Bio-inspired Autonomic Structures: a middleware for Telecommunications Ecosystems

Manzalini, Antonio; Minerva, Roberto; Moiso, Corrado

doi:10.1007/978-0-387-09753-4_1

Cited by 4 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, bio-inspired autonomic networks have been developed, by applying some biological principle to network management [36]. In the artificial intelligence area several techniques coming from Bayesian probability, evolutionary computation or machine learning areas can be employed.…”

Section: Cognitive Mechanismsmentioning

confidence: 99%

“…In the last decade, the increasing complexity of network infrastructures has raised up the conceptual need of building networks with autonomic management, see [36]. By following the same approach proposed for enabling Autonomic Network Management (ANM), i.e., by using intelligent mechanisms, such as bio-inspired techniques, or more generally particle methods in the application plane of an SDN architecture, we do believe it is possible to enhance the security that is achievable in a Software Defined Network, using only the controller functionalities.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cognition: A Tool for Reinforcing Security in Software Defined Networks

Tantar

Palattella

Avanesov

et al. 2014

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Abstract. Security is one of the most important requirements for networks and serious concerns for network providers and users. Software Defined Networking offers to network managers new opportunities for deploying efficient security mechanisms. By means of applications and controller functionalities, SDN is able to provide a highly reactive network security monitoring, to perform comprehensive traffic analysis, and to enforce fine-grained dynamic access policies. In the present work, we show how such security mechanisms can be further enforced by applying cognitive functions at the SDN application plane. The proposed approach that finds its foundation into the control loops applied in Autonomic Managers Networks (AMNs), can efficiently enable secure and safe SDN scenarios.

show abstract

Section: Cognitive Mechanismsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Cognition: A Tool for Reinforcing Security in Software Defined Networks

Tantar

Palattella

Avanesov

et al. 2014

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

show abstract

“…Swarm and evolutionary algorithms are getting increased research attention from signal processing and communication society as an alternative to conventional optimization techniques. The key advantages of the swarm and evolutionary algorithms are that they are capable of yielding a near-optimal solution to several non-deterministic polynomial-time (NP) hard problems 9,10 and the availability of different algorithms for solving the problems. In MBM-mMIMO, ML performs an exhaustive search over an exponentially large search space 11 for symbol detection problem, and consequently, the detection becomes an NP-hard optimization problem.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Update strategy: The proposed detection algorithm first computes the initial solution x (0) and performs user-ordering as defined in Equation (10). At each parent node, the proposed detection algorithm chooses the child node corresponding to transmit symbol vector of the next user in the sequence O.…”

Section: Edge Weight and Cost Functionmentioning

confidence: 99%

Graph traversal aided detection in uplink MBM massive MIMO based on socio‐cognitive knowledge of swarm optimization

Datta

Bhatia

Mandloi

et al. 2021

Int J Communication

View full text Add to dashboard Cite

Summary Media‐based modulation (MBM) plays a crucial role in enhancing the spectral efficiency and energy efficiency of massive MIMO (mMIMO) systems for 5G and beyond wireless communications. In MBM, multiple parasitic elements, also termed as radio frequency (RF) mirrors, are placed near the transmit antennas for generating different channel fade realizations. These realizations are obtained by ON/OFF switching of RF mirrors. One of those channel fade realizations is selected (using a part of the incoming information bits) for transmitting a part of the information bits utilizing a symbol chosen from the conventional constellation set (using another part of the incoming information bits). Transmission of a symbol through one of the available channel realizations constitutes a sparse transmit vector for each user in MBM‐mMIMO. The sparse nature of transmitted symbols from multiple users and inter‐user interference makes the symbol detection in uplink MBM‐mMIMO challenging. Therefore, in this article, the problem of symbol detection in MBM‐mMIMO is analyzed from a graph‐theoretical point of view, and a graph‐traversal aided low‐complexity symbol detection algorithm is proposed inspired by socio‐cognitive learning of swarm optimization. Also, the convergence characteristic of the proposed technique is investigated theoretically. Further, an analytical expression of upper bound on bit error rate performance is derived and corroborated through simulations. Viability and robustness of the proposed technique are also justified through simulations over state‐of‐art detection techniques, under both perfect and imperfect channel state information scenarios.

show abstract