Pisit Boonsrimuang

The capture of a target spacecraft by a chaser is an on-orbit docking operation that requires an accurate, reliable, and robust object recognition algorithm. Vision-based guided spacecraft relative motion during close-proximity maneuvers has been consecutively applied using dynamic modeling as a spacecraft on-orbit service system. This research constructs a vision-based pose estimation model that performs image processing via a deep convolutional neural network. The pose estimation model was constructed by repurposing a modified pretrained GoogLeNet model with the available Unreal Engine 4 rendered dataset of the Soyuz spacecraft. In the implementation, the convolutional neural network learns from the data samples to create correlations between the images and the spacecraft’s six degrees-of-freedom parameters. The experiment has compared an exponential-based loss function and a weighted Euclidean-based loss function. Using the weighted Euclidean-based loss function, the implemented pose estimation model achieved moderately high performance with a position accuracy of 92.53 percent and an error of 1.2 m. The in-attitude prediction accuracy can reach 87.93 percent, and the errors in the three Euler angles do not exceed 7.6 degrees. This research can contribute to spacecraft detection and tracking problems. Although the finished vision-based model is specific to the environment of synthetic dataset, the model could be trained further to address actual docking operations in the future.

The mobile traffic continuously grows at a rapid rate driven by the widespread use of wireless devices. Along with that, the demands for higher data rate and better coverage lead to increase in power consumption and operating cost of network infrastructure. The concept of heterogeneous networks (HetNets) has been proposed as a promising approach to provide higher coverage and capacity for cellular networks. HetNet is an advanced network consisting of multiple kinds of base stations, i.e., macro base station (MBS), and small base station (SBS). The overlay of many SBSs into the MBS coverage can provide higher network capacity and better coverage in cellular networks. However, the dense deployment of SBSs would cause an increase in the power consumption, leading to a decrease in the energy efficiency in downlink cellular networks. Another technique to improve energy efficiency while reducing power consumption in the network is to introduce sleep control for SBSs. This paper proposes cell throughput based sleep control which the cell capacity ratio for the SBSs is employed as decision criteria to put the SBSs into a sleep state. The simulation results for downlink communications demonstrate that the proposed scheme improves the energy efficiency, compared with the conventional scheme.

The major drawback of orthogonal frequency division multiplexing (OFDM) is a large peak-to-average power ratio (PAPR). To overcome the PAPR problem, many PAPR reduction methods have been proposed. Partial transmit sequence (PTS) method is a well known method to reduce the PAPR statistics of an OFDM signal. The PTS method can reduce the PAPR by controlling the multiplied weighting factor of sub-block both data sub carrier and clusters. The PTS method can improve PAPR reduction performance very well as increasing a number of weighting factors and clusters. However, the number of side information is proportionally increasing as increased number of weighting factors or/and clutters, which are required to inform the side information to the receiver for recovering the original data. In this paper, we propose the new weighting factor which can improve the PAPR reduction performance better than conventional PTS method. The proposed method doesn't have any additional number of side information and system complexity. The proposed method can achieve both a better PAPR performance and BER performance with keeping the same number of side information when compare with the conventional PTS method. This paper presents various computer simulation results to verify the effectiveness of proposed method.

In highly mobile environments, signal quality of discrete Fourier transform spreading-orthogonal frequency division multiplexing (DFTS-OFDM) with a frequency domain equalization method would be degraded due to the occurrence of inter-channel interference (ICI). To solve this problem, this paper proposes an iterative based time domain equalization (TDE) method with a time domain channel impulse response (CIR) estimation for DFTS-OFDM signals. The salient features of the proposed method are the use of a time domain training sequence (TS) in the estimation of CIR instead the conventional pilot subcarriers and to employ the TDE method with a maximum likelihood (ML) estimation instead of the conventional one-tap minimum mean square error frequency domain equalization (One-Tap MMSE- FDE) method. This paper also proposes a low complexity iterative based TDE method using the properties of a symmetric banded CIR transfer matrix for solving the simultaneous equations instead of direct calculation of an inverse matrix. It also presents various simulation results in highly mobile environments to demonstrate the effectiveness of the proposed iterative based TDE with CIR estimation for the TS inserted DFTS-OFDM signals compared to the conventional One-Tap MMSE-FDE method.

In highly time-varying fading channel, the Discrete Fourier Transform Spreading Orthogonal Frequency Division Multiplexing (DFTS-OFDM) signal would be damaged significantly by the inter-channel interference (ICI) due to the loss of orthogonality among subcarriers which leads a fatal degradation of bit error rate (BER) performance. To solve this problem, this paper proposes a time domain equalization (TDE) technique in conjunction with a time domain channel impulse response (CIR) estimation method for the DFTS-OFDM signal without using a guard interval (GI). The features of proposed method is to employ a time domain training sequence (TS) both for the estimation of time domain CIR at every sampling time and for removing the inter-symbol interference (ISI) incurred in the multipath fading channel. The proposed method also employs the TDE with a maximum likelihood (ML) estimation method in the demodulation of received time domain signal at every symbol instead of using the conventional Minimum Mean Square Error-Frequency Domain Equalization (MMSE-FDE) method. This paper presents various simulation results to demonstrate the effectiveness of proposed demodulation method for the DFTSOFDM signal without GI as comparing with the conventional MMSE-FDE and TDE methods both of using GI under highly mobile environments.

The future satellite communication systems are required to support the higher transmission data rate for providing multimedia services by employing the efficient modulation method such as multilevel QAM. However, the employment of conventional single carrier transmission method with multi-level QAM which has larger PAPR (Peak to Averaged Power Ratio) would cause the fatal degradation of signal quality due to the non-linear amplifiers located at the earth station and satellite. To overcome this problem, this paper proposes broadband satellite communication systems by using the multilevel QAM-OFDM technique with IDAR (Improved Decision Aided Reconstruction) method, which is designed for satellite channel. In the IDAR method, the characteristics of non-linear amplifiers are required to be known at the receiver for mitigating the intermodulation noise. This paper also proposes the estimation method for AM-AM and AM-PM conversion characteristics of non-linear amplifiers by using low PAPR preamble symbols. The various computer simulations are conducted in this paper to verify the effectiveness of proposed system in the non-linear satellite channel.