Mobile and Broadband Access Networks Lab session OPNET: UMTS - Part 2 Background information

Transcription

1 Mobile and Broadband Access Networks Lab session OPNET: UMTS - Part 2 Background information Abram Schoutteet, Bart Slock 1 UMTS Practicum CASE 2: Soft Handover Gain 1.1 Background The macro diversity gain is responsible for the increase in system performance caused by soft handovers. The soft handover gain can be easily illustrated using the following example. As a radio connection is supported by two or more radio links during the soft handover, it is possible to use links with lower quality as long as the combined effect is sufficient to provide the requested quality of the radio connection. Imagine a desired BLER of 1/ When using two bearers with a BLER of only 1/100 this can be achieved, as a block error will only occur if both links fail. This will happen with a probability of 1/100 * 1/100 = 1/ Lowering the link quality enables the transmission power to be kept at lower levels resulting in less interference in the system. On the other hand every soft handover results in one or more extra connections in the radio network layer, consuming additional resources and causing interference. In the downlink direction an additional base station transmitting at relatively high power levels causes the total amount of power present in the system to rise significantly. In the uplink direction a mobile station in soft handover also transmits two or more signals - which are received by two or more different Node B s. Hence the interference present in the system increases although not as significantly as in the downlink direction. This is due to the higher power levels used by the transmitter in the base station. Typically in the downlink data is transmitted at values up to some Watts whereas mobile terminals transmit at around 125mW. Soft handovers do not only consume extra radio resources but multiple-link connections require also more orthogonal codes and consume extra resources in the Node B and the RNC. Also the signalling capacity over the Iub interface between the Node B and the RNC in the UTRA network part is more extensively used when multiple links are allowed. For efficient resource usage it is therefore necessary to allow the appropriate amount of soft handovers in the system to maximise the soft handover gain while keeping the increase in interference due to multiple link connections below acceptable levels. Another advantage of soft handover is its soft capacity. CDMA networks are not hard capacity limited. This means that additional users cannot be hard blocked due to a 1

2 lack of timeslots or shortage in the number of copper wires available, as is the case in for example GSM and POTS networks respectively. Instead of being hardware limited, CDMA networks are interference limited. This means that every additional user will gradually degrade the noise figure in the system until the network is fully loaded. The interferencedetermined behaviour of soft handover networks makes it possible for loaded cells to borrow capacity from surrounding cells with lower traffic density. 1.2 Exercise 1 : Soft Handover Gain - Power The scenario The scenario should look like Figure 1. This scenario consists of a simple network consisting of three Node Bs, connected to a single RNC. One user moves along a straight-line path at a constant speed. Figure 1: Scenario for the SHO gain - Power Exercise In this exercise, we investigate how the use of soft handover enables the transmission power to be kept at lower levels resulting in less interference in the system. We will determine the soft handover threshold value for which the average uplink transmission power is still lower than the average uplink transmission power for hard handover. 1.3 Exercise 2 : Soft Handover Gain - Throughput 2 2 of 6

3 1.3.1 The scenario The scenario consists of a central cell surrounded by 6 adjacent cells, all connected to the same RNC. A high number of static users (14) was distributed in the hexagonal cell coverage area of the central Node B. As other cells surround the central cell it is very likely that users in the loaded cell - especially those located near the cell edge area - will have connections with the Node Bs from the adjacent cells. The 6 mobile terminals, one in each of the surrounding cells, move towards the coverage area of the central cell during the simulation. Each of the mobile terminals (21 in total) try to set up traffic flows of the four UMTS traffic classes (conversational, streaming, interactive and background). The configuration of the Sample load of OPNET is used Exercise In this exercise, we investigate how soft handover networks make it possible for loaded cells to borrow capacity from surrounding cells with lower traffic density. 3 3 of 6

4 2 UMTS Practicum CASE 3: Coverage and Quality of Service 2.1 Exercise 1: Capacity and Coverage Background UMTS makes use of CDMA (Code Division Multiple Access) as its multiple access scheme in its radio access layer. This scheme offers a better performance against radio signal interference and multi-path fading then for example TDMA or FDMA. The number of users per cell is also not hard limited by a fixed number of time slots or frequency bands. CDMA systems are commonly known as spread spectrum systems because to transmit the information, a larger amount of bandwidth than would normally be used, is utilized. This reduces the average power spectral density of the signal and hence reduces generated interference. In contrary to TDMA/FDMA systems, all users are transmitting in the same frequency band, simultaneously. By assigning different spreading codes to all users, distinction can be made between them. As the different codes have a very small - theoretically zero - cross correlation it is possible to despread the signal at the receiver and retrieve the desired original data signal for each user even when very low transmitter powers are used. This results in a much more efficient use of the frequency available. Power control algorithms are vital for operating a WCDMA system. To counter the nearfar effect it is essential to equalize the received power at the base station for all mobile terminals transmitting in the same frequency band. Closed (inner) loop power control: CL power control algorithms are the main means to counter the uplink near-far effect. The fast power control algorithm works as shown in Figure 2. Every 667ms (1/1500 Hz) the base station compares the estimated SIR (Signal to Interference Ratio) of each mobile station s signal, with a SIR target value. If the measured SIR is higher than the target SIR, the base station will command the MS to power down; in the other case the base station sends a power up command. The basic step size to which the user adjusts its transmit power - following received TPC commands - is ±1dB or ±2dB with an accuracy of ±0,5dB up to a maximum transmit power. The SIR target value used in the CL power control method is provided by the outer loop power control algorithm. Outer loop power control: This fairly simple algorithm sets the E b /N 0 target for the fast (closed loop) power control described in the previous section. This method aims at maintaining the quality of communication, while preventing capacity waste and using as low power as possible. With a frequency varying between 10 and 100 Hz, the received and the desired quality of both uplink and downlink SIR are compared. If the received quality is better than the quality that has to be achieved, the SIR target is decreased; in the other case the SIR target is increased. 4 4 of 6

5 Figure 2: Inner loop power control Figure 3: Outer loop power control The scenario In the first scenario we consider a simple network consisting of a Node B, connected to a single RNC. One user moves along a straight-line path at a constant speed. The scenario should look like Figure Exercise In this exercise, we investigate how the load on a cell influences the coverage of the cell. We will examine the interference and coverage of the system. 5 5 of 6

6 Contact persons Figure 4: Scenario with 5 fixed nodes Abram Schoutteet (abram.schoutteet@intec.ugent.be) Bart Slock (bart.slock@intec.ugent.be) Responsible teachers Prof. Piet Demeester (piet.demeester@intec.ugent.be) Prof. Ingrid Moerman (ingrid.moerman@intec.ugent.be) 6 6 of 6