Hepatocellular carcinoma (HCC) is a major cause of cancer-related mortality, and microwave ablation (MWA) is commonly used for patients ineligible for surgical resection. A critical challenge following MWA is the assessment of the ablative margin, which is complicated by non-diffeomorphic deformations introduced by thermal effects during the procedure. This paper proposes a Multi-sequence Distance-guided Complementary Network (MDCNet) that utilizes multi-sequence MRI to quantify the extent of tumor contraction after MWA. To account for the differential contraction responses of liver parenchyma and tumor tissue, we propose a novel distance-aware mask transformation strategy. This method explicitly models the spatial attenuation of MWA energy and approximates the influence of liver parenchyma's linear elastic response on tumor shrinkage, thereby enhancing the spatial adaptiveness of feature weighting. To capture the distinct structural characteristics of liver tissue emphasized by different MRI sequences and to leverage their complementary information, a gated channel fusion module is introduced to dynamically integrate features from delayed-phase and T2-weighted images. To validate the practical effectiveness of our proposed method, we evaluate the ablative margins of 115 HCC patients using a fine-tuned TransMorph model that incorporated tumor contraction predictions generated by MDCNet, and compare the results with radiologist 2D assessments. The registration method enhanced with MDCNet improved tumor deformation accuracy and achieved a higher Youden Index in detecting incomplete ablations. Moreover, MDCNet provides interpretable predictions, thereby facilitating clinical decision support.

Compact UAV systems, while advancing delivery and surveillance, pose significant security challenges due to their small size, which hinders detection by traditional methods. This paper presents a cost-effective, unsupervised UAV detection method using spatial-temporal sequence processing to fuse multiple LiDAR scans for accurate UAV tracking in real-world scenarios. Our approach segments point clouds into foreground and background, analyzes spatial-temporal data, and employs a scoring mechanism to enhance detection accuracy. Tested on a public dataset, our solution placed 4th in the CVPR 2024 UG2+ Challenge, demonstrating its practical effectiveness. We plan to open-source all designs, code and sample data for the research community @ github.com/lianghanfang/UnLiDAR-UAV-Est.

This paper revisits the longstanding challenge of coordinating database systems with general-purpose OS interfaces, such as POSIX, which often lack tailored support for DB requirements. Existing approaches to this DB-OS co-design struggle with limited design space, security risks, and compatibility issues. To overcome these hurdles, we propose a new co-design approach leveraging virtualization to elevate the privilege level of DB processes. Our method enables database systems to fully exploit hardware capabilities via virtualization, while minimizing the need for extensive modifications to the host OS kernel, thereby maintaining compatibility. We demonstrate the effectiveness of our approach through two novel virtual memory mechanisms tailored for database workloads: (1) an efficient snapshotting mechanism that captures memory snapshots at millisecond intervals for in-memory databases and HTAP workloads, and (2) a streamlined in-kernel buffer pool design. We introduce Libdbos , a lightweight guest kernel implementing these mechanisms. Our evaluations highlight significant improvements in latency and efficiency compared to existing snapshotting and buffer pool designs, underscoring the potential of the approach.

With the advantages of maneuverability and low cost, Unmanned Aerial Vehicles (UAVs) are widely deployed in mobile edge computing as micro servers to provide computing service. However, tasks usually require a large amount of energy and have strict time constraints, while the battery energy and endurance of UAVs are limited. Therefore, energy consumption and delay have become key issues in such architectures. To address this issue, an Entropy Normalized Soft Actor-Critic (ENSAC) computation offloading algorithm is proposed in this paper, aiming to minimize the weighted sum of task offloading delay and energy consumption. In ENSAC, we formulate the task offloading problem as a Markov Decision Process (MDP). Considering the non-convexity, high-dimensional state space, and continuous action space of this problem, the ENSAC algorithm fully combines deviation strategy and maximum entropy reinforcement learning, and designs a system utility function under entropy normalization as a reward function, thus ensuring fairness in weighted energy consumption and delay. What’s more, ENSAC algorithm also considers UAV trajectory planning, task offloading ratio, and power allocation in the UAV-assisted MEC system. Therefore, compared with previous methods, ENSAC algorithm has stronger stability, better exploration performance, and can handle more complex environments and larger action space. Finally, extensive experiments demonstrate that, in both energy-saving and delay-sensitive scenarios, the ENSAC algorithm can quickly converge to the optimal solution while maintaining stability. Compared with four benchmark algorithms, it reduces the total system cost by 52.73%.

Most neural network pruning methods, such as filter-level and layer-level prunings, prune the network model along one dimension (depth, width, or resolution) solely to meet a computational budget. However, such a pruning policy often leads to excessive reduction of that dimension, thus inducing a huge accuracy loss. To alleviate this issue, we argue that pruning should be conducted along three dimensions comprehensively. For this purpose, our pruning framework formulates pruning as an optimization problem. Specifically, it first casts the relationships between a certain model’s accuracy and depth/width/resolution into a polynomial regression and then maximizes the polynomial to acquire the optimal values for the three dimensions. Finally, the model is pruned along the three optimal dimensions accordingly. In this framework, since collecting too much data for training the regression is very time-costly, we propose two approaches to lower the cost: 1) specializing the polynomial to ensure an accurate regression even with less training data; 2) employing iterative pruning and fine-tuning to collect the data faster. Extensive experiments show that our proposed algorithm surpasses state-of-the-art pruning algorithms and even neural architecture search-based algorithms.

In recent years, supervised person re-identification (re-ID) models have received increasing studies. However, these models trained on the source domain always suffer dramatic performance drop when tested on an unseen domain. Existing methods are primary to use pseudo labels to alleviate this problem. One of the most successful approaches predicts neighbors of each unlabeled image and then uses them to train the model. Although the predicted neighbors are credible, they always miss some hard positive samples, which may hinder the model from discovering important discriminative information of the unlabeled domain. In this paper, to complement these low recall neighbor pseudo labels, we propose a joint learning framework to learn better feature embeddings via high precision neighbor pseudo labels and high recall group pseudo labels. The group pseudo labels are generated by transitively merging neighbors of different samples into a group to achieve higher recall. However, the merging operation may cause subgroups in the group due to imperfect neighbor predictions. To utilize these group pseudo labels properly, we propose using a similarity-aggregating loss to mitigate the influence of these subgroups by pulling the input sample towards the most similar embeddings. Extensive experiments on three large-scale datasets demonstrate that our method can achieve state-of-the-art performance under the unsupervised domain adaptation re-ID setting.

As an instance-level recognition problem, re-identification (re-ID) requires models to capture diverse features. However, with continuous training, re-ID models pay more and more attention to the salient areas. As a result, the model may only focus on few small regions with salient representations and ignore other important information. This phenomenon leads to inferior performance, especially when models are evaluated on small inter-identity variation data. In this paper, we propose a novel network, Erasing-Salient Net (ES-Net), to learn comprehensive features by erasing the salient areas in an image. ES-Net proposes a novel method to locate the salient areas by the confidence of objects and erases them efficiently in a training batch. Meanwhile, to mitigate the over-erasing problem, this paper uses a trainable pooling layer P-pooling that generalizes global max and global average pooling. Experiments are conducted on two specific re-identification tasks (i.e., Person re-ID, Vehicle re-ID). Our ES-Net outperforms state-of-the-art methods on three Person re-ID benchmarks and two Vehicle re-ID benchmarks. Specifically, mAP / Rank-1 rate: 88.6% / 95.7% on Market1501, 78.8% / 89.2% on DuckMTMC-reID, 57.3% / 80.9% on MSMT17, 81.9% / 97.0% on Veri-776, respectively. Rank-1 / Rank-5 rate: 83.6% / 96.9% on VehicleID (Small), 79.9% / 93.5% on VehicleID (Medium), 76.9% / 90.7% on VehicleID (Large), respectively. Moreover, the visualized salient areas show human-interpretable visual explanations for the ranking results.

An energy-efficient power control algorithm with max-min fairness for energy harvesting ad hoc networks is studied in this paper. Specially, each node of the network has an ability of harvesting energy from external environment, e.g., thermal or solar etc. We aim to maximize the minimum individual energy efficiency while guaranteeing the energy causality constraint and quality of service (QoS) of each node. The fairness-aware energy-efficient power control problem is formulated as a fractional nonconvex problem. To make the formulated problem tractable, we transform the original problem into an equivalent subtractive-form problem via Dinkelbach approach. A combinational framework of successive convex approximation (SCA) and geometric programming (GP) is further adopted to obtain a near optimal solution of the power control problem. Simulation results show that the proposed algorithm can provide greater fairness in energy harvesting ad hoc networks.

This paper investigates the resource allocation problem in full-duplex (FD) relay-assisted device-to-device (D2D) communications underlaying a multi-carrier cellular network. Under residual self-interference (SI), we aim to maximize the energy efficiency (EE) of D2D links via subcarrier assignment and power allocation while satisfying the quality of service (QoS) requirements of both D2D and cellular links. To solve the formulated EE maximization problem, we propose an iterative algorithm based on nonlinear fractional programming, where power allocation and subcarrier assignment are performed sequentially in each iteration, to handle the nonconvexity of the original problem. It is shown that the power allocation problem fits in D.C. (difference of two concave functions) programming while the subcarrier assignment corresponds to a job assignment problem whose optimal solution can be found in polynomial time. Finally, simulation results demonstrate that the proposed algorithm converges rapidly and achieves higher EE of D2D communications than existing schemes.

This paper investigates the energy-efficient resource allocation problem for the energy harvesting-powered device-to-device (D2D) communications underlaying cellular networks under the harvest-then-transmit protocol. Our objective is to maximize the energy efficiency (EE) of D2D communications via joint time allocation and power control while satisfying the quality of service (QoS) requirement of cellular user (CU) and energy causality constraint. Exploiting the non-linear fractional programming, an iterative resource allocation scheme based on the Dinkelbach method is proposed to solve the formulated nonconvex optimization problem. For the optimization problem in each iteration, we decompose it into two subproblems: the power control subproblem and the time allocation subproblem. It is shown that the former subproblem fits in D.C. (difference of two concave functions) programming while the latter subproblem corresponds to a linear programming problem which can be efficiently solved. Simulation results validate the EE performance of our proposed scheme.

This paper is mainly to deal with the problem of long-term sparse channel estimation of coherent signals for mm-wave hybrid beamforming systems. The paper focuses on long-term beamforming, and it is based on large-scale fading channel directions but not the instantaneous small-scale fading gain, which reduces the computational complexity greatly. Moreover, the method the paper proposed couples 2-D beamspace MUSIC with the modified spatial smoothing with the advantages of processing coherent signals. It is also a compatible method which can be applied to the incoherent and coherent signals simultaneously. The computer simulation results verify the effectiveness and superiority of the proposed method when the signals are coherent.

This paper considers the blind estimation of the carrier frequency offset (CFO) and sample frequency offset (SFO) in orthogonal frequency division multiplexing (OFDM) system. Conventional methods require periodically transmitted pilot tones in frequency domain or identical training symbols in preamble. To achieve a blind estimation scheme, a joint maximum-likelihood (ML) estimator is proposed by applying another FFT within one symbol duration. Its simplified closed-form solution is then derived to reduce computational complexity. Theoretical analysis and simulation results confirm that the proposed low-complexity scheme achieves almost the same performance as ML estimation and outperforms other blind methods.

In this paper, a Stackelberg game based power control framework is proposed for interference coordination among device-to-device (D2D) communications and spectrum-sharing cellular networks. The cross-tier interference from D2D communications is controlled by the cellular base station (CBS) through setting a price for the shared subchannels. Based on the cross-tier interference price, each D2D pair adaptively updates its power strategy subsequently. A distributed power control algorithm with limited coordination among D2D pairs is proposed and guaranteed to converge to a Karush- Kuhn-Tucker (KKT) point of the optimization problem at D2D pairs's side by introducing a new co-tier price factor. At the CBS's side, a price update algorithm is designed ably according to cross-tier interference measurement. Simulation results show that the proposed algorithms can achieve higher sum rate for D2D communications when compared with that of the non-cooperative power allocation algorithm.

This paper considers the optimal energy-efficient transmission policy in time-sharing multiuser systems with a hybrid energy harvesting transmitter. The system operation energy is supplied by a constant energy source which is driven by a non-renewable resource and an energy harvester which harvests energy from its surrounding environment. Our goal is to maximize the energy-efficient transmission of multiuser systems considering the allocation of power and transmission time jointly in an off-line manner. The original nonconvex objective function is transformed into convex optimization problem via the fractional programming approach. Furtherly, we solve the convex problem by Lagrange dual decomposition method. Simulation results show that the proposed energy-efficient transmission policy has a better performance than the transmission policy which decomposes optimization problem into two parts (power allocation, time allocation) to solve iteratively.

In this paper, system throughput and user fairness for single-antenna user based on zero-forcing precoding are investigated with total power constraint (TPC) and per antenna power constraint (PAPC), respectively. By analyzing fairness of users signal to noise ratios (SNRs), the closed-form solution of fairness is obtained, and the tradeoff between system throughput and user fairness is presented. For multiple-antenna user, receive antenna selection (RAS) by maximizing channel norm is exploited to greatly reduce computational complexity. Simulation results show that different power constraints have a significant impact on system throughput and user fairness, and the throughput by considering user fairness decreases compared to that by considering maximum system throughput.

In this paper, we investigate the problem of maximizing the weighted sum energy efficiency (WS-EE) in downlink OFDMA small cell network. To solve the problem in a distributed manner and reduce the complexity, we decompose the resource blocks (RBs) allocation and the power control into two subproblems. For the subproblem of RBs allocation, the optimal RBs allocation strategy is obtained based on rate maximization given any feasible network power profile. For the subproblem of power control, a novel distributed pricing based algorithm is developed with limited inter-small cell coordination. Specifically, each small cell base station (SBS) updates its current interference information and broadcasts it to the other SBSs. With this collected information, the optimal power control strategy of each SBS is derived. Simulation results show that the proposed algorithm outperforms the existing algorithms in terms of the WS-EE.