In beyond 3G systems, one of the important factors is to address Radio access technology (RAT) selection and load balancing between heterogeneous networks to ensure high spectral efficiency in an era where spectral resources are at a premium. This work aims to address the feasibility of utilizing WiFi as complementary service for HSDPA, to prevent quality of service deterioration in the event of network overload in HSDPA during the busy period. The proposed RAT selection algorithm is based on the load of each system, and the results show that the outage probability can be improved by up to 45% relative to a stand-alone HSDPA system.

The purpose of this paper is to present a custom-made cross-layer network simulator for wireless IEEE 802.11e in a multihop environment. A previous work of our research team created a simulator just for the physical plus MAC layers, which in this work was adapted to model the transmission of a packet from a source to a destination using intermediate nodes. Some initial tests were performed using a simple routing algorithm, and the results obtained from these tests indicate that the system is capable of deliver the packets regardless of the source/destination, and successfully calculates several metrics such as end-to-end delay and number of packets lost.

European projects like Simulation of Enhanced UMTS Access and Core Network (IST-SEACORN) proposed a set of enhancements to UMTS. By considering the deployment scenarios and tele-traffic parameters from the Vehicular scenario of SEACORN, a simulator was produced to extract conclusions about blocking and handover failure probabilities. Simulation, which consider the burstiness of traffic, were performed for different cases, from single- to multi-service situations, and from absence to presence of mobility. Besides quality of service results in the air interface, including blocking and handover failure probabilities, the simulator allows for extracting conclusions about the validation of the Bernoulli/ Poisson/Pascal model for the computation of the ONOFF blocking probability, the ratio between the number of call rejected at the beginning of an ON period and the total number of bursts generated during a session. As the theoretical and the simulation results agree, the validation was achieved.

This chapter starts by presenting the Stanford University Interim (SUI) and modified Friis propagation models. Although the SUI model is being recommended for WiMAX, the comparison between the model and experimental results show that, in our environment, at 3.5 GHz, the modified Friis model with gama = 3 fits better the measurement values. From the analyses of the signal-to-noise-plusinterference ratio, SNIR, interference-to-noise ratio and reuse pattern, it is found that both noise and interference present a strong limitation to the performance of fixed WiMAX, mainly for higher order modulation and coding schemes (MCSs). In general terms, the use of sectorization in fixed WiMAX enables to reduce the reuse pattern while considering sub-channelisation allows for improvement on the coverage. The reduction of the reuse pattern directly corresponds to an increase in the system capacity but the improvement in the coverage range (through subchannelisation) can also allow for an improvement in UL system capacity, as adaptive MCS are used. Two different approaches are considered for graphical cellular planning, and the district of Covilha˜ is considered as a case study. On the one hand, one considered a GIS based WiMAX planning tool conceived by considering coverage issues, frequency reuse, and the impact of the different classes of service. On the other, WinpropTM is used as it distinguishes among different MCS in the graphical presentation of the results. Both tools consider the information coming from the digital terrain profile. The GIS functionalities allow for appropriately adjusting azimuth and tilt of antennas. This cellular planning exercises confirm the results of theoretical analysis, where different crowns are achieved for the coverage with each MCS (corresponding to a given range of values for SNIR), for the maximum physical throughput, and for the “best server” cells. The frequency radio resources should be considered as the most valuable resourceduring the planning of wireless broadband access networks. As a rule, spectral efficiency needs to be optimized by using several advanced techniques, corresponding to an optimization from the cost-benefit point of view.

In this work we simulate the ad hoc mode of IEEE 802.11e for routing optimisation. We simulate the behaviour of routing algorithms at the network layer by using a custom-made cross-layer network simulator developed by our team, which simultaneously considers the physical and Medium Access Control (MAC) layers. Although the simulator also supports the infrastructure mode, in this paper we focus on the ad hoc feature which was introduced by the authors.We opted for the simulator approach over the theoretical analysis, butwe also present amathematical model for IEEE 802.11 ad hoc networks. Some initial tests were performed by using a simple routing algorithm (to evaluate the behaviour of the system in terms of selection of the path between a source and a destination, and the correctness of the calculated metrics,which include end-to-end delay, packets lost, packets delivered), but more advanced cross-layer design solutions were also tested. When information from the physical and MAC layers is used as an input to the routing algorithm, improvements are achieved in the performance of the network. Several functions were compared and the algorithm that privileges shorter links accounting with the metric “collision rate” achieves the best results. When compared with a standard routing solution, this cross-layer approach allows to increase the number of packets delivered, while not significantly affecting the end-to-end delay of the packets.

A multi-service traffic model is proposed, and its validation is achieved by using event-based simulation results which consider the burstiness of traffic. By considering the deployment scenarios and tele-traffic parameters from the Vehicular scenario of IST-SEACORN, a simulator was produced to extract conclusions about blocking and handover failure probabilities. Simulations, were performed for different cases, from single- to multi-service situations, and from absence to presence of mobility. Besides results for the quality of service in the air interface, including blocking and handover failure probabilities for several types of traffic, the simulator allows for extracting conclusions about the validation of the Bernoulli/Poisson/Pascal model for the computation of the on/off blocking probability, the ratio between the number of session rejected at the beginning of an on period and the total number of bursts generated during a session. In the single-service case, the theoretical and the experimental values of on/off blocking probability are close to each other. An almost perfect concordance between theoretical and simulation values is verified when the average sojourn time in cells is equal to the average holding time. In the multi-service case, although the results are not exactly coincident a coherent behaviour is achieved for an average traffic per user up to 0.06 Erl.

An overview of All-IP Enhanced Universal Mobile Telecommunication System, E-UMTS, service needs in the business city centre, BCC, scenario is first presented. Then, E-UMTS traffic generation and activity models are described and characterised. System level simulations are carried out and the enhanced performance is demonstrated based in a single quality parameter, which simultaneously accounts for call blocking and handover failure probabilities. End-to-end delays do not present a limitation. By considering a grade of service of 1% for the quality parameter, and different hypothesis for costs and prices, an optimum coverage distance is obtained around ~200-425 m, which maximises the supported throughput per km2. However, results for the profit in percentage indicates that coverage distances in the range 395-425 m should be used in BCC.

Optimization of IEEE 802.11e MAC protocol performance is done by modifying several parameters left open in the standard, like buffer size and acknowledgement policies. By using the developed event-driven simulator, results were extracted with single and multi-service. With only a single service, higher values for the goodput are obtained for video and background (BK) traffics in comparison with the voice (VO) application. Besides, the number of supported user is higher for voice. With mixtures of traffic, when the number of station is small the goodput is lower for VO. However, for higher number of station, the lowest values for the goodput occur for background traffic. By using the Block Acknowledgement (ACK) procedure, for video and BK traffics, the capacity is improved when the number of station is equal or higher than 16 and 12, respectively. From a detailed analysis of the video application, a reduction of more than 40% was achieved in the delay. However, a clear trend is not identified for the optimum block size.

In this paper Enhanced UMTS offices scenarios and classes of services up to wideband are taken into account. A model is proposed for optimising Enhanced UMTS based in costs and revenues. A system level simulator is used to obtain the blocking probability, and other QoS measures, e.g., handover failure probability and delay. Using these results, one obtains the system capacity, i.e., the supported fraction of active users and throughput for a given grade of service. The profit (in percentage) was obtained, and the optimum (most profitable) cell radius was found. A higher number of pico-cells (with a cell radius around 34 m) will be a profitable solution for the optimisation of network planning. This will also allow for increasing system capacity and reducing prices.

This work provides the specification of a MAC layer simulator that evaluates the service quality in WiFi, and presents initial results as a stand-alone simulator. The simulator will account for inter-working with WiMAX and HSDPA in the context of the IT-MOTION tool. Higher throughputs are found for the FTP service and for the video service, mainly because the frames transmitted in these services are longer than the ones for the voice application. However, the number of supported users is higher for voice. This simulator will allow for tuning-up several parameters like the ones related to how to use BlockACK, normal ACK, and NO ACK policies.