Repository logo

Browsing by Author "Luo, Jie, committee member"

Now showing 1 - 7 of 7
  • Thumbnail Image
    ItemOpen Access
    Autonomous trucks as a scalable system of systems: development, constituent systems communication protocols and cybersecurity
    (Colorado State University. Libraries, 2024) Elhadeedy, Ahmed, author; Daily, Jeremy, advisor; Chong, Edwin, committee member; Papadopoulos, Christos, committee member; Luo, Jie, committee member
    Driverless vehicles are complex to develop due to the number of systems required for safe and secure autonomous operation. Autonomous vehicles embody the definition of a system of systems as they incorporate several systems to enable functions like perception, decision-making, vehicle controls, and external communication. Constituent systems are often developed by different vendors globally which introduces challenges during the development process. Additionally, as the fleet of autonomous vehicles scales, optimization of onboard and off-board communication between the constituent systems becomes critical. Autonomous truck and trailer configurations face challenges when operating in reverse due to the lack of sensing on the trailer. It is anticipated that sensor packages will be installed on existing trailers to extend autonomous operations while operating in reverse in uncontrolled environments, like a customer's loading dock. Power Line Communication (PLC) between the trailer and the tractor cannot support high bandwidth and low latency communication. Legacy communications use powerline carrier communications at 9600 baud, so upfitting existing trailers for autonomous operations will require adopting technologies like Ethernet or a wireless harness between the truck and the trailer. This would require additional security measures and architecture, especially when pairing a tractor with a trailer. We proposed tailoring the system of systems Model for autonomous vehicles. The model serves as the governing framework for the development of constituent systems. It's essential for the SoS model to accommodate various development approaches that are used for hardware, and software such as Agile, or Vee models. Additionally, a queuing model for certificates authentication compares the named certificate approach with the traditional approach. The model shows the potential benefits of named certificates when the autonomous vehicles are scaled. We also proposed using named J1939 signals to reduce complexities and integration efforts when multiple on-board or off-board systems request vehicle signals. We discuss the current challenges and threats on autonomous truck-trailer communication when Ethernet or a wireless harness is used, and the impact on the Electronic Control Unit (ECU) lifecycle. In addition to using Named Data Networking (NDN) to secure in-vehicle and cloud communication. Named Data Networking can reduce the complexity of the security of the in-vehicle communication networks where it provides a networking solution with security by design.
  • Thumbnail Image
    ItemOpen Access
    Data mining and spatiotemporal analysis of modern mobile data
    (Colorado State University. Libraries, 2019) Fang, Luoyang, author; Yang, Liuqing, advisor; Jayasumana, Anura P., committee member; Luo, Jie, committee member; Wang, Haonan, committee member
    Modern mobile network technologies and smartphones have successfully penetrated nearly every aspect of human life due to the increasing number of mobile applications and services. Massive mobile data generated by mobile networks with timestamp and location information have been frequently collected. Mobile data analytics has gained remarkable attention from various research communities and industries, since it can broadly reveal the human spatiotemporal mobility patterns from the individual level to an aggregated one. In this dissertation, two types of spatiotemporal modeling with respect to human mobility behaviors are considered, namely the individual modeling and aggregated modeling. As for individual spatiotemporal modeling, location privacy is studied in terms of user identifiability between two mobile datasets, merely based on their spatiotemporal traces from the perspective of a privacy adversary. The success of user identification then hinges upon the effective distance measures via user spatiotemporal behavior profiling. However, user identification methods depending on a single semantic distance measure almost always lead to a large portion of false matches. To improve user identification performance, we propose a scalable multi-feature ensemble matching framework that integrates multiple explored spatiotemporal models. On the other hand, the aggregated spatiotemporal modeling is investigated for network and traffic management in cellular networks. Traffic demand forecasting problem across the entire mobile network is first studied, which is considered as the aggregated behavior of network users. The success of demand forecasting relies on effective modeling of both the spatial and temporal dependencies of the per-cell demand time series. However, the main challenge of the spatial relevancy modeling in the per-cell demand forecasting is the uneven spatial distribution of cells in a network. In this work, a dependency graph is proposed to model the spatial relevancy without compromising the spatial granularity. Accordingly, the spatial and temporal models, graph convolutional and recurrent neural networks, are adopted to forecast the per-cell traffic demands. In addition to demand forecasting, a per-cell idle time window (ITW) prediction application is further studied for predictive network management based on subscribers' aggregated spatiotemporal behaviors. First, the ITW prediction is formulated into a regression problem with an ITW presence confidence index that facilitates direct ITW detection and estimation. To predict the ITW, a deep-learning-based ITW prediction model is proposed, consisting of a representation learning network and an output network. The representation learning network is aimed to learn patterns from the recent history of demand and mobility, while the output network is designed to generate the ITW predicts with the learned representation and exogenous periodic as inputs. Upon this paradigm, a temporal graph convolutional network (TGCN) implementing the representation learning network is also proposed to capture the graph-based spatiotemporal input features effectively.
  • Thumbnail Image
    ItemOpen Access
    Sensing, communications and monitoring for the smart grid
    (Colorado State University. Libraries, 2012) Duan, Dongliang, author; Yang, Liuqing, advisor; Scharf, Louis L., committee member; Luo, Jie, committee member; Song, Rui, committee member
    With the increasing concern for environmental factors, reliability, and quality of service, power grids in many countries are undergoing revolution towards a more distributed and flexible "smart grid." In the development of the envisioned smart grid, situational awareness takes a fundamental role for a number of crucial advanced operations, such as power flow scheduling, dynamic pricing, energy management, wide area control, wide area protection etc. To fulfill the mission of situational awareness across various entities in the grid, more advanced sensing, communications and monitoring techniques need to be introduced to the existing power grid. In this research, we will first address the issue of battery power efficiency (BPE) in a wireless sensor network (WSN) which is essential for the sensing system lifetime. We show that the BPE can be improved either by selecting a more battery-power-efficient modulation format or by developing a cooperative communications scheme. Then, to transmit the sensed data over the scarse wireless bandwidth, we adopt cognitive radio as a possible solution. To enable the cognitive radio communication, we aim at improving both the reliability and efficiency of the overall system via cooperative spectrum sensing. With these fundamental communication capabilities available for the sensed data, we then investigate wide area power grid monitoring based on synchronized measurements from newly developed devices such as phasor measurement units (PMUs), mode meters and so on. In addition, an optimal fusion technique is studied as a good foundation for detection in wireless sensor networks, with application to event detection in the power grid.
  • Thumbnail Image
    ItemOpen Access
    Simultaneous wireless information and power transfer (SWIPT) in cooperative networks
    (Colorado State University. Libraries, 2019) Wang, Dexin, author; Yang, Liuqing, advisor; Chong, Edwin K. P., committee member; Luo, Jie, committee member; Wang, Haonan, committee member
    In recent years, the capacity and charging speed of batteries have become the bottleneck of mobile communications systems. Energy harvesting (EH) is regarded as a promising technology to significantly extend the lifetime of battery-powered devices. Among many EH technologies, simultaneous wireless information and power transfer (SWIPT) proposes to harvest part of the energy carried by the wireless communication signals. In particular, SWIPT has been successfully applied to energy-constrained relays that are mainly or exclusively powered by the energy harvested from the received signals. These relays are known as EH relays, which attract significant attention in both the academia and the industry. In this research, we investigate the performance of SWIPT-based EH cooperative networks and the optimization problems therein. Due to hardware limitations, the energy harvesting circuit cannot decode the signal directly. Power splitting (PS) is a popular and effective solution to this problem. Therefore, we focus on PS based SWIPT in this research. First, different from existing work that employs time-switching (TS) based SWIPT, we propose to employ PS based SWIPT for a truly full-duplex (FD) EH relay network, where the information reception and transmission take place simultaneously at the relay all the time. This more thorough exploitation of the FD feature consequently leads to a significant capacity improvement compared with existing alternatives in the literature. Secondly, when multiple relays are available in the network, we explore the relay selection (RS) and network beamforming techniques in EH relay networks. Assuming orthogonal bandwidth allocation, both single relay selection (SRS) and general relay selection (GRS) without the limit on the number of cooperating relays are investigated and the corresponding RS methods are proposed. We will show that our proposed heuristic GRS methods outperform the SRS methods and achieve very similar performance compared with the optimal RS method achieved by exhaustive search but with dramatically reduced complexity. Under the shared bandwidth assumption, network beamforming among EH relays is investigated. We propose a joint PS factor optimization method based on semidefinite relaxation. Simulations show that network beamforming achieves the best performance among all other cooperative techniques. Finally, we study the problem of power allocation and PS factor optimization for SWIPT over doubly-selective wireless channels. In contrast to existing work in the literature, we take the channel variation in both time and frequency domains into consideration and jointly optimize the power allocation and the PS factors. The objective is to maximize the achievable data rate with constraints on the delivered energy in a time window. Since the problem is difficult to solve directly due to its nonconvexity, we proposed a two-step approach, named joint power allocation and splitting (JoPAS), to solve the problem along the time and frequency dimensions sequentially. Simulations show significantly improved performance compared with the existing dynamic power splitting scheme. A suboptimal heuristic algorithm, named decoupled power allocation and splitting (DePAS), is also proposed with significantly reduced computational complexity and simulations demonstrate its near-optimum performance.
  • Thumbnail Image
    ItemOpen Access
    Spanning sensor resource management
    (Colorado State University. Libraries, 2018) Krakow, Lucas W., author; Chong, Edwin K. P., advisor; Burns, Patrick, committee member; Pezeshki, Ali, committee member; Luo, Jie, committee member
    This paper presents multiple applications of sensor resource management. The general focus entails two chapters on adaptive estimation of time-varying sparse signals and three chapters exploring autonomous control of unmanned aerial vehicles (UAVs) sensor platforms employed for target tracking. All of the included applications are posed as decision control problems formulated in the rigorous framework of a partially observable Markov decision process (POMDP) and solution methods based on Bellman's equation are exercised, generating adaptive control policies for action selections in the given scenarios. Specifically, the rollout optimization method is administered in the cases of signal estimation under the objective of maximizing the information gain about the unknown sparse signal. For the UAV sensor platform control, nominal belief-state optimization (NBO) is employed for control selection for optimizing objectives including target-tracking error, surveillance performance and fuel efficiency. The empirical studies in each investigation present evidence that non-myopic solution methods, accounting for both the immediate and future costs of the current action choices, provide performance gains for these scenarios.
  • Thumbnail Image
    ItemOpen Access
    Sparse representations in multi-kernel dictionaries for in-situ classification of underwater objects
    (Colorado State University. Libraries, 2017) Hosseini, Somayeh, author; Pezeshki, Ali, advisor; Azimi-Sadjadi, Mahmood R., advisor; Chong, Edwin, committee member; Luo, Jie, committee member; Kirby, Michael, committee member
    The performance of the kernel-based pattern classification algorithms depends highly on the selection of the kernel function and its parameters. Consequently in the recent years there has been a growing interest in machine learning algorithms to select kernel functions automatically from a predefined dictionary of kernels. In this work we develop a general mathematical framework for multi-kernel classification that makes use of sparse representation theory for automatically selecting the kernel functions and their parameters that best represent a set of training samples. We construct a dictionary of different kernel functions with different parametrizations. Using a sparse approximation algorithm, we represent the ideal score of each training sample as a sparse linear combination of the kernel functions in the dictionary evaluated at all training samples. Moreover, we incorporate the high-level operator's concepts into the learning by using the in-situ learning for the new unseen samples whose scores can not be represented suitably using the previously selected representative samples. Finally, we evaluate the viability of this method for in-situ classification of a database of underwater object images. Results are presented in terms of ROC curve, confusion matrix and correct classification rate measures.
  • Thumbnail Image
    ItemOpen Access
    Spectrum efficiency for future wireless communications
    (Colorado State University. Libraries, 2015) Yu, Bo, author; Yang, Liuqing, advisor; Luo, Jie, committee member; Morton, Yu, committee member; Wang, Haonan, committee member
    Spectrum efficiency has long been at the center of wireless communication research, development, and operation. Today, it is even more so with the explosive popularity of mobile internet, social networks, and smart phones that are more powerful than our desktops not long ago. As a result, there is an urgent need to further improve the spectrum efficiency in order to provide higher wireless data capacity. To respond to this demand, the 3rd Generation Partnership Project (3GPP) standardized the radio interface specifications for the next generation mobile communications system, called Long Term Evolution (LTE), in Release 8 specifications in 2008. Then the development continued and an enhanced LTE radio interface called LTE-Advanced (LTE-A) was standardized in Release 10 specifications in 2011. In order to ensure the sustainability of 3GPP radio access technologies over the coming decade, 3GPP standardization will need to continue identifying and providing new solutions that can respond to the future challenges. In this research, we investigate the potential technologies for further spectrum efficiency enhancement in the future steps of the standardization. One key direction is the further enhancement of local area technologies, which play a more and more important role in complementing the wide area networks. Specifically, we investigate two promising techniques for spectrum efficiency improvement in a macro-assisted small cell architecture, called the Phantom cell, which is proposed by DOCOMO. One is the possibility of dynamic allocation of subframes to uplink (UL) or downlink (DL) in time-division duplexing (TDD), called `Dynamic TDD'. The other is the more dynamic and flexible 3-dimensional (3D) beamforming which is facilitated by the adoption of active antenna systems (AAS) in BSs. In addition, full-duplex transmission and cooperative communication are two promising techniques known to enhance the spectrum efficiency of wireless communications. We focus on applying full-duplex in cooperative relaying networks and investigating the optimal resource allocation (both power and relay location) for full-duplex decode-and-forward (DF) relaying systems for spectrum efficiency enhancement.