University of Colorado Denver, Denver, CO, U.S.
I'm an assistant professor at University of Colorado Denver. My research mainly focuses on cyber-physical systems, IoTs, and mobile computing. Before joining UCD, I worked as a research fellow at The University of Michigan at Ann Arbor, MI, USA, with Prof. Kang G. Shin, as a Research Scientist at Singapore University of Technology and Design, Singapore, with Dr. Yu (Jason) Gu, and as a research assistant at University of Victoria, Canada, with Prof. Jianping Pan. I'm a senior member of IEEE and a member of ACM.
I'm looking for students/postdocs to explore the cyber-physical world together. Contact me if you are interested!
The best way to reach is through my email at: email@example.com, or just stop by my office at Lawrence St. Center, LW816.
University of Colorado Denver, Denver, CO, U.S.
University of Michigan, Ann Arbor, MI, U.S.
Singapore University of Technology and Design, Singapore
University of Victoria, Victoria, BC, Canada
Ph.D. in Computer Science
Nankai University, Tianjin, China
B.Eng. in Computer Science
Tianjin University, Tianjin, China
My current research mainly focues on cyber-physical systems and Internet-of-Things, such as energy/power systems and ground/space vehicles. We exploit methodologies such as physical modeling and data analytics to diagnose/predict/optimize system performance. In the past, I made research contributions to mobile computing as well as wireless communications and networks. My research is in close collaboration with commercial companies such as General Motors and Microsoft.
Please click on each project to find more details.
In this project, we aim to improve the reliability and safety of vehcile systems by designing and implementing novel cyber-physical solutions for on-board diagnostics. This project is in close collaboration with General Motors, SETI Institute, Microsoft Research, and the HyberLynx team at CUDenver. This project is partly supported by the Comcast Media and Technology Center at CUDenver.
Related Publications at: MobiCom'19.
The prevalence of battery-powered systems such as electric vehicles, smartphones, and IoT devices has made batteries crucial to everyone’s daily life and business. Battery health, however, degrades over time, not only decreasing system reliability such as unexpected system shutoffs, but also causing overheating/gassing which, in turn, increases safety risks such as thermal runaway or even battery fire/explosion. To address these problems, we must monitor, prognose, and optimize battery health throughout the physical system life. However, existing battery management systems (BMSes) are usually treated as complementary system components attached/embedded to/in batteries, and are unable to make optimal health management decisions adaptively based on system dynamics or user requirements. Our approach tightly integrates the cyber (battery management software) and the physical (sensing of battery state) to enable significant improvements in battery life and performance. Specifically, we will develop R-AWARE, a recovery period-assisted battery health management that schedules system operation while considering both system/user requirements and battery health. R-AWARE will improve battery health via relaxation-aware battery scheduling of battery charging/discharging, and recovery-based thermal control. It will advance the science of CPS by uncovering a thorough understanding of battery recovery and exploiting it via a recovery-aware scheduler during system operation. This project is partly supported by NSF under CNS-1739577.
Related Publications at: ICCPS'19, TEC'18, MobiSys'17, TMC'17, ICCPS'17, ICCPS'16.
Recent progress in battery technology has made it possible to use batteries to power various physical platforms, such as ground/air/water vehicles. These platforms require hundreds/thousands of battery cells to meet their power and energy needs. Of these, automobiles, locomotives, and unmanned air vehicles (UAVs) face the most stringent environmental challenges. In particular, and of special importance to the automotive industry, is the transition from conventional powertrains to (plug-in) hybrid and electric vehicles (EVs), all of which are subject to environmental and operational variations. Current state-of-the-art still needs significant improvements in the architecture and algorithms of battery management before achieving the desired levels of efficiency and performance. To meet this need, we aim to (i) design a dynamically reconfigurable energy storage system to withstand harsh internal and external stresses; (ii) develop cell-level thermal management algorithms; (iii) develop efficient, dependable charge and discharge scheduling algorithms; (iv) develop comprehensive, diagnostic/prognostic algorithms with system parameters adjusted for making optimal decisions; and (v) build a testbed, implement and evaluate the proposed architecture and algorithms on the testbed. This project is supported by NSF under CNS-1446117.
Related Publications at: TOSN'18, TSG'18, TCPS'17, TSG'16, RTSS'16, e-Energy'16, ICCPS'15, ICCPS'14, RTSS'13.
In this project, we aim to effectively exploit the limited mobility in wireless sensor networks --- e.g., via robot, drones, and human beings --- to improve the information gathering in the network and replenish the energy supply of nodes thereof. We achieve these objectives with a combination of mathematical modeling, algoroihtm design, and system implementation.
Related Publications at: TVT'19, TVT'16, TMC'15, TITS'15, TMC'14, MOBIHOC'14, INFOCOM'14, TMC'13, INFOCOM'12.
Charging mobile devices “fast” has been the focus of both industry and academia, leading to the deployment of various fast charging technologies. However, existing fast charging solutions are agnostic of users’ available time for charging their devices, causing early termination of the intended/- planned charging. This, in turn, accelerates the capacity fading of device battery and thus shortens the device operation. In this paper, we propose a novel user-interactive charging paradigm, called iCharge, that tailors the device charging to the user’s real-time availability and need. The core of iCharge is a relaxation-aware (R-Aware) charging algorithm that maximizes the charged capacity within the user’s available time and slows down the battery’s capacity fading. iCharge also integrates R-Aware with existing fast charging algorithms via a user-interactive interface, allowing users to choose a charging method based on their availability and need. We evaluate iCharge via extensive laboratory experiments and field-tests on Android phones, as well as user studies. R-Aware is shown to slow down the battery fading by more than 36% on average, and up to 60% in extreme cases, when compared to existing fast charging algorithms. This slowdown of capacity fading translates to, for instance, an up to 2-hour extension of the LTE time for a Nexus 5X phone after its use for 2 years, according to our trace-driven analysis of 976 device charging cases of 7 users over 3 months.
Insucient support of electric current sensing on commodity mobile devices leads to inaccurate estimation of their battery’s stateof-health (SoH), which, in turn, shuts them off unexpectedly and accelerates their battery fading. In this paper, we design V-BASH, a new battery SoH estimation method based only on their voltages and is compatible to commodity mobile devices. V-BASH is inspired by the physical phenomenon that the relaxing battery voltages correlate to battery SoH. Moreover, it is enabled on mobile devices with a common usage battery of most users frequently taking a long time to charge their devices. The design of V-BASH is guided by 2, 781 empirically collected relaxing voltage traces with 19 mobile device batteries. We evaluate V-BASH using both laboratory experiments and eld tests on mobile devices, showing a <6% error in SoH estimation.
Cell imbalance commonly found in large battery packs degrades their capacity delivery, especially for cells connected in series where the weakest cell dominates their overall capacity. In this paper, we present a case study of exploiting system reconfiguration to mitigate the cell imbalance in battery packs. Specifically, instead of using all the cells in a battery pack to support the load, selectively skipping cells to be discharged may actually enhance the pack’s capacity delivery. Based on this observation, we propose CSR, a Cell Skipping-assisted Reconfiguration algorithm that identifies the system configuration with (near)-optimal capacity delivery. We evaluate CSR using large-scale emulation based on empirically collected discharge traces of 40 Lithium-ion cells. CSR is shown to achieve close-to-optimal capacity delivery when the cell imbalance in the battery pack is low and improve the capacity delivery by up to 94% in case of high imbalance.
Lithium-ion cells are widely used in various platforms, such as electric vehicles (EVs) and mobile devices. Complete and fast charging of cells has always been the goal for sustainable system operation. However, fast charging is not always the best solution, especially in view of a new finding that cells need to rest/relax after being charged with high current to avoid accelerated capacity fading. Fast charging for its typical Charge-and-Go scenario does not allow this needed relaxation. In this paper, we propose ∗-Aware, a novel charging algorithm which maximizes the charged capacity within the user-specified available charging time (i.e., user-awareness) while ensuring enough relaxation (i.e., cell-awareness). We motivate and evaluate ∗-Aware via extensive measurements over 10 months. ∗-Aware is shown to improve the charged capacity by 6.9–50.5% over other charging algorithms that also ensure relaxation, and by almost 3x in some extreme cases. Furthermore, ∗-Aware slows down the capacity fading by 49.55% when compared to fast charging.
Unbalanced battery cells are known to significantly degrade the performance and reliability of a large-scale battery system. In this paper, we exploit emerging reconfigurable battery packs to mitigate the cell imbalance via the joint consideration of system reconfigurability and State-of-Health (SoH) of cells. Via empirical measurements and validation, we observe that a significantly larger amount of capacity can be delivered when cells with similar SoH levels are connected in series during discharging, which in turn extends the system operation time. Based on this observation, we propose two SoH-aware reconfiguration algorithms focusing on fully and partially reconfigurable battery packs, and prove their (near) optimality. We evaluate the proposed SoH-aware reconfiguration algorithms using both experiments and simulations. The algorithms are shown to deliver about 10–30% more capacity than SoH-oblivious configuration approaches.
Different from energy harvesting which generates dynamic energy supplies, the mobile charger is able to provide stable and reliable energy supply for sensor nodes, and thus enables sustainable system operations. While previous mobile charging protocols either focus on the charger travel distance or the charging delay of sensor nodes, in this work we propose a novel Energy Synchronized Charging (ESync) protocol, which simultaneously reduces both of them. Observing the limitation of the Traveling Salesman Problem (TSP)-based solutions when nodes energy consumptions are diverse, we construct a set of nested TSP tours based on their energy consumptions, and only nodes with low remaining energy are involved in each charging round. Furthermore, we propose the concept of energy synchronization to synchronize the charging requests sequence of nodes with their sequence on the TSP tours. Experiment and simulation demonstrate ESync can reduce charger travel distance and nodes charging delay by about 30% and 40% respectively.
Recently, much research effort has been devoted to employing mobile chargers for energy replenishment of the robots in robotic sensor networks. Observing the discrepancy between the charging latency of robots and charger travel distance, we propose a novel tree-based charging schedule for the charger, which minimizes its travel distance without causing the robot energy depletion. We analytically evaluate its performance and show its closeness to the optimal solutions. Furthermore, through a queue-based approach, we provide theoretical guidance on the setting of the remaining energy threshold at which the robots request energy replenishment. This guided setting guarantees the feasibility of the tree-based schedule to return a depletion-free charging schedule. The performance of the tree-based charging schedule is evaluated through extensive simulations. The results show that the charger travel distance can be reduced by around 20%, when compared with the schedule that only considers the robot charging latency.
Large-scale battery packs with hundreds/thousands of battery cells are commonly adopted in many emerging cyberphysical systems such as electric vehicles and smart micro-grids. For many applications, the load requirements on the battery systems are dynamic and could significantly change over time. How to resolve the discrepancies between the output power supplied by the battery system and the input power required by the loads is key to the development of large-scale battery systems. Traditionally, voltage regulators are often adopted to convert the voltage outputs to match loads’ required input power. Unfortunately, the efficiency of utilizing such voltage regulators degrades significantly when the difference between supplied and required voltages becomes large or the load becomes light. In this paper, we propose to address this problem via an adaptive reconfiguration framework for the battery system. By abstracting the battery system into a graph representation, we develop two adaptive reconfiguration algorithms to identify the desired system configurations dynamically in accordance with real-time load requirements. We extensively evaluate our design with empirical experiments on a prototype battery system, electric vehicle driving trace-based emulation, and battery discharge trace-based simulations. The evaluation results demonstrate that, depending on the system states, our proposed adaptive reconfiguration algorithms are able to achieve 1× to 5× performance improvement with regard to the system operation time.
The introduction of mobile elements in wireless sensor networks creates a new dimension to reduce and balance the energy consumption for resource-constrained sensor nodes; however, it also introduces extra latency in the data collection process due to the limited mobility of mobile elements. Therefore, how to arrange and schedule the movement of mobile elements throughout the sensing field is of ultimate importance. In this paper, the online scenario where data collection requests arrive progressively is investigated, and the data collection process is modeled as an M/G/1/c–NJN queuing system, where NJN stands for nearest-job-next, a simple and intuitive service discipline. Based on this model, the performance of data collection is evaluated through both theoretical analysis and extensive simulation. The NJN discipline is further extended by considering the possibility of requests combination (NJNC). The simulation results validate our analytical models and give more insights when comparing with the first-come-first-serve (FCFS) discipline. In contrast to the conventional wisdom of the starvation problem, we reveal that NJN and NJNC have a better performance than FCFS, in both the average and more importantly the worst cases, which gives the much needed assurance to adopt NJN and NJNC in the design of more sophisticated data collection schemes for mobile elements in wireless ad hoc sensor networks, as well as many other similar scheduling application scenarios.
Cell imbalance, a notorious but widely found issue, degrades the performance and reliability of large battery packs, especially for cells connected in series where their overall capacity delivery is dominated by the weakest cell. In this paper, we exploit the emerging reconfigurable battery packs to mitigate the cell imbalance via the joint consideration of system reconfigurability and State-of-Health (SoH) of cells. Via empirical measurements and validation, we observe that more capacity can be delivered when cells with similar SoH are connected in series during discharging. Based on this observation, we propose two SoHaware reconfiguration algorithms focusing on fully and partially reconfigurable battery packs, and prove their (near) optimality in capacity delivery. We evaluate the proposed reconfiguration algorithms analytically, experimentally, and via emulations, showing 10–60% improvement in capacity delivery when compared with SoH-oblivious approaches, especially when facing severe cell imbalance.
Cell imbalance in large battery packs degrades their capacity delivery, especially for cells connected in series where the weakest cell dominates their overall capacity. In this article, we present a case study of exploiting system reconfigurations to mitigate the cell imbalance in battery packs. Specifically, instead of using all the cells in a battery pack to support the load, selectively skipping cells to be discharged may actually enhance the pack’s capacity delivery. Based on this observation, we propose CSR, a Cell Skipping-assisted Reconfiguration algorithm that identifies the system configuration with (near)-optimal capacity delivery. We evaluate CSR using large-scale emulation based on empirically collected discharge traces of 40 lithium-ion cells. CSR achieves close-to-optimal capacity delivery when the cell imbalance in the battery pack is low and improves the capacity delivery by about 20% and up to 1x in the case of a high imbalance.
Significant research has been devoted to reduce the energy consumption of mobile devices, but how to increase their energy supply has received far less attention. Moreover, reducing the energy consumption alone does not always extend the device operation time due to a unique battery property — the capacity it delivers hinges critically upon how it is discharged. In this paper, we propose B-MODS, a novel design of battery-aware mobile data service on mobile devices. B-MODS constructs battery-friendly discharge patterns utilizing the recovery effect so as to increase the capacity delivered from batteries while meeting data service requirements. We implement B-MODS as an application layer library on the Android platform. Our experiments with diverse mobile devices under various application scenarios have shown that B-MODS increases the capacity delivery from the battery by up to 49.5%, with which an increase in the user-perceived data service utilities of up to 28.6% is observed.
Large-scale lithium-ion battery packs are widely adopted in systems such as electric vehicles and energy backup in power grids. Due to factors such as manufacturing difference and heterogeneous discharging conditions, cells in the battery pack may have different statuses, such as diverse voltage levels. This cell diversity is commonly known as the cell imbalance issue. For the charging of battery packs, the cell imbalance not only early on terminates the charging process before all cells are fully charged, but also leads to different desired charging currents among cells. In this paper, based on the advancement in recon- figurable battery systems, we demonstrate how to utilize system reconfigurability to mitigate the impact of cell imbalance on an efficient charging process. With the proposed reconfigurationassisted charging (RAC), cells in the system are categorized according to their real-time voltages, and the charging process is performed in a category-by-category manner. To charge cells in a given category, a graph-based algorithm is presented to charge cells with their desired charging currents, respectively. We evaluate RAC through both experiments and simulations. The results demonstrate that the RAC increases the capacity charged into cells by about 25% and yields a dramatically reduced variance.
Recently, adopting mobile energy chargers to replenish the energy supply of sensor nodes in wireless sensor networks has gained increasing attention from the research community. Different from energy harvesting systems, the utilization of mobile energy chargers is able to provide more reliable energy supply than the dynamic energy harvested from the surrounding environment. While pioneering works on the mobile recharging problem mainly focus on the optimal offline path planning for the mobile chargers, in this work, we aim to lay the theoretical foundation for the on-demand mobile charging (DMC) problem, where individual sensor nodes request charging from the mobile charger when their energy runs low. Specifically, in this work, we analyze the on-demand mobile charging problem using a simple but efficient Nearest-Job-Next with Preemption (NJNP) discipline for the mobile charger, and provide analytical results on the system throughput and charging latency from the perspectives of the mobile charger and individual sensor nodes, respectively. To demonstrate how the actual system design can benefit from our analytical results, we present two examples on determining the essential system parameters such as the optimal remaining energy level for individual sensor nodes to send out their recharging requests and the minimal energy capacity required for the mobile charger. Through extensive simulation with real-world system settings, we verify that our analytical results match the simulation results well and the system designs based on our analysis are effective.
Mobility-assisted data collection in sensor networks creates a new dimension to reduce and balance the energy consumption for sensor nodes. However, it also introduces extra latency in the data collection process due to the limited mobility of mobile elements. Therefore, how to schedule the movement of mobile elements throughout the field is of ultimate importance. In this paper, the on-demand scenario where data collection requests arrive at the mobile element progressively is investigated, and the data collection process is modelled as an M=G=1=c-NJN queuing system with an intuitive service discipline of nearest-job-next (NJN). Based on this model, the performance of data collection is evaluated through both theoretical analysis and extensive simulation. NJN is further extended by considering the possible requests combination (NJNC). The simulation results validate our models and offer more insights when compared with the first-come-first-serve (FCFS) discipline. In contrary to the conventional wisdom of the starvation problem, we reveal that NJN and NJNC have better performance than FCFS, in both the average and more importantly the worst cases, which offers the much needed assurance to adopt NJN and NJNC in the design of more sophisticated data collection schemes, as well as other similar scheduling scenarios.
The introduction of mobile elements has created a new dimension to reduce and balance the energy consumption in wireless sensor networks. However, data collection latency may become higher due to the relatively slow travel speed of mobile elements. Thus, the scheduling of mobile elements, i.e., how they traverse through the sensing field and when they collect data from which sensor, is of ultimate importance and has attracted increasing attention from the research community. Formulated as the traveling salesman problem with neighborhoods (TSPN) and due to its NP-hardness, so far only approximation and heuristic algorithms have appeared in the literature, but the former only have theoretical value now due to their large approximation factors. In this paper, following a progressive optimization approach, we first propose a combine-skip-substitute (CSS) scheme, which is shown to be able to obtain solutions within a small range of the lower bound of the optimal solution. We then take the realistic multirate features of wireless communications into account, which have been ignored by most existing work, to further reduce the data collection latency with the multirate CSS (MR-CSS) scheme. Besides the correctness proof and performance analysis of the proposed schemes, we also show their efficiency and potentials for further extensions through extensive simulation.
I work closely with students of cross-discipline backgrounds at CUDenver.
Abstract: Modern connected and autonomous vehicles (CAVs) are equipped with an increasing number of Electronic Control Units (ECUs) that produce large amounts of data. The data is exchanged between ECUs via an in-vehicle network. Furthermore, CAVs do not only have physical interfaces, but also increased data connectivity to the Internet via their Telematic Control Units (TCUs) which make them accessible remotely just like mobile phones. As a result, an increasing number of attack vectors make vehicles an attractive target for hackers. Automotive cyber-security research is a relatively novel field which tries to respond to constantly rising threats with countermeasures. In this talk, we will give a primer on cyber-security in the automotive domain as well as discuss privacy concerns of the big data generated by cars.
Abstract: For the first time ever, we have more people living in urban areas than in rural areas. Based on this inevitable urbanization, the research in my group is aimed at addressing sustainability challenges related to urban mobility (e.g., energy consumption and traffic congestion) by data-driven modeling and applications with a Cyber-Physical-Systems (CPS) approach in the vision of Smart Cities. In this talk, I will focus on mobility modeling and resultant applications based on large-scale cross-domain CPS, e.g., cellular networks, payment systems, social networks, and transportation systems (including electric vehicles, taxis, buses, subway, private vehicles, Ubers). I will first show how cross-domain CPS systems can be collaboratively utilized to capture real-time urban mobility by a set of model integration techniques. Then I will show how the captured mobility can be used to design various urban mobile services to close the “loop”, from urban-scale ridesharing to for-hire vehicle dispatching, electric toll collection management, electric-vehicle charging recommendation, and emergency response under mobility anomaly. Finally, I will present some research challenges related to future cross- domain CPS in the context of the smart cities research.
Abstract: Autonomous multi-UAV networks will revolutionize the design of systems solutions for civil, entertainment, public services, security and other critical application areas. At NEC Labs, we are particularly interested in the application of such multi-UAV networks to public safety networks. Our goal is simple: how do we design, implement and deploy a reliable and flexible multi-UAV network that can meet the demands of first-responder networks? In this talk, I will introduce SkyLiTE, our multi-UAV network that is designed for public safety applications. SkyLiTE is one of the first efforts at a fully autonomous, untethered multi-UAV network that can be deployed on demand in challenging scenarios to achieve public safety mission objectives. Our SkyLiTE prototype consists of three important components: SkyHAUL, which is a high bandwidth, millimeter wave backhaul network that achieves 1Gbps of bandwidth across the entire multi-UAV network; SkyCORE, a UAV-optimized, lightweight LTE Evolved Packet Core network; and SkyRAN, a full-fledged LTE radio access network. SkyLiTE is deployed either to complement and augment existing terrestrial cellular networks in public safety scenarios, or as a standalone LTE network in areas not covered by fixed infrastructure.
Abstract: Mobile navigation services are used by billions of users around globe today. While GPS spoofing is a known threat, it is not yet clear if spoofing attacks can truly manipulate road navigation systems. In this work, we explore the feasibility of a stealthy manipulation attack against road navigation systems. The goal is to trigger the fake turn-by-turn navigation to guide the victim to a wrong destination without being noticed. Our key idea is to slightly shift the GPS location so that the fake navigation route matches the shape of the actual roads and trigger physically possible instructions. To demonstrate the feasibility, we perform controlled measurements by implementing a portable GPS spoofer and testing on real cars. The complete attack is validated by extensive trace-driven simulation and real-world driving tests. Deceptive user studies using a driving simulator also show that 95% of the participants follow the navigation to the wrong destination without recognizing the attack.
Abstract: Vision processing is a key component in automated driving systems. Most vision processing algorithms today, however, are not designed for safety-critical real-time applications. Considering these algorithms are typically both data- and computation-intensive and require advanced hardware platform, applying them to automated driving brings new architecture challenges. In this talk, I will discuss some architecture challenges and potential solutions to support the vision processing for real-time vehicle controls.
Abstract: Graphic processing units (GPUs) have seen wide-spread use in several computing domains as they have the power to enable orders of magnitude faster and more energy-efficient execution of many applications. Unfortunately, it is not straightforward to reliably adopt GPUs in many safety-critical cyber-physical systems that require predictable timing correctness, one of the most important tenets in certification required for such systems. A key example is the advanced automotive system where timeliness of computations is an essential requirement of correctness due to the interaction with the physical world. In this talk, I will describe several system-level and algorithmic challenges and our developed solutions on ensuring predictable timing correctness in GPU-accelerated systems.
Abstract: During the past decade, we are moving swiftly towards a mobile-centered world. This thriving mobile ecosystem builds upon the interplay of three important parties: the mobile user, OS, and app. These parties interact via designated interfaces many of which are newly invented for or introduced to the mobile platform. Nevertheless, as these new ways of interactions arise in the mobile ecosystem, what is enabled by these communication interfaces often violates the expectations of the communicating parties. This shakes the foundation of the mobile ecosystem and results in significant security and privacy hazards. In this talk, we describe our attempts to fill this gap by: 1.) securing the conversations between trusted parties, 2.) regulating the interactions between partially trusted parties, and 3.) defending the communications between untrusted parties. First, we deal with the case of two opposing parties, mobile OS and app, and analyze the Inter-Process Communication protocol (Binder) between them. We found that the OS is frequently making unrealistic assumptions on the validity (sanity) of transactions from apps, thus creating significant security hazards. We analyzed the root cause of this emerging attack surface and secured this interface by developing effective precautionary testing framework and runtime diagnostic tool. Then, we study the deficiency of how existing mobile user interact with app, a party he can only partially trust. We found that in the current mobile ecosystem, information about the same user in different apps can be easily shared and aggregated, which clearly violates the conditional trust mobile user has on each app. We address this issue by providing an OS-level extension that allows the user to track and control, during runtime, the potential flow of his information across apps. Last, we elaborate on how to secure the voice interaction channel between two trusted parties, mobile user and OS. The open nature of the voice channel makes applications that depend on voice interactions, such as voice assistants, difficult to secure and exposed to various attacks. We solve this problem by proposing the first system that provides continuous and usable authentication for voice commands. It takes advantage of the neck-surface acceleration to filter only those commands that originate from the voice of the owner.