György Dán | KTH Royal Institute of Technology (original) (raw)
Papers by György Dán
arXiv (Cornell University), May 15, 2017
Eurasip Journal on Wireless Communications and Networking, Sep 24, 2022
International Workshop on Peer-to-Peer Systems, Feb 25, 2008
INSIGHT
ABSTRACTNetwork softwarization has revolutionized the architecture of cellular wireless networks.... more ABSTRACTNetwork softwarization has revolutionized the architecture of cellular wireless networks. State‐of‐the‐art container based virtual radio access networks (vRAN) provide enormous flexibility and reduced life‐cycle management costs, but they also come with prohibitive energy consumption. We argue that for future AI‐native wireless networks to be flexible and energy efficient, there is a need for a new abstraction in network softwarization that caters for neural network type of workloads and allows a large degree of service composability. In this paper we present the NeuroRAN architecture, which leverages stateful function as a user facing execution model, and is complemented with virtualized resources and decentralized resource management. We show that neural network based implementations of common transceiver functional blocks fit the proposed architecture, and we discuss key research challenges related to compilation and code generation, resource management, reliability and s...
2017 IEEE International Conference on Smart Grid Communications (SmartGridComm)
IEEE Transactions on Vehicular Technology
2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm), 2018
2016 28th International Teletraffic Congress (ITC 28), 2016
IEEE Transactions on Control of Network Systems, 2022
Lecture Notes in Computer Science, 2019
Recently, there has been substantial progress in the formal understanding of how caching resource... more Recently, there has been substantial progress in the formal understanding of how caching resources should be allocated when multiple caches each deploy the common LRU policy. Nonetheless, the role played by caching policies beyond LRU in a networked setting where content may be replicated across multiple caches and where channels are unreliable is still poorly understood. In this paper, we investigate this issue by first analyzing the cache miss rate in a system with two caches of unit size each, for the LRU, and the LFU caching policies, and their combination. Our analytical results show that joint use of the two policies outperforms LRU, while LFU outperforms all these policies whenever resource pooling is not optimal. We provide empirical results with larger caches to show that simple alternative policies, such as LFU, provide superior performance compared to LRU even if the space allocation is not fine tuned. We envision that fine tuning the cache space used by such policies may...
IEEE Transactions on Mobile Computing, 2021
2020 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm), 2020
IEEE Journal on Selected Areas in Communications, 2019
IEEE Transactions on Instrumentation and Measurement, 2019
IEEE Transactions on Communications, 2018
The best effort service offered by the Internet is notsatisfactory for the transmission of loss a... more The best effort service offered by the Internet is notsatisfactory for the transmission of loss and delay sensitivedata, such as real-time voice and video. In order to provideservice guarantees at ...
The Internet has rapidly evolved from being a scientific experiment to a commercial network conne... more The Internet has rapidly evolved from being a scientific experiment to a commercial network connecting millions of hosts that carries traffic generated by a large amount of applications with divers ...
In this paper we propose an analytical model to evaluate the end-to-end loss and delay characteri... more In this paper we propose an analytical model to evaluate the end-to-end loss and delay characteris-tics for live multicast streaming. We consider multiple-tree architectures and push based transmission. We give an asymptotic bound on the performance of large overlays, and present a necessary condition for the overlay to be stable. Based on the model we show how the overlay's loss characteristics is influenced by the number of distribution trees, the error control solutions used, the allocation of the bandwidth resources of the peers between the trees and the number of nodes in the overlay. Based on our results, we propose a tree structure that improves the stability of the overlay. We use the model to show how the structure of the overlay and the available bandwidth influence the delay characteristics of the data distribution and the playout buffer requirements.
arXiv (Cornell University), May 15, 2017
Eurasip Journal on Wireless Communications and Networking, Sep 24, 2022
International Workshop on Peer-to-Peer Systems, Feb 25, 2008
INSIGHT
ABSTRACTNetwork softwarization has revolutionized the architecture of cellular wireless networks.... more ABSTRACTNetwork softwarization has revolutionized the architecture of cellular wireless networks. State‐of‐the‐art container based virtual radio access networks (vRAN) provide enormous flexibility and reduced life‐cycle management costs, but they also come with prohibitive energy consumption. We argue that for future AI‐native wireless networks to be flexible and energy efficient, there is a need for a new abstraction in network softwarization that caters for neural network type of workloads and allows a large degree of service composability. In this paper we present the NeuroRAN architecture, which leverages stateful function as a user facing execution model, and is complemented with virtualized resources and decentralized resource management. We show that neural network based implementations of common transceiver functional blocks fit the proposed architecture, and we discuss key research challenges related to compilation and code generation, resource management, reliability and s...
2017 IEEE International Conference on Smart Grid Communications (SmartGridComm)
IEEE Transactions on Vehicular Technology
2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm), 2018
2016 28th International Teletraffic Congress (ITC 28), 2016
IEEE Transactions on Control of Network Systems, 2022
Lecture Notes in Computer Science, 2019
Recently, there has been substantial progress in the formal understanding of how caching resource... more Recently, there has been substantial progress in the formal understanding of how caching resources should be allocated when multiple caches each deploy the common LRU policy. Nonetheless, the role played by caching policies beyond LRU in a networked setting where content may be replicated across multiple caches and where channels are unreliable is still poorly understood. In this paper, we investigate this issue by first analyzing the cache miss rate in a system with two caches of unit size each, for the LRU, and the LFU caching policies, and their combination. Our analytical results show that joint use of the two policies outperforms LRU, while LFU outperforms all these policies whenever resource pooling is not optimal. We provide empirical results with larger caches to show that simple alternative policies, such as LFU, provide superior performance compared to LRU even if the space allocation is not fine tuned. We envision that fine tuning the cache space used by such policies may...
IEEE Transactions on Mobile Computing, 2021
2020 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm), 2020
IEEE Journal on Selected Areas in Communications, 2019
IEEE Transactions on Instrumentation and Measurement, 2019
IEEE Transactions on Communications, 2018
The best effort service offered by the Internet is notsatisfactory for the transmission of loss a... more The best effort service offered by the Internet is notsatisfactory for the transmission of loss and delay sensitivedata, such as real-time voice and video. In order to provideservice guarantees at ...
The Internet has rapidly evolved from being a scientific experiment to a commercial network conne... more The Internet has rapidly evolved from being a scientific experiment to a commercial network connecting millions of hosts that carries traffic generated by a large amount of applications with divers ...
In this paper we propose an analytical model to evaluate the end-to-end loss and delay characteri... more In this paper we propose an analytical model to evaluate the end-to-end loss and delay characteris-tics for live multicast streaming. We consider multiple-tree architectures and push based transmission. We give an asymptotic bound on the performance of large overlays, and present a necessary condition for the overlay to be stable. Based on the model we show how the overlay's loss characteristics is influenced by the number of distribution trees, the error control solutions used, the allocation of the bandwidth resources of the peers between the trees and the number of nodes in the overlay. Based on our results, we propose a tree structure that improves the stability of the overlay. We use the model to show how the structure of the overlay and the available bandwidth influence the delay characteristics of the data distribution and the playout buffer requirements.