Infotech Oulu Annual Report 2014 - Networking (NET)

Professor Savo Glisic, Professor Mika Ylianttila and Dr. Jussi Haapola
Department of Communications Engineering and the Centre for Wireless Communications, University of Oulu
savo.glisic(at)ee.oulu.fi, mika.ylianttila(at)ee.oulu.fi, jussi.haapola(at)ee.oulu.fi
http://www.infotech.oulu.fi/net

 

Background and Mission

The mission of the Networking Group (NET) is to conduct high level research and provide leading solutions in the field of wireless networks as well as the high level graduate and postgraduate education in this filed. NET is internationally perceived as a forerunner in its field, and a valued partner for research cooperation which results in being the member of EU network of excellency in wireless communications, WiFiUS program within the Finland-US cooperation as well as participant of a number of EU funded projects.  

Its success is due to the capability to react fast to the changes taking place in the operational environment, as well as to the needs expressed by its research partners. As a provider of high-quality university training, NET aims at producing theses and dissertations, and peer-reviewed publications of the highest rank and extensive set of postgraduate courses within the PhD program for networking.

 

Scientific Progress

The Networking group (NET) focusing mainly on “Future Wireless Internet” consists of three teams: Networking, Future Wireless Internet and Sensor Network Architectures, and Sensor Networking. The group carries our research in the field of 5G/6G network architectures, spectra management, networks economics and security, as well as the future Internet applications, as portrayed in Figure 1.

Large scale network architectures

# Multihop/medical ICT

# Heterogeneous (macro, small, femto, WLAN, ...)

# Cognitive/semi cognitive

# Multioperator

# DNA networks

# Security

# SDN (Active Networks)

# Network virtualization

# Green networks

# LEN (Low Exposure Networks)

# IoPT/Complex Networks

# Large scale network architectures

# Wireless Internet/instead of wireless access to the Internet

# Robust design of Wireless Networks

# Multioperator Spectra Management, sharing / Spectra
   trading

 

Figure 1. Research focus of NET research group.

 

Scientific problems in which group is engaged in include network optimization on the 2.5 (MAC), 3 and 4 layers; development of secure network architectures; development of efficient sensor network architectures; and efficient protocols on the application layer.

For postgraduate training NET group provides following courses

1. Mobile Telecommunication Systems
2. Communication Networks 1
3. Communication Networks 2
4. Basics of optimization

Postgraduate Courses

1. (7cr/12cp) Topology Control & Graph theory
2. (7cr/12cp) Cognitive Networks & Game theory
3. (7cr/12cp) Networks & Convex optimization theory
4. (7cr/12cp) QoS Management & Queuing theory
5. (7cr/12cp) Multiple Access & Markov Chain Theory
6. (7cr/12cp) Wireless Networks Information Theory
7. (7cr/12cp) Network Protocol Design
8. (7cr/12cp) Radio Resource Management
9. (7cr/12cp) Networks Architectures
    (Sensor Networks, Ad Hoc Networks, Mesh,     
     Mobile Networks, Cellular Networks/LTE,
     WLAN/WiMAX/IEEE 802.XX)
10. (7cr/12cp) Networks Security
11. (7cr/12cp) Cooperative and opportunistic networking
12. (7cr/12cp) Wireless Applications
13. (7cr/12cp) Sensor Networks
14. (7cr/12cp) Internet Economic
15. (7cr/12cp) Advanced course on Networks Optimization
16. (7cr/12cp) Advanced Routing & Network Coding
17. (7cr/12cp) Wireless Internet Core Network
18. (7cr/12cp) Networks Connectivity
19. (7cr/12cp) Bio Inspired Paradigms in Wireless Networks
20. (7cr/12cp) Large scale networks
21. (7cr/12cp) Complex Networks Theory

 

Networking

The group is working on developing new networking paradigms for future wireless networks with the goal of convincing the community and all the relevant players in the field of the necessity to support networking science in order to enable further progress in the wireless business. One of the major contributions is new dynamic network architectures, referred to as DNA networks and microeconomics developed for this type of networks. In this concept, certain classes of wireless terminals can be turned temporarily into an access point any time while connected to the Internet. This creates a Dynamic Network Architecture (DNA) since the number and location of these APs vary in time. In this work, we developed a framework to optimize different aspects of this architecture. First, the dynamic AP association problem is addressed with the aim to optimize the network by choosing the most convenient APs to provide the QoS levels demanded by the users with the minimum cost. Then, an economic model is developed to compensate the users for serving as APs and thus, augmenting the network resources. The users’ security investment is also taken into account in the AP selection. A pre-clustering process of the DNA is proposed to keep the optimization process feasible in a high dense network. To dynamically reconfigure the optimum topology and adjust it to the traffic variations, a new specific encoding of genetic algorithm (GA) is developed. Numerical results show that GA can provide the optimum topology up to two orders of magnitude faster than exhaustive search for network clusters and the improvement significantly increases with the cluster size.

Figure 2 illustrates the adaptation of the networks when changes in the number of terminals N and the number of access point M, denoted as DNA(N,M) network, vary in time. The performance of the network is quantified in terms of the network utility which is proportional to the network throughput and inversely proportional to the power consumption. 

Figure 2. Dynamic Topology and Architecture reconfiguration scenarios.

 

Another example of significant result is work in the area of cognitive networks.

Cognitive radio technology enables secondary users (SUs) to opportunistically use the vacant licensed spectrum and significantly improves the utilization of spectrum resource. Traditional architectures for cognitive radio networks (CRNs), such as cognitive cellular networks and cognitive ad hoc networks, impose energy-consuming cognitive radios to SUs’ devices for communication and cannot efficiently utilize the spectrum harvested from the primary users (PUs). To enhance the spectrum and energy efficiencies of CRNs, we have designed a new architecture, which is called the Cognitive Capacity Harvesting network (CCH). In CCH, a collection of relay stations (RSs) with cognitive capability are deployed as shown in Figure 3 to facilitate the accessing of SUs. In this way, the architecture not only removes the requirement of cognitive radios from SUs and reduces their energy consumption, but also increases frequency reuse and enhances spectrum efficiency. In view of the importance of the RSs on the improvement of spectrum and energy efficiencies, in this work, we developed the RS placement strategy in CCH. A cost minimization problem is mathematically formulated under the spectrum and energy efficiency constraints. Considering the NP hardness of the problem, we designed a framework of heuristic algorithms to compute the near-optimal solutions. Extensive simulations show that the proposed algorithms outperform the random placement strategy and the number of required RSs obtained by our algorithms is always within 2 times of that in the optimal solution.

Figure 3. Cognitive network architecture.

 

Significant results have been also obtained in the area of spectrum trading. Spectrum trading creates more accessing opportunities for secondary users (SUs) and economically benefits the primary users (PUs). However, it is challenging to implement spectrum trading in multi-hop cognitive radio networks (CRNs) due to harsh cognitive radio (CR) requirements on SUs’ devices, uncertain spectrum supply from PUs and complex competition relationship among different CR sessions. Unlike the per-user based spectrum trading designs in previous studies, in this work, we developed a novel session based spectrum trading system, spectrum clouds, in multi-hop CRNs. In spectrum clouds, illustrated in Figure 4 we introduce a new service provider, secondary service provider (SSP), to facilitate the accessing of SUs without CR capability and harvest uncertain spectrum supply. The SSP also conducts spectrum trading among CR sessions w.r.t. their conflicts and competitions. Leveraging a 3-dimensional (3-D) conflict graph, we mathematically describe the conflicts and competitions among the candidate sessions for spectrum trading. Given the rate requirements and bidding values of candidate trading sessions, we formulate the optimal spectrum trading into the SSP’s revenue maximization problem under multiple cross-layer constraints. In view of the NP-hardness of the problem, we develop heuristic algorithms to pursue feasible solutions. Through extensive simulations, we show that the solutions found by our algorithms are close to the optimal one.

Figure 4. Spectrum clouds.

 

Future wireless internet and sensor network architectures

Future Internet team has focused on the topics related to secure and trustworthy wireless internet and working in close cooperation with Centre for Internet Excellency (CIE). In the Internet applications, the goal is to plan a networking architecture which is highly scalable, adaptable, energy efficient and secure. These new application areas present new requirements for wireless networking.  Solutions has been demonstrated on the existing platforms within projects such as SIGMONA and CONVINCE.

The SIGMONA project, SDN Concept in Generalized Mobile Network Architectures, studied network architectures and functions for evolution of the LTE/EPC (3GPP) mobile networks. The project applied the latest networking and computing technologies and architectures onto the LTE/EPC mobile network. The project aimed at evaluation, specification and validation of a Software Defined Mobile Network concept designed onto the software defined networking (SDN), network virtualization and cloud computing principles. The project provided an insight into the feasibility and opportunities of such network concepts, as well as evaluated the limits of performance and scalability of the new technologies applied on mobile broadband networks. New opportunities for traffic, resource and mobility management were studied. New challenges on network security and solutions for those were addressed. Network virtualization solutions in the mobile transport networks, as well as effects on the network monitoring and network management solutions are relevant, too. Use cases, such as content delivery were considered.

CONVINcE addresses the challenge of reducing the power consumption in IP-based video networks with an end-to-end approach. The idea of the CONVINcE (Consumption OptimizatioN in VIdeo Networks) project grew out of two observations. The first one is that the Internet's carbon footprint would exceed by 2020 those of air travel by a factor of two. The second observation is that video drives the Internet traffic: The sum of all forms of IP video will be 80% to 90% of total IP traffic in 2017. Furthermore, should people not be convinced by the necessity to reduce carbon dioxide emission, they will listen to the economic aspect of the problem: The price of electricity is increasing and will still increase in the coming years. The power consumption characteristics are studied and optimized along the media path from the Head End, where content is encoded, to the terminals, where the content is consumed, embracing of course the CDN (Content Distribution Network) and the core access networks between them. The partners’ efforts will concentrate on architectures, hardware and software design, protocols and basic technologies in the devices. In parallel to these activities focused on optimizing the power consumption in each part of the system, the project will run transversal activities on “Software best practices & Eco-design” and “Power & QoE measurements”. The results of the project will be visible through demonstrators targeting tests of which conclusions will be made public in order to disseminate the best practices in the domain.

Sensor networking

The main research topics of the group consider urban critical infrastructure monitoring in the specific topics of smart energy grids, municipal water distribution, and disaster prediction, detection, and recovery monitoring. There are three major solution types considered: long-range (up to 20 km single-hop) wireless sensor networks, hybrid multihop sensor-public telecommunications infrastructure networks, and device-to-device communication networks using ad hoc LTE communications. All of the topics share common challenges that are need for longevity where it is impractical to service sensor nodes; harsh communications propagation environment in urban areas due to location of the sensor nodes and potentially long distances; need to cover large areas with thousands of nodes that have a drastically changing communications pattern from low-rate periodic measurement reporting to highly delay-sensitive critical message delivery; and the communications environment is susceptible to significant fluctuation due to temporal interference and obstructions in urban environment. In addition, the themes involve real public or private operators whose core business is not communications. Hence, using standardised solutions is a requirement to build networks.

Figure 5 depicts some of the ways in which, e.g. smart grid communications may be made resilient to temporary equipment malfunction, failure, or outside interference. The research challenges are how to design communications algorithms that can successfully deliver well over 99% of generated traffic even if large parts of supporting infrastructure fails and provide very low latencies, down to few tens of ms in case of emergency messages. Such scale sensor networks are impractical to implement, therefore, professional network simulation tools are used and verified with analytical approaches, such as Markov models. The group is also actively participating in communications standardisation in the IEEE and ETSI fora.

Figure 5. Example of device-to-device and hybrid sensor-LTE communications topology in smart grids.

 

Doctoral Theses

Namal, S (2014) Enhanced communication security and mobility management in small-cell networks. Acta Universitatis Ouluensis, Technica C 508.

 

Personnel

professors

2

postdoctoral researchers

1

doctoral students

7

total

10

person years for research

9

 

 

External Funding

Source

EUR

Tekes

370 000

international

121 000

total

  491 000

 

Selected Publications

[1] C.W. Patterson, A.B. MacKenzie, S. Glisic, B. Lorenzo, J. Roning, L.A. DaSilva, An economic model of subscriber offloading between Mobile Network Operators and WLAN operators, International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt), 2014 12th, Page(s): 444 – 451.

[2] A.S. Shafigh, B. Lorenzo, S. Glisic, J. Perez-Romero, L.A. DaSilva, A.B. MacKenzie, J. Roning, A Framework for Dynamic Network Architecture and Topology Optimization IEEE/ACM Transactions on Networking, 2015.

[3] M. Li, S. Salinas, P. Li, X. Huang, Y. Fang, S. Glisic, Optimal Scheduling for Multi-Radio Multi-Channel Multi-Hop Cognitive Cellular Networks, IEEE Transactions on Mobile Computing, Volume: 14, Issue: 1, 2015, Page(s): 139 – 154.

[4] M.F. Hanif, L.-N. Tran, M. Juntti, S. Glisic, On Linear Precoding Strategies for Secrecy Rate Maximization in Multiuser Multiantenna Wireless Networks, IEEE Transactions on  Signal Processing,Volume: 62, Issue: 14, 2014, Page(s): 3536 – 3551.

[5] Ming Li, Pan Li, Xiaoxia Huang, Yuguang Fang, S. Glisic, Energy Consumption Optimization for Multihop Cognitive Cellular Networks, IEEE Transactions on  Mobile Computing,Volume: 14, Issue: 2, 2015, Page(s): 358 – 372.

[6] Miao Pan, Pan Li, Yang Song, Yuguang Fang, P. Lin, S. Glisic, When Spectrum Meets Clouds: Optimal Session Based Spectrum Trading under Spectrum Uncertainty, IEEE Journal on  Selected Areas in Communications, Volume: 32,  Issue: 3, 2014, Page(s): 615 – 627.

[7] P. Porambage, C. Schmitt, P. Kumar, A. Gurtov, M. Ylianttila, "PAuthKey: A Pervasive Authentication Protocol and Key Establishment Scheme for Wireless Sensor Networks in Distributed IoT Applications," International Journal of Distributed Sensor Networks, Article ID 357430, 2014.

[8] P. Kumar, M. Ylianttila, A. Gurtov, S-G Lee and H-J. Lee, "An Efficient and Adaptive Mutual Authentication Framework for Heterogeneous Wireless Sensor Network-Based Applications", Sensors 2014, 14(2), 2732-2755.

Last updated: 2.6.2016