In this paper, we consider domain-adaptive imitation learning with visual observation, where an agent in a target domain learns to perform a task by observing expert demonstrations in a source domain. Domain adaptive imitation learning arises in practical scenarios where a robot, receiving visual sensory data, needs to mimic movements by visually observing other robots from different angles or observing robots of different shapes. To overcome the domain shift in cross-domain imitation learning with visual observation, we propose a novel framework for extracting domain-independent behavioral features from input observations that can be used to train the learner, based on dual feature extraction and image reconstruction. Empirical results demonstrate that our approach outperforms previous algorithms for imitation learning from visual observation with domain shift.

This paper proposes a new sequential model learning architecture to solve partially observable Markov decision problems. Rather than compressing sequential information at every timestep as in conventional recurrent neural network-based methods, the proposed architecture generates a latent variable in each data block with a length of multiple timesteps and passes the most relevant information to the next block for policy optimization. The proposed blockwise sequential model is implemented based on self-attention, making the model capable of detailed sequential learning in partial observable settings. The proposed model builds an additional learning network to efficiently implement gradient estimation by using self-normalized importance sampling, which does not require the complex blockwise input data reconstruction in the model learning. Numerical results show that the proposed method significantly outperforms previous methods in various partially observable environments.

In this paper, we propose a robust imitation learning (IL) framework that improves the robustness of IL when environment dynamics are perturbed. The existing IL framework trained in a single environment can catastrophically fail with perturbations in environment dynamics because it does not capture the situation that underlying environment dynamics can be changed. Our framework effectively deals with environments with varying dynamics by imitating multiple experts in sampled environment dynamics to enhance the robustness in general variations in environment dynamics. In order to robustly imitate the multiple sample experts, we minimize the risk with respect to the Jensen-Shannon divergence between the agent’s policy and each of the sample experts. Numerical results show that our algorithm significantly improves robustness against dynamics perturbations compared to conventional IL baselines.

Using airborne full-waveform LiDAR metrics derived by 3-D tree segmentation, this study estimated single tree's diameter at breast height (DBH) and stem volume (STV). Four regression models were used, including multilinear regression and three up-to-date regression models (i.e., least square boosting trees regression, random forest, and ε-support vector regression) from the machine learning field. This study aimed to comparatively evaluate these regression models in predicting DBH and STV at single-tree level and find some clues to regression model's selection. The study sites were located in the Bavarian Forest National Park, Germany, a mixed temperate mountain forest. Our comparisons were performed across different tree species types (coniferous and deciduous) and foliage conditions (leaf-on/leaf-off seasons). The importance of predictor variables was also examined. Experimental results revealed that the best accuracy from machine learning methods outperformed the multilinear model by 1.5 cm for DBH and 0.18 m3for STV in terms of rmse. Through comparative analysis, our work provided some clues to the performance variation of regression models for extracting 3-D tree parameters.

A new digital predistortion (DPD) technique based on envelope feedback is proposed for linearization of power amplifiers (PAs). Unlike conventional DPD techniques that need frequency down converters (FDCs) in the feedback path to recover the complex envelope of the PA output, the proposed technique does not need an FDC. Instead it employs two envelope detectors, estimating the envelope of the PA output and that of the difference signal between the PA input and the PA output. It is shown that the complex envelope of the PA output can be estimated from the two envelope feedbacks if the sign of the phase distortion of the AM-PM characteristic remains the same for all PA input magnitudes. Simulation results show that the performance of the proposed DPD is comparable to the conventional DPD employing an FDC.

In this study, we developed the Allometric Scaling and Resource Limitations (ASRL) model by using the best GLAS tree heights to optimize the ASRL. At first, we obtained the best metric of GLAS tree heights by comparing with LVIS tree heights in six sites. Then, the best metric GLAS tree heights were separately used to optimize ASRL model and test the accuracy of prediction heights from optimized ASRL model in sites scale and country scale. Validation result showed that predicted tree heights from optimized ASRL model had high accuracy.

A new predistortion technique for dynamic spectrum allocation systems such as cognitive radio (CR) is proposed. The system model considered in this paper occupies a small band at a time, but the center frequency can be changed in the wide range of frequency. In this scenario, the front-end filter may not eliminate the harmonics of the power amplifier (PA) output. The proposed PD reduces the spectral regrowth of the fundamental signal at the carrier frequency (ωο) and removes the harmonics (2ωο, 3ωο, ⋯) at the same time. The proposed PD structure is composed of multiple predistorters (PDs) centered at integer multiples of ωο. The PD at ωοis for removing spectral regrowth of the fundamental signal, and the others are for harmonic reduction. In the proposed PD structure, parameters of PDs are found jointly. Simulation results show that the spectral regrowth can be reduced by 20dB, and the 2nd and 3rd harmonics can be reduced down to -70dB from the power of the fundamental signal.

Abstract In this paper, we evaluate the options for a viable security architecture for the 3GPP Home (evolved) NodeB (H(e)NB) and present the trade‐off between the level of security provided and the associated cost of a H(e)NB device. Low cost of the H(e)NB device is one of the key factors to achieve high market penetration and the major contributor to the cost of the H(e)NB device is the security architecture. It is of interest to analyse the security challenges, security requirements and security mechanisms for (i) H(e)NB authentication, (ii) provisioning of security credentials, (iii) backhaul link protection, (iv) location verification and (v) user access control in the H(e)NB. It is required that the H(e)NB architecture mitigates the threats foreseen by opting a comprehensive security mechanisms amenable to both the operators and the subscribers, while satisfying the acceptable level of security required. Copyright © 2009 John Wiley & Sons, Ltd.

A new partial transmit sequence (PTS) technique for orthogonal frequency division multiplexing (OFDM) signals which does not require side information, is proposed. In the proposed method, the transmitter applies PTS to reduce the peak-to-average power ratio (PAPR) and does not send the side information on the PTS phase factors to the receiver. Based on the OFDM symbol structure having pilot subcarriers, the receiver jointly estimates the channel and PTS phase factors from the pilot subcarriers. The proposed technique inherently causes bit error rate (BER) loss depending on the PTS subgroup size. However, simulation results demonstrate that if the PTS subgroup size is carefully chosen, the proposed method can reduce the PAPR significantly with a negligible BER loss.

In this paper, a new amplify-and-forward (AF) relay systems using superimposed pilot signals is proposed for frequency selective fading channel environments. In the proposed scheme, the relay superimposes an additional pilot sequence on the received signal from the source, and forwards the combined signal to the destination. The superimposed pilot signal in this way enables the destination node to estimate the channel between the relay and destination (R-D), and this channel information is used to process the received signal at the destination optimally; the noise correlation due to the frequency selective fading in the R-D channel is whitened. The optimal power ratio of the superimposed pilot to the incoming signal at the relay is obtained via simulations, and it is shown under frequency selective channel environments that the proposed scheme yields better bit-error- rate (BER) performance than the conventional method where all relay power is allocated only to the incoming signal.

A new polynomial predistortion technique for linearizing nonlinear power amplifier is introduced. The proposed predistorter is based on direct learning, whereas most other techniques are based on the indirect learning architecture. The resulting structure of the proposed method is simpler than the existing ones. Computer simulation shows that the proposed method is robust to the initial conditions and attains comparable performance to other methods based on the indirect learning architecture under normal conditions.