API-level fuzzing has become increasingly important for discovering subtle bugs in modern software, yet generating effective fuzzing harnesses remains a complex and error-prone task. Existing approaches often rely on limited consumer code or shallow program analysis, which fail to capture deep API semantics and interdependencies, resulting in poor coverage and high false positive rates. Recent methods incorporating Large Language Models (LLMs) have improved harness generation by leveraging pretrained knowledge, but they still struggle with hallucinations and lack domain-specific understanding.

Spectrum sharing problem of the cognitive radio networks (CRN) with non-stationarity of secondary users (SUs) traffic demands, which could undermine the performances of traditional dynamic spectrum access (DSA) methods, is addressed in this study. The SU traffic arrival pattern is modelled with Poisson process and a SU-traffic-aware multi-agent deep reinforcement learning (MADRL) method is proposed to optimize the spectrum sharing among SUs. Each agent firstly learns the traffic patterns of both local SU and primary users (PUs) to avoid collisions and to improve quality of service (QoS) of SU. Moreover, a priori-rule-assisted DRL algorithm is designed to accelerate agent's training process by enforcing each agent to execute preset actions under specific states. Simulation results show that the proposed method achieves significant improvement on SU QoS while maintains low SU-PU collision probability.

Wireless powered Internet of Things (WPIoT) has gained great concern due to its benefits of high deployment flexibility and low maintenance overhead. The minimum delay broadcast scheduling problem is very critical for many applications of WPIoT. However, traditional broadcast scheduling algorithms assume that Internet of Things (IoT) devices always possess sufficient energy to support data transmission or reception, which does not hold in WPIoT with the special feature of using the store-charge-and-forward communication mode. Furthermore, existing solutions rely heavily on the interference-avoiding technology to handle the signal interference problem, and overlook the powerful interference processing capability of the successive interference cancellation (SIC) technology. To efficiently resolve this problem, this article proposes a delay-efficient energy-aware broadcast scheduling algorithm called EABS. EABS algorithm incorporates a novel broadcast link scheduling method by fully considering the special feature of WPIoT and efficiently utilizing the advantage of the SIC technology to significantly improve the broadcast delay. Extensive experiments based on a real-world dataset are conducted to evaluate the performance of our algorithm, and the results demonstrate the better performance of our algorithm than the baseline algorithms.

We consider the network-side mobile localization problem in future 5G and beyond wireless networks with distributed multi-antenna base stations (BSs). For this application, we propose a semi-supervised multi-point channel charting (SS-MPCC) framework, which consists of (i) collaborative collection of channel state information (CSI) and other side-information by distributed BSs; (ii) local CSI feature extraction and self-learning of a dissimilarity metric, and (iii) global graph construction and constrained manifold learning. We show that side-information from routine network operations, including timestamps, channel qualities, and a small set of labeled samples, can be exploited to construct a consistent global graph. The graph is then mapped to a 2D channel chart using constrained manifold learning for localization purposes. We evaluate the performance of SS-MPCC in a simulated urban outdoor scenario with realistic user motion. Our results show that SS-MPCC achieves a mean localization error of 5.6 m with only 10% of labeled CSI samples. SS-MPCC does not require accurate synchronization among multiple BSs and is promising for future cellular localization.

To perform multi-user multiple-input and multipleoutput transmission in millimeter-wave (mmWave) cellular systems, the high-dimensional channels need to be estimated for designing the multi-user precoder. Conventional grid-based Compressed Sensing (CS) methods for mmWave channel estimation suffer from the basis mismatch problem, which prevents accurate channel reconstruction and degrades the precoding performance. This paper formulates mmWave channel estimation as an Atomic Norm Minimization (ANM) problem. In contrast to grid-based CS methods which use discrete dictionaries, ANM uses a continuous dictionary for representing the mmWave channel. We consider a continuous dictionary based on sub-sampling in the antenna domain via a small number of radio frequency chains. We show that mmWave channel estimation using ANM can be formulated as a Semidefinite Programming (SDP) problem, and the channel can be accurately estimated via off-the-shelf SDP solvers in polynomial time. Simulation results indicate that ANM can achieve much better estimation accuracy compared to grid-based CS, and significantly improves the spectral efficiency provided by multi-user precoding.

We consider two-hop Device-to-device (D2D) relaying in multi-cell downlink networks. D2D relaying is envisioned to be a promising cell coverage extension technique, which can provide improved cell-edge performance without a dense infrastructure deployment. Relaying complicates the resource allocation and interference management in multi-cell networks. We first study the aggregate co-channel interference characteristics when D2D relaying is applied. A fluid network model is used to analyze the inter-cell interference in a multi-cell network with a minimum inter-base station distance. We develop a model for capturing the interaction between relaying decisions made in the own cell and inter-cell interference created to other cells. We investigate network steady state, and optimize key parameters for network-level management. Both simulation and analysis results are provided to help to understand the performance of D2D relaying.

We consider a user-centric network-level coordination architecture for 5G heterogeneous Radio Access Networks (RANs), based on RAN softwarization and a centralized coordination framework. We describe the RAN as a set of logical RAN entities, related to cells in a Heterogeneous Network (HetNet), under the control of a central coordination entity. This description allows the creation of Network Functions (NFs) with an abstracted view of the network. We describe a centralized coordination framework, and then develop a NF for InterCell Interference Coordination (ICIC) in a 5G HetNet, optimizing the radio resource usage at network-level. We construct a Network Graph to abstract the problem of resource allocation and cell offloading, with the NF seeking for an optimal solution based on this abstraction. Simulations are performed in a HetNet scenario with a Tabu Search algorithm. Results show the feasibility of performing network-level coordination through a modular NF, with an abstracted view of the network.

We discuss the motivation and potential use-cases of self-organizing networks (SON) in 5G networks. Disruptive technologies and features of 5G networks include millimeter wave (mmW), massive multiple-input multiple output (MIMO), and cloud-radio access network (C-RAN). These have ramifications on SON aspects of the networks. We discuss several new SON use-cases and problems pertinent to emerging 5G technologies. The discussed 5G SON use-cases include spectrum management and sharing, user association, multi-radio access technology (RAT) optimization, and directional cell search for mmW networks. We then investigate directional cell search in detail, and consider a network graph based approach for self-organized beam assignment in mmW 5G networks. Simulations results in a realistic Manhattan environment demonstrate the benefits of proposed approach, in terms of improved user signal-to-interference plus noise ratios to potential handover beams, thereby resulting in better directional cell discovery.

Gait is usually a crucial indicator for recognizing and evaluating the progression of Parkinson's disease (PD). To assess gait variability in the elderly with PD in a continuous and natural way, we pay more attention to their feet pressure variability while walking, and try to obtain the varying patterns of center of pressure (CoP). In this paper, we propose a framework based on the bimodal distribution probability density functions (BDPDFs) to recognize gait patterns of PD patients. The result evaluated with the 10-fold cross-validation method demonstrates that the Bagging classifier is able to provide classification precision of 82.4% with AUC of 88.3%. The features and the classifiers used in the present study can also be used to study the effect of dopamine and rehabilitation in PD patients.