Spectrogram factorization methods such as Non-Negative Matrix Factorization (NMF) are frequently used as a way to separate individual sound sources from complex sound mixtures. More recently, they have also been used as a first stage for the automatic transcription of polyphonic music. The problem of sound source separation is different (but related) to the problem of automatic music transcription. The output of the first is the separated audio signals corresponding to each sound source, whereas the output of the second is a symbolic representation/music score that encodes the discrete pitches/notes that are played and when they are played. Many variations of factorization methods have been proposed. Two important design choices are the way spectra are represented and what distance measures are used to compare them in the optimization used for factorization. A common assumption has been that a variant that yields better signal separation will result in better automatic transcription. In this work, we investigate experimentally this question and show that this relationship is not necessarily true.

This paper presents a technique to transform high-effort voices into breathy voices using adaptive pre-emphasis linear prediction (APLP). The primary benefit of this technique is that it estimates a spectral emphasis filter that can be used to manipulate the perceived vocal effort. The other benefit of APLP is that it estimates a formant filter that is more consistent across varying voice qualities. This paper describes how constant pre-emphasis linear prediction (LP) estimates a voice source with a constant spectral envelope even though the spectral envelope of the true voice source varies over time. A listening experiment demonstrates how differences in vocal effort and breathiness are audible in the formant filter estimated by constant pre-emphasis LP. APLP is presented as a technique to estimate a spectral emphasis filter that captures the combined influence of the glottal source and the vocal tract upon the spectral envelope of the voice. A final listening experiment demonstrates how APLP can be used to effectively transform high-effort voices into breathy voices. The techniques presented here are relevant to researchers in voice conversion, voice quality, singing, and emotion.

In this paper we propose a novel approach to find a global correspondence between two images by maximizing mutual information in the presence of large scale changes and rotations. Our approach combines the local descriptors with global search. We have tested our method on various test images and compared the matching performance with the SIFT descriptors. The experimental results show that the proposed local descriptor and graph-based search provides robust point matching for scale, rotation and illumination changes

In this paper, a configurable coding scheme is proposed and analyzed with respect to computational complexity and distortion (C-D). The major coding modules are analyzed in terms of computational C-D in the H.263 video coding framework. Based on the analyzed data, operational C-D curves are obtained through an exhaustive search, and the Lagrangian multiplier method. The proposed scheme satisfies the given computational constraint independently of the changing properties of the input video sequence. A technique to adaptively control the optimal encoding mode is also proposed. The performance of the proposed technique is compared with a fixed scheme where parameters are determined by off-line processing. Experimental results demonstrate that the adaptive approach leads to computation reductions of up to 19%, which are obtained with test video sequences and compared to the fixed, while the peak signal-to-noise ratio degradations of the reconstructed video are less than 0.05 dB.

Most standard video coders have high computational complexity in order to achieve high performance in terms of compression ratio. In this paper, a configurable coding scheme is analyzed with respect to the computational complexity and distortion. The major coding modules are analyzed in terms of computational complexity and distortion in the H.263 video coding framework. Based on the analyzed data, operational C-D curve is obtained through an exhaustive search and Lagrangian multiplier method. It is shown that an optimally chosen operational mode makes a significant difference.

We present a general framework to quantify the data throughput capabilities of a wireless communication system when it combines: (1) multiple transmit signals; (2) adaptive modulation for each signal; and (3) adaptive array processing at the receiver. We assume a noise-limited environment, corresponding to either an isolated cell or a multicell system whose out-of-cell interference is small compared with the thermal noise. We focus on the user data throughput, in bits per second/Hertz (bps/Hz), and its average over multipath fading, which we call the user spectral efficiency. First, an analysis method is developed to find the probability distribution and mean value of the spectral efficiency over the user positions and shadow fadings, both as a function of user distance from its serving base station and averaged over the cell coverage area. We assume fading conditions and receiver processing that lend themselves to closed-form analysis. The resulting formulas are simple and straightforward to compute, and they provide a number of valuable insights. Next, we run Monte Carlo simulations, both to confirm the analysis and to treat cases less amenable to simple analysis. A key contribution of this paper is a simple formula for the mean spectral efficiency in terms of the propagation exponent, mean signal-to-noise ratio at the cell boundary, number of antennas, and type of coding. Under typical propagation conditions, the mean spectral efficiency using three transmit and three receive antennas ranges from 19.2 bps/Hz (uncoded) to 26.8 bps/Hz (ideally coded), highlighting the potential benefits of multiple transmissions combined with adaptive techniques. This is much higher than the spectral efficiencies for a link using a single transmitter and a threefold receive diversity under the same conditions, where the range is from 8.77 bps/Hz to 11.4 bps/Hz. Moreover, the latter results are not nearly as practical to achieve, as they can for large signal constellations that would be highly vulnerable to impairments.

Using the MPEG-4 advanced audio coder (AAC) music as an example of streaming applications, we investigate the improvement of error performance for the streaming service by link adaptation and power control techniques in an enhanced general packet radio services (EGPRS) cellular network. A low packet error rate and variability are essential in providing a short error-burst length so that error concealment techniques can be effectively applied to music packets. We study the effects of a combined link adaptation and power control scheme (referred to as the error-based scheme) for achieving a target error rate and reducing error variability. By simulation, we compare the error performance of the error-based scheme at both the EGPRS block and AAC frame level with another adaptation algorithm (referred to as the throughput-based scheme) with a goal of maximizing overall network throughput. It is found that when offered with a similar traffic load, the former scheme can provide noticeable improvement of music quality over the throughput-based scheme. To achieve a similar AAC frame error rate, our results also show that the error-based scheme can increase the link throughput over the throughput-based scheme by 66.7% in one of our examples. These results reveal that by aiming at required error targets and thus reducing error variability, the error-based scheme for link adaptation and power control are helpful in improving quality and capacity for streaming applications.

We investigate the high spectral efficiency capabilities of a cellular data system that combines the following: 1) multiple transmit signals, each using a separately adaptive modulation; 2) adaptive array processing at the receiver; and 3) aggressive frequency reuse (reuse in every cell). We focus on the link capacity between one user and its serving base station, for both uncoded and ideally coded transmissions. System performance is measured in terms of average data throughput, where the average is over user location, shadow fading, and fast fading. We normalize this average by the total bandwidth, call it the mean spectral efficiency, and show why this metric is a useful representation of system capability. We then quantify it, using simulations, to characterize multiple-input multiple-output systems performance for a wide variety of channel conditions and system design options.

We describe a simulation study of a cellular system using multiple-input multiple-output (MIMO) antenna techniques along with adaptive modulation and aggressive frequency reuse. We show, for the case of 3 transmit and 3 receive antennas, how much MIMO systems outperform systems with receive-diversity-only when noise dominates. When co-channel interference from surrounding cells dominates, the differences shrink, as do the absolute numbers. We quantify these reductions for the specific cases studied, and discuss further areas of research.

The capacity of multiple input, multiple output wireless channels is computed for Ricean channels. The novelty is a geometrical (ray-tracing) interpretation of the MIMO channel capacity formula to find array geometries which greatly enhance channel capacity compared to SISO systems.

The use of multiple element antennas-both transmitting and receiving-has great potential for increasing the capacity of wireless systems. The maximum capacity is computed for an "edge-excited" cell with 3 base sites and a subscriber terminal with a 3 element array, for line-of-sight channels. In an attempt to approach this capacity, we show that the use of innovative signal processing at the receiver (blind source separation technique) seems to allow the system to achieve a significant portion of this great capacity promised by information theory.

The capacity of multiple input, multiple output (MIMO) wireless channels is computed for Ricean channels. The novelty is a geometrical (ray-tracing) interpretation of the MIMO channel capacity formula to find array geometries which greatly enhance channel capacity compared to single input-single output (SISO) systems.

The design of a high speed (>150 Mb/s) wireless local area network (WLAN), requires that many factors be considered, including technical, economic, and regulatory. A major technical factor is the channel response behavior (multipath) in the indoor environment as a function of the frequency band, building type and radio system architecture. The consequences of designing indoor wireless systems with directional antennas at one or both ends of a line-of-sight (LOS) link are investigated. We determine how narrow the beamwidth must be so that the maximum data rate is not limited by multipath effects. For such beamwidths, simple unequalized two-level frequency shift keying (FSK) or phase shift keying (PSK) modems can be used in place of the more costly and complex "anti-multipath" modems, and data rates above 1 Gb/s may be achieved. The channel impulse response in an empty room is estimated using geometrical optics, observing that with directional antennas, multipath rays must arrive from the same direction as the LOS ray. The link outage probability is then estimated as a function of the antenna beamwidth, and guidelines are established for the selection of the frequency band and antenna placement. Experiments using a 19-GHz 622-Mb/s binary phase shift keying (BPSK) link and 15/spl deg/ beamwidth horn antennas in an office building with plaster walls and large metallized windows have demonstrated error-free performance on both LOS and non-LOS (NLOS) links.

In this paper, we develop a new multistage detector to approach the optimal solution of the detection problem in a synchronous code division multiple access (CDMA) system. The transformation modifies the diagonal elements of the Hessian matrix of the quadratic likelihood function and brings the continuous minimum of the transformed function as close to the optimal solution as possible. The computational complexity is essentially linear with the number of users, except that a few computations of a quadratic function are needed.

For the CCITT recommendations on frame synchronization (as enhanced in Jones and Al-Subbagh, 1985), the performance in the presence of jamming, interference or other disturbance is improved by using bit erasure information. The acquisition time and misalignment detection times are reduced and the false exit time is increased, with only a minor reduction in the already large false entry time.>

We consider high-speed wireless indoor links in which narrowbeam antennas are used to limit multipath delay spread. For this kind of link, the performances of simple 2- and 4-level modulation schemes are analyzed. The amplitudes and phases of all multipath rays are assumed to be static (no time variation), as expected for fixed links in an indoor environment. We consider coherent PSK with data-directed and pilot tone carrier recovery, differential PSK, and coherent and noncoherent forms of 2-level ASK. Our objective is to find techniques for which factors other than multipath limit the attainable bitrate. Thus for each scheme, analysis is used to identify regions of the multipath parameter set over which data eye closure occurs. Results are given for the 2- and 3-ray multipath channels which can arise in an indoor environment with narrowbeam antennas. Specifically, we assume two parallel walls approximately perpendicular to the direct ray and a narrowbeam antenna at one or both ends of the link. The 2-level schemes are found to be quite robust with respect to 2-ray multipath, and can mostly tolerate a second ray amplitude equal to that of the direct ray. The 4-level schemes are sensitive to 2-ray multipath, showing eye closure at many phases with a second ray power about 7.5 dB below the first (direct) ray. A similar sensitivity is found for 2-level schemes with 3-ray multipath. Finally, 4-level schemes with 3-ray multipath are significantly more sensitive, showing eye closure with a second ray power about 11 db below the direct ray. Thus, considerations of simplicity and robustness favor 2-level PSK schemes (coherent with pilot tone, or differential) for narrowbeam high speed transmission indoors.>

The consequences of designing indoor wireless systems with directional antennas at one or both ends of a line-of-sight link are investigated. The authors determine how narrow the beamwidth must be so that the maximum data rate is not limited by multipath effects. For such beamwidths, simple unequalized 2-level FSK or PSK modems can be used in place of the more costly and complex a 'anti-multipath' modems, and data rates above 1 Gbit/sec may be achieved. The channel impulse response in an empty room is estimated using geometrical optics, observing that with directional antennas, multipath rays must arrive from the same direction as the line of sight (LOS) ray. The link outage probability is then estimated as a function of antenna beamwidth, and guidelines are established for the selection of frequency band and antenna placement. Experiments using a 19-GHz 622-Mb/s BPSK link and 15 degree beamwidth horn antennas in an office building with plaster walls and large metallized windows have demonstrated error-free performance on both LOS and non-LOS links.>

1 GHz complex impulse response data in mountainous terrain is measured at closely spaced locations, and is processed as data from a synthetic aperture array. Experimental data from linear and crossed arrays with 50 or 100 elements is considered. The direction of arrival for each delayed component is identified, and contour plots of the receiver power at bearings and distances are produced. These contour plots closely match the topography of the region, and clearly indicate that the strongest mountain reflections come from the steepest slopes. These results are used to establish a relationship between the mountain reflection coefficients and the topography, thus making it possible to invert the problem and estimate the impulse response (multipath delay profile) in mountainous terrain directly from topographical map data. Such estimates can help to select cell site locations and antenna configurations to minimize the delay spread.>

A digital PLL bit synchronizer with variable loop bandwidth for rapid acquisition and good tracking performance is proposed, and its performance analyzed using Markov chain techniques. Results are presented for the distributions of acquisition time and time to first bit slip in terms of state transition probabilities. For burst mode data, results for the timing error and bit error rate as a function of the preamble bit number are obtained. All results are evaluated by repeated matrix products and verified by simulation. Comparison of the variable bandwidth DPLL to a fixed bandwidth DPLL shows significantly faster acquisition for a given tracking performance.>

A second-order DPLL with time-varying loop gains is applied to the symbol synchronization of burst mode data signals. An algorithm to control the DPLL loop gains is derived from adaptive Kalman filtering theory. Simulation results for the variable gain DPLL compared to a fixed gain DPLL demonstrate the improved acquisition performance.>

Although the optimal detection of signals in a K-user CDMA synchronous system has been proved to be NP-hard, some suboptimal methods do exist to approximate the optimal one. The authors develop a new multistage detector using the transformation proposed by Carter (1984) to approach the optimal solution. The transformation modifies the diagonal elements of the hessian matrix of the quadratic function corresponding to the optimal receiver and brings the continuous minimum of the transformed function as closely to the optimal solution as possible. The convergence of the new multistage algorithms is studied. The computational complexity is essentially linear with the number of users except that a few computations of a quadratic function are needed. The numerical results for the detection of synchronous CDMA signals are shown to be very encouraging.>

The probability of successful frame sync in packet transmission using bit erasure information is compared to the same probability when erasure information is ignored. For a given bound on the false alarms, a significant improvement is found on additive white Gaussian noise (AWGN) channels if an erasure zone is added to the receiver. An improvement is also found for channels in which the presence of jamming or other interference is detected and erased by the receiver. An expression for the probability of frame synchronization in packet transmission is derived as a function of the bit error and erasure probability, the amount of overlap, and the number of bit errors/erasures which are tolerated in the frame sync word.>

A digital phase-locked loop (DPLL) bit synchronizer that tracks the zero crossings of a bandlimited binary signal is discussed. The synchronizer reduces pattern jitter or self noise with a compensation signal in the synchronizer feedback loop without using a prefilter. Analytical results are derived for the timing jitter variance (additive noise and self noise) of the synchronizer. Computer simulations and laboratory measurements are shown to verify the effectiveness of the pattern jitter compensation techinque for a synchronizer operating with both spectral raised cosine signaling pulses as well as for signaling pulses generated by a realizable filter network. Implementation of the pattern jitter compensation method in an adaptive synchronizer structure for applications where a priori knowledge of the signaling pulse shape is not available is also discussed.>