Chi Chung Ko

Perspective rectification of document images using fuzzy set and morphological operations

In this paper, we deal with the problem of document image rectification from image captured by digital cameras. The improvement on the resolution of digital camera sensors has brought more and more applications for non-contact text capture. Unfortunately, perspective distortion in the resulting image makes it hard to properly identify the contents of the captured text using traditional optical character recognition (OCR) systems. We propose in this work a new technique, which is capable of removing perspective distortion and recovering the fronto-parallel view of text with a single image. Different from reported approaches in the literature, the image rectification is carried out using character stroke boundaries and tip points (SBTP), which are extracted from character strokes based on multiple fuzzy sets and morphological operators. The algorithm needs neither high-contrast document boundary (HDB) nor paragraph formatting (PF) information. Experimental results show that our rectification process is fast and robust.

A fine granular scalable to lossless audio coder

This paper presents Advanced Audio Zip (AAZ), a fine grained scalable to lossless (SLS) audio coder that has recently been adopted as the reference model for MPEG-4 audio SLS work. AAZ integrates the functionalities of high-compression perceptual audio coding, fine granular scalable audio coding, and lossless audio coding in a single framework, and simultaneously provides backward compatibility to MPEG-4 Advanced Audio Coding (AAC). AAZ provides the fine granular bit-rate scalability from lossy to lossless coding, and such a scalability is achieved in a perceptually meaningful way, i.e., better perceptual quality at higher bit-rates. Despite its abundant functionalities, AAZ only introduces negligible overhead in terms of lossless compression performance compared with a nonscalable, lossless only audio coder. As a result, AAZ provides a universal yet efficient solution for digital audio applications such as audio archiving, network audio streaming, portable audio playing, and music downloading which were previously catered for by several different audio coding technologies, and eliminates the need for any transcoding system to facilitate sharing of digital audio contents across these application domains

Divisible load scheduling on single-level tree networks with buffer constraints

Scheduling a divisible load on a heterogeneous single-level tree network with processors having finite-size buffers is addressed. We first present the closed-form solutions for the case when the available buffer size at each site is assumed to be infinite. Then we analyze the case when these buffer sizes are of finite size. For the first time in the domain of DLT (divisible load theory) literature, the problem of scheduling with finite-size buffers is addressed. For this case, we present a novel algorithm, referred to as incremental balancing strategy, to obtain an optimal load distribution. Algorithm IBS adopts a strategy to feed the divisible load in a step-by-step incremental balancing fashion by taking advantage of the available closed-form solutions of the optimal scheduling for the case without buffer size constraints. Based on the rigorous mathematical analysis, a number of interesting and useful properties exhibited by the algorithm are proven. We present a very useful discussion on the implications of this problem on the effect of sequencing discussed in the literature. Also, the impact of Rule A, a rule that obtains a reduced optimal network to achieve optimal processing time by eliminating a redundant set of processor-link pairs, is also discussed. Numerical examples are presented.

Efficient partitioning and scheduling of computer vision and image processing data on bus networks using divisible load analysis

Abstract We investigate the data partitioning, distribution, and scheduling problem for minimizing the total processing time of computer vision and image processing (CVIP) data on bus networks. Using the recently evolved divisible load paradigm (DLT) [V. Bharadwaj, D. Ghose, V. Mani, T.G. Robertazzi, Scheduling divisible loads in parallel and distributed systems, IEEE Computer Society Press, Los Almitos, California, 1996] for processing loads that are computationally intensive, we design and analyze a scheduler that optimally partitions the CVIP data and assigns it to the processors in the network in such a way that the total processing time is a minimum. In addition to the transmission delay in the network, we consider all the overhead components that penalize the time performance in the problem formulation. With this formulation, we derive closed-form solutions for the optimal processing time when the CVIP data distribution follows a fixed sequence. We then derive a necessary and sufficient condition for the existence of an optimal processing time. We then prove an optimal sequence theorem that identifies a sequence that gives rise to an optimal processing time among all possible load distribution sequences, whenever such a sequence of load distribution exists. The performance of the strategy proposed is also analyzed with respect to speed-up and processor utilization or efficiency metrics. Several illustrative examples are shown for the ease of understanding.

A new scheme for PAPR reduction in OFDM systems with ICI self-cancellation

Frequency offset in orthogonal frequency division multiplexing (OFDM) systems leads to loss of orthogonality among subcarriers, so that intercarrier interference (ICI) occurs. A promising ICI self-cancellation scheme based on a data allocation of (X/sub k/, X/sub k+1/=X/sub k/) has been proposed to deal with this problem. However, this scheme degrades the performance in peak-to-average power ratio (PAPR). In this paper, to lessen this impairment, a simple data allocation of (X/sub k/, X/sub k+1/=X/sub k/*) is proposed to control both ICI and PAPR at the same time. As a result, the PAPR of an OFDM with the presented scheme is reduced by half when compared with that of ICI self-cancellation scheme. Furthermore, the average carrier-to-interference ratio (CIR) of the presented scheme is 12 dB higher than that of a normal OFDM system.

Decentralized bit, subcarrier and power allocation with interference avoidance in multicell OFDMA systems using game theoretic approach

This paper addresses the adaptive allocation of subcarrier, bit and power resources in the downlink of multicell OFDMA systems, where base stations allocate radio resources in a decentralized manner. Our algorithm uses discrete bit-loading values which is more applicable to the practical systems. Noncooperative game theoretic approach is adopted and we first only include the transmit power and throughput in the payoff function of each base station. Using the optimal solutions as the benchmark, the performance of this noncooperative game is obtained through computer simulations. Our simulation results show that in the multicell OFDMA systems with frequency reuse factor of one, such payoff function provides inadequate utilization efficiency due to the presence of high co-channel interference. We next incorporate the number of allocated subcarriers in the payoff function of each base station. We show that the new payoff function avoid neighboring base station from transmitting using the same subcarrier when interference is large. This resembles the use of interference avoidance technique in the algorithm, and can achieve better resource utilization when compared to the water-filling counterpart.

A New Blind Joint Timing and Frequency Offset Estimator for OFDM Systems Over Multipath Fading Channels

A blind joint timing and frequency synchronization algorithm is derived for orthogonal frequency-division multiplexing (OFDM) signals transmitting over multipath fading channels. The proposed estimator requires neither the knowledge of the signal-to-noise ratio (SNR) nor the power delay profile of the fading channels, except that the maximum delay spread L should be known and should be shorter than the guard interval. In the proposed estimator, the transmitted time-domain OFDM samples are modeled as an independent Gaussian random sequence, and the received time-domain OFDM samples are partitioned into a few subsets in such a way that the neighboring entries in each subset are uncorrelated. This is achieved by picking out OFDM samples that are spaced L samples apart into each subset. By modeling the channel gains as unknown deterministic variables, a joint symbol timing and frequency offset (FO) estimation algorithm is derived based on these subsets. Simulation results show that, in terms of lock-in probability, the proposed blind ST estimator achieves better performance than the estimators studied by Van De Beek and Speth , as well as the approximate maximum-likelihood estimator described by Lee and Cheun when the SNR is sufficiently high. For FO estimation, in terms of mean square error (MSE), the proposed estimator outperforms the estimators studied by Van De Beek and Speth at medium and high SNR. In terms of bit error rate (BER), the proposed estimator is superior to the estimators studied by Van De Beek and Speth Furthermore, the simulation results show that the performance of the proposed estimator is tolerant to the variation in the number of delay paths. This means that the proposed estimator will still adequately perform, even when the number of delay paths is time varying and is not estimated at the receiver. Finally, simulation results show that the use of virtual subcarriers has negligible performance loss in the proposed estimator.

Multiple-Access Interference Suppression for Interleaved OFDMA System Uplink

In this paper, the issue of multiple-access interference (MAI) suppression for the uplink in an interleaved orthogonal frequency-division multiple-access (OFDMA) system is investigated. In such a system, a carrier frequency offset (CFO) disrupts the orthogonality between the subcarriers and gives rise to MAI among users. Based on the signature vector formulated for each user, we propose a novel detector that performs MAI suppression before CFO compensation and fast Fourier transform (FFT) demodulation. Subspace zero-forcing and minimum mean square error (MMSE) techniques are then developed to suppress MAI. The proposed scheme is shown to become almost MAI free, provided that the CFO estimation is accurate enough. From the obtained simulation results, the proposed scheme is also found to be able to enhance the system performance at low complexity.

On the Modeling of a Non-Cooperative Multicell OFDMA Resource Allocation Game with Integer Bit-Loading

We address the adaptive allocation of subcarrier, bit and power resources among the base stations (BSs) of a downlink multi-cell OFDMA system using the non-cooperative game theoretic approach. Unlike the information theoretic approach where continuous bit-loading is used, our algorithm allocates integer number of bits to each subcarrier. Using a utility function with linear pricing on both the transmit bit rate and power, we first show that for a simple two-cell system with single user and fixed modulation, a Nash equilibrium (NE) solution always exists. However, for games formulated on more complicated systems such as with more than two cells, with more than one user in a cell, or when adaptive modulation is used, we observed that the existence of NE cannot be guaranteed. If NE points exist, the game can be played repeatedly with each BS (the player) selects its best strategy until the solution converges, or known as the myopic game. Based on the framework of potential games with coupled constraints, we propose an algorithm which can achieve a stable solution among the players despite that NE solution to the originally problem does not necessarily exist. The convergence of the proposed algorithm is verified by simulation.

A Linear Programming Solution to the Subcarrier-and-Bit Allocation of Multiclass Multiuser OFDM Systems

Adaptive subcarrier-and-bit allocation in orthogonal frequency division multiplexing (OFDM) systems can significantly improve spectral efficiency and reduce power consumption. The combinatorial optimization on the power, bit and subcarrier allocation in multiclass multiuser OFDM systems normally involved nonlinear objective functions and constraints, resulting in highly complex and computation intensive optimization problems. In this paper, we propose a method to linearize and simplify the optimization problem formulated on subcarrier-and-bit allocation in multiclass multiuser OFDM systems. With a drastically reduced complexity, the converted linear optimization problem can be solved in close to real time and the solutions found are optimal, as no relaxation is made during the linearization. Different approximate relationships between symbol error rate (SER) and bit error rate (BER) are also used to investigate the calculation on the required transmit power.