NEURAL STRATEGIES TO HANDLE ROUTING IN COMPUTER NETWORKS

1993 ◽  
Vol 04 (03) ◽  
pp. 269-289 ◽  
Author(s):  
S. CAVALIERI ◽  
A. DI STEFANO ◽  
O. MIRABELLA

In this paper, the authors adopt a neural approach to deal with the problem of routing in a packet switching network. The aim is to define a routing strategy which will combine the advantages of both the centralized and the distributed approaches. The neural approach presented is based on the idea of inserting a neural network (N/N) into each node in the computer network which will be responsible for computing the route between its node and the immediately adjacent one. Two distributed routing solutions are presented in the paper based on an optimizing network and a mapping network. The routing obtainable and the implementation resources needed for the two approaches are evaluated. Finally, the performance offered by the neural strategies proposed is compared with that offered by classical distributed and centralized routing solutions. As a parameter of merit, the effect of overloading caused by the additional traffic present in each solution is used.

Author(s):  
Jiankun Hu

he history of computer networks can be traced back to the early 1960s, when voice-grade telephone networks dominated the communication networks. With the increasing importance of computers, as well as the ever-increasing expense of centralized mainframe computers that were growing in size, there was a need to decentralize computer systems. This trend also highlighted the need to connect computers together, by means of computer networks, so that their capacity could be shared among geographically distributed users. Unlike the circuit switching telephone networks, where voice is transmitted at a constant rate between sender and receiver, the traffic in computer networks tends to be bursty. To meet the requirements of data communications, people began to invent more efficient and robust networks, i.e., packet switching networks. The first published work on packet-switching techniques was that of Leonard Kleinrock (Kleinrock, 1961, 1964). The first packet-switching computer network called ARPANET was developed in 1969 and then became the ancestor of today’s public Internet (Kurose & Ross, 2001).


Informatics ◽  
2020 ◽  
Vol 17 (3) ◽  
pp. 78-86
Author(s):  
Ya. V. Bubnov ◽  
N. N. Ivanov

The paper proposes effective method of computer network protection from data exfiltration by the system of domain names. Data exfiltration by Domain Name System (DNS) is an approach to conceal the transfer of confidential data to remote adversary using data encapsulation into the requesting domain name. The DNS requests that transfer stolen information from a host infected by malicious software to an external host controlled by a malefactor are considered. The paper proposes a method of detecting such DNS requests based on text classification of domain names by convolutional neural network. The efficiency of the method is based on assumption that domain names exploited for data exfiltration differ from domain names formed from words of natural language. To classify the requests in convolutional neural network the use of character embedding for representing the string of a domain name is proposed. Quality evaluation of the trained neural network used for recognition of data exfiltration through domain name system using ROC-analysis is performed.The paper presents the software architecture used for deployment of trained neural network into existing infrastructure of the domain name system targeting practical computer networks protection from data exfiltration. The architecture implies creation of response policy zones for blocking of individual requests, classified as malicious.


Author(s):  
Guillermo Agustín Ibáñez Fernández

A computer network consists of computers that communicate via any physical media through a network formed by links and nodes, the nodes being the computers. Computer networks have evolved along their short history. Computer networks have changed drastically in mission and implementation from the early projects supported by the Advanced Research Projects Agency (ARPA) and from other organizations, tracing back the origins to 1962. The ARPA network (ARPANET) consisted initially of a small set of nodes at research centres and universities, connected with links at 56 kbps across the United States. ARPANET was the core of the early Internet, a network for research centres and universities. Computer networks are based on the concept of packet switching within a shared communication medium, as opposite to circuit switching, the dominant paradigm for the precedent telegraph and telephone networks. In 1968 Paul Baran proposed a network system based on nodes that forward datagrams or packets from different users over a common line between computer systems from origin to destination. The packet switching paradigm provides resiliency of network against network node failures, the independent routing of datagrams per node makes possible that the datagrams reach their destination even in presence of multiple node failures. Computer networks hardware and communication protocols evolved through time: the Network Control Protocol (NCP) evolved to the four layer (1978) TCP/IP protocol stack. TCP/IP became dominant against the complex seven layer Open Systems Interconnection (OSI) stack proposed (1977) by International Standard Organization (ISO), too complex for implementation and interoperability. A view of the Internet Engineering Task Force (IETF) and Institute of Electrical and Electronic Engineers (IEEE) protocols stack is shown in Figure 1. The role of IP protocol as simple interconnection protocol between networks using dissimilar LAN technologies (Token Ring, Token Bus, ATM LANE, Ethernet, Wi-Fi) has been essential to build the Internet. However, with the widespread deployment of high performance/cost Ethernet and its self compatibility (10/100/1 Gigabit/10 Gigabit), Ethernet is becoming more and more the interconnecting technology, and Metro Ethernet Services are offered by network service providers. Computer networks are based on communication standards. The reference standardization organizations for computer networks, but not the only ones, are the Internet Engineering Task Force, which is the standard organization for Internet Protocols, and the Institute of Electrical and Electronics Engineers (IEEE) LAN MAN Group (IEEE 802) that elaborates and approves standards on the lower layers of protocol stack. The International Standards Organizations (ISO) and the International Telecommunications Union are also key organizations for computer networks, given the convergence of telecommunication and data networks. The evolution of computer networks affects all protocol layers, including the application layer. The development of increasingly sophisticated computer applications that seamlessly integrate data and communications, creating a new virtual space where actors interact is the objective of “social” applications and collaborative software.


Author(s):  
Guillermo Agustín Ibáñez Fernández

A computer network consists of computers that communicate via any physical media through a network formed by links and nodes, the nodes being the computers. Computer networks have evolved along their short history. Computer networks have changed drastically in mission and implementation from the early projects supported by the Advanced Research Projects Agency (ARPA) and from other organizations, tracing back the origins to 1962. The ARPA network (ARPANET) consisted initially of a small set of nodes at research centres and universities, connected with links at 56 kbps across the United States. ARPANET was the core of the early Internet, a network for research centres and universities. Computer networks are based on the concept of packet switching within a shared communication medium, as opposite to circuit switching, the dominant paradigm for the precedent telegraph and telephone networks. In 1968 Paul Baran proposed a network system based on nodes that forward datagrams or packets from different users over a common line between computer systems from origin to destination. The packet switching paradigm provides resiliency of network against network node failures, the independent routing of datagrams per node makes possible that the datagrams reach their destination even in presence of multiple node failures.


2010 ◽  
pp. 2343-2352
Author(s):  
Guillermo Agustín Ibáñez Fernández

A computer network consists of computers that communicate via any physical media through a network formed by links and nodes, the nodes being the computers. Computer networks have evolved along their short history. Computer networks have changed drastically in mission and implementation from the early projects supported by the Advanced Research Projects Agency (ARPA) and from other organizations, tracing back the origins to 1962. The ARPA network (ARPANET) consisted initially of a small set of nodes at research centres and universities, connected with links at 56 kbps across the United States. ARPANET was the core of the early Internet, a network for research centres and universities. Computer networks are based on the concept of packet switching within a shared communication medium, as opposite to circuit switching, the dominant paradigm for the precedent telegraph and telephone networks. In 1968 Paul Baran proposed a network system based on nodes that forward datagrams or packets from different users over a common line between computer systems from origin to destination. The packet switching paradigm provides resiliency of network against network node failures, the independent routing of datagrams per node makes possible that the datagrams reach their destination even in presence of multiple node failures.


Sign in / Sign up

Export Citation Format

Share Document