In this paper, a design alternative for guaranteeing short-term QoS in the E-PON (Ethernet-Passive Optical Network) OLT (Optical Line Terminal) node is studied. A scheduling algorithm called Short-term QoS Deficit Round Robin (SQ-DRR) is proposed to guarantee tunable deterministic QoS constraints for multimedia applications over E-PON. The major appealing aspect of the scheduler is that it guarantees delay constraint for short-term aggregate burst traffic violating pre-contracted descriptors at the same time without loosing long-term fairness. We then evaluate the scheduler performance with and without admission control scheme under non-stationary long-range dependence (LRD) traffic. The simulation results indicate that the SQ-DRR performs well in dynamic burst traffic conditions.

This paper proposes an efficient two-phase optimization approach for a compact W-band double-plane stepped rectangular waveguide filter design, which combines genetic algorithms (GAs) with the simplified transmission-line model and full-wave analysis. Being more efficient and robust than the gradient-based method, the approach can lead to a compact waveguide filter design. Numerical results show that the resultant waveguide filter design with 4 sections (total length 19.6 mm) is sufficient to meet the design goal and provides comparable performance to that with 8 sections (total length 35.6 mm) by the Chebyshev synthesis approach. Based on the present approach, nineteen compact high-pass waveguide filters have been implemented and measured at the W-band with satisfactory performance.

In recent years, there has been an increased focus on the mechanics of information transmission in spiking neural networks. Especially the Noise Shaping properties of these networks and their similarity to Delta-Sigma Modulators has received a lot of attention. However, very little of the research done in this area has focused on the effect the weights in these networks have on the Noise Shaping properties and on post-processing of the network output signal. This paper concerns itself with the various modes of network operation and beneficial as well as detrimental effects which the systematic generation of network weights can effect. Also, a method for post-processing of the spiking output signal is introduced, bringing the output signal more in line with conventional Delta-Sigma Modulators. Relevancy of this research to industrial application of neural nets as building blocks of oversampled A/D converters is shown. Also, further points of contention are listed, which must be thoroughly researched to add to the above mentioned applicability of spiking neural nets.

In this letter, a new computation method for the noise parameters of a linear noisy two-port network is introduced. A new error function, which considers noise figure and source admittance error simultaneously, is proposed to estimate the four noise parameters. The global optimization of the error function is searched directly by using a genetic algorithm.

This paper describes a method for separating a target sound from other noise arriving in a single direction when the target cannot, therefore, be separated by directivity control. Microphones are arranged in a line toward the sources to form null sensitivity points at given distances from the microphones. The null points exclude non-target sound sources on the basis of weighting coefficients for microphone outputs determined by blind source separation. The separation problem is thereby simplified to instantaneous separation by adjustment of the time-delays for microphone outputs. The system uses a direct (i.e. non-iterative) algorithm for blind separation based on second-order statistics, assuming that all sources are non-stationary signals. Simulations show that the 2-microphone system can separate a target sound with separability of more than 40 dB for the 2-source problem, and 25 dB for the 3-source problem when the other sources are adjacent.

It is well known that the trellises of lattices can be employed to decode efficiently. It was proved in [1] and [2] that if a lattice L has a finite trellis under the coordinate system , then there must exist a basis (b1,b2,,bn) of L such that Wi=span() for 1in. In this letter, we prove this important result in a completely different method, and give an efficient method to compute all bases of this type.

An algorithm for the sound projection using multiple sources is presented. The source strength vector is obtained by using a fast estimation approach instead of the conventional eigenvalue decomposition (EVD) method. The computation load is therefore greatly reduced, which makes the algorithm more efficient in practical applications.

Personal Communication Network (PCN) is an emerging wireless network that promises many new services for the telecommunication industry. The high speed backbone network (ATM or WDM) is one possible approach to provide broadband wireless transmission with PCN's using the ATM switching networks for interconnection of PCN cells. The wireless ATM backbone network design problem is that of allocating backbone links among ATM switches to reduce the effects of terminal mobility on the performance of ATM-based PCN's. In this paper, the wireless ATM backbone network design (WABND) problem is formulated and studied. The goal of the WABND is to minimize the location update cost under constraints. Since WABND is NP-hard, a heuristic algorithm and a genetic algorithm are proposed to solve it. These algorithms are used to find the close-to-optimal solution. Simulated results show that the proposed algorithms are able to achieve good performance.

This paper describes an estimation of the computational and memory complexities of Greengard-Rokhlin's Fast Multipole Algorithm (GRFMA). GRFMA takes a quad tree structure and six calculation processes. We consider a perfect a-ary tree structure and the number of floating-point operations for each calculation process. The estimation for both complexities shows that the perfect quad tree is the best and the perfect binary tree is the worst. When we apply GRFMA to the computation of realistic problems, volume scattering are the best case and surface scattering are the worst case. In the worst case, the computational and memory complexities of GRFMA are O(Llog2 L) and O(Llog L), respectively. The computational complexity of GRFMA is higher than that of the multilevel fast multipole algorithm.

In this paper, the 1-D real-valued discrete Gabor transform (RDGT) proposed in our previous work and its relationship with the complex-valued discrete Gabor transform (CDGT) are briefly reviewed. Block time-recursive RDGT algorithms for the efficient and fast computation of the 1-D RDGT coefficients and for the fast reconstruction of the original signal from the coefficients are then developed in both the critical sampling case and the oversampling case. Unified parallel lattice structures for the implementation of the algorithms are studied. And the computational complexity analysis and comparison show that the proposed algorithms provide a more efficient and faster approach for the computation of the discrete Gabor transforms.

Numerical Stability of the Finite Element/Finite Difference Time Domain Hybrid algorithm is dependent on the hybridization mechanism adopted. A framework is developed to analyze the numerical stability of the hybrid time marching algorithm. First, the global iteration matrix representing the hybrid algorithm following different hybridization schemes is constructed. An analysis of the eigenvalues of this iteration matrix reveals the stability performance of the algorithm. Thus conclusions on the performance with respect to numerical stability of the different schemes can be arrived at. Further, numerical experiments are carried out to verify the conclusions based on the stability analysis.

In many electromagnetic field problems, matrix equations were always deduced from using the method of moment. Among these matrix equations, some of them might require a large amount of computer memory storage which made them unrealistic to be solved on a personal computer. Virtually, these matrices might be too large to be solved efficiently. A fast algorithm based on a Toeplitz matrix solution was developed for solving a bordered Toeplitz matrix equation arising in electromagnetic problems applications. The developed matrix solution method can be applied to solve some electromagnetic problems having very large-scale matrices, which are deduced from the moment method procedure. In this paper, a study of a computationally efficient order-recursive algorithm for solving the linear electromagnetic problems [Z]I = V, where [Z] is a Toeplitz matrix, was presented. Upon the described Toeplitz matrix algorithm, this paper derives an efficient recursive algorithm for solving a bordered Toeplitz matrix with the matrix's major portion in the form of a Toeplitz matrix. This algorithm has remarkable advantages in reducing both the number of arithmetic operations and memory storage.

This paper presents new encoding methods for the binary genetic algorithm (BGA) and new converting methods for the real-coded genetic algorithm (RCGA). These methods are developed for the specific case in which some parameters have to be searched in wide ranges since their actual values are not known. The oversampling effect which occurs at large values in the wide range search are reduced by adjustment of resolutions in mantissa and exponent of real numbers mapped by BGA. Owing to an intrinsic similarity in chromosomal operations, the proposed encoding methods are also applied to RCGA with remapping (converting as named above) from real numbers generated in RCGA. A simple probabilistic analysis and benchmark with two ill-scaled test functions are carried out. System identification of a simple electrical circuit is also undertaken to testify effectiveness of the proposed methods to real world problems. All the optimization results show that the proposed encoding/converting methods are more suitable for problems with ill-scaled parameters or wide parameter ranges for searching.

Routing protocols are a critical component in IP networks. Among these, the Open Shortest Path First (OSPF) has been a widely used routing protocol in IP networks for some years. Beside dedicated hardware, a great interest on routing systems based on open software is raising among Internet Service Providers. Many open source implementations of this protocol have been developed, among which GNU Zebra is one of the most complete. In this paper we perform a study of the performances of the Shortest Path First computation in GNU Zebra, as prescribed by the Internet Engineering Task Force, and we provide a comparison between a Cisco 2621 access router and a PC-based router equipped with routing software GNU Zebra. Moreover we describe a set of modifications made on the GNU Zebra code in order to optimize some processes, whose algorithms were not efficient and whose experimental measures had showed a lack of optimization, thus finally obtaining performances better than the one measured on commercial systems.

MBF (Microwave Beam Forming) antennas are beam forming antennas that perform pattern control in RF, for a low-cost design suitable for mobile terminals. An MBF antenna has only a single output port, since this antenna consists of an array antenna, microwave phase shifters, and a power combiner. Because of this simple configuration, MBF antennas cannot adopt conventional beam forming algorithms that require both phase and amplitude control, and signal observation of each antenna element. In this paper, mutual coupling matrix estimation and null forming methods are presented for MBF antennas. It is shown that the mutual coupling matrix can be estimated by changing the antenna weight instead of signal observation of each antenna element. It is also shown that phase-only null forming, including mutual coupling effect, can be done by the optimum phase perturbations. Numerical and experimental results show the performance of these algorithms.

This paper gives an efficient algorithm to compute addition in Jacobian of C34 curves, aiming at C34 curve cryptosystems. Using C34 curves for cryptosystems has two advantages. The first is safety and the second is the short size of the base field. In the paper, we modify the addition algorithm of for Cab curves in the specific manner to C34 curves. We classify all of the forms of the Groebner bases of ideals involved in the algorithm and eliminate the use of Buchberger algorithm from it. Our resulting algorithm computes the addition in Jacobian of C34 curves in about 3 times amount of computation of the one in elliptic curves, when the sizes of groups are set to be the same.

This paper presents a technique for designing a dielectric Electronically Steerable Parasitic Array Radiator (Espar) antenna to achieve miniaturization of the conventional Espar antenna. The antenna's size is reduced by immersing the central active element in a dielectric cylinder, mounting the surrounding planar parasitic elements at the circumference of the cylinder, and decreasing the radius of the ground skirt to that of the parasitic elements. An example of a polycarbonate (εr = 2.9 + j0.006) Espar antenna operating at 2.484 GHz is optimised by using a genetic algorithm in conjunction with an FEM-based cost function. The designed antenna generates a half-power beam width of 78and a main lobe that elevates at an angle of only 5from the horizontal plane. The designed antenna is also fabricated and measured. Good agreement between the measurement and simulation results is obtained. We reduce the size of the designed Espar antenna to 1/8 the size of its conventional counterpart while achieving a 12improvement in half-power beam width.

The recent sight system requires high stabilization functions for the longer range of observation and the higher kill probability. To this end, it is necessary to compensate rotational disturbances which are not stabilized with the conventional 2-axes stabilization system. This paper proposes a simple method on the rotational motion estimation for the stabilization of the sight system.

A speaker identification system based on wavelet transform (WT) derived from codebook design using fuzzy c-mean algorithm (FCM) is proposed. We have combined FCM and the vector quantization (VQ) algorithm to avoid typical local minima for speaker data compression. Identification accuracies of 94% were achieved for 100 Mandarin speakers.

This paper presents a distributed task assignment algorithm in a logical unidirectional ring, which guarantees that almost all tasks are assigned to servers with the first come first served (FCFS) policy without a global clock. A task assignment for a process is obtained in the time period needed for a message to circle the ring. This time period is almost optimal for a unidirectional ring. The FCFS policy is very important in terms of task fairness and can also avoid starvation and provide an efficient response time. Simulation results show that the algorithm generally works better than conventional task assignment or load balancing schemes with respect to both mean response time and task fairness.