Wireless service provider backbone network evolution with packet technologies

Third-generation (3G) wireless networks provide for both increased circuit-switched voice capacity and high-speed packet data capabilities which promise greater access to rich multimedia content delivered to handheld or mobile devices. New data-centric high-bandwidth services will evolve the backbone to a data/Ethernet-dominated network. With such an evolution, the problem facing wireless service providers is how to leverage and evolve the existing SONET/SDH infrastructure to transmit Ethernet, as well as deliver legacy and next generation services in the metro backbone network. In this paper, we compare two evolution alternatives based on different choices of technology, namely Ethernet over SONET (EoS) and IP over MPLS over SONET (IPoMPLSoS). The comparisons cover network architecture, design, QoS, reliability and financial analysis. The paper also presents a case study based on a typical service provider offering CDMA 3G 1X services. We present two cases: one for a Greenfield operator deploying the metro core, and the other for an existing operator upgrading the core to carry Ethernet or MPLS traffic. Key insights and a summary of the study results are included as well

During a large-scale natural disaster or unexpected events, the terrestrial telecommunications become stretched, overloaded or totally destroyed in several cases. In these scenarios, a robust and capable emergency communication system is required to guarantee efficient communication to public safety agencies during the emergency phases. Airborne communication network systems have been recently studied for the provision of wireless communication services in those scenarios and it is a promising candidate for rapidly deployable and resilient emergency networks. However, the choice of communication technologies from aerial platforms is a challenging issue due to their payload capacity constraints. In this paper, we present a realistic Hybrid Aerial-Terrestrial Network (HATN) architecture based on the deployment of 4G-LTE Remote Radio Head (RRH) embedded on Helikite platforms and investigate their applicability and performance characteristics in emergency and unexpected scenarios.We present an adapted simulation model for the analysis of HATN. Our results confirm that the proposed architecture is able to provide high-capacity IP data HATN suitable for many forms of multimedia delivery.

The recent advances in wireless mesh communications technologies have given the rise to evolve maritime wireless mesh networks (MWMNs). Maritime users require more high data rate applications, where the cost associated with the conventional satellite and the legacy systems that provide low data rate applications have posed major limitations for the emergence of MWMNs. Besides, the network architecture and protocol design remain big challenges in coping with the different ship mobility patterns and with the numerous navigational scenarios in the maritime environment. Most of the research studies [1], [2] have focused on the development of maritime communications technologies, where less attention has been considered to provide a realistic and reliable framework to implement a network that can satisfy maritime users' requirements. This paper provides a detailed research study (an in-depth overview) on maritime communications technologies which can be deployed for implementing reliable and resilient maritime wireless mesh networks. Comprehensive guidelines are outlined to easily understand and critically assess the different deployed maritime communications networks, and identify the milestones in the process of developing MWMNs. Wireless service providers will use the outcome of this research research in expanding their communications infrastructure to the maritime environment.

Recent events have shown that in the aftermath of an unexpected incident, communication infrastructures play an important role in supporting critical services. Airborne communication networks have been recently studied for the provision of wireless communication services and it is a promising candidate for rapidly deployable and resilient emergency networks. However, the choice of communication technologies from Aerial platforms is a challenging issue and depends on a variety of factors including platform payload capacity, coverage and capacity requirements, to name a few. In this paper, we investigate the performance of 4G LTE-WiFi multimode base stations deployed on airborne platforms which provides coverage for first responders during emergencies. We present an adapted simulation model for the analysis of hybrid aerial-terrestrial systems and study the impact of platform elevation and mobility on the cell coverage and channel capacity. Performance analysis with a platform deployment of a single Aerial Base Station (eNodeB) corroborates that airborne units with 4G communication capabilities are very promising candidates for robust communication links during emergency relief operations.

On spectrum sensing in cognitive radio CDMA networks with beamforming

Summary form only given. Telecommunication networks are evolving from closed systems with limited, standardized services, to open systems which will allow great creativity in building and deploying new services. These systems will heavily leverage Internet technology in an effort to create this open environment. This evolution is being aggressively pursued by wireless service providers (WSPs). Along with the benefits of these networks come increasingly high risks of a variety of attacks that may compromise security. Current, so called second generation (2G) wireless telecommunication networks are implemented using standardized control protocols for user and device authentication, mobility management, session control and services control. These networks are closed in the sense that control messages are exchanged on a private packet-switched network based on the signaling system No. 7 standards. Because of their closed nature, there are few successful attacks on these networks. The next, so called third generation (3G) wireless telecommunication networks are migrating towards IP technology, with the ultimate goal being an all-IP network. Standards for these systems, called the IP multimedia subsystem (IMS) are being defined by the third generation partnership projects (3GPP and 3GPP2). These networks will use IP for transport of information, and Internet protocols such as the session initiation protocol (SIP) and mobile IP, for session control and mobility management. These networks open the possibility for IP-based services and must interwork with 2G networks. Because new services will be introduced in the IP-domain of these networks, new attacks on 3G networks are possible. Because IP networks are more accessible than SS7 networks, the control portion of the 3G networks is now more vulnerable to attack. These attacks may be remote denial of service attacks, or attacks that target the integrity of specific services. The means of the attack may vary depending on the interworking model used and the service being offered. In this talk we discuss the different security risks in IP-based 3G networks, different attack types, and the trade-offs of high performance, open network architectures versus secure network infrastructure.

For code-division multiple-access (CDMA) cellular communications synchronous operation offers significant advantages over asynchronous operation. This feature has been already included in second-generation systems, such as the CdmaOne based on the IS-95 standard. Nonetheless, synchronous operation does not seem to be assessed as a basic system option for third-generation systems. This is mainly due to the contrasted choice of having global positioning system (GPS) receivers located at any of the base transceiver stations (BTSs) to perform station synchronization. However, the existing CdmaOne interface permits time transfer between the various BTSs and a simple time difference measurement technique can provide an internal, network-based means for time synchronization of base stations clocks. This paper considers a previously proposed approach based on the pilot strength measurements message delivered by every mobile station (MS), combined with round trip delay measurements that BTSs perform on the MS transmissions. We consider some possible algorithms to reduce time misalignments, and continuously phase-lock the clocks at different BTSs. Having developed an appropriate analytical/simulation approach, in the environment of a CDMA network, the paper evaluates and compares the performance of two suitable centralized architecture configurations, i.e., master-slave synchronization (MSS) and mutual synchronization (MUS). Both network architectures turn out to be applicable and provide suitable timing performance. Finally, the paper provides a preliminary discussion on some related issues, such as GPS avoidance, synchronization of BTSs inside buildings, tunnels, and subways and hybrid MSS/MUS synchronization network architectures.

Traffic flow forecasting is an essential task of an intelligent transportation system (ITS), closely related to intelligent transportation management and resource scheduling. Dynamic spatial-temporal dependencies in traffic data make traffic flow forecasting to be a challenging task. Most existing research cannot model dynamic spatial and temporal correlations to achieve well-forecasting performance. The multi-head self-attention mechanism is a valuable method to capture dynamic spatial-temporal correlations, and combining it with graph convolutional networks is a promising solution. Therefore, we propose a multi-head self-attention spatiotemporal graph convolutional network (MSASGCN) model. It can effectively capture local correlations and potential global correlations of spatial structures, can handle dynamic evolution of the road network, and, in the time dimension, can effectively capture dynamic temporal correlations. Experiments on two real datasets verify the stability of our proposed model, obtaining a better prediction performance than the baseline algorithms. The correlation metrics get significantly reduced compared with traditional time series prediction methods and deep learning methods without using graph neural networks, according to MAE and RMSE results. Compared with advanced traffic flow forecasting methods, our model also has a performance improvement and a more stable prediction performance. We also discuss some problems and challenges in traffic forecasting.

IP over WDM has emerged as a winning network architecture to facilitate today's IP centric Internet. Although the total amount of bandwidth demands keeps increasing, there still exists a huge gap between the increasing full-wavelength capacity and the existing low-rate IP traffic. Efficiently grooming low-rate traffic stream to high-capacity lightpaths helps to alleviate the bandwidth mismatch, effectively improve the network resource utilization, and reduce the overall network cost. In this paper, we study traffic grooming problem in IP over WDM framework. The problem of minimizing the total number of transmitters and receivers in the networks is formulated as an integer linear programming (ILP) optimization problem. The ILP formulation produces optimal solutions for static traffic demands. However, applying the ILP solution to large and realistic networks is not practical due to its prohibitively large computation time. We therefore develop a heuristic algorithm, namely traffic aggregation algorithm. A significant improvement on wavelength utilization is observed in our experiments when we use our traffic aggregation algorithm. I. INTRODUCTION The popularity of the Internet and Internet protocol (IP)- based Intranet is promising enormous growth in data traffic originating from hosts that are IP endpoints. This growth is being fueled by various applications such as those driven by World-Wide-Web (WWW) and by the indirect impact of increased computing power and storage capacity at the end systems. The advent of new services with increasing intelligence and the corresponding bandwidth demands are further adding to the traffic growth. New access technologies such as Asymmetric Digital Subscriber Line (ADSL), High- bit-rate Digital Subscriber Line (HDSL), and fiber to the home (FTTH) would remove the access bottlenecks and enforce an even faster growth of demand on the backbone network. These changing trends have led to a fundamental shift in traffic patterns and the traffic is mostly due to data communications. In the past, the amount of data traffic on carrier networks was smaller than the voice-centric traffic. Therefore, the carrier networks were designed to primarily support voice traffic, and the data traffic was transmitted using the voice channels. Now, the core networks are being designed primarily for data traffic with voice support at the edges. The voice can be carried in the core networks using voice-over-IP or similar paradigms. To meet these growing demands, WDM will continue to emerge as a dominating trend for use in backbone networks. The research reported in this paper is funded in part by the National Science Foundation under grant ANI-0087746, grant ANI-0434872 and grant ANI- 0323374. Architectures will include to satisfy the need for better quality of service (QoS), protection, and availability guarantees in IP networks. WDM significantly increases the fiber capacity utilization by dividing the available bandwidth into non-overlapping wavelength channels, supporting connections between the two end nodes by establishing an all-optical channel, namely lightpath, which allows use of different formats, bit-rates, and protocol transparency. At the same time, most network designers believe that IP is going to be the common traffic convergence layer in communication networks. Consequently, IP over WDM has been envisioned as the winning combination of the network architecture.

Multiple access schemes for Cognitive Radio networks: A survey

This paper presents an experimental deployment of a multi-channel wireless backbone network (WBN) with an Open Flow-based traffic management framework. In a WBN, a set of APs, each of which uses a single and different channel, is connected by Ethernet and thus constructs a VAP handling multiple channels. To flexibly control traffic flow transmitted over multiple channels, we also implement a traffic management framework with a simple channel utilization method based on the Open Flow control. We then deploy a 6-hop WBN in a real environment and provide the Internet access service to conference attendees. Since the control messages are inherently transmitted with the introduction of Open Flow, the way of isolating between control plane and data plane will become critical issue. To solve this issue, we employ an additional AP, which constructs a dedicated control network for Open Flow. The results show that the proposed WBN can attain the network capacity of about 25 Mbps in 6-hop, thereby providing significant throughput performance for various applications including video streaming.

A dynamic CDMA network for multicore systems

Since applications can generate both streaming and best effort traffics, there is then a need for an on-chip network to provide QoS for streaming traffic and to guarantee packets delivery without loss for best effort traffic. Instead of handling both streaming and best-effort traffic in a circuit or packet switched network, which often leads to a complex design, we propose a hybrid router combining circuit and packet switching with bus architecture. It efficiently and separately handles streaming and best effort traffics. Code Division Multiple Access "CDMA" is used in circuit switching and bus to improve resources utilization. The proposed router architecture has been synthesized in both ASIC 65 nm and FPGA. Synthesis results show that a practical network on chip can be built using the proposed router architecture.

In response to the issue of low color contrast against similar backgrounds and occlusion from clustered growth during oblate jujube harvesting, this paper proposes a novel network architecture, termed BEA-Net, which is designed from the ground up for real-time, robust detection in agronomic environments. The proposed method integrated the Bottleneck Transformer (BoT-CTR3), which incorporated multi-head self-attention, enhancing the local feature extraction ability and improving the detection baseline. Subsequently, the hybrid attention mechanism of Efficient Multi-scale Channel Attention (EMCA) was embedded into the backbone network, which improved color feature discrimination under similarly colored backgrounds. Finally, the Adaptively Spatial Feature Fusion (ASFF) module was integrated into the detection head, strengthening the model’s contextual information extraction to compensate for information loss in occluded regions. Experimental results demonstrated that the improved network achieved precision, recall, F1-score, and mean average precision (mAP) of 94%, 92.2%, 93%, and 97.8% respectively—surpassed the baseline network by 3.2, 3.6, 3.0, and 2.1 percentage points. Comprehensive ablation studies validated the contribution of each component, while comparative evaluations against 14 state-of-the-art single-stage detectors confirmed BEA-Net's superior performance. To verify generalizability, the algorithm was tested on our self-built crisp persimmon dataset, achieving 96.9% mAP and 97% F1-score—represented improvements of 2.3 and 4.0 percentage points respectively. Furthermore, deployment on three distinct Android mobile devices achieved an optimal inference time of 50 ms. Benchmarking against YOLOv5 variants (n, s, m) confirmed this performance fully satisfies the accuracy and real-time requirements of automated picking systems, significantly enhancing the algorithm's practical application value in real-world scenarios.

In recent years, convolution neural networks have significant progress on keypoint detection. However, lots of approaches require multiple upsampling of small featuremaps to produce the final output. The huge information loss in the upsampling process is the key to restrict the accuracy. This work presents a new network architecture, which is designed to solve the problem of information loss in the process of upsampling and feature fusion compared with the previous approaches. We replace the normal convolution in the backbone network by using dilated convolution with stride of one to keep featuremaps’ size consistent, which avoids the multi-times upsampling during prediction. We also explore the feature fusion methods and propose a feature fusion block (AFB), which improves the accuracy and accelerates the convergence of model with the help of multi-scale resampling and attention mechanism. Excellent results are achieved on MsCOCO 2017 dataset and Ali apparel keypoints dataset. The code will be released for further research.