Lighting constrained visible-light communications are expected as indoor communications of next generation. In lighting constrained visible-light communications, lighting devices are used not only for illuminating rooms but also for optical wireless communications. For lighting constrained visible-light communications, we have been proposed a variable N-parallel code-shift-keying (VN-CSK) using a modified prime sequence code (MPSC). The VN-CSK system using MPSC has not only a suppression function for reducing co-channel interference from neighboring lighting devices, but also a function for keeping constant data transmission regardless of dimming control. In this paper, the bit error rate (BER) of the VN-CSK system using MPSC is derived under an indoor visible-light communication channel by theoretical analysis. Moreover, we evaluate the BER performance for the brightness level (dimming control stage).

In this paper, the optimum combination of optical pseudo-noise (PN) code and modulation scheme to achieve high total data transmission rate is presented. Moreover, the bit error rate (BER) performance of a wireless OCDMA system using chip-level detection is evaluated through theoretical analysis in the multi-user case. It is shown that, in a wireless OCDMA system with chip-level detection, the total data transmission rate of a multi-pulse pulse position modulation (MPPM) systems with optical pseudo-noise code generated by an M-sequence is better than that of an MPPM system with optical orthogonal code and that of an MPPM system with an extended prime code sequence. Moreover, the total data transmission rate of an MPPM/SIK system using modified pseudo orthogonal M-sequence sets can achieve more than 1.0 [bit/chip].

For high transmission efficiency, good modulation schemes are expected. This paper focuses on the enhancement of the modulation scheme of free space optical turbo coded system. A free space optical turbo coded system using a new signaling scheme called hybrid PPM-OOK signaling (HPOS) is proposed and investigated. The theoretical formula of the bit error rate of the uncoded HPOS system is derived. The effective information rate performances (i.e. channel capacity) of the proposed HPOS turbo coded system are evaluated through computer simulation in free space optical channel, with weak, moderate, strong scintillation. The performance of the proposed HPOS turbo coded system is compared with those of the conventional OOK (On-Off Keying) turbo coded system and BPPM (Binary Pulse Position Modulation) turbo coded system. As results, the proposed HPOS turbo coded system shows the same tolerance capability to background noise and atmospheric turbulence as the conventional BPPM turbo coded system, and it has 1.5 times larger capacity.

In this paper, N-CSK (N parallel Codes Shift Keying) using modified pseudo orthogonal M-sequence sets (MPOMSs) to realize the parallel combinatory spread spectrum (PC/SS) communication system for the optical communications is proposed. Moreover, the upper bound of data transmission rate and the bit error rate (BER) performance of this N-CSK system using the chip-level detection are evaluated through theoretical analysis by taking into account the scintillation, background-noise, avalanche photo-diode (APD) noise, thermal noise, and signal dependence noise. It is shown that the upper bound of data transmission rate of the proposed system is better than those of OOK/CDM and SIK/CDM. Moreover, the upper bound of data transmission rate of the proposed system can achieve about 1.5 [bit/chip] when the code length of MPOMS is 64 [chip].

Probabilistic frequent itemset mining, which discovers frequent itemsets from uncertain data, has attracted much attention due to inherent uncertainty in the real world. Many algorithms have been proposed to tackle this problem, but their performance is not satisfactory because handling uncertainty incurs high processing cost. To accelerate such computation, we utilize GPUs (Graphics Processing Units). Our previous work accelerated an existing algorithm with a single GPU. In this paper, we extend the work to employ multiple GPUs. Proposed methods minimize the amount of data that need to be communicated among GPUs, and achieve load balancing as well. Based on the methods, we also present algorithms on a GPU cluster. Experiments show that the single-node methods realize near-linear speedups, and the methods on a GPU cluster of eight nodes achieve up to a 7.1 times speedup.