Cross-layer Information Exchange Research Articles

TTIS-YOLO: a traffic target instance segmentation paradigm for complex road scenarios

Abstract The instance segmentation of traffic targets in complex road scenes is one of the most challenging tasks in autonomous driving. Unlike the bounding box localization for object detection and the category perception mask for semantic segmentation, instance segmentation requires accurate identification of each object under each category and more precise segmentation and positioning of these target objects. Although instance segmentation has apparent advantages, methods, for instance segmentation in complex road scenes, still need to be discovered. In this paper, we proposed an efficient instance segmentation method TTIS-YOLO(Traffic Target Instance Segmentation - YOLO) based on YOLOV5-7.0 for traffic object segmentation of complex road scenes. Our main work is as follows: to propose a MECSP(Multiscale Efficient Cross Stage Partial Network) module, which has fewer parameters, better cross-layer information exchange, and feature representation capabilities. Propose an Efficient Bidirectional Cross Scale Connection optimization method that enables the network to perform more detailed and efficient feature fusion without losing original information, refining the mask flow. WIoU Loss is used as the loss function of positioning and segmentation, and the positioning performance of the model is effectively improved through the strategy of dynamically allocating gradient gains. Experiments have shown that our proposed TTIS-YOLO outperforms baseline models and other mainstream instances segmentation algorithms such as Mask RCNN, YOLACT, SOLO, and SOLOV2 with the highest segmentation accuracy and fastest inference speed. Our proposed TTIS-YOLO-S achieves the best balance between segmentation accuracy and inference speed. Compared to the baseline model, the AP50 and recall values on the Cityscapes validation set increased by 1.7% and 0.9%, respectively, with a parameter reduction of 20.6%, inference speed of 78.1fps on GeForce RTX 3090Ti. Meanwhile, TTIS-YOLO-L achieved the highest segmentation accuracy, with an AP50 value of 27%, and the model parameter quantity decreased by 35.4% compared to the baseline model.

Pairwise Guided Multilayer Cross-Fusion Network for Bird Image Recognition

Bird identification is the first step in collecting data on bird diversity and abundance, which also helps research on bird distribution and population measurements. Most research has built end-to-end training models for bird detection task via CNNs or attentive models, but many perform unsatisfactorily in fine-grained bird recognition. Bird recognition tasks are highly influenced by factors, including the similar appearance of different subcategories, diverse bird postures, and other interference factors such as tree branches and leaves from the background. To tackle this challenge, we propose the Progressive Cross-Union Network (PC-Net) to capture more subtle parts with low-level attention maps. Based on cross-layer information exchange and pairwise learning, the proposed method uses two modules to improve feature representation and localization. First, it utilizes low- and high-level information for cross-layer feature fusion, which enables the network to extract more comprehensive and discriminative features. Second, the network incorporates deep semantic localization to identify and enhance the most relevant regions in the images. In addition, the network is designed with a semantic guidance loss to improve its generalization for variable bird poses. The PC-Net was evaluated on an extensively used birds dataset (CUB-200-2011), which contains 200 birds subcategories. The results demonstrate that the PC-Net achieved an impressive recognition accuracy of 89.2%, thereby outperforming maintained methods in bird subcategory identification. We also achieved competitive results on two other datasets with data on cars and airplanes. The results indicated that the PC-Net improves the accuracy of diverse bird recognition, as well as other fine-grained recognition scenarios.

Heterogeneous network flow and Petri nets characterize multilayer complex networks

Interacting subsystems are commonly described by networks, where multimodal behaviour found in most natural or engineered systems found recent extension in form of multilayer networks. Since multimodal interaction is often not dictated by network topology alone and may manifest in form of cross-layer information exchange, multilayer network flow becomes of relevant further interest. Rationale can be found in most interacting subsystems, where a form of multimodal flow across layers can be observed in e.g., chemical processes, energy networks, logistics, finance, or any other form of conversion process relying on the laws of conservation. To this end, the formal notion of heterogeneous network flow is proposed, as a multilayer flow function aligned with the theory of network flow. Furthermore, dynamic equivalence is established with the framework of Petri nets, as the baseline model of concurrent event systems. Application of the resulting multilayer Laplacian flow and flow centrality is presented, along with graph learning based inference of multilayer relationships over multimodal data. On synthetic data the proposed framework demonstrates benefits of multimodal flow derivation in critical component identification. It also displays applicability in relationship inference (learning based function approximation) on multimodal time series. On real-world data the proposed framework provides, among others, multimodal flow interpretation of U.S. economic activity, uncovering underlying empirical steady state probability distribution, as well as inherent network (economic) robustness.

Block Multi-Dimensional Attention for Road Segmentation in Remote Sensing Imagery

High-resolution remote sensing image (RSI) segmentation is a relatively mature application in various deep learning projects. In this study, aiming at slender objects in road RSIs, BMDANet combines cross-layer information exchange and block multi-dimensional attention (BMDA) module and optimizes road feature extraction by using multi-dimensional information to construct a global attention module. The experimental results based on the Ottawa road dataset show that our algorithm improved the recognition results of the road in RSI, and excelled the existing RSI road segmentation algorithm and reached the state-of-the-art. In addition, based on comparative experiments, the addition of the BMDA module to different algorithms can effectively improve the accuracy of the algorithm. It has proven the effectiveness and embedding of our BMDA module in RSI road segmentation algorithms.

Trade-off between spectral efficiency and normalized energy in Ad-hoc wireless networks

With the ever-increasing deployment of wireless communication technologies, effective management of scarce power and radio spectrum resources is of primary concern. The available bandwidth spectrum should be utilized in the most efficient and controlled way for practical economic growth of industries in conformity with the evolving demand of energy for powering wireless devices. In the proposed optimization scheme, energy management and limited spectrum sharing are distributively implemented for handling quality of service provisioning and potential resource allocation across multiple nodes with inherent operational capability deficiencies. We employ cross-layer information exchange and convex optimization techniques to simultaneously achieve more efficient radio spectrum usage and optimal energy consumption in ad-hoc wireless networks with distributed scenarios. We consider the time-invariant additive white Gaussian noise channel and the time-varying Rayleigh fading channel to study the trade-off between the two network design objectives of achieving improved spectrum efficiency and minimizing the energy overhead in data routing paradigm. The original problem is transformed into an equivalent convex optimization problem through logarithmic processing to obtain the approximate global optimal solution. Moreover, the robustness and efficiency of the proposed framework is evaluated in large-scale set ups to provide the scalability analysis on both the employed performance objectives. The developed optimization scheme is compared with the machine learning models previously proposed in literature by employing the computational and/or time complexity metrics. Finally, the effectiveness of the proposed optimization model is substantiated through the simulation comparison results with the existing schemes in terms of various key performance parameters such as throughput, energy efficiency, and average bit errors.

The Right Time: Reducing Effective End-to-End Delay in Time-Slotted Packet-Switched Networks

Modern access network technologies like Long Term Evolution (LTE) and High Speed Packet Access (HSPA) use time-slotting mechanisms to optimize resource sharing and overall network performance. In time-slotted networks, the one-way delay of all packets in a packet stream depends on the absolute point in time when the first packet of the stream is sent. With appropriate feedback signals, applications can exploit this effect to reduce their effective end-to-end delay. Time-critical applications such as real-time sensor data acquisition or voice-over-IP (VoIP) communications can shift their acquisition interval in order to adapt to the network timing. Information about the actual time-slotting periods can be gathered by active network measurements or through implementation of cross-layer information exchange. In this paper, a method is proposed to determine the optimum send time for particular destinations and to support applications in adjusting their send time accordingly. Theoretical findings are supported by the offline analysis of measurement data and by a proof-of-concept implementation that confirms the feasibility and effectiveness of the proposed solution in operational LTE and HSPA networks.

DVB-RCS 네트워크에서 Cross-layer 및 네트워크 코딩 기반 PEP 연구

본 논문에서는 DVB-RCS 네트워크에서 cross-layer와 네트워크 코딩 기반 PEP 기술을 제안한다. 제안한 PEP 기술에서는 TCP와 링크 단 간 위성 자원 할당 정보 교환을 제공할 수 있는 cross-layer 정보 교환 기법을 제안하고, 할당된 위성 자원 정보와 NC-R (Network-Coded Redundancy) 패킷 전송률에 의한 TCP 혼잡제어윈도우 튜닝 기법을 제안한다. 시뮬레이션 결과 제안한 PEP 기술을 사용하면 TCP에서 사용가능 한 자원을 조사하기 위한 지연 없이 바로 최대 이용 가능한 TCP 혼잡제어윈도우 값을 사용할 수 있는 것을 살펴 볼 수 있었다. 또한 제안한 PEP 기법에서는 TCP 혼잡제어윈도우 튜닝 기법과 NC-R 패킷 전송을 통하여 패킷 손실이 발생할 수 있는 이동형 단말 환경에서도 불필요한 위성 자원 사용 없이 TCP 데이터 처리율을 향상 할 수 있는 것을 볼 수 있었다.

Cross-Layer Aided Energy-Efficient Opportunistic Routing in Ad Hoc Networks

Most of the nodes in ad hoc networks rely on batteries, which requires energy saving. Hence, numerous energy-efficient routing algorithms have been proposed for solving this problem. In this paper, we exploit the benefits of cross-layer information exchange, such as the knowledge of the Frame Error Rate (FER) in the physical layer, the maximum number of retransmissions in the Medium Access Control (MAC) layer and the number of relays in the network layer. Energy-consumption-based Objective Functions (OF) are invoked for calculating the end-to-end energy consumption of each potentially available route for both Traditional Routing (TR) and for our novel Opportunistic Routing (OR), respectively. We also improve the TR and the OR with the aid of efficient Power Allocation (PA) for further reducing the energy consumption. For the TR, we take into account the dependencies amongst the links of a multi-hop route, which facilitates a more accurate performance evaluation than upon assuming the links that are independent. Moreover, two energy-efficient routing algorithms are designed based on Dijkstra's algorithm. The algorithms based on the energy OF provide the theoretical bounds, which are shown to be close to the bound found from exhaustive search, despite the significantly reduced complexity of the former. Finally, the end-to-end throughput and the end-to-end delay of this system are analyzed theoretically and a new technique of characterizing the delay distribution of OR is proposed. The simulation results show that our energy-efficient OR outperforms the TR and that their theoretical analysis accurately matches the simulation results.

Distributed Media-Aware Rate Allocation for Video Multicast Over Wireless Networks

A unified optimization framework for rate allocation among multiple video multicast sessions sharing a wireless network is presented. Our framework applies to delivery of both scalable and non-scalable video streams. In both cases, the optimization objective is to minimize the total video distortion of all peers without incurring excessive network utilization. Our system model explicitly accounts for heterogeneity in wireless link capacities, traffic contention among neighboring links, as well as different video rate-distortion (RD) characteristics. The proposed distributed rate allocation scheme leverages cross-layer information exchange between the media access control and application layers to achieve fast convergence at the optimal, media-aware allocation. Performance of the proposed media-aware rate allocation protocol is compared against a heuristic scheme based on TCP-friendly rate control (TFRC). In network simulations of standard-definition video streaming over single or multiple multicast trees, the proposed scheme consistently achieves higher overall video quality than the TFRC-based heuristics. When delivering scalable streams, the flexibility of per-peer rate adaptation inside each multicast tree yields a further slight improvement in overall video quality over multicast of non-scalable streams.

Distributed Media-Aware Rate Allocation for Wireless Video Streaming

When multiple video streams share a wireless network, careful rate allocation is needed to prevent congestion, as well as to balance the video qualities among the competing streams. In this paper, we present a distributed media-aware rate allocation protocol, and evaluate its performance for streaming of high-definition video over 802.11-based wireless home networks. Our optimization framework accommodates heterogeneity in wireless link speeds and video rate-distortion characteristics, as well as traffic contention among neighboring links. The goal of the protocol is to minimize the total video distortion of all participating streams while limiting network utilization. It relies on cross-layer information exchange between video rate controllers at the end hosts and link state monitors at the relaying wireless nodes. Results from various network simulations confirm that the media-aware allocation outperforms transmission control protocol-friendly rate control in terms of average video quality and fairness among the streams.

A unified framework for distributed video rate allocation over wireless networks

When multiple video streams share a wireless network, careful rate allocation is needed to prevent congestion, as well as to balance the video qualities among the competing streams. In this paper, we present a unified optimization framework for video rate allocation over wireless networks. Our framework applies to both unicast and multicast sessions, and accommodates both scalable and non-scalable streams. The optimization objective is to minimize the total distortion of all video streams without incurring excessive network utilization. Our system model explicitly accounts for heterogeneity in wireless link capacities, traffic contention among neighboring links, as well as different video rate-distortion (RD) characteristics. The proposed distributed media-aware rate allocation scheme leverages cross-layer information exchange between the MAC and application layers to achieve fast convergence at the optimal allocation. We evaluate performance of the proposed scheme for streaming of high-definition (HD) and standard-definition (SD) video sequences over 802.11-based wireless networks, both in unicast and multicast scenarios. The scheme consistently outperforms conventional TCP-Friendly Rate Control (TFRC) in terms of overall video quality, and achieves more balanced qualities among the streams.

Cross-layer distributed joint power control and scheduling for delay-constrained applications over CDMA-based wireless ad-hoc networks

In this paper, we propose a cross-layer distributed power control and scheduling protocol for delay-constrained applications over mobile CDMA-based ad hoc wireless networks. Herein, we propose a novel scheme where power control is employed to combat delay occurring on multi-hop wireless ad hoc networks via cross-layer information exchange. Based on that, a distributed power control and scheduling protocol is proposed to control the incurred delay as well as the multiple access interference (MAI). Unlike other previous work on power control and scheduling, we also investigate the impact of Doppler spread upon the system performance, and provide a robust system which employs a combination of power control, and coding/interleaving to combat the effects of Doppler spread by exploiting the time diversity when the Doppler spread gets large. Hence, our proposed approach can function appropriately over a wide range of channel conditions.

Cross-layer design of ad hoc networks for real-time video streaming

Cross-layer design breaks away from traditional network design where each layer of the protocol stack operates independently. We explore the potential synergies of exchanging information between different layers to support real-time video streaming. In this new approach information is exchanged between different layers of the protocol stack, and end-to-end performance is optimized by adapting to this information at each protocol layer. We discuss key parameters used in the cross-layer information exchange along with the associated cross-layer adaptation. Substantial performance gains through this cross-layer design are demonstrated for video streaming.