Search

In this article, we establish a novel framework for a multi-user mobile edge computing (MEC) network in which a set of users with high downlink rate demands and a set of users with intensive computation tasks can collaborate to achieve a mutually-beneficial scenario such that completion time of the tasks is reduced and the base station (BS) can send more information at a higher rate to the downlink users. Specifically, by leveraging non-orthogonal multiple access (NOMA) for uplink and downlink traffic, the user with the computation task can offload shares of the computation task to the edge cloud and the downlink user. At the same time, this user forwards the information it receives from the BS to the downlink user. In this set up, we jointly optimize the communication resources, computational resources at the edge cloud and user devices, pairings among the two sets of users, the shares of computation tasks, and relay bits to minimize the total task completion time while satisfying downlink users’ incentive requirements. For a network with a single computation demanding user and a single downlink user, the optimal solution to the problem is provided. For a network with multiple users, the problem is non-convex and computationally challenging. Hence, we propose an efficient, low complexity algorithm that utilizes the bottleneck matching algorithm, convex optimization, and the block coordinate descent scheme to obtain a locally-optimal solution. Simulation results demonstrate that, as compared with the state-of-the-art, the total task completion time is greatly reduced (32%–51%), and a large computational energy savings at the edge cloud (38%–55%) is achieved. Simultaneously, the downlink users’ rates improve compared to the orthogonal transmission. [ABSTRACT FROM AUTHOR]

Night time vehicle detection and tracking using traditional visible light cameras is a challenging task due to the limited visibility. The current state-of-the-art systems treat the vehicles at night time as paired vehicle headlights or taillights, with no ability to determine the contour of the vehicle or its spatial occupancy. Therefore, this paper proposes the first night time framework that combines the vehicle headlights and taillights to localize the vehicle contours. This new framework includes a novel multi-camera vehicle representation that groups and reconstructs vehicle headlights and taillights following mutual geometric distances between different vehicle components. This novel vehicle contour representation successfully removes duplicated vehicle lights and also compensates for the missing vehicle lights in the detection process. Eventually, vehicle headlight alignment and contour adjustment are used to further refine the vehicle contours. The proposed multi-camera system considers typical four-wheel vehicles, e.g., cars and SUVs, in the monitoring and might not be able to handle large trucks, e.g., 18-wheelers. The experiments are conducted on night time traffic videos under various scenarios and the proposed system attains an average of 0.896 in Multiple Object Tracking Accuracy (MOTA) and an average of 0.904 in Jaccard Coefficient (JC), which indicates 19.2% and 15.9% increases over the baseline system. The night time traffic datasets used in this paper are available at https://github.com/JustMeZXX/Intelligent-Night-Time-Traffic-Surveillance [ABSTRACT FROM AUTHOR]

At ultra-low altitudes, an unmanned aerial vehicle (UAV) can act as a personal base station where it communicates only with one user. User equipment (UE) can be inside the pocket of a user or near their chest while facing or facing-away from the flying base station. In these scenarios, the wireless channel can go through different fading levels, depending on the UAV's location, user orientation, the location of the UE near the user's body, and the frequency of the transmitted signal. In this work, we provide measurement results and investigate how the human body affects the Air-to-Ground (AtG) channel characteristics under various use cases of holding a device. These channel characteristics include the average signal strength, shadowing, multipath, and the Rician K-factor. We target three different use cases of holding the device: Near-Chest Facing, In-pocket Facing, and Near-Chest Facing-away from the transmitting UAV. We perform this study at carrier frequencies of 900 MHz and 2.5 GHz and in Line-of-Sight conditions. First, we conduct a set of baseline experiments to understand ground-to-ground and AtG channels in free space. Second, we conduct AtG experiments with the user holding the device and show that the human body can induce either gains or losses compared to free space, depending on its relative orientation to the UAV. Third, we find that there are two distinct regions of operation, one in which the channel characteristics are mainly affected by the UAV and another that is dominated by the user's body. Fourth, we address the time-varying nature of the K-factor and analyze the user's body impact on its mean and standard deviation. We find that if a user changes their orientation to face away from the transmitting UAV, their body can cause extreme fluctuations in the K-factor value over time and reduce its mean value by an average and a maximum of 6.8 and 15 dB, respectively. Finally, to demonstrate the impact of our findings on the design of deployment strategies of UAVs, we considered how the human body and the relative UAV hovering position can affect the physical layer security in a UAV-assisted network. We show that the secrecy rate for a UAV-based network can be heavily influenced by the human body orientation relative to the UAV hovering location, consequently resulting in a different optimal deployment strategy compared to existing schemes that employ free-space pathloss models that neglect such human-induced effects. [ABSTRACT FROM AUTHOR]

Accurate channel state information (CSI) is essential to increasing throughput in multi-input, multi-output (MIMO) systems with digital beamforming. CSI can be acquired by channel estimation and reported via feedback mechanisms. While the training and feedback overhead are typically proportional to the number of antennas, uplink measurements can be utilized to predict downlink CSI assuming perfect channel reciprocity in a time division duplex (TDD) mode. However, many works make the assumption that channels are perfectly reciprocal, which is often not the case in practice due to poor channel estimation and physical channel asymmetry. In this work, we investigate the key challenges in channel feedback performance, including TX-RX imbalance, channel coherence, and interference for mobile systems. We evaluate IEEE 802.11ac-based implicit and explicit feedback schemes with both emulated and in-field MIMO channels, particularly in regards to drone-based transmissions. Our analysis with channel emulation shows that implicit feedback is susceptible to channel reciprocity errors, while explicit feedback is more sensitive to Doppler effects. We propose a hybrid feedback mechanism that increases the throughput by 32% over conventional feedback methods. We additionally evaluate the impact of frequency offset asymmetry on the performance of distributed mobile systems. Our in-field experiments demonstrate that explicit feedback can provide better throughput improvement than implicit feedback in highly-mobile air-to-ground channels. Since our study spans many critical frequency bands, these results serve as a fundamental step towards understanding the impacts of asymmetric factors of channel reciprocity for drone-based beamforming systems. [ABSTRACT FROM AUTHOR]

Nonlinear channel codes are employed to protect two information fields that are related to the daylight saving time (DST) and leap second notifications within the enhanced WWVB broadcast format. The general codebook design criteria and the construction of these codes using integer linear program (ILP) is formulated. Specifically, two different variations are presented: i) minimizing the union bound of probability of word error (PWE), which is a tighter bound but has high computational complexity and ii) minimizing an analytical closed form approximation to the PWE, which has much lower complexity but is a looser bound on the exact PWE. By leveraging the highly non-uniform a-priori probabilities of the different messages being transmitted in these two fields, the proposed nonlinear codes provide coding gains of about 2–3 dB in the presence of additive white Gaussian noise (AWGN). When compared to checksum codes with the same rate, the nonlinear codes offer performance advantages of about 1–2 dB for the DST notification field. In the presence of impulsive noise, simulations and analysis show that the probability of word error when employing the proposed nonlinear code is lower by many orders of magnitude compared with that of the uncoded system. [ABSTRACT FROM AUTHOR]

A vast array of potential applications is emerging for drones and other devices to collaborate from disaster relief and search and rescue missions to smart agriculture and IoT systems. As drones move across multiple altitudes, they must have the ability to communicate across any direction in a three-dimensional (3D) space. However, due to the heterogeneous nature of the drone body and its interaction with the mounted antennas, different antenna positions on the drone can result in variations in the radiation pattern. While there have been a fair number of airborne communication works, few consider the role that antenna positioning has on the resulting transmission along the azimuth and elevation planes. In this work, we study the effects of the drone body and various antenna placements on the radiation pattern and fading of drone-based channels. Through systematic anechoic chamber and in-field measurements, we show that the drone body alters the radiation pattern of the mounted antennas, rendering the common assumption of a constant azimuth radiation pattern invalid. In addition, the body increases polarization mixing of drone-based channels, resulting in significant degradation of the cross-polarization discrimination (XPD). Hence, we propose using effective radiation pattern and XPD values instead of relying on measurements and/or assumptions that disregard drone-antenna interaction. We then analyze the shadowing and losses associated with the drone body for many antenna setups at various elevation angles and show that when mounted on the opposite side from the ground transmitter, shadowing increases with relatively-higher drone elevations. To account for these body-induced effects, we introduce rotational loss that results in better prediction results of the large-scale fading behavior compared to conventional models that neglect these body effects. Then, we analyze the small-scale fading for various antenna setups and show that the Rician K-factor is strongly dependent on elevation for polarization-matched vertical links, while it is approximately flat for cross-polarized links. To do so, we conduct a set of drone-to-drone (DtD) experiments at high altitudes with no surrounding reflectors and compare results to those obtained by our ground-to-drone (GtD) measurements. We find that, while at low elevations the ground can reduce the K-factor by 10 dB, at higher elevations, small-scale fading is dominated by the antennas, not the ground. [ABSTRACT FROM AUTHOR]

Channel emulation is widely used to control wireless channels and replay scenarios for repeatable experimentation and evaluation. Characterizing the propagation of massive Multiple-Input Multiple-Output (MIMO) systems, is challenging due to the rapid scaling of the number of channels needed to create representative scenarios as well as accurately modeling complex interdepedencies that exist within the channel matrix. Nonetheless, having the ability to emulate massive MIMO channels is critical for the development and testing of next-generation cellular protocols and algorithms. To achieve such scale for massive MIMO channel emulation solutions, in this paper, we study the tradeoff in channel emulation fidelity versus the hardware resources consumed using both analytical modeling and FPGA-based implementations. To reduce the memory footprint of our design, we optimize our channel emulation using an iterative structure to generate geometry-based channels at scale. In addition, the effects of varying the channel update rate and word length selection are evaluated in the paper and can significantly improve the implementation efficiency on an FPGA. [ABSTRACT FROM AUTHOR]

In this paper, we develop a new capacity-approaching code, namely, parallel-concatenated (PC)-Low Density Parity Check (LDPC) convolutional code that is based on the parallel concatenation of trellis-based quasi-cyclic LDPC (TQC-LDPC) convolutional codes. The proposed PC-LDPC convolutional code can be derived from any QC-LDPC block code by introducing the trellis-based convolutional dependency to the code. The capacity-approaching PC-LDPC convolutional codes are encoded through parallel concatenated trellis-based QC recursive systematic convolutional (RSC) encoder (namely, QC-RSC encoder) that is also proposed in this paper. The proposed PC-LDPC convolutional code and the associated encoder retain a fine input granularity on the order of the lifting factor of the underlying block code. We also describe the corresponding trellis-based QC maximum a posteriori probability (namely, QC-MAP) decoder that efficiently decodes the PC-LDPC convolutional code. Performance and hardware implementation results show that the PC-LDPC convolutional codes with the QC-MAP decoder have two times lower complexity for a given bit-error-rate (BER), signal-to-noise ratio, and data rate, than conventional QC-LDPC block codes and LDPC convolutional codes. Moreover, the PC-LDPC convolutional code with the QC-MAP decoder outperforms the conventional QC-LDPC block codes by more than 0.5 dB for a given BER, complexity, and data rate and approaches Shannon capacity limit with a gap smaller than 1.25 dB. This low decoding complexity and the fine granularity make it feasible to efficiently implement the proposed capacity-approaching PC-LDPC convolutional code and the associated trellis-based QC-MAP decoder in next generation ultra-high data rate mobile systems. [ABSTRACT FROM PUBLISHER]

Super resolution (SR) for real-life video sequences is a challenging problem due to complex nature of the motion fields. In this paper, a novel blind SR method is proposed to improve the spatial resolution of video sequences, while the overall point spread function of the imaging system, motion fields, and noise statistics are unknown. To estimate the blur(s), first, a nonuniform interpolation SR method is utilized to upsample the frames, and then, the blur(s) is(are) estimated through a multi-scale process. The blur estimation process is initially performed on a few emphasized edges and gradually on more edges as the iterations continue. Also for faster convergence, the blur is estimated in the filter domain rather than the pixel domain. The high-resolution frames are estimated using a cost function that has the fidelity and regularization terms of type Huber–Markov random field to preserve edges and fine details. The fidelity term is adaptively weighted at each iteration using a masking operation to suppress artifacts due to inaccurate motions. Very promising results are obtained for real-life videos containing detailed structures, complex motions, fast-moving objects, deformable regions, or severe brightness changes. The proposed method outperforms the state of the art in all performed experiments through both subjective and objective evaluations. The results are available online at http://lyle.smu.edu/~rajand/Video_SR/. [ABSTRACT FROM AUTHOR]