Suppose there are two correlated sources, one producing classical data and the other producing mixed states with commuting density operators. We combine the two sources into one and call it a classical-quantum source. In this paper, we consider the problem of distributed compression of the classical-quantum source. This is a special case of the classical-quantum Slepian-Wolf problem addressed by Devetak and Winter [1]. Suppose Alice (resp. Bob) receives classical (resp. quantum) data from the classical-quantum source. Alice and Bob, who cannot communicate with each other, want to send their received data to Charlie at the minimum communication cost. We consider two scenarios, the blind and visible scenarios, for Bob's quantum data compression. We formulate our problem in terms of classical information theory and then using some results in classical information theory, we show partial results on the achievable rate region for asymptotically lossless distributed compression of the classical-quantum source in the blind and visible scenarios.

The wiretap channel with causal/noncausal side information at the transmitter is an extension of the original wiretap channel of Wyner to the multi-user case and has been investigated by several authors. Although much effort has been made to establish the secrecy capacity, it is still elusive even if the wiretap channel is physically degraded. In this paper, for a physically degraded wiretap channel, we show a lower bound on the secrecy capacity for the causal case and an upper bound for the noncausal case. We also show that under a certain condition, the secrecy capacity for the causal case is the same as that for the noncausal case. To illustrate the usefulness of our theoretical result, we introduce the concept of a symmetric wiretap channel with state, for which the above condition is easily checked, and we give some examples in which the secrecy capacity can be computed numerically.

A visual cryptography scheme (VCS) is an encryption method for images that does not need a computer to decode a secret image. In this paper we propose a simple, progressive VCS which is constructed from a traditional threshold VCS. Our progressive VCS has better decoded image quality and flexibility than the Fang-Lin scheme, a previous progressive VCS. We also present a block-wise progressive VCS which has jigsaw puzzle like decryption, and an XOR-based progressive VCS without pixel expansion. We give some experimental results which show that our scheme is superior to the Fang-Lin scheme with respect to the decoded image quality and the decoding speed controllability.

Wyner has shown in his seminal paper on (discrete memoryless) wiretap channels that if the channel between the sender and an eavesdropper is a degraded version of the channel between the sender and the legitimate receiver, then the sender can reliably and securely transmit a message to the receiver, while the eavesdropper obtains absolutely no information about the message. Later, Leung-Yan-Cheong and Hellman extended Wyner's result to the case where the noise is white Gaussian. In this paper we extend the white Gaussian wiretap channel to the colored Gaussian case and show the finite block length secrecy capacity of colored Gaussian wiretap channels. We also show an asymptotic lower bound on the secrecy capacity of a specific colored Gaussian wiretap channel for which optimum power allocation can be found by water filling.

In this paper, we present a new class of low-density MDS array codes for tolerating double disk failures in disk arrays. The proposed MDS array code has lower encoding and decoding complexity than the EVENODD code of Blaum et al

In this paper we present a class of low-density MDS array codes that have sparsest parity-check matrices and fast encoding and erasure decoding algorithms. Proposed codes are a modification of the MDS array codes of Blaum et al. in 1996 and can be encoded and decoded by XOR operations with lower complexity than the MDS array codes of Feng et al. in 2005

In this paper we propose an algebraic construction of efficiently encodable irregular LDPC codes. The proposed irregular LDPC codes have not only an efficient encoding algorithm but also guaranteed minimum distances. Simulation results show that the proposed codes perform well compared to randomly constructed irregular LDPC codes

Low-density parity-check (LDPC) codes are one of the most promising next generation error correcting codes and many investigations shows that LDPC codes suitable for many hardware implementation. Although randomly constructed LDPC codes are usually encoded by using generator matrix, this method requires quadratic time complexity and is not easy to implement. This work presents the encoding of array-type LDPC codes and a special class of Sridhara-Fuja-Tanner (SFT) codes by division circuits as cyclic codes, which are very easy to implement.