to reduce the overheads of channel contention and frequent handshaking via adaptively arbitrating and scheduling the transmission sequence of STAs. That is, the proposed APS makes AP be able to defer the ACK frames replied to the STAs, which still have more data in their buffers, in order to temporarily terminate their subsequent contention accesses. A terminated STA is permitted to transmit data frame again only when it receives the ACK frame replied from AP. Using the ACK-based polling mechanism, the overheads of channel contention and frequent handshaking are reduced and the network goodput is improved. Using the ACK frames, the APS can be further enhanced to support the quality of service (QoS) for various multimedia applications. Simulation results demonstrate that the APS with enhanced QoS function is able to efficiently cope with various transmission requirements in multimedia WLANs.

Synchronous wavelength reusing/sharing protocols for nonregenerative WDM dual buses

The dual-bus optical network with nonregenerative stations has become popular. In general, the nonregenerative station implementation is better than the regenerative station implementation in media access control delay, cost, and susceptibility of station failures, etc. In this paper, two novel cycle-based protocols, named as synchronous wavelength sharing protocol (SWSP) and synchronous wavelength reusing/sharing protocol (SWRP), are proposed for nonregenerative wavelength-division-multiplexing dual-bus networks. The simulation results show that the network throughput and delay of SWRP are better than those of SWSP, when the numbers of wavelengths and tunable receivers are both limited, and/or the network traffic load is heavy.

Cells splitting/merging strategy to reduce call blocking probability in ATM networks

In ATM networks, call admission control (CAC) is employed during the call setup phase to determine whether a virtual channel or virtual path connection request can be accepted or rejected. Most of CACs will reject an incoming call if no path can be found with enough bandwidth from source to destination. However, at the moment of rejecting a call, the total remaining network bandwidth may be large enough to service it. In this paper, we define the multiple path selection problem in ATM networks and propose an algorithm to find the minimum number of paths. Based on the found paths, two cell splitting/merging strategies are introduced. The splitting/merging operation is based on the sequence number within each cell. To guarantee that the cell stream sequence received at the destination is the same as that delivered at the source, the maximum number of consecutive cells that can be dropped on a split path is analyzed. Moreover, a simple buffer replacement scheme is proposed to improve the tolerance of burst cell loss.