Aspects of cellular planning for E-UMTS (Enhanced UMTS) are addressed, and results for system capacity are obtained, as a function of the coverage distance. E-UMTS traffic generation and activity models are described and characterised, based on population and service penetration values. An urban scenario was defined with a selection of relevant multi-rate bursty applications. By using a System Level Simulator results were obtained for the end-to-end delay, and for blocking and handover failure probabilities. By using these results, models for the supported fraction of active users and throughput were obtained, as a function of the cell coverage distance. When the coverage distance decreases from 600 to ~200 m, the supported throughput per km2 increases from ~1Mb/s to ~5 Mb/s. By using trisectorial antennas with one amplifier a maximum throughput per BS of 600kb/s was obtained.

Optimisation of the cellular planning process for multi-service Enhanced UMTS in business city centre environments is addressed by using results for system capacity as an input for a cost/revenue function. Based in a definition of simulations scenarios, traffic behaviour, and Quality of Service aspects, a System Level Simulator is used for obtaining system capacity results in the business city centre scenario when overload conditions are considered. Optimum values of the cell coverage distance that maximise the profit per unit length were found. For prices of 5 and 7 Euro cents/MB, optimum values of 80 and 150 % occur around 240m, respectively.

After 3G mobile and wireless systems take-up, the research community is now directing its interest towards unified ways of looking at system design, optimization, and quality of service (QoS) issues to satisfy the requirements of next generation multi-service mobile and wireless IP-based networks. Their implementation requires the development of IP QoS architecture and mechanisms encompassing the various layer of the OSI Model to cope with the new requirements. This work identifies the requirements on latency (link radio), latency (end-to-end), bit error rate, data rates, and traffic characterisation for mobile multimedia network beyond 3G. Characterisation parameters were identified and values were suggested for their range of variation, in the context of cross-layer design.

The analysis of the impact of voice-over-IP into HSPA and LTE multi-service traffic is a challenge. The validation of a Bernoulli-Poisson-Pascal multi-service traffic model in the presence of mobility is explored for different values for the number of channels. A mixture of voice (VOI) and video telephony (VTE) is considered. In a previous work, a critical situation was assumed where the total number of channels was exactly the same as needed by four VTE sessions. However, when few extra channels are added there is a significant reduction on VTE on/off blocking probability. With this few extra resources, VOI can now be easily supported. Since it is supported on top of a system saturated with VTE sessions, Pbonoff decreases 4.6 times for VTE. Since the number of resources is not critical anymore, the range of values of the offered traffic for which the theoretical model agrees with simulation results becomes wider. There is no significative difference between the multi-service traffic results for the seven cell geometry and the roundabout (or ring) cellular topologies in the presence of mobility.

An overview of Enhanced UMTS (E-UMTS) deployment scenarios and supported services is presented, based on the views of nowadays-relevant players. The influence of mobility in E-UMTS is discussed and a set of mobility models is given. Then, a set of scenarios is described, that put together the most important components of E-UMTS deployment situations. These include expected population density, mobility characteristics, and expected usage of service mix, for each environment. A number of nearly thirty applications, divided into in five groups (Sound, High Interactive Multimedia, Narrowband, Wideband, Broadband), is considered.

An overview of E-UMTS deployment scenarios and supported services is presented, based on the views of currently relevant players. Deployment and mobility scenarios include expected population density, and usage of service mix, for each environment. Nearly thirty applications are considered. However, a reduced set of applications is needed for simulations purposes, and scenarios were defined with a selection of the most relevant applications. E-UMTS traffic generation and activity models, based on population and service penetration values, are also described and characterised. ON/OFF states have been characterised by appropriate statistical distributions and parameters.

Based in a definition of simulations scenarios, traffic behaviour, and QoS (quality of service) aspects, a SLS (system level simulator) is used for obtaining system capacity results for urban and business city centre (BCC) scenarios, when overload conditions are considered. Optimum values of the cell coverage distance that maximise the profit per unit length (in BCC) or by unit area (urban) were found. For prices of 5 and 7 Euro cents/MB, while in urban scenarios, optimum values of 20 and 70 %, respectively, were found for profit (for coverage distances of 217 and 621 m), in BCC, optimum values occur around 240 m, and are 80 and 150 %, respectively.

A multi-service traffic model is proposed, and its validation is achieved by using event-based simulation results. In the single-service case, theoretical and experimental results for ON-OFF blocking probability are close to each other, and there is an almost perfect concordance between theoretical and simulation values when the average sojourn time in cells is equal to the average holding time. In the multi-service case, the behaviour is not exactly the same but a coherent behaviour is achieved for an average traffic per user up to 0.10 Erl.

Terceira geração (3G) é o termo genérico utilizado para a geração de sistemas móveis de telecomunicações imediatamente posterior ao GSM. Os sistemas 3G procuram garantir melhores serviços para a transmissão de voz, textos e dados. O Sistema Universal de Telecomunicações Móveis (UMTS  Universal Mobile Telecommunications System) é parte do sistema global de comunicações de terceira geração da União Internacional de Telecomunicações (ITU). A 3G foi baseada na norma GSM actual, porém evoluída, estendida e melhorada de forma a incluir novas especificações para a interface ar mais adequada para os serviços de alta velocidade e multimédia. Este sistema permite que os utilizadores da segunda geração, GSM, migrem facilmente para os serviços de terceira geração. O UMTS padrão pertence à terceira geração de redes móveis. O objectivo primordial da 3G é a realização de uma estrutura universal capaz de suportar tanto os serviços existentes como os futuros. A 3G está normalizada pelas sociedades 3G Partnership Projects: 3GPP (na Europa) e 3GPP2 (nos EUA). Os conceitos tecnológicos para os sistemas 3G continuam em desenvolvimento e proporcionaram já a definição da geração 3.5, em que se inclui o UMTS Melhorado (E-UMTS), uma rede totalmente IP (All-IP) que pretende assegurar maior capacidade, flexibilidade na rede nuclear de suporte (core network) e utilização de serviços integrados suplementares. Dada a complexidade das redes UMTS e E-UMTS, nem sempre é possível desenvolver modelos analíticos simples para projectar e planear o sistema. Para ultrapassar essas dificuldades recorre-se frequentemente a simulações, que se baseiam num desenho topológico existente ou planeado, numa estimativa da distribuição do tráfego e numa variedade de serviços com múltiplos requisitos de qualidade de serviço (QoS), o que constitui uma grande diferença relativamente às redes de segunda geração.

A multi-service traffic model is presented, and its validation is achieved in a vehicular scenario by using event-based simulation results. In the single-service case, theoretical and experimental results for ON-OFF blocking probability are close to each other, and there is an almost perfect concordance between theoretical and simulation values when the average sojourn time in cells is equal to the average holding time. The challenges presented by the multi-service case are identified, and a question related with a possible correspondence between blocking/handover failure probabilities and ON-OFF blocking probability are presented.