35 TR 101 274 V1.1.1 (1998-06)

Table 14: System sensitivity as a function of modulation and coding schemes and data bit rates with ROH equal 16 kbit/s

When FSK modulation is used, the system sensitivity is shown in table 8.

9.3Cellular deployment performance for FH-CDMA system

As stated in subclause 4.5, cellular deployment means that a certain number of frequencies are reused in other cells (or sectors) properly distanced. This means that a certain amount of co-channel interference has to be taken into account during link budget calculation. The presence of co-channel interference causes an increment of the power threshold level because it acts like an additional noise power at the input of the receiver. In this context, relation (4) can be evaluated by adding a contribution for each cell (or sector) working at the same frequency on the basis of subclause 4.5 rules. Supposing the interference as a gaussian noise, a general figure on power threshold level with co-channel interference can be evaluated by means of the following equation:

At any rate, the best way to evaluate this figure is by means of measured sensitivity curves provided by the manufacturers.

10Impact on performance and capacity due to CRS and TS antenna

In subclause 4.5 a very simple methodology has been provided in order to calculate the overall C/I for a given frequency reuse environment. That methodology is based on the assumption that all the users are located in the worst interfering site. However, this hypothesis could lead to a pessimistic C/I estimation and, in addition, does not show how the TS antenna influences the C/I related to the different frequency reuse environment. In order to investigate these issues it has become necessary to perform simulations: the user traffic in each of the six cells surrounding the cell under analysis has been modelled in order to estimate the interference statistics.

TS antennas have been considered: figure 5 shows a typical antenna characteristics for 1 GHz to 3 GHz bands, and figure 6 show a typical antenna characteristics for 10,5 GHz band. In addition, three frequency reuse environment has been considered with six surrounding cells as shown, for M = 1, in figure 2. In particular, the following assumptions have been made:

•six surrounding interfering cells;

•P-MP system with 32 64 kbit/s channels for each cell;

•uniform users distribution over the interfering cells;

•CRS with omnidirectional antenna;

•poisson traffic model;

ETSI

•mean call duration = 100 s;

•traffic per cell with GOS = 1 %;

•perfect power control and all interfering links with the same modulation scheme; On the basis of the previous assumptions, the following case studies have been considered:

•access schemes: TDMA, DS-CDMA, FDMA and FH-CDMA;

•M = 1, M = 3 and M = 4;

•antenna patterns as in figure 5 and figure 6;

•propagation exponent for the useful links: γ = 2 ;

•propagation exponent for the interfering links: γ = 2 and γ = 3 .

The results for several case studies are shown in figures 7 to 14 in terms of the cumulative distribution functions of the overall C/I. In particular, each curve shows, for a given C/I value, the probability of having a lower C/I value. In other words, for a given C/I, each curve provides, for the relevant case study, the percentage of time in which a lower C/I value could be expected. As an example, a C/I with a probability equal to 10-3 means that we could expect to have a lower C/I for about 45 minutes during a month with 31 days.

In addition it could be easily found that the usage of a more directive antenna (see figure 6) makes it possible to gain 2 dB on the overall C/I for TDMA, FDMA, FH-FDMA and 4 dB for DS-CDMA.

In order to explain better how to use the results provided by the figures below, two examples shall be analysed: one for DS-CDMA and one for TDMA access scheme.

Cellular deployment for Pseudo Random DS-CDMA can be analysed by means of relation (27). Looking at figure 7, we can read a C/I value of -4,5 dB for a M = 1 and with a probability equal to 10-3. By using this value in (27) instead of the term:

It is possible to find out a degradation in capacity lower than 10 %. It is worth noting that by means of relation (28), the relevant figures for Pseudo Random DS-CDMA can be used also for analysing Orthogonal DS-CDMA cellular deployment.

As far the TDMA, FDMA and FH-CDMA access schemes are concerned, cellular deployment can be analysed by means of relation (20). Looking at figure 13, we can read a C/I value of 16,8 dB for a M = 4 and with a probability equal to 10-3. Using this value in (20) instead of the term (C/I)tot , it is possible to find out the feasibility of this cluster size, even with uncoded transmission, for the area of deployment in which the interfering link have a propagation exponent

γ = 3 (e.g. rural application).

ETSI

Figure 5: RPE for 1 GHz to 3 GHz bands

Figure 6: RPE for 10,5 GHz band

ETSI

Figure 7: C/I cumulative distribution function for Pseudo Random DS-CDMA with antenna in figure 5 and interfering propagation exponent γ = 2

Figure 8: C/I cumulative distribution function for Pseudo Random DS-CDMA with antenna in figure 6 and interfering propagation exponent γ = 2

ETSI

Figure 9: C/I cumulative distribution function for Pseudo Random DS-CDMA with antenna in figure 5 and interfering propagation exponent γ = 3

Figure 10: C/I cumulative distribution function for Pseudo Random DS-CDMA with antenna in figure 6 and interfering propagation exponent γ = 3

ETSI

Figure 11: C/I cumulative distribution function for TDMA, FDMA and FH-CDMA with antenna in figure 5 and interfering propagation exponent γ = 2

Figure 12: C/I cumulative distribution function for TDMA, FDMA and FH-CDMA with antenna in figure 6 and interfering propagation exponent γ = 2

ETSI

Figure 13: C/I cumulative distribution function for TDMA, FDMA and FH-CDMA with antenna in figure 5 and interfering propagation exponent γ = 3

Figure 14: C/I cumulative distribution function for TDMA, FDMA and FH-CDMA with antenna in figure 6 and interfering propagation exponent γ = 3

ETSI