Technical Challenges

Figure 1 shows the tasks accomplished by the cooperation and functionality of the CMONs and CSCIs and the technical challenges that OneFIT will aim to overcome, in order to realize the opportunistic network concept. The technical challenges are related to the tasks: suitability determination, creation, maintenance and handling of forced terminations.

Related to the suitability determination are the following technical challenges:

  • Design, development and standardization of discovery procedures enabling communication end-points to discover other devices (which can be part of the infrastructure-less segment of an opportunistic network), infrastructure nodes and other communication end-points. IEEE procedures, using for example beacons, as well as 3GPP mobility management procedures and the 3GPP "Access Network Discovery and Selection Function" (ANDSF) are candidate to be used as a basis and to be expanded by OneFIT.

  • Determination of nodes that are candidate for introduction in the opportunistic network. This requires advanced profile management functionality, for obtaining quantitative information and knowledge/ experience on the behaviour and capabilities of a node, for instance in terms of mobility, traffic (and therefore, spare capacity), overall status (e.g., energy, processing, storage capabilities), etc. This information, and knowledge and experience will be the foundation for deciding whether the node can be candidate for participation in the opportunistic networks.

  • Identification and potential generation of spectrum opportunities. The work will not be about sensing algorithms and strategies. Spectrum opportunity identification and generation will be controlled by the infrastructure and will be agnostic with respect to the way that the information was collected and stored in a spectrum database. The intelligence will be attributed to the CSCI, and will cover: (i) machine learning techniques (for instance, Bayesian networks, reinforcement learning, and other) in order to enable the development of knowledge on the spectrum usage in given time periods and areas; (ii) information and knowledge management means for quickly and efficiently, in terms of signalling, identifying spectrum opportunities; (iii) opportunity generation, i.e., rearrangement of the spectrum assignment in order to release bands that can be temporarily allocated for the formation of the opportunistic network. Interest will also be placed on virtual bandwidth aggregation methods, i.e., the creation of channels through the aggregation of potentially disjoint bands.

  • Interference scenarios resulting from the opportunistic network. Pertinent results will be obtained from simulations (conducted off-line) and will be exploited for assessing the impact and deciding on the suitability of the opportunistic network concept, in the particular cases encountered.

The technical challenge that is related to the creation of the opportunistic network is the solution of the following optimization problem: Given information and knowledge on the context of operation, the profiles and the policies, find the best, feasible configuration of the opportunistic network.

The configuration of an opportunistic network consists of the:

  • Selection of nodes that will participate (subset of the set of candidate ones)

  • Selection of spectrum that will be used for the communication of the nodes

  • Determination of the routing pattern for the interconnection of the nodes.

The problem can be solved in a divide and conquer (e.g., one phase dedicated to each of the points above) or a more aggregate manner. An aggregate solution approach is to derive cost factors (e.g., related to the spectrum bands that can be used, the candidate nodes that can be selected, etc.) from the inputs of the particular instance encountered, and to provide them to a suitably modified routing (e.g., [1 ], [2]) or general optimization algorithm (e.g., [3], [4]). This approach can yield a suite of fully distributed or less distributed solution strategies. Comprehensive studies of the behaviour of the strategies will be conducted. The provision of distributed solutions is the main focus of this challenge, in conjunction with the provision of some less-distributed solutions, which can be used as benchmarks.

Related to the maintenance of the opportunistic network are the following challenges:

  • How to control the QoS offered to the applications that are served by the opportunistic network. This involves monitoring mechanisms and procedures for overcoming degradations, i.e., control of the admission of communication end-points, congestion control and load balancing among the nodes of the opportunistic network. A solution can offer valuable insight in whether Future Internet applications can be adequately served through opportunistic networks, and especially, their infrastructure-less segment.

  • Reconfiguring the network because of changes in the nodes (due to mobility, faults and other potential reasons), the spectrum or routing conditions.

Related to the handling of forced terminations is the challenge of trying to preserve the provision of applications as much as possible.

Meeting aforementioned challenges and providing usable technical solutions will pave the way for innovative models of application provision in the Future Internet.

Figure 1. Role accomplished by the CMONs and CSCIs and technical challenges that will be addressed by the project in order to realize the opportunistic network concept (click image to enlarge)

[1] A. H. Badis, A. Khaldoun Al Agha, "CEQMM: a complete and efficient quality of service model for MANETs", in Proc. of the 3rd ACM international workshop on Performance evaluation of wireless ad hoc, sensor and ubiquitous networks (PE-WASUN'06), Terromolinos, Spain, 2006.
[2] Rocket project Website,
[3] D. Bertsekas, "Network Optimization: Continuous and Discrete Models", Athena Scientific, 1998.
[4] D. Bertsekas, "Linear network optimization, algorithms and codes", MIT Press, 1991.