Finding structure in audio for music information retrieval

This paper presents a brief overview of recent approaches to two problems in music information retrieval: query by example and automated source separation. It describes the challenges inherent in musical query by example and the systems that take very different approaches to the problem. The paper also explores approaches to source separation in a musical context, focusing on systems that take distinct approaches to the problem. Finally, it comments about the future directions for digital signal processing (DSP) research in the context of music information retrieval.     

Routing Correlated Data with Fusion Cost in Wireless Sensor Networks

In this paper, we propose a routing algorithm called minimum fusion Steiner tree (MFST) for energy efficient data gathering with aggregation (fusion) in wireless sensor networks. Different from existing schemes, MFST not only optimizes over the data transmission cost, but also incorporates the cost for data fusion, which can be significant for emerging sensor networks with vectorial data and/or security requirements. By employing a randomized algorithm that allows fusion points to be chosen according to the nodes' data amounts, MFST achieves an approximation ratio of 5/4log(k + 1), where k denotes the number of source nodes, to the optimal solution for extremely general system setups, provided that fusion cost and data aggregation are nondecreasing against the total input data. Consequently, in contrast to algorithms that only excel in full or nonaggregation scenarios without considering fusion cost, MFST can thrive in a wide range of applications     

Novel method for simultaneous formation of wires and vias of a printed circuit board using nanoporous body

A new type of a flexible printed circuit board with landless vias is developed using a novel method called interconnection via nanoporous structure (INPS). This method can make wires and vias of the printed circuit board simultaneously by a single photo-exposure process. A new photo-induced selective plating method was used to impregnate a nanoporous substrate with copper, and a new photomask was designed, which constitutes of a completely vacant large hole for via and aggregation patterns of very fine holes for wire. Because of the simple process, the INPS board is characterized by landless vias and very fine circuit. Owing to the structure, it is also characterized by flexibility and detachable wires.     

Reduced-rate retransmissions for spread-spectrum packet radio multimedia networks

A reduced-rate retransmission (RRR) scheme is proposed for improving the throughput performance of spread-spectrum packet radio networks. The scheme takes advantages of the available multi-rate scalable source coding techniques. It assumes that several versions of a data packet with different sizes (number of information bits) are available. The transmission of a packet starts from its full-size version. If the full-size version is not correctly received, its half-size version is used in the retransmission. If further retransmissions are needed, the quarter-size version and so on are used. The shrunk packets are transmitted either in a minislot if the processing gain is kept the same, or occupying a slot duration by increasing the processing gain proportionally. In both cases, the effective signal to interference ratio for a packet is increased. As a result, the system throughput is improved. Theoretical and numerical results are provided in this paper which illustrate the throughput improvement. Another advantage of the proposed RRR scheme is that the packet-size reduction provides finer granules for link adaptation. Therefore, it is especially suitable for multimedia applications for which codes of variable rate for the source data are available and which can tolerate gracefully degraded quality of service. The performance of the proposed scheme in fading channels is also addressed.     

Global stability of Internet congestion controllers with heterogeneous delays

In this paper, we study the problem of designing globally stable, scalable congestion control algorithms for the Internet. Prior work has primarily used linear stability as the criterion for such a design. Global stability has been studied only for single node, single source problems. Here, we obtain conditions for a general topology network accessed by sources with heterogeneous delays. We obtain a sufficient condition for global stability in terms of the increase/decrease parameters of the congestion control algorithm and the price functions used at the links.     

Parametric effects on the transient Voltage performance of a DC superconducting cable

The objective of this work was to determine the significant parameters of a 3-GW 200-kV dc superconducting cable system which influence the transient voltage distribution in the various parts of the cable. The cable system consists of four coaxial metallic cylinders. It was found that the dielectric constant and the electrical resistivity of the soil significantly affect the severity of the transient voltages; lower dielectric constant and higher resistivity of the soil will increase the magnitudes of the transient voltages by increasing the earth-return impedance. It was also found that the effect of the conductor internal impedances of the cable is insignificant. Shorting the coaxial cylinders of the cryogen flow and the cryostat will lessen the severity of the transient voltages. Grounding the second, third, and fourth cylinders at regular intervals with low-impedance grounding impedance will also improve the transient performance of the cable. More research is needed to evaluate these procedures.     

Precision of quantization of the hall conductivity in a finite-size sample: Power law

A microscopic calculation of the conductivity in the integer quantum Hall effect (IQHE) mode is carried out. The precision of quantization is analyzed for finite-size samples. The precision of quantization shows a power-law dependence on the sample size. A new scaling parameter describing this dependence is introduced. It is also demonstrated that the precision of quantization linearly depends on the ratio between the amplitude of the disorder potential and the cyclotron energy. The data obtained are compared with the results of magnetotransport measurements in mesoscopic samples.     

Sigma-delta modulators operating at a limit cycle

A new type of sigma-delta modulator that operates in a special mode named limit-cycle mode (LCM) is proposed. In this mode, most of the SDM building blocks operate at a frequency that is an integer fraction of the applied sampling frequency. That brings several very attractive advantages: a reduction of the required power consumption per converted bandwidth, an immunity to excessive loop delays and to digital-analog converter waveform asymmetry and a higher tolerance to clock imperfections. The LCMs are studied via a graphical application of the describing function theory. A second-order continuous time SDM with 5 MHz conversion bandwidth, 1 GHz sampling frequency and 125 MHz limit-cycle frequency is used as a test case for the evaluation of the performance of the proposed type of modulators. High level and transistor simulations are presented and compared with the traditional SDM designs.     

Design methodologies for high-performance noise-tolerant XOR-XNOR circuits

Scaling down to deep submicrometer (DSM) technology has made noise a metric of equal importance as compared to power, speed, and area. Smaller feature size, lower supply voltage, and higher frequency are some of the characteristics for DSM circuits that make them more vulnerable to noise. New designs and circuit techniques are required in order to achieve robustness in presence of noise. Novel methodologies for designing energy-efficient noise-tolerant exclusive-OR-exclusive- NOR circuits that can operate at low-supply voltages with good signal integrity and driving capability are proposed. The circuits designed, after applying the proposed methodologies, are characterized and compared with previously published circuits for reliability, speed and energy efficiency. To test the driving capability of the proposed circuits, they are embedded in an existing 5-2 compressor design. The average noise threshold energy (ANTE) is used for quantifying the noise immunity of the proposed circuits. Simulation results show that, compared with the best available circuit in literature, the proposed circuits exhibit better noise-immunity, lower power-delay product (PDP) and good driving capability. All of the proposed circuits prove to be faster and successfully work at all ranges of supply voltage starting from 3.3 V down to 0.6 V. The savings in the PDP range from 94% to 21% for the given supply voltage range respectively and the average improvement in the ANTE is 2.67X.     

Protecting local access telecommunications networks: Toward a minimum-cost solution

The problem of designing fibre-optic networks for local-access telecommunications generates (at least) three non-trivial subproblems. In the first of these subproblems one must determine how many fibre-optic cables (fibres) are required at either end of a street. In the next subproblem a minimum-cost network must be designed to support the fibres. The network must also provide distinct paths from either end of the street to the central exchange(s). Finally, the fibre-optic cables must be placed in protective covers. These covers are available in a number of different sizes, allowing some flexibility when covering each section of the network. In this paper we describe a dynamic programming (DP) formulation for finding a minimum-cost (protective) covering for the network (the third of the subproblems). This problem is a generalised set covering problem with side constraints and is further complicated by the introduction of fixed and variable welding costs. The DP formulation selects covers along each arc (in the network), but cannot exactly model the fixed costs and so does not guarantee optimality. We also describe an integer programming (IP) formulation for assessing the quality of the DP solutions. The cost of the networks constructed by the IP model is less than those designed using the DP model, but the saving is not significant for the problems examined (less than 0.1%). This indicates that the DP model will generally give very good solutions. Furthermore, as the problem dimensions grow, DP gives significantly better solution times than IP.