Even the artificial intelligence (AI) has been widely used and significantly changed our life, deploying the large AI models on resource limited edge devices directly is not appropriate. Thus, the model split inference is proposed to improve the performance of edge intelligence (EI), in which the AI model is divided into different sub-models and the resource-intensive sub-model is offloaded to edge server wirelessly for reducing resource requirements and inference latency. Unfortunately, with the sharp increasing of edge devices, the shortage of spectrum resource in edge network becomes seriously in recent years, which limits the performance improvement of EI. Refer to the NOMA-based edge computing (EC), integrating non-orthogonal multiple access (NOMA) technology with split inference in EI is attractive. However, the NOMA-based communication aspect and the influence of intermediate data transmission fail to be considered properly in model split inference of EI in previous works, and the sophistication in resource allocation caused by NOMA scheme makes it further complicated. Thus, the Effective Communication and Computing resource allocation algorithm is proposed in this paper for accelerating the split inference in NOMA-based EI, shorted as ECC. Specifically, the ECC takes the energy consumption and the inference latency into account to find the optimal model split strategy and resource allocation strategy (subchannel, transmission power, computing resource). Since the minimum inference delay and energy consumption cannot be satisfied simultaneously, the gradient descent (GD) based algorithm is adopted to find the optimal tradeoff between them. Moreover, the loop iteration GD approach (Li-GD) is developed to reduce the complexity of the GD algorithm caused by parameter discretization. The key idea of Li-GD is that: the initial value of the$i\mathrm {th}$layer’s GD procedure is selected from the optimal results of the former$(i-1)$layers’ GD procedure whose intermediate data size is the closest to$i\mathrm {th}$layer. Additionally, the properties of the proposed algorithms are investigated, including convergence, complexity, and approximation error. The experimental results demonstrate that the performance of ECC is much better than that of the previous studies.

Online Knowledge Distillation (OKD) is designed to alleviate the dilemma that the high-capacity pre-trained teacher model is not available. However, the existing methods mostly focus on improving the ensemble prediction accuracy from multiple students (a.k.a. branches), which often overlook the homogenization problem that makes student model saturate quickly and hurts the performance. We assume that the intrinsic bottleneck of the homogenization problem comes from the identical branch architecture and coarse ensemble strategy. We propose a novel Adaptive Hierarchy-Branch Fusion framework for Online Knowledge Distillation, termed AHBF-OKD, which designs hierarchical branches and adaptive hierarchy-branch fusion module to boost the model diversity and aggregate complementary knowledge. Specifically, we first introduce hierarchical branch architectures to construct diverse peers by increasing the depth of branches monotonously on the basis of target branch. To effectively transfer knowledge from the most complex branch to the simplest target branch, we propose an adaptive hierarchy-branch fusion module to create hierarchical teacher assistants recursively, which regards the target branch as the smallest teacher assistant. During the training, the teacher assistant from the previous hierarchy is explicitly distilled by the teacher assistant and the branch from the current hierarchy. Thus, the important scores to different branches are effectively and adaptively allocated to reduce the branch homogenization. Extensive experiments demonstrate the effectiveness of AHBF-OKD on different datasets, including CIFAR-10/100 and ImageNet 2012. For example, on ImageNet 2012, the distilled ResNet-18 achieves Top-1 error of 29.28\%, which significantly outperforms the state-of-the-art methods. The source code is available at https://github.com/linruigong965/AHBF.

Extended targets tracking in sea clutter by the video data of radar has gained a special interest recently. The existing methods are not working well with close targets. Misdetection would arise once the close targets are regarded as a large target or a clutter region. Therefore, a rain algorithm is proposed for this problem by separating the measurements of different targets into clusters. The rain algorithm is inspired by the movement of raindrops fall in the mountains. Raindrops would converge into rivers and the mountains would be separated into individual peaks by the rivers. This phenomenon is available to develop a new algorithm providing what is called ‘rain algorithm’. In the rain algorithm, the raindrops are designed to find the boundary plots between the closely targets. The rivers are regarded as the boundary of targets and each peak equals a target. The rain algorithm is integrated into the multilevel threshold plot fusion method here and the extra computation is affordable. Both real data and simulated data are performed with the proposed method and several recent methods. The results showcase that the proposed method outperforms the others under various sea conditions.

In recent years, multi‐extended‐targets detection in sea clutter has gained a special interest. Dynamic programming based track‐before‐detect (DP‐TBD) algorithm is used to detect extended targets in video data of high resolution radars. Two innovations are presented in this work. First one is a novel partition method to cluster targets into well separate groups for the problem of high‐dimensional maximisation. Second one is a novel merit function specifically designed for extended targets for the problem of target extended. Both the principle of this novel DP‐TBD method specifically for extended targets and its detail implementation are presented. Then, both the real data and simulated data are performed. The comparison of the results shows that compared with particle filter based track before detect algorithm, the proposed method obtains better performance in detection rate, position error and false alarm rate. Meanwhile, far less calculation is spent with the novel partition method. Therefore, it can safely conclude that the proposed method is very practical under various sea conditions in the real world.