Shijia Pan

BOES: Building Occupancy Estimation System using sparse ambient vibration monitoring

In this paper, we present a room-level building occupancy estimation system (BOES) utilizing low-resolution vibration sensors that are sparsely distributed. Many ubiquitous computing and building maintenance systems require fine-grained occupancy knowledge to enable occupant centric services and optimize space and energy utilization. The sensing infrastructure support for current occupancy estimation systems often requires multiple intrusive sensors per room, resulting in systems that are both costly to deploy and difficult to maintain. To address these shortcomings, we developed BOES. BOES utilizes sparse vibration sensors to track occupancy levels and activities. Our system has three major components. 1) It extracts features that distinguish occupant activities from noise prone ambient vibrations and detects human footsteps. 2) Using a sequence of footsteps, the system localizes and tracks individuals by observing changes in the sequences. It uses this tracking information to identify when an occupant leaves or enters a room. 3) The entering and leaving room information are combined with detected individual location information to update the room-level occupancy state of the building. Through validation experiments in two different buildings, our system was able to achieve 99.55% accuracy for event detection, less than three feet average error for localization, and 85% accuracy in occupancy counting.

Indoor Person Identification through Footstep Induced Structural Vibration

Person identification is crucial in various smart building applications, including customer behavior analysis, patient monitoring, etc. Prior works on person identification mainly focused on access control related applications. They achieve identification by sensing certain biometrics with specific sensors. However, these methods and apparatuses can be intrusive and not scalable because of instrumentation and sensing limitations. In this paper, we introduce our indoor person identification system that utilizes footstep induced structural vibration. Because structural vibration can be measured without interrupting human activities, our system is suitable for many ubiquitous sensing applications. Our system senses floor vibration and detects the signal induced by footsteps. Then the system extracts features from the signals that represent characteristics of each persons gait pattern. With the extracted features, the system conducts hierarchical classification at an individual step level and then at a trace (i.e., collection of consecutive steps) level. Our system achieves over 83% identification accuracy on average. Furthermore, when the application requires different levels of accuracy, our system can adjust confidence level threshold to discard uncertain traces. For example, at a threshold that allows only most certain 50% traces for classification, the identification accuracy increases to 96.5%.

Polaris: getting accurate indoor orientations for mobile devices using ubiquitous visual patterns on ceilings

Ubiquitous computing applications commonly use digital compass sensors to obtain orientation of a device relative to the magnetic north of the earth. However, these compass readings are always prone to significant errors in indoor environments due to presence of metallic objects in close proximity. Such errors can adversely affect the performance and quality of user experience of the applications utilizing digital compass sensors. In this paper, we propose Polaris, a novel approach to provide reliable orientation information for mobile devices in indoor environments. Polaris achieves this by aggregating pictures of the ceiling of an indoor environment and applies computer vision based pattern matching techniques to utilize them as orientation references for correcting digital compass readings. To show the feasibility of the Polaris system, we implemented the Polaris system on mobile devices, and field tested the system in multiple office buildings. Our results show that Polaris achieves 4.5° average orientation accuracy, which is about 3.5 times better than what can be achieved through sole use of raw digital compass readings.

PANDAA: physical arrangement detection of networked devices through ambient-sound awareness

Future ubiquitous home environments can contain 10s or 100s of devices. Ubiquitous services running on these devices (i.e. localizing users, routing, security algorithms) will commonly require an accurate location of each device. In order to obtain these locations, existing techniques require either a manual survey, active sound sources, or estimation using wireless radios. These techniques, however, need additional hardware capabilities and are intrusive to the user. Non-intrusive, automatic localization of ubiquitous computing devices in the home has the potential to greatly facilitate device deployments. This paper presents the PANDAA system, a zero-configuration spatial localization system for networked devices based on ambient sound sensing. After initial placement of the devices, ambient sounds, such as human speech, music, foot- steps, finger snaps, hand claps, or coughs and sneezes, are used to autonomously resolve the spatial relative arrangement of devices using trigonometric bounds and successive approximation. Using only time difference of arrival measurements as a bound for successive estimations, PANDAA is able to achieve an average of 0.17 meter accuracy for device location in the meeting room deployment.

Characterizing Wave Propagation to Improve Indoor Step-Level Person Localization using Floor Vibration

The objective of this paper is to characterize frequency-dependent wave propagation of footstep induced floor vibration to improve robustness of vibration-based occupant localization. Occupant localization is an essential part of many smart structure applications (e.g., energy management, patient/customer tracking, etc.). Exist- ing techniques include visual (e.g. cameras and IR sensors), acoustic, RF, and load-based approaches. These approaches have many deployment and operational requirements that limits their adaptation. To overcome these limitations, prior work has utilized footstep-induced vibrations to allow sparse sensor configuration and non-intrusive detection. However, frequency dependent propagation characteristics and low signal-to-noise ratio (SNR) of footstep-induced vibrations change the shape of the signal. Furthermore, estimating the wave propagation velocity for forming the multilateration equations and localizing the footsteps is a challenging task. They, in turn, lead to large errors of localization. In this paper, we present a structural vibration based indoor occupant localization technique using improved time-difference-of-arrival between multiple vibration sensors. In particular we overcome signal distortion by decomposing the signal into frequency components and focusing on high energy components for accurate indoor localization. Such decomposition leverages the frequency-specific propagation characteristics and reduces the effect of low SNR (by choosing the components of highest energy). Furthermore, we develop a velocity calibration method that finds the optimal velocity which minimizes the localization error. We validate our approach through field experiments in a building with human participants. We are able to achieve an average localization error of less than 0.21 meters, which corresponds to a 13X reduction in error when compared to the baseline method using raw data.

SugarTrail: Indoor navigation in retail environments without surveys and maps

A system that helps people navigate in indoor environments on a fine-grained level can enable a variety of pervasive computing applications in retail environments. Existing indoor navigation systems rely on extensive RF tagging surveys and accurate floor plans. These prerequisites are often impractical in indoor environments. In this paper, we present SugarTrail, a system for indoor navigation assistance in retail environments that minimizes the need for active tagging and does not require existing maps. By leveraging the structured movement patterns of shoppers in retail store environments, the system provides higher accuracy than existing radio finger-printing approaches. With minimal setup and active user participation, the system automatically learns user movement pathways in indoor environments from radiofrequency and magnetic signatures. These pathways are clustered and used to automatically build a navigable virtual roadmap of the environment. We present results from a campus testbed and from actual radio measurements collected in an operational supermarket to show that SugarTrail system can navigate users with a success rate of > 85% and an average accuracy of 0.7m.

Headio: zero-configured heading acquisition for indoor mobile devices through multimodal context sensing

Heading information becomes widely used in ubiquitous computing applications for mobile devices. Digital magnetometers, also known as geomagnetic field sensors, provide absolute device headings relative to the earths magnetic north. However, magnetometer readings are prone to significant errors in indoor environments due to the existence of magnetic interferences, such as from printers, walls, or metallic shelves. These errors adversely affect the performance and quality of user experience of the applications requiring device headings. In this paper, we propose Headio, a novel approach to provide reliable device headings in indoor environments. Headio achieves this by aggregating ceiling images of an indoor environment, and by using computer vision-based pattern detection techniques to provide directional references. To achieve zero-configured and energy-efficient heading sensing, Headio also utilizes multimodal sensing techniques to dynamically schedule sensing tasks. To fully evaluate the system, we implemented Headio on both Android and iOS mobile platforms, and performed comprehensive experiments in both small-scale controlled and large-scale public indoor environments. Evaluation results show that Headio constantly provides accurate heading detection performance in diverse situations, achieving better than 1 degree average heading accuracy, up to 33X improvement over existing techniques.

Occupant traffic estimation through structural vibration sensing

The number of people passing through different indoor areas is useful in various smart structure applications, including occupancy-based building energy/space management, marketing research, security, etc. Existing approaches to estimate occupant traffic include vision-, sound-, and radio-based (mobile) sensing methods, which have placement limitations (e.g., requirement of line-of-sight, quiet environment, carrying a device all the time). Such limitations make these direct sensing approaches difficult to deploy and maintain. An indirect approach using geophones to measure floor vibration induced by footsteps can be utilized. However, the main challenge lies in distinguishing multiple simultaneous walkers by developing features that can effectively represent the number of mixed signals and characterize the selected features under different traffic conditions. This paper presents a method to monitor multiple persons. Once the vibration signals are obtained, features are extracted to describe the overlapping vibration signals induced by multiple footsteps, which are used for occupancy traffic estimation. In particular, we focus on analysis of the efficiency and limitations of the four selected key features when used for estimating various traffic conditions. We characterize these features with signals collected from controlled impulse load tests as well as from multiple people walking through a real-world sensing area. In our experiments, the system achieves the mean estimation error of ±0.2 people for different occupant traffic conditions (from one to four) using k-nearest neighbor classifier.

Step-level person localization through sparse sensing of structural vibration

We describe a step-level indoor localization system which uses the ground vibration induced by human footsteps. Indoor localization is important for various smart building applications, including resources arrangement optimization, patient/customer tracking, etc. Geophones are used to measure the ground vibrations and time difference of arrival (TDoA) for different sensors are used to solve the multilateration localization problem. The advantages of this system include its sparsity and also its stability over time. Lesser dependency on instrument people is another upside of this system. The results of pilot tests show that this system can be successfully used for indoor localization.

Robust Occupant Detection Through Step-Induced Floor Vibration by Incorporating Structural Characteristics

The objective of this paper is to present an occupant detection method through step-induced structural vibration. Occupant detection enables various smart building applications such as space/energy management. Ambient structural vibration monitoring provides a non-intrusive sensing approach to achieve that. The main challenges for structural vibration based occupant footstep detection include that (1) the ambient structural vibration noise may overwhelm the step-induced vibration and (2) there are various other impulse-like excitations that look similar to footstep excitations in the sensing environment (e.g., door closing, chair dragging, etc.), which increase the false alarm rate for occupant detection. To overcome these challenges, a two-stage step-induced signal detection algorithm is developed to (1) incorporate the structural characteristics by selecting the dominant frequencies of the structure to increase the signal-to-noise ratio in the vibration data and thus improve the detection performance and (2) perform footstep classification on detected events to distinguish step-induced floor vibrations from other impulse excitations. The method is validated experimentally in two different buildings with distinct structural properties and noise characteristics, Carnegie Mellon University (CMU) campus building and Vincentian Nursing Home deployments in Pittsburgh, PA. The occupant footstep detection F1 score shows up to 4X reduction in detection error compared to traditional thresholding method.