Sha Tao

Secure and Privacy-Preserving Smartphone-Based Traffic Information Systems

Increasing smartphone penetration, combined with the wide coverage of cellular infrastructures, renders smartphone-based traffic information systems (TISs) an attractive option. The main purpose of such systems is to alleviate traffic congestion that exists in every major city. Nevertheless, to reap the benefits of smartphone-based TISs, we need to ensure their security and privacy and their effectiveness (e.g., accuracy). This is the motivation of this paper: We leverage state-of-the-art cryptographic schemes and readily available telecommunication infrastructure. We present a comprehensive solution for smartphone-based traffic estimation that is proven to be secure and privacy preserving. We provide a full-blown implementation on actual smartphones, along with an extensive assessment of its accuracy and efficiency. Our results confirm that smartphone-based TISs can offer accurate traffic state estimation while being secure and privacy preserving.

A Power-Efficient Continuous-Time Incremental Sigma-Delta ADC for Neural Recording Systems

This paper presents an analog-to-digital converter (ADC) dedicated to neural recording systems. By using two continuous-time incremental sigma-delta ADCs in a pipeline configuration, the proposed ADC can achieve high-resolution without sacrificing the conversion rate. This two-step architecture is also power-efficient, as the resolution requirement for the incremental sigma-delta ADC in each step is significantly relaxed. To further enhance the power efficiency, a class-AB output stage and a dynamic summing comparator are used to implement the sigma-delta modulators. A prototype chip, designed and fabricated in a standard 0.18 μm CMOS process, validates the proposed ADC architecture. Measurement results show that the ADC achieves a peak signal-to-noise-plus-distortion ratio of 75.9 dB over a 4 kHz bandwidth; the power consumption is 34.8 μW, which corresponds to a figure-of-merit of 0.85 pJ/conv.

Securing smartphone based ITS

GPS-equipped smartphones present several advantages for data acquisition in Intelligent Transportation Systems (ITS), compared to solutions that require a new communication infrastructure. However, there are still significant challenges to meet before deployment. Traffic information and location samples must be collected in a secure manner, to not jeopardize the system operation. Equally important, users must be assured about their privacy, notably the protection of information on their whereabouts. To address this two-fold problem, we propose extending the Generic Bootstrapping Architecture (GBA) with anonymous authentication. Identity and location information are protected and separated, and location samples cannot be linked to each other and to any specific user. Thus, our scheme protects users even in the case of a compromised ITS server. Initial evaluation results indicate the feasibility of our approach with off-the-self mobile platforms.

An IIP2 digital calibration technique for passive CMOS down-converters

The IIP2 requirement in fully integrated direct-conversion receivers using FDD duplexing is prohibitively high and demands the use of an external filter in order to attenuate the leakage from the transmitter. This paper presents a digital calibration technique for passive CMOS down-converters that allows a direct conversion receiver achieve the requirements without external filtering. A Least-Mean-Square optimization algorithm is used in order to reduce the low-frequency second-order intermodulation product. The algorithm controls the digital calibration structures at the biasing of the passive mixer and adapts them until the second order intermodulation drops below the noise level. The method is tested by calibrating a 1.2-V 65nm CMOS passive mixer targeting UMTS/LTE applications at several corner conditions including worst case mismatches in the switching pairs.

Hybrid vehicle positioning and tracking using mobile phones

Due to the pervasive deployments of mobile communication technologies, vehicle positioning and tracking by locating the drivers mobile phones has become feasible. However, no single positioning method can provide decent tradeoff between accuracy and coverage. To address this issue, we propose a Kalman filter-based hybrid method which can track the mobile phones traveling on-board vehicles. The proposed method combines coordinates collected by assisted global positioning system (A-GPS) mobile phones and location estimates calculated from observed time difference of arrival (OTDOA) measurements. Numerical results demonstrate the effectiveness of this hybrid scheme in a simulated vehicular scenario.

Smartphone-based traffic information system for sustainable cities

Traffic Information Systems (TISs) can play a significant role towards creating sustainable cities through improved traffic conditions. The collection of reliable and rich information with low cost is paramount. The use of smartphones carried by individuals for future implementations of TISs present several advantages compared to traditional solutions. This demo integrates our results from previous work addressing challenges on traffic estimation for urban road networks and on security and privacy protection for such TISs.

MVL-PUFs: multiple-valued logic physical unclonable functions

Summary Physical unclonable functions (PUFs) are promising hardware security primitives suitable for protecting resource-constrained devices. In this paper, we propose to use multiple-valued logic (MVL) for implementing hardware-efficient PUF integrated circuits. We show that by extracting device mismatch in either current-mode or voltage-mode MVL comparators, the proposed PUF circuits can generate unique and reliable chip identifiers. In order to stabilize PUF responses, we utilize multiple thresholds of MVL comparators, whose outputs are selected and combined according to the sensed temperature. To reduce power and further enhance reliability, the PUF circuits are biased in the weak-inversion region. Evaluation results show that the proposed MVL-PUFs are unique and reliable over a wide temperature range. In addition, they significantly improve the energy efficiency of the state-of-the-art PUFs. Copyright

A comparative design study of continuous-time incremental sigma-delta ADC architectures

This paper presents a comparative design study of continuous-time CT incremental sigma-delta IΣΔ ADCs, which can expand another dimension of the IΣΔ ADC world that is dominated by discrete-time implementations. Several CT IΣΔ ADC architectures are introduced and analyzed aiming to reduce the modulators sampling frequency and consequently the power dissipation. Based on the analytical results, three CT IΣΔ ADCs are selected to be examined, implemented, and tested. The three ADC prototypes, fabricated in a standard 0.18-m CMOS technology, demonstrate competitive figure-of-merits in terms of power efficiency compared to the state-of-the-art counterparts. Copyright

Vehicle location using wireless wide area network

The ability to locate the position of a mobile device has emerged as a key facility of existing and future generation mobile systems. Many value added location based services have been enabled by this feature. Due to the pervasive deployments of mobile communication technologies, vehicle positioning by locating the drivers mobile devices has become feasible. Moreover, it potentially displaces systems that were designed specifically for vehicles. This paper first investigates the standard mobile location methods and exploits this information in intelligent transportation systems, especially in vehicle location. Hybrid solutions are then proposed based on the well-established and standardized location methods.

Design considerations for pipelined continuous-time incremental Sigma-Delta ADCs

This paper addresses design considerations for power-efficient pipelined continuous-time (CT) incremental Sigma-Delta (IΣΔ) ADC architectures. By pipelining identical CT IΣΔ ADC stages, the proposed architecture provides the design freedom coming from both the pipeline ADC and the IΣΔ ADC. In searching for a low-power solution given a target resolution, different configurations are examined analytically and simulated using behavioral models. For further power reduction, power-efficient circuits are proposed to implement the active blocks in each configuration. Based on the architecture-level analysis, a configuration that leads to minimum power-area consumption is chosen and implemented as a test-case using the proposed circuit blocks. Post-layout simulations show that the test-case ADC, with 3.2-kHz bandwidth, achieves a peak SNDR of 82.5-dB while dissipating a total power of 18.27-μW.