Multi-behavior recommendation enriches user preference modeling by incorporating diverse auxiliary interactions. However, most existing methods simply treat interactions without target behaviors as absolute negative feedback. This strategy ignores an important intermediate state known as user hesitation, where users exhibit strong intent but fail to complete the final conversion due to various reasons. Consequently, models cannot distinguish true disinterest from intended but hesitant behavior, which introduces substantial noise into preference modeling. To address this issue, we propose a novel framework named Calibrated Hesitation Analysis for Multi-Behavior Recommendation via Mutual Information Maximization (CHARM). Specifically, we aggregate auxiliary behaviors that lead to successful conversions into latent intent representations and train an inference network by maximizing the mutual information between these intents and observed target behaviors. We then apply this network to auxiliary behaviors without conversion, under the assumption that the conversion had occurred, in order to infer latent conversion probabilities and identify high-intent hesitation candidates. Furthermore, to distinguish genuine hesitation from interaction termination caused by competing item choices, we design a competitor substitution penalty strategy to refine hesitation confidence scores. Finally, the calibrated hesitation set is incorporated into the recommendation process to improve ranking quality. Extensive experiments on three real-world datasets demonstrate that CHARM consistently outperforms existing state-of-the-art methods. The source code is available at https://github.com/city59/CHARM.

Multi-Behavior recommendation captures fine-grained user intents by jointly modeling multiple interaction behaviors, significantly improving recommendation accuracy and diversity. However, existing methods often overlook complex semantic relations between items and entities. Moreover, constructing separate subgraphs for different interaction types frequently results in sparse graph structures, which exacerbates the long-tail problem. To address these challenges, we propose Long-tail Inductive Feature Transfer (LIFT), a method for knowledge-aware multi-behavior recommendation. To mitigate uneven node distribution in graph structures, we introduce a knowledge transfer-based feature reconstruction mechanism. Specifically, we first drop a portion of the neighbors of head nodes to construct proxy representations for tail nodes, training a reconstructor on the tail proxy to reconstruct the original head node features. The trained reconstructor is then used to backfill missing neighbor information for tail nodes, thereby achieving a more balanced feature distribution across nodes. Furthermore, we integrate multi-behavior and semantic contrastive learning to jointly optimize the representations. Extensive experiments on four datasets demonstrate that LIFT outperforms state-of-the-art methods, with further analysis validating its uniformity in representation learning. The source code is available at:https://github.com/city59/LIFT.

Contrastive learning (CL) has demonstrated exceptional capability in extracting supervised signals and mitigating noise, increasingly attracting interest for its application in knowledge graph-enhanced (KG) recommendation. Nevertheless, existing approaches predominantly rely on a single augmentation strategy to construct contrastive views, often failing to effectively balance feature perturbation with semantic preservation. To overcome this limitation, we propose motif-guided multiview contrastive learning (MMCL), a unified framework for contrastive augmentation. MMCL leverages diverse view generation strategies across the user–item bipartite graph and the KG, perturbing graph structures while preserving essential semantics to the greatest extent possible. Specifically, for collaborative signals, we devise a data augmentation mechanism to model the interaction dynamics between users and items. For semantic information, we introduce a novel graph augmentation technique by constructing an interactive motif KG, enabling semantic contrastive learning within localized views. Furthermore, joint enhancement of graph and interaction data allows the model to capture global topological features of nodes. MMCL integrates contrastive learning at both local and global levels, effectively embedding local semantics and global topology into the learned representations. Extensive experiments conducted on three publicly available datasets reveal that MMCL outperforms advanced methods, particularly in scenarios with sparse interactions and noisy KG, while significantly alleviating popularity bias in recommendation.

In multiaccess edge computing (MEC) systems, service migration has been extensively applied to ensure service quality by migrating services to follow mobile users. The existing migration methods mainly focus on optimizing service response latency and migration costs by predicting user’s movements. However, some malicious adversaries can learn auxiliary knowledge, i.e., users’ mobility model and service migration trajectory, and launch location inference attacks to infer user locations. This leads to serious personal security threats, like malvertising, fraud and kidnapping. In this article, we propose a location privacy-aware service migration method to against adversaries’ location inference attacks in multiuser MEC systems. First, we adopt an entropy-based location privacy metric to accurately measure user’s location privacy leakage risk. Then, we formulate the service migration progress as a joint optimization problem that minimizes service response latency and location privacy leakage risk. To cope with interuser interference, we developed a multiagent soft actor–critic (MASAC) algorithm to help users collaboratively make service migration decisions. Finally, simulations based on real-world user movement trajectories were conducted to demonstrate the superiority of the proposed method. Evaluation and analysis results showed that our proposed method can effectively protect user location privacy while maintaining a low service response latency.

The emergence of a news recommendation system can effectively improve the news reading experience of users. The most important task of the system is how to learn news and user representations accurately. In the process of learning news representations, most of the current research works do not fully utilize the news features, which makes it difficult to learn more comprehensive news representations. Most research work only learns user representations from a single perspective, which may not be sufficient to learn diverse and dynamic user representations. Therefore, we propose a news recommendation system with a multiview graph convolutional network (NRMG). It contains two parts: news representation and user representation. The knowledge–content collaboration network is adapted to learn news representations from news content and entities, while the multiview graph convolutional network (GCN) is utilized to learn user representations from the user’s click history. The advantage of the NRMG system is that we not only expand the available features by constructing a subclass knowledge graph (KG), but also effectively improve the ability of the news recommendation system to accurately learn news and user representations. Experimental results on the real dataset MIcrosoft News Dataset (MIND) show that the NRMG achieves a 2.25% improvement in area under the receiver operating characteristic (ROC) curve (AUC) value compared with state-of-the-art methods.

In this paper, we present PAssTrack, a Wi-Fi based passive human tracking system which is adapted to the practical antenna spacing of most commodity Wi-Fi access points (APs) and achieves accurate tracking results. We mainly enhance our system in the following four aspects. Firstly, we modify 2D MUSIC algorithm for estimating parameters including angle of arrival (AoA) and relative time of flight (rToF) of dynamic human reflection path. And we analyze the superiority of our algorithm compared with the state-of-the-art solutions. Secondly, we leverage the estimated rToF and the continuity of AoA to resolve the angle ambiguity caused by antenna spacing larger than half of the wavelength. Thirdly, we optimize the mesh model based on Fresnel zone theory to a dual antenna version for fine-grained velocity estimation. Lastly, we introduce hologram for target localization in order to compensate the errors of estimated path parameters using velocity estimates, and reduce the impact of outliers through kernel density estimation (KDE). The experiments are conducted in two real-world indoor environments, and the results demonstrate the advantages of PAssTrack in aspects of better tracking accuracy than the state-of-the-arts, easy and effective calibration, robustness under the case of blocked transceivers and adaptiveness to different antenna spacing.

In this article, we present a work using a smartphone with an off-the-shelf WiFi router for human activity recognition with various scales. The router serves as a hotspot for transmitting WiFi packets. The smartphone is configured with customized firmware and developed software for capturing WiFi channel state information (CSI) data. We extract the features from the CSI data associated with specific human activities, and utilize the features to classify the activities using machine learning models. To evaluate the system performance, we test 20 types of human activities with different scales including seven small motions, four medium motions, and nine big motions. We recruit 60 participants and spend 140 hours for data collection at various experimental settings, and have 36 000 data points collected in total. Furthermore, for comparison, we adopt three distinct machine learning models, including convolutional neural networks (CNNs), decision tree, and long short-term memory. The results demonstrate that our system can predict these human activities with an overall accuracy of 97.25%. Specifically, our system achieves a mean accuracy of 97.57% for recognizing small-scale motions that are particularly useful for gesture recognition. We then consider the adaptability of the machine learning algorithms in classifying the motions, where CNN achieves the best predicting accuracy. As a result, our system enables human activity recognition in a more ubiquitous and mobile fashion that can potentially enhance a wide range of applications such as gesture control, sign language recognition, etc.

A two-untrusted-relay transmission scheme is proposed with lossy-forward (LF) relaying being utilized. Two untrusted relays are located between one source and one common destination and there is no direct link between the source and the destination, which referred to as diamond relaying network. Since LF relaying allows intra-link errors and always forwards the decoded information sequences from the relays to the destination, it realizes the reliable and secure transmission of decoder-and-forward based relay networks with untrusted relays. Reliable-and-secure probability is derived to evaluate the performance of the untrusted relay networks, which represents the probability that the destination can recover the original message sent from the source whereas the relays cannot. We find that as long as the contributions of the transmit power of source and relays are balanced, a certain degree of reliable-and-secure probability can be held, even without the support from a friendly jamming signal.

Nowadays, Internet of Things (IoT) based smart home system is equipped with a large number of smart devices, such as smart speakers and cameras, which can greatly facilitate users to control and automate their home environment. However, recent studies have shown that smart home system is at great risk of privacy leakage. Especially, external attackers can infer user privacy information by passively sniffing encrypted smart home network traffic. Traditional methods mainly focus on sniffing WiFi devices but pay less attention to other commodity devices such as Zigbee and Bluetooth (BLE). In this paper, we focus on inferring fine-grained sensitive details about users using diverse commodity devices. We apply deep learning techniques to infer users' behaviors through identifying and localizing smart home devices being used due to the excellent performance of deep learning in many fields. Specifically, we first pre-process encrypted device traffic, select valid features, and use Convolutional Neural Networks (CNN) for device identification. In addition, we extract the Received Signal Strength Indicator (RSSI) from the frame information of traffic packets and employ Sparse Autoencoder (SAE) to extract stable and distinguishable high-dimensional features for RSSI measurement. Features are fed into a Multilayer Perceptron (MLP) to predict the device's localization. In this way, we can infer human activity by identifying and localizing the devices being used. Extensive experiment results show that our work can achieve a mean position estimation error of 1.34m even in an unseen environment, outperforming other common- used localization algorithms based on RSSI fingerprints.

We analyze the outage performance of a lossy forwarding (LF) relaying system with the simultaneous wireless information and power transfer (SWIPT) capability. In the system of LF with SWIPT (LF-SWIPT), a source broadcasts its message to both a relay and a destination. A relay node with SWIPT functionality harvests energy and decodes information from the source signal. The energy is split into two parts for information processing and message forwarding, respectively. For information processing, the relay attempts to decode the incoming source signal and forms an estimate. Unlike the existing decode-and-forward SWIPT system (DF-SWIPT), the estimate is always forwarded using the harvested energy. The destination performs joint decoding to recover the message with the signals received from both the source node and the relay node. We derive the outage probability for the LF-SWIPT system based on the theorem ofsource coding with side information. The simulation results demonstrate that the proposed system achieves significant gains (around 1–2 dB) compared to the DF-SWIPT system. We further evaluate the impact of the distance and the power splitting (PS) strategy on the system performance using simulations. Finally, we build an optimization algorithm on the PS ratio by maximizing the admissible region from the theoretical perspective.

For large-scale Internet of Things (IoT), backscatter communication is a promising technology to reduce power consumption and simplify deployment. However, due to the variable excitation source (ES) signal strength and time-varying channel condition, backscatter communication lacks stability, along with limited communication range as a few meters. Adaptive date rate (ADR) is beneficial to solve such issues, but is burdensome when implement on the capability limited tags. In this paper, we design a system named FreeBack with rate adaptation in backscatter communication. Our modulation approach is denoted as Adaptive Chirp-OOK where the ES recursively generates chirp signal, and the tags reflect the chirp signal with the On-Off Key modulation. According to channel symmetry, the tags perform rate adaption only based on the received ES signal strength instead of feedback from receiver. Such adaptation method enables the receiver to successfully decode signal through the time-varying channel, even for signal under the noise floor. We have implemented the prototype system based on the USRP platform. Extensive experiment results demonstrate the effectiveness of the proposed system. Our system provides valid ES-tag distance up to 27m, which is 7× as compared with normal backscatter system. FreeBack significantly increases the backscatter communication stability, by supporting data rate adaptation ranges from 0. 33kbps to 1. 2Mbps, and guaranteeing the bit error rate (BER) below 1%.

The WiFi backscatter communications offer ultralow power and ubiquitous connections for IoT systems. Caused by the intermittent-nature of the WiFi traffics, state-of-the-art WiFi backscatter communications are not reliable for backscatter link or simple for the tag to do the adaptive transmission. In order to build reliable WiFi backscatter communications, we present GuardRider, a WiFi backscatter system that enables backscatter communications to improve the quality of service (QoS). The key contribution of GuardRider is an optimization algorithm of designing RS codes to follow the statistical knowledge of WiFi traffics and adjust backscatter transmission. With GuardRider, the reliable baskscatter link is guaranteed and a backscatter tag is able to adaptively transmit information without heavily listening to the excitation channel, by taking QoS into account. We built a hardware prototype of GuardRider using a customized tag with FPGA implementation. Both the simulations and field experiments verify that GuardRider could achieve notably gains in bit error rate and frame error rate, which are a hundredfold reduction in simulations and around 99% in filed experiments. Our system is able to achieve around 700 kbps throughput.

The ever-increasing number of IoT devices in our surrounding environment bring us tremendous amount of opportunities but also challenges including limited battery life, low computational capability and scalability of multiple access. Recent advances in backscatter communication have enabled ubiquitous IoT devices to communicate in a cost-and power-efficient way. However, most of the proposed backscatter solutions nowadays focus on the single tag paradigm, i.e., multiple tags do not transmit simultaneously and thus the solutions have difficulties to scale with a large number of tags. This work presents CBMA, a backscatter system that enables multiple concurrent backscatter tags to communicate reliably and efficiently. For the first time, we demonstrate that multiple tags can backscatter concurrently and efficiently with novel impedance-based power control at the tag, and can be successfully decoded with commodity WiFi devices without affecting the existing WiFi communication. We present the design details of CBMA and build a prototype with off-the-shelf WiFi devices and FPGA. The CBMA system achieves a 10-tag bit rate of 8Mbps while supporting a communication distance up to 10m. Compared to single-tag solutions, CBMA improves the backscatter throughput by more than 10× even in challenging indoor scenarios with rich multipath and interference.

Finger gesture input is emerged as an increasingly popular means of human-computer interactions. In this paper, we propose iPand, an acoustic sensing system that enables finger gesture input on the skin, which is more convenient, user-friendly and always accessible. Unlike previous works, which implement gesture input with dedicated devices, our system exploits passive acoustic sensing to identify the gestures, e.g. swipe left, swipe right, pinch and spread. The insight of our system is that specific gesture emits unique friction sound, which can be captured by the microphone embedded in wearable devices. We capture these acoustic signals and extract the features by using bandpass filters and short-time Fourier Transform. The offline convolutional neural network is adopted to recognize the gestures. iPand is implemented and evaluated using COTS smartphones and smartwatches. Experiment results show that iPand can achieve the recognition accuracy of 89%, 83% and 78% in three daily scenarios (i.e., library, lab and cafe), respectively. Particularly, our system supports multi-touch function where 2-4 fingers are enabled for more efficient and expressive gesture input, and its average accuracy for individual finger gesture reaches up to 83% within 12 gestures.

The binary chief executive officer (CEO) problem with an arbitrary number of agents is considered in this paper. A scheme which separates the reconstruction of observations and the final decision of a common source is assumed. Hence, we first derive the outer bound for the rate- distortion region by providing the converse proof of a binary multiterminal source coding problem which is the key to solve the binary CEO problem. The distortion of the binary CEO problem is then determined by the Poisson binomial process based on the using majority voting logic for the final decision. The rate-distortion behavior of the binary CEO problem is then analyzed based on the outer bound by solving a convex optimization problem. It is found that the distortion decreases until it converges to a certain level, as the sum rate and/or the number of agents increases.

A two-node binary chief executive officer (CEO) problem is investigated. Noise-corrupted versions of a binary sequence are forwarded by two nodes to a single destination node over orthogonal additive white Gaussian noise (AWGN) channels. We first reduce the binary CEO problem to a binary multiterminal source coding problem, of which an outer bound for the rate-distortion region is derived. The distortion function is then established by evaluating the relationship between the binary CEO and multiterminal source coding problems. A lower bound approximation on the Hamming distortion (HD) is obtained by minimizing a distortion function subject to constraints obtained based on the source-channel separation theorem. Encoding/decoding algorithms using concatenated convolutional codes and a joint decoding scheme are used to verify the lower bound on the HD. It is found that the theoretical lower bounds on the HD and the computer simulation-based bit error rate performance curves have the same tendencies. The differences in the threshold signal-to-noise ratio between the theoretical lower bounds and those obtained by simulations are around 1.5 dB in AWGN channel. The theoretical lower bound on the HD in block Rayleigh fading channel is also evaluated by performing Monte Carlo simulation.

In this paper, we theoretically analyze the outage probability of decode-and-forward (DF) relaying system allowing intra-link errors (DF-IE), where the relay always forwards the decoder output to the destination regardless of whether errors are detected after decoding in the information part or not. The results apply to practical fading scenarios where all the links between the nodes suffer from independent block Rayleigh fading. The key idea of DF-IE system is that the data sequence forwarded by the relay is highly correlated with the original information sequence sent from the source, and hence with a proper joint decoding technique at the destination, the correlation knowledge can well be exploited to improve the system performance. We analyze this problem in the information theoretical framework of correlated source coding. Using the theorems for lossy source-channel separation and for source coding with side information, the exact outage probability is derived. It is then shown that the exact expression can be reduced to a simple, yet accurate approximation by replacing the theorem for source coding with side information by the Slepian-Wolf theorem. Compared with conventional DF relaying where relay keeps silent if errors are detected after decoding, DF-IE can achieve even lower outage probability. Moreover, by allowing intra-link errors, the optimal position of the relay is found to be exactly the midpoint between the source and destination. Results of the simulations are provided to verify the accuracy of the analytical results.