Doi: 10. 1016/S0305-0548(03)00202-8
Download 490.36 Kb. Pdf ko'rish
|
10.1016@S0305-05480300202-8
Computers & Operations Research 32 (2005) 59–86 www.elsevier.com/locate/dsw Cost-optimal topology planning of hierarchical access networks Istv(an G(odor ∗ , G(abor Magyar Ericsson Research, Trac Analysis and Network Performance Laboratory, P.O. Box 107, H-1300 Budapest 3, Hungary Abstract This paper deals with the problem of cost-optimal hierarchical topology planning for telecommunications access networks. The generalized problem is computationally hard and only its substantially simpli/ed variants have been studied before. We propose a heuristic algorithm relying on iterative problem decomposition, clustering methods and local optimization. The main idea of the approach is to combine clustering and local optimization operators and to apply the combined operators iteratively within the levels of the hierarchy. By an extensive empirical analysis it is shown that the presented method e3ciently solves the underlying optimization problem for instances of practical size, furthermore, it is robust to handle 4exibility in the problem statement and in the cost function. ? 2003 Elsevier Ltd. All rights reserved. Keywords: Network planning; Topology optimization; Location science; Hierarchical concentrator location; Mobile telecommunications; UMTS 1. Introduction This paper deals with cost-optimal topology planning of hierarchical telecommunication access networks. The main planning task is to aggregate the user tra3c in a multi-level tree-like fashion by the use of intermediate concentrator nodes in a cost-optimal way. We note that the problem statement is general, therefore the proposed algorithms can be applied to several related problems in the areas of facility location, hierarchical location-allocation, etc. Current communication networks are typically separated into two major parts. One part is the core network that consists of high-capacity switching equipment interconnected by a mesh of high-bandwidth links. The other part is the access network that carries tra3c between user terminals and the core network nodes. The core network interconnects di?erent areas of the access ∗ Corresponding author. 0305-0548/$ - see front matter ? 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0305-0548(03)00202-8 60 I. G-odor, G. Magyar / Computers & Operations Research 32 (2005) 59–86 network and interfaces towards external networks. The two parts are typically planned and operated separately. In this paper, we deal with the access network. Cost-optimal planning of access networks is a critical issue in designing today’s communications networks, especially in the case of mobile networks (for example, GSM or UMTS, see Section 4 ). The typically smaller link capacities in an access network are more sensitive to improper dimen- sioning than the capacities of large core network links; precise planning of links is crucial in order to meet the quality of service requirements. In addition, access network represents a major part of the total network cost, considering only their sheer size. The size and complexity of mobile access networks is becoming enormous making manual planning a di3cult and time-consuming task. The solution obtained by algorithmic optimization is superior to that of manual planning. Quick algo- rithmic optimization allows the evaluation and comparison of di?erent deployment strategies; this is not possible in case of manual network planning within the very short timeframes. A very important requirement of the optimization procedure is technology-neutrality, as technology can change very rapidly in the telecommunication environment. Cost-based access network optimization algorithms play a central role in network planning software tools; their quality and e3ciency have an important in4uence on the usefulness of the whole software package. Our problem is to plan an access network, which consists of a number of trees. There are several constraints on these access trees making the problem NP-hard. Therefore, we do not expect to /nd the globally optimal solution, but we rely on heuristic algorithms. The main idea of our approach is to combine clustering and local optimization operators and to iterate the combined operators within the levels of the hierarchy. The rest of the paper is organized as follows. Section 2 de/nes the optimization problem. Section 3 brie4y investigates the related work in the /eld. Section 4 presents an application example from the telecommunications sector; this environment will be used when presenting computational results. Section 5 describes our algorithms, with the detailed pseudocodes in the Appendix A . Section 6 demonstrates the robustness and e3ciency of our algorithms through an ex- tensive empirical comparison. Section 7 contains the conclusions and outlines further open questions. 2. Problem denition 2.1. Network model Fig. 1 gives a simpli/ed illustration of the network model that is used in the case of mobile access network planning. The access network is modeled by a directed graph and it consists of aggregation trees that collect the tra3c demands of base station nodes. The aggregation trees have a speci/ed depth limit and there are certain degree constraints for intermediate concentrator nodes in each tree level. In each aggregation tree, the tra3c 4ows through special concentrator nodes upwards to the root node through transmission links. The concentrator nodes are selected from the nodes, therefore the concentrators have their own tra3c demands as well. 2.2. Input Let N = {1; : : : ; n} denote the set of nodes. Each input node i ∈ N is associated with trac T i ∈ T to be aggregated. Note that the tra3c can be any kind of complex data, we only require I. G-odor, G. Magyar / Computers & Operations Research 32 (2005) 59–86 61 Download 490.36 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling