We propose a novel pre-equalizer for multi-level IQ-modulation signals. The pre-equalizer is composed of multiple binary analog finite impulse response (FIR) filters which can realize smaller integrated-circuit (IC) chip size. In the proposed design scheme, required complex-number operations are implemented by a butterfly construction of real-number FIR filters. The analog FIR filters were designed using 28 nm fully depleted silicon on insulator (FD-SOI) based complementary metal-oxide-semiconductor (CMOS) circuits. The performance was investigated using numerical simulation of a 40-Gb/s 16-ary quadrature amplitude modulation (16-QAM) standard single-mode fiber (SSMF) transmission system. The simulated CMOS circuits successfully compensated the 16-QAM signals distorted by chromatic dispersion. The error vector magnitude (EVM) of the transmitted 16-QAM signals was improved from 27% to 12%.

The aim of the study is to explore the possibility of the use of Meteor Burst Communications (MBCs) in equatorial regions. We installed the master and the remote stations at Yogyakarta, Java Island and Jimbaran, Bali Island, Indonesia, respectively. As a preliminary experimental result, we confirmed that some packet transmissions between the two stations were achieved through meteor burst channels.

To implement K-user multi-rate multi-power transmission in non-orthogonal multiple access systems with successive interference cancellation, a repeat accumulator code serially concatenated with a spreading is employed for each user to implement a variable, low-rate coding. A joint rate and power optimization (RPO) is proposed to maximize the sum rate with error free decoding. Numerical results show that our proposed coding-spreading scheme with joint RPO, supporting the multi-rate transmission with the same structure of encoder, approaches the Shannon limit.

Low computational complexity spectrum sensing based on cyclostationarity for multiple receive antennas is proposed. The proposed technique does not calculate certain test statistics at all receive antennas, as opposed to conventional techniques that perform calculations at all receive antennas. Therefore, a low computational complexity can be achieved for sensing. Numerical examples verify that the proposed technique can obtain favorable sensing performance, even with the low computational complexity.