The transmission capacity of single-mode single-core fibers is approaching the limit of about 100 Tbps owing to nonlinearity and the fiber fuse, which limit the input signal power. To avoid the so-called capacity crunch and dramatically increase the transmission capacity, a new approach utilizing space-division multiplexing (SDM) and mode-division multiplexing (MDM) has been proposed and demonstrated. In these new multiplexing technologies, a problem that is unprecedented in conventional single-mode fibers, namely, the crosstalk between transmission channels, arises. Although the input signal channels are mixed through at the output end, the signal channel can be identified using multiple-input multiple-output (MIMO) signal processing technology at the output end. However, even using the MIMO digital signal processing technology, the problem that the computation time increases with the number of channels remains, resulting in increased latency, and so the number of channels is limited. To design a transmission system using the SDM and MDM technologies, the crosstalk should be analyzed precisely. However, the origin and behavior of crosstalk are different in SDM using multicore fibers (MCFs) and in MDM using few-mode fibers (FMFs). The behavior of crosstalk in single-mode MCFs is predictable to some extent by statistical analysis, and the system can be designed by considering the results of the analysis. On the other hand, the behavior of crosstalk of FMFs is less predictable. Since the mode launched at the input end is not the eigenmode, mode discrimination or accurate mode demultiplexing is difficult using a conventional mode demultiplexer. In addition, the eigenmode itself of FMFs is not always the hybrid mode predicted by the conventional theory but sometimes a linearly polarized (LP) mode, contradicting the conventional theory. In other words, the demultiplexed signal always involves crosstalk regardless of the transmission distance, and the quantity of crosstalk cannot be analyzed statistically. Therefore, the crosstalk and its behavior are unpredictable. In spite of these unpredictable phenomena, the signal channel can be identified using the MIMO signal processing as in single-mode MCFs. This means that the MDM technology using FMFs is established in the engineering or inductive logic sense, but still involves unexplained phenomena in the scientific or deductive logic sense. In this review, we discuss the predictable behavior of crosstalk in single-mode MCFs and also the unpredictable behavior of crosstalk in FMFs.
This letter presents a novel approximate adder that reduces energy and power consumption by leveraging a simplified lower-part approximation. The proposed scheme reduces hardware costs while providing an acceptable accuracy performance. Implemented in a 32-nm CMOS technology, the proposed adder achieves area and power reductions of 67% and 91%, respectively, compare to a conventional adder. In terms of energy, it improves the power-delay and energy-delay products by 13.1% and 17.0%, respectively, compared to the other approximate adders considered herein. In addition, when adopted in a digital image processing application, the proposed adder shows a very promising output quality compared to that produced by an exact adder while providing excellent energy efficiency.
To alleviate the danger of foreign object debris on runways, Frequency Modulated Continuous Wave linear cell radar based detection systems have been evaluated in the literature. This radar uses a 90 GHz millimeter wave and radio-over-fiber (RoF) to achieve high detection accuracy and low cost. This paper reports the experimental results of interference mitigation experiments at Kuala Lumpur International Airport. False images formed due to interference can be suppressed by performing simple signal processing on more than one radar image in which the transmission delay in RoF links is discrepant.