Deep Neural Network (DNN) testing is one of the common practices to guarantee the quality of DNNs. However, DNN testing in general requires a significant amount of test inputs with oracle information (labels), which can be challenging and resource-intensive to obtain. To relieve this problem, we propose DATIS, a distance-aware test input selection approach for DNNs. Specifically, DATIS adopts a two-step approach for selecting test inputs. In the first step, it selects test inputs based on improved uncertainty scores derived from the distances between the test inputs and their nearest neighbor training samples. In the second step, it further eliminates test inputs that may cover the same faults by examining the distances among the selected test inputs. To evaluate DATIS, we conduct extensive experiments on 8 diverse subjects, taking into account different domains of test inputs, varied DNN structures, and diverse types of test inputs. Evaluation results show that DATIS significantly outperforms 15 baseline approaches in both selecting test inputs with high fault-revealing power and guiding the selection of data for DNN enhancement.

Multiple input multiple output (MIMO) relay channels has great application potentials in wireless communication systems and therefore attracts both academia and industrial attentions recently. In this paper, we consider the MIMO relay channels where both source node and relay node are equipped with multiple antennas and the destination node has a single antenna. We address the optimal beamforming design when decode-and-forward half-duplex relaying protocol is deployed. Specifically, we develop an efficient algorithm to determine the optimal beamforming vector at the source when transmitting to relay and destination at the same time. Based on the exact channel capacity formulation, the proposed algorithm can achieve arbitrarily high accuracy with low computational complexity.

Dual-cast transmission appears naturally in network coding-based two-way relay systems, where the relay terminal needs to send network coded information to both end users simultaneously. In this paper, we investigate the optimal beamforming design for wireless dual-cast channels. Specifically, we present an efficient closed-form solution of the optimal beamforming vectors and their achievable maximal dual-cast rate for both orthogonal and nonorthogonal user channel cases. Comparison with conventional non-network-coding based schemes shows that the proposed design of dual-cast beamforming with network coding can best explore the capacity benefit of multiple antennas at the relay terminal.