John G. Markoulidakis

Evaluation of location area planning scenarios in future mobile telecommunication systems

In third generation mobile telecommunication systems, signalling requirements due to location updating and paging are expected to be remarkable, mainly because of the huge number of mobile subscribers. Location area planning is characterised by the trade-off between the number of location updates and the amount of paging signalling that the network has to deal with. Location area planning should be based on criteria which guarantee that signalling load will be kept under tolerable levels. Various approaches for location area planning in a city environment—the worst case environment—are discussed in this paper. The simplest approach is the use of heuristic algorithms for approximating the optimum location area configuration. Then more complex scenarios which are based on geographical criteria (population distribution, highway topology, etc.) are investigated. Finally, user mobility characteristics are taken into account in the most complex scenarios. All scenarios are applied over a simulation model representing a city area, providing us with a very clear view of their performance characteristics.

Traffic model for third generation cellular mobile telecommunication systems

Traffic analysis for third generation mobile telecommunication systems should take into account a variety of services (e.g., voice, data, video) and “environments” (private, public outdoor, public indoor) as well as the user mobility behavior. The analytical traffic model presented in this paper incorporates all the above mentioned features. The model utilizes a set of simplifying assumptions regarding the distribution of certain random time intervals (e.g., call duration and “cell residence time”) and the handover arrival process. The core of the model focuses on the estimation of the cell border crossing rate and the time it takes a busy mobile user to leave a cell area. This allows for the estimation of the handover rate and the call duration within a cell, which in turn, utilizing an iterative method, allows for the calculation of the offered traffic load per cell. The major advantages of the model are: (a) the simple closed form solutions, (b) its independence from the applied radio resource management scheme, and (c) its accuracy. The latter is validated via a simulation tool, which accommodates different cell-layout scenarios over a geographical area, representing a ‘typical’ city center, modeled as a Manhattan grid.

Performance bounds of a multiple-step paging strategy in future universal mobile telecommunication systems

In future mobile telecommunications, due to the huge number of users and the specific functions that support mobility, a significant amount of signaling load will have to be carried by the finite capacity of the radio link. Hence, methods aiming at radio link signaling load reduction are welcome. In this paper we propose and analyze a method that saves paging signaling load by exploiting information related to the terminal location during the most recent interaction between the terminal and the network. The penalty paid is extra processing power and extra paging delays. An analytical model is developed so as to describe the performance versus traffic intensity and mobility conditions. The performance of the proposed paging scheme is investigated, and it is shown that the method operates well even in the worst case, which is the high user mobility conditions. Possible extensions of the method, which exploit information related to the mobility degree of each individual user or information characterizing the mobility conditions in a certain location area, are also proposed.

Periodic attachment in future mobile telecommunications

In mobile telecommunications the network needs to monitor the terminal status, i.e., switched on, switched off, busy, idle. Once a mobile terminal is switched on, it automatically informs the network accordingly (attachment action). Regarding the way that the network identifies that a terminal is switched off (detachment), a procedure called periodic attachment is proposed: the terminal (when switched on) periodically reports to the network its active status. If the network detects silence for more than an agreed time period it considers the terminal switched off. Paging signalling savings can be achieved if the network is informed early that a mobile terminal is switched off. However, periodic attachment requires access to the network database system, which might cause a bottleneck. In this paper the performance characteristics of periodic attachment and the induced signalling load are evaluated as influenced by the selection of the time-out period. A constant time-out period for the whole day can cause a large amount of database transactions especially during low traffic hours. A method for dynamic time-out period selection to fit the varying conditions during the day is proposed and evaluated. >

The average number of handovers per call in satellite UMTS systems

Satellite systems in future mobile communications will be integrated with terrestrial cellular systems, aiming at worldwide coverage. We develop mobility models for the estimation of the average number of handovers per call in satellite systems. This parameter is crucial especially for low Earth orbit (LEO) systems, where the increased number of handovers affects the quality of service experienced by their users. The proposed mobility models, apart from providing indicative values for the number of handovers in geostationary and LEO systems, also provide means to identify the cell orientation, which minimizes this parameter in LEO systems.

Terminal detach scenarios in future mobile telecommunications

In mobile telecommunications the network should monitor the terminal status, i.e.,switched on, switched off, busy. Once a mobile terminal is switched on, it automatically informs the network about its status (attachment action). However, regarding the way that the network can identify that a terminal is switched off (detachment), there are three alternatives: (a)Periodic detachment scenario. The terminal (when switched on) periodically informs the network about its state. If the network detects that there is no contact with the terminal for more than the agreed time period, it detaches the terminal, (b)Detach on unsuccessful paging. At the instance of a call terminating to a mobile terminal and after a number of unanswered paging messages, the terminal is detached. (c)Detachment action scenario. The moment that the mobile terminal is switched off, it informs the network about its status. Scenarios (a) and (c) claim that paging signaling savings can be achieved when the network is early informed that a mobile terminal is switched off. However, these scenarios require access to the network database, which is expected to be a bottleneck. On the other hand, scenario (b) requires no access to the database but does not guarantee paging signaling savings. In this paper we model the behavior of a terminal so as to compare the above scenarios in terms of performance. Note that paging signaling and database transactions are not the same cost.