Planning umts base station location: optimization models with power control and algorithms

Abstract

Classical coverage models, adopted for second-generation cellular systems, are not suited for planning universal mobile telecommunication system (UMTS) base station (BS) location because they are only based on signal predictions and do not consider the traffic distribution, the signal quality requirements, and the power control (PC) mechanism. In this paper, we propose discrete optimization models and algorithms aimed at supporting the decisions in the process of planning where to locate new BSs. These models consider the signal-to-interference ratio as quality measure and capture at different levels of detail the signal quality requirements and the specific PC mechanism of the wideband CDMA air interface. Given that these UMTS BS location models are nonpolynomial (NP)-hard, we propose two randomized greedy procedures and a tabu search algorithm for the uplink (mobile to BS) direction which is the most stringent one from the traffic point of view in the presence of balanced connections such as voice calls. The different models, which take into account installation costs, signal quality and traffic coverage, and the corresponding algorithms, are compared on families of small to large-size instances generated by using classical propagation models.Index Terms-Code-division multiple access (CDMA), optimization algorithms, optimization models, planning, power control (PC), universal mobile telecommunication system (UMTS).

Radio Planning and Optimization of W-CDMA Systems

“…In previous work [2,4] we discussed the importance of power control mechanisms in UMTS network planning. Indeed, the SIR depends on the received powers which in turn depend on the transmitted ones: therefore, these powers have to be adjusted by the PC mechanism so as to minimize interference and guarantee quality.…”

“…Since even the UMTS BS location problem restricted to uplink direction is NPhard [2], we have developed Greedy Randomized Adaptive Search Procedures (GRASP) and Tabu Search algorithms to find good approximate solutions in a reasonable amount of time. GRASP is a simple heuristic in which a randomized greedy procedure is applied a predefined number of times and the best solution found is returned.…”

Coverage planning for optimizing HSDPA performance and controlling R99 soft handover

“…The total power available in cell i is denoted by P tot i . Similar to many previous work on UMTS optimization [2], the network service area is modeled by a regular grid of pixels J = {1, . .…”

“…Research on engineering HSDPA networks ranges from analysis of physicaland link-layer capacity [8,16], resource sharing and admission control strategies [4,21], scheduling policies [6,10,15,18], to performance consideration in network dimensioning and planning [3,20,26,28]. The last topic is becoming increasingly important as the HSDPA service grows rapidly in scale, and it leads to new optimization problems in view of the current literature (e.g., [2,19]). …”

WiMAX Network Planning Using Adaptive-Population-Size Genetic Algorithm

“…When a BS has a capacity limit, the problem is called capacitated; otherwise, it is uncapacitated. Amaldi et al [2] study the problem in uncapacitated UMTS cellular networks by formulating the problem as an Integer Program (IP), and resort to randomized greedy search and tabu search. In Yu et al [11], a two-tier assignment variant is considered to model 802.16j MMR with the assumptions that the RSs and BSs are uncapacitated.…”