scholarly journals Federated learning over WiFi: Should we use TCP or UDP?

2021 ◽  
Author(s):  
Vineeth S
Keyword(s):  
Packet Loss ◽  
Distributed Learning ◽  
Learning Paradigm ◽  
Limited Bandwidth ◽  
Tcp Protocol ◽  
Network Connections ◽  
Model Aggregation ◽  
Centralized Model

Federated learning is a distributed learning paradigm where a centralized model is trained on data distributed over a large number of clients, each with unreliable and relatively slow network connections. The client connections typically have limited bandwidth available to them when using networks such as 2G, 3G, or WiFi. As a result, communication often becomes a bottleneck. Currently, the communication between the clients and server is mostly based on TCP protocol. In this paper, we explore using the UDP protocol for the communication between the clients and server. In particular, we develop UDP-based algorithms for gradient aggregation-based federated learning and model aggregation-based federated learning. We propose methods to construct model updates in case of packet loss with the UDP protocol. We present a scalable framework for practical federated learning. We conduct experiments over WiFi and observe that the UDP-based protocols can lead to faster convergence than the TCP-based protocol -- especially in bad networks. Code available at the repository: \url{https://github.com/vineeths96/Federated-Learning}.

Virtualized Fog Network with Load Balancing for IoT based Fog-to-Cloud

2020 ◽  
Vol 4 (3) ◽  
Author(s):  
Istabraq M. Al-Joboury ◽  
Emad H. Al-Hemiary
Keyword(s):  
Load Balancing ◽  
Packet Loss ◽  
Fog Computing ◽  
Transport Protocol ◽  
Network Simulator ◽  
Huge Number ◽  
Message Queue ◽  
Limited Bandwidth ◽  
Cloud Architecture ◽  
Fog Network

Fog Computing is a new concept made by Cisco to provide same functionalities of Cloud Computing but near to Things to enhance performance such as reduce delay and response time. Packet loss may occur on single Fog server over a huge number of messages from Things because of several factors like limited bandwidth and capacity of queues in server. In this paper, Internet of Things based Fog-to-Cloud architecture is proposed to solve the problem of packet loss on Fog server using Load Balancing and virtualization. The architecture consists of 5 layers, namely: Things, gateway, Fog, Cloud, and application. Fog layer is virtualized to specified number of Fog servers using Graphical Network Simulator-3 and VirtualBox on local physical server. Server Load Balancing router is configured to distribute the huge traffic in Weighted Round Robin technique using Message Queue Telemetry Transport protocol. Then, maximum message from Fog layer are selected and sent to Cloud layer and the rest of messages are deleted within 1 hour using our proposed Data-in-Motion technique for storage, processing, and monitoring of messages. Thus, improving the performance of the Fog layer for storage and processing of messages, as well as reducing the packet loss to half and increasing throughput to 4 times than using single Fog server.

Energy Consumption Optimization by Increasing The Processor Speed of Moving Communication Devices in Transport Layer Protocol

2019 ◽  
Vol 16 (2) ◽  
pp. 30
Author(s):  
Fakhrur Razi ◽  
Ipan Suandi ◽  
Fahmi Fahmi
Keyword(s):  
Energy Consumption ◽  
Mobile Devices ◽  
Mobile Device ◽  
Packet Loss ◽  
Video Data ◽  
Transmission Delay ◽  
Transport Layer ◽  
Tcp Protocol ◽  
Transport Layer Protocol ◽  
Processor Speed

The energy efficiency of mobile devices becomes very important, considering the development of mobile device technology starting to lead to smaller dimensions and with the higher processor speed of these mobile devices. Various studies have been conducted to grow energy-aware in hardware, middleware and application software. The step of optimizing energy consumption can be done at various layers of mobile communication network architecture. This study focuses on examining the energy consumption of mobile devices in the transport layer protocol, where the processor speed of the mobile devices used in this experiment is higher than the processor speed used in similar studies. The mobile device processor in this study has a speed of 1.5 GHz with 1 GHz RAM capacity. While in similar studies that have been carried out, mobile device processors have a speed of 369 MHz with a RAM capacity of less than 0.5 GHz. This study conducted an experiment in transmitting mobile data using TCP and UDP protocols. Because the video requires intensive delivery, so the video is the traffic that is being reviewed. Energy consumption is measured based on the amount of energy per transmission and the amount of energy per package. To complete the analysis, it can be seen the strengths and weaknesses of each protocol in the transport layer protocol, in this case the TCP and UDP protocols, also evaluated the network performance parameters such as delay and packet loss. The results showed that the UDP protocol consumes less energy and transmission delay compared to the TCP protocol. However, only about 22% of data packages can be transmitted. Therefore, the UDP protocol is only effective if the bit rate of data transmitted is close to the network speed. Conversely, despite consuming more energy and delay, the TCP protocol is able to transmit nearly 96% of data packets. On the other hand, when compared to mobile devices that have lower processor speeds, the mobile devices in this study consume more energy to transmit video data. However, transmission delay and packet loss can be suppressed. Thus, mobile devices that have higher processor speeds are able to optimize the energy consumed to improve transmission quality.Key words: energy consumption, processor, delay, packet loss, transport layer protocol

scholarly journals The Yellow Active Queue Management Algorithm in ICN Routers Based on the Monitoring of Bandwidth Competition

Electronics ◽  
2021 ◽  
Vol 10 (7) ◽  
pp. 806
Author(s):  
Li Zeng ◽  
Hong Ni ◽  
Rui Han
Keyword(s):  
Packet Loss ◽  
Service Providers ◽  
Exclusion Process ◽  
Active Queue Management ◽  
Existence Condition ◽  
Queue Management ◽  
Limited Bandwidth ◽  
Management Algorithm ◽  
Asymmetric Simple Exclusion

Deploying the active queue management (AQM) algorithm on a router is an effective way to avoid packet loss caused by congestion. In an information-centric network (ICN), routers not only play a role of packets forwarding but are also content service providers. Congestion in ICN routers can be further summarized as the competition between the external forwarding traffic and the internal cache response traffic for limited bandwidth resources. This indicates that the traditional AQM needs to be redesigned to adapt to ICN. In this paper, we first demonstrated mathematically that allocating more bandwidth for the upstream forwarding flow could improve the quality of service (QoS) of the whole network. Secondly, we propose a novel AQM algorithm, YELLOW, which predicts the bandwidth competition event and adjusts the input rate of request and the marking probability adaptively. Afterwards, we model YELLOW through the totally asymmetric simple exclusion process (TASEP) and deduce the approximate solution of the existence condition for each stationary phase. Finally, we evaluated the performance of YELLOW by NS-3 simulator, and verified the accuracy of modeling results by Monte Carlo. The simulation results showed that the queue of YELLOW could converge to the expected value, and the significant gains of the router with low packet loss rate, robustness and high throughput.

scholarly journals QoS measurement of VoIP codec usage on limited bandwidth network over UDP-based VPN

2020 ◽  
Vol 12 (1) ◽  
pp. 13-17
Author(s):  
Panca Susanto
Keyword(s):  
Packet Loss ◽  
Sound Quality ◽  
Communication Quality ◽  
Voice Communication ◽  
Limited Bandwidth ◽  
Delay Jitter ◽  
Opinion Score ◽  
Tunneling Method ◽  
Encryption Method ◽  
The Voice

The usage of internet protocol for voice communication is widely used and more efficient rather than an analog signal. However, there is no security guaranteed on IP-based voice communication. The voice payload can be easily tapped or even manipulated. In the case of improving the security aspect, communication quality should be also considered. VoIP requires sufficient bandwidth to get proper communication quality. The ITU-T released a standard unit of communication quality, known as Mean Opinion Score (MOS) which is made from the subjective judgments of some individuals. However, MOS method takes time and is expensive. In this research, we measure VoIP communication which is secured by using VPN and build a tool for analyzing the voice packet between communication peers. The tool has capabilities to measure delay, jitter, and packet loss. Since VoIP has a QoS standard by ITU-T, the usage of VPN for security purpose needs to be considered. The sound quality might be decreased due to the addition of header for tunneling method, as well as the additional delay when the encryption processing is carried out. We used 3 types of codec: a-Law, GSM, and iLBC which will be passed on 4 types of bandwidth (256, 128, 64, 32 kbps) through the UDP-based VPN that use 3 types of encryption method (3-DES, Blowfish, AES).

scholarly journals FUZZY LOGIC BASED IMPROVED MEDIUM ACCESS CONTROLLER FOR CONGESTION CONTROL AND PREDICTION THE SYSTEMS IN WSNS

2021 ◽  
Vol 9 (2) ◽  
pp. 425-432
Author(s):  
M. Sri Lakshmi, Et. al.
Keyword(s):  
Fuzzy Logic ◽  
Packet Loss ◽  
Bat Algorithm ◽  
High Volume ◽  
Poor Quality ◽  
Volume Data ◽  
Energy Aware ◽  
Medium Access ◽  
Fuzzy Logic Systems ◽  
Limited Bandwidth

in resource-constrained networks, particularly those with limited bandwidth to manage high-volume data transmission, network congestion is a major issue, resulting in poor quality of service, including packet loss and delay throughput. Due to self-contained batteries that limit sensor node lifetime, this issue is important in wireless sensor networks (WSNs) with limitations and restrictions, such as limited processing power, memory, and transmission. By determining a path that avoids congested highways, the network can be extended. As a result, we present a WSN route determination architecture that is congestion-aware. The architecture is divided into three stages: In a top-down hierarchical structure, the first path is created. Energy-aware assisted routing for route derivation. Exponential smoothing is used to forecast congestion, but the final parameters for route determination are not taken into account. We use fuzzy logic systems to evaluate proper weights for a variety of factors, including shop count, remaining energy, buffer occupancy, and forwarding rate, as well as a bat algorithm to optimize the weight over the membership functions. Eventually proposed model shows the high throughput, low packet loss, save energy, and extending network lifetime.                       

Effects of Packet-Loss and Long Delay Cycles on the Performance of the TCP Protocol in Wireless Networks

2011 ◽  
pp. 346-374
Author(s):  
Hussein Al-Bahadili ◽  
Haitham Y. Adarbah
Keyword(s):  
Wireless Networks ◽  
Packet Loss ◽  
Analytical Models ◽  
Network Simulator ◽  
Tcp Performance ◽  
Utilization Factor ◽  
Tcp Protocol ◽  
Transport Control ◽  
User Friendly ◽  
Tcp Reno

Many analytical models have been developed to evaluate the performance of the transport control protocol (TCP) in wireless networks. This chapter presents a description, derivation, implementation, and comparison of two well-known analytical models, namely, the PFTK and PLLDC models. The first one is a relatively simple model for predicting the performance of the TCP protocol, while the second model is a comprehensive and realistic analytical model. The two models are based on the TCP Reno flavor, as it is one of the more popular implementations on the Internet. These two models were implemented in a user-friendly TCP performance evaluation package (TCP-PEP). The TCP-PEP was used to investigate the effect of packet-loss and long delay cycles on the TCP performance measured in terms of sending rate, throughput, and utilization factor. The results obtained from the PFTK and PLLDC models were compared with those obtained from equivalent simulations carried-out on the widely used NS-2 network simulator. The PLLDC model provides more accurate results (closer to the NS-2 results) than the PFTK model.

ROHC Tunneling Protocol for Source Specific Multicast Architecture

2012 ◽  
Vol 462 ◽  
pp. 881-885
Author(s):  
Muhammad Imran Sarwar ◽  
Tat Chee Wan ◽  
Pantea Keikhosrokiani ◽  
Sureswaran Ramadass
Keyword(s):  
Packet Loss ◽  
Wireless Mesh Network ◽  
Network Performance ◽  
Distributed Applications ◽  
Mesh Network ◽  
Bandwidth Efficiency ◽  
Video Streams ◽  
Limited Bandwidth ◽  
Wireless Mesh ◽  
Packet Header

Advances in IP-based multimedia applications, coupled with the evolution of wireless networks results in a need for effective multimedia transmission and delivery of real-time content such as voice and video streams. Multicast has the potential to optimize network performance and support scalable distributed applications without compromising performance, efficiency, and QoS but introduces new challenges to transmit multicast packet in the backbone especially when incorporated into wireless mesh network with limited bandwidth. We propose ROHC enabled tunnel for source specific multicast architecture that will compress packet header and aggregate them to together in a frame to save bandwidth and address the packet loss. Our ROHC enabled tunneling protocol will act as SSM channel (S, G) and is designed to be deployed in the backbone network. In this paper, we analyze and discuss the how to address the bandwidth efficiency and packet loss with respect to our proposed tunneling protocol.

scholarly journals Implementation of Hotspot Network for Internal Campus Communications Utilizing Smartphone and Free Software

2017 ◽  
Vol 1 (1) ◽  
pp. 33
Author(s):  
Pande Ketut Sudiarta ◽  
I Putu Ardana
Keyword(s):  
Packet Loss ◽  
Circuit Switching ◽  
Voice Communication ◽  
Campus Environment ◽  
Telephone Communication ◽  
Mobile Phone Technology ◽  
Limited Bandwidth ◽  
Telephone Exchange ◽  
User Data

VoIP (voice over internet protocol) telephone communication using a data network. There is a change in switching from circuit switching technology into packet switching. Phone exchange can now use the Personal Computer equipped VoIP applications. Even the development of mobile phone technology to make VoIP communication can be performed utilizing the Smartphone. VoIP applications commonly use such as WhatsApp, Line etc. However, this application will cut the user data packets and often the quality is not satisfactory due to limited bandwidth and location of the remote server. During this time, Udayana University campus at several locations has been equipped with Hotspot network is mostly used to connect to the Internet. Hotspot network can also be used for voice communications with VoIP technology by adding a VoIP server. This concept not raises communication costs and should produce sound quality will be better because of the close location of the server. Because that researchers need to develop a model of telephone communication network utilizing Smartphone hotspot and students to be able to communicate in a campus environment. Method of this research is to develop a network utilizing hotspots and VoIP telephone exchange using the mini PC installed software FreePBX. On the side of the Smartphone using the free soft phone application and for aircraft used FXS analog phone as a codec. Tests will be performed for the communication between Smartphone and Smartphone to FXS terms of QOS and MOS produced. The results obtained, if the latency and packet loss have a value corresponding to the Real Time Protocol (RTP), the obtained MOS appropriate for the codec used while with the same codec if the value of packet loss and latency results are high then MOS obtained be small or less good quality. So the quality of VoIP is highly dependent on the quality of the signal obtained hotspot. In general, VoIP communication using a Smartphone connected to the server on the network hotspot mini pc can be used as a voice communication on campus.

scholarly journals Evaluasi Kualitas Layanan Jaringan Komputer pada Jaringan Komputer STIKOM Dinamika Bangsa Jambi

2019 ◽  
Vol 8 (1) ◽  
pp. 66-77
Author(s):  
Yudi Novianto ◽  
Abdul Harris ◽  
Lola Yorita Astri
Keyword(s):  
Quality Of Service ◽  
Service Quality ◽  
Data Management ◽  
Computer Networks ◽  
Packet Loss ◽  
Internet Services ◽  
Network Service ◽  
Network Connections ◽  
Qos Parameters

STIKOM Dinamika Bangsa Jambi is an institution engaged in education. It's utilize computer networks to carry out institutional and other data management activities. Network service quality / Quality of Service (QOS) is used to measure the level of performance of internet network connections which aims to improve the quality of internet services for institutions. The method used is the action reseach where in this method before evaluating the quality of the existing network, first step in the diagnosis, planning and retrieving the output of the Axence NetTools5 application. The Quality of Service (QOS) parameters that will be seen through this application include delay, packet loss, bandwidth (throughput). Evaluation is done by comparing the catch with Tiphon's standard where for the delay output obtained between 0-1 ms so that based on the tiphon standard includes a very good category. For packet loss the output obtained is 0% and includes a very good category. As for throughput, the output obtained is at least 75.5287%, so it can be concluded that the quality of output on the network is good.

Sign in / Sign up

Export Citation Format

Share Document