In this paper, a coded convolutional spreading codedivision multiaccess with cyclic prefix (CS-CDMA/CP) system in conjunction with iterative receivers is proposed for synchronous downlink transmission over frequency selective fading channels and its performance is compared with the uncoded CS-CDMA/CP employing maximum-likelihood sequence estimation (MLSE) receiver as well as with the conventional coded direct sequence CDMA (DS-CDMA) employing (maximal ratio combining-) MRC-RAKE. Contrary to the uncoded CSCDMA/ CP system which suffers from performance degradation at low signal-to-noise power ratios (SNRs), it is demonstrated that the proposed scheme can attain almost the full path diversity gain over all SNRs with single-user complexity. Compared with the coded DS-CDMA system with single-user receivers, the proposed system is free from multiuser interference (MUI) whenever the multipath delay of the channel does not exceed the designed maximum delay. It is also demonstrated, moreover, that the proposed system shows a robust performance even if the channel length exceed the designed maximum delay slightly. Because of the subblock structure of CS-CDMA/CP, the proposed system can utilize time-diversity effects, when the channel subjects fast fading, and this greatly improves the resultant performance. Finally, it is shown that the receiver complexity of the proposed system is compared quite favorably with the iterative multiuser receivers for coded DS-CDMA systems of comparable performance.

We propose a method of predicting the number of requests for video titles in an on-demand video delivery system considering their time dependency. To handle many heterogeneous video requests in a video streaming delivery network system using video servers, it is effective to introduce video caching systems in the network system. For the cache system design in video delivery network systems, an effective cache algorithm for predicting how many requests will occur in the future is needed. Since almost all such proposed algorithms are for WWW systems, they do not consider the time dependency of the number of requests, so they are not applicable to a video system with time-dependent requests because a customer in the system watches one video title just one time. In our method, the number of requests is represented by a function of time by considering how a customer obtains information about a video and requests it and by considering that a customer requests each video only once. We have developed a method of calculating the coefficients of the function from the series of the number of requests and for estimating the number of requests in the future. We evaluated the convergence and calculation time. The results show that the calculations converge and the calculation time is 31 ms. They also show that the calculation cost is acceptable for a general video delivery system.

Summary form only given. M. Crochemore et al. introduced a two-pass lossless data compression scheme called data compression using antidictionaries (DCA) (see Proc. IEEE, vol.88, no.11, p.1756-68, 2000). DCA finds some words (minimal forbidden words) that never appear in the text to be compressed, and chooses a subset of minimal forbidden words as an antidictionary (AD). We develop improved DCAs in two ways. (1) We propose an improved DCA using, instead of the AD, predictable dictionaries made up of the words whose last symbols are uniquely determined by their prefixes. The good properties of the original DCA hold in this extension. (2) We present an algorithm to construct the AD (or predictable dictionary) by using a suffix array of the text. We can reduce the amount of memory necessary to construct an AD by using the suffix array. Moreover, by applying a suffix array instead of a suffix tree, we can exploit the relationship between the suffix array and the BWT (Burrows-Wheeler transform), giving us the chance of choosing between DCA and the Burrows-Wheeler compression algorithm (BWCA).

A necessary and sufficient condition for sampling in the general framework of shift invariant spaces is derived. Then this result is applied, respectively, to the regular sampling and the perturbation of regular sampling in shift invariant spaces. A simple necessary and sufficient condition for regular sampling in shift invariant spaces is attained. Furthermore, an improved estimate for the perturbation is derived for the perturbation of regular sampling in shift invariant spaces. The derived estimate is easy to calculate, and shown to be optimal in some shift invariant spaces. The algorithm to calculate the reconstruction frame is also presented.

From the Paley-Wiener 1/4-theorem, the finite energy signal f(t) can be reconstructed from its irregularly sampled values f(k+/spl delta//sub /spl kappa//) if f(t) is band-limited and sup/sub /spl kappa//|/spl delta//sub /spl kappa//|<1/4. We consider the signals in wavelet subspaces and wish to recover the signals from its irregular samples by using scaling functions. Then the method of estimating the upper bound of sup/sub /spl kappa//|/spl delta//sub /spl kappa//| such that the irregularly sampled signals can be recovered is very important. Following the work done by Liu and Walter (see J. Fourier Anal. Appl., vol.2, no.2, p.181-9, 1995), we present an algorithm which can estimate a proper upper bound of sup/sub /spl kappa//|/spl delta//sub /spl kappa//|. Compared to Paley-Wiener 1/4-theorem, this theorem can relax the upper bound for sampling in some wavelet subspaces.