Daqing Zhang

Bio: Daqing Zhang is an academic researcher from Peking University. The author has contributed to research in topics: Context (language use) & Mobile computing. The author has an hindex of 67, co-authored 331 publications receiving 16675 citations. Previous affiliations of Daqing Zhang include Institut Mines-Télécom & Institute for Infocomm Research Singapore.

iCROSS: toward a scalable infrastructure for cross-domain context management

Abstract: With the development of pervasive computing techniques, the world will be filled with interconnected context-aware domains (e.g., homes, offices, hospitals, etc.). While the previous studies focused solely on the management of contexts produced in a single domain, in this paper we discuss the challenges to be addressed for cross-domain context management. By analyzing the requirements from several scenarios, we identify two context producer---consumer patterns in multi-domain environments. Furthermore, to deal with the mobile entity problem raised in cross-domain context sharing, a transparent query mechanism that enables applications to obtain context information about mobile entities from remote domains is proposed. Two prototype applications--smart home and community services in a smart campus--have been developed to demonstrate the key features and usefulness of cross-domain context management. Initial experiments have also been conducted to evaluate the performance of our system.

Location Independent Vital Sign Monitoring and Gesture Recognition Using Wi-Fi

Abstract: Recent years have witnessed the rapid progress of Wi-Fi based contactless sensing. Compared to traditional wearable based approaches, Wi-Fi sensing does not require the target to wear any sensors and is able to capture rich context information of human target in a non-intrusive manner. Though promising, one major issue hindering the adoption of Wi-Fi sensing is the location and orientation dependence of the performance, i.e., if the human target changes the location or orientation, the sensing performance may degrade significantly. This chapter delves into this issue, analyzes the factors affecting the sensing performance and presents solutions to addressing this issue, moving Wi-Fi sensing one step closer towards real-life deployment.

Tunable narrow-band single-channel add-drop integrated optical filter with ultrawide FSR

Abstract: Abstract Free-spectral-range (FSR)-free optical filters have always been a critical challenge for photonic integrated circuits. A high-performance FSR-free filter is highly desired for communication, spectroscopy, and sensing applications. Despite significant progress in integrated optical filters, the FSR-free filter with a tunable narrow-band, high out-of-band rejection, and large fabrication tolerance has rarely been demonstrated. In this paper, we propose an exact and robust design method for add-drop filters (ADFs) with an FSR-free operation capability, a sub-nanometer optical bandwidth, and a high out-of-band rejection (OBR) ratio. The achieved filter has a 3-dB bandwidth of < 0.5 nm and an OBR ratio of 21.5 dB within a large waveband of 220 nm, which to the best of our knowledge, is the largest-FSR ADF demonstrated on a silicon photonic platform. The filter exhibits large tunability of 12.3 nm with a heating efficiency of 97 pm/mW and maintains the FSR-free feature in the whole tuning process. In addition, we fabricated a series of ADFs with different periods, which all showed reliable and excellent performances.

Polarization-independent electro-absorption optical modulator based on trapezoid polymer-graphene waveguide

Abstract: A polarization-independent electro-absorption modulator based on the trapezoid polymer-graphene waveguide (PGW) was proposed. The modulator was constructed on a trapezoid polymer waveguide, and the insulting dielectric spacer sandwiched in the two graphene layers was placed on the surface of the trapezoid polymer waveguide core. The simulation results show that by applying different gate voltages on the graphene layers, effective mode index of the TE and TM modes in the PGW can realize the similar changes in the C-band, which provides the possibility for realizing the polarization-independent modulation. Here, through simulation and optimization, the presented modulator has a compact size and the extinction ratio (ER) of 37 dB can be achieved by reasonably setting working points of “OFF” and “ON” states with 800 μm long active graphene length. The ER variation between the two operating modes is about 0.46 dB. The corresponding power consumption of the modulator is about 23.6 pJ/bit.

Context Aware Computing for The Internet of Things: A Survey

Abstract: As we are moving towards the Internet of Things (IoT), the number of sensors deployed around the world is growing at a rapid pace. Market research has shown a significant growth of sensor deployments over the past decade and has predicted a significant increment of the growth rate in the future. These sensors continuously generate enormous amounts of data. However, in order to add value to raw sensor data we need to understand it. Collection, modelling, reasoning, and distribution of context in relation to sensor data plays critical role in this challenge. Context-aware computing has proven to be successful in understanding sensor data. In this paper, we survey context awareness from an IoT perspective. We present the necessary background by introducing the IoT paradigm and context-aware fundamentals at the beginning. Then we provide an in-depth analysis of context life cycle. We evaluate a subset of projects (50) which represent the majority of research and commercial solutions proposed in the field of context-aware computing conducted over the last decade (2001-2011) based on our own taxonomy. Finally, based on our evaluation, we highlight the lessons to be learnt from the past and some possible directions for future research. The survey addresses a broad range of techniques, methods, models, functionalities, systems, applications, and middleware solutions related to context awareness and IoT. Our goal is not only to analyse, compare and consolidate past research work but also to appreciate their findings and discuss their applicability towards the IoT.

Membership Inference Attacks Against Machine Learning Models

Abstract: We quantitatively investigate how machine learning models leak information about the individual data records on which they were trained. We focus on the basic membership inference attack: given a data record and black-box access to a model, determine if the record was in the model's training dataset. To perform membership inference against a target model, we make adversarial use of machine learning and train our own inference model to recognize differences in the target model's predictions on the inputs that it trained on versus the inputs that it did not train on. We empirically evaluate our inference techniques on classification models trained by commercial "machine learning as a service" providers such as Google and Amazon. Using realistic datasets and classification tasks, including a hospital discharge dataset whose membership is sensitive from the privacy perspective, we show that these models can be vulnerable to membership inference attacks. We then investigate the factors that influence this leakage and evaluate mitigation strategies.