In this chapter, we investigate the outage performance for the FFR-based TDD-OFDMA system in the full loading case. Note that the full loading case yields higher probability of generating the cross-slot interference. The arriving MSs are assumed to be uniformly distributed in a cell and request the time slots for uplink and downlink within a TDD-frame duration during a period of serving time. Let the numbers of uplink and downlink slots in cell 1 is symmetric traffics (△DL1 :△U L1 = 7 : 7) and the interfering cell 2 has vary downlink/uplink ratios. That is, CTS=-6 ∽ 6. We analyze and simulate the success probability versus the inner radius for various CTS region between cell 1 and cell 2. We demonstrate how the inner radius be designed to maximize spectrum utilization subject to the required success probability.
Flat-Fading Channel
In Fig. 4.5, the success probability is evaluated according to the equation (4.8) for various CTS environments. As shown in the figure, this analysis method can compute the success probability for the FFR-based TDD-OFDMA system. The little inaccuracy may be caused by limitation simulation time. While size of inner region is small, the success probability can be maintained higher than link reliability requirement (> 0.9). However, the larger
Table 4.1: Simulation Parameters in Two-Cell Environment.
Parameter Value
Number of Macrocells (I) 2
Number of time slots for TDD-frame (N ) 14
ΔDL:ΔU Lin cell 1 7:7
Radius of Macrocell 1000 m
Carrier Frequency (fc) 2500 MHz
System Bandwidth (W ) 10 MHz
BS/MS Antenna Omni-directional
BS/MS Transmit Power (Pbs/Pms) 43dBm /27 dBm Noise Power Spectrum Density (N0) -174 dBm/Hz Standard Deviation between BS to BS (σbs) 3dB Standard Deviation between MS to BS (σms) 10dB
Pathloss Exponent Factor (α) 4
Outage Threshold (γth) 0 dB
link reliability 0.9
FFT size (M ) 1024
Sub-carrier Bandwidth (Wsub) 10.938 kHz
Number of Null Sub-carriers 184
Number of Pilot Sub-carriers 120
Number of Data Sub-carriers 720
Number of Sub-channels (m) 40
Sub-carriers in each sub-channel 18
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4.5: Effects of asymmetric traffic ratios on the size of inner region under the flat-fading channel in FFR-based TDD-OFDMA systems.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4.6: Effects of asymmetric traffic ratios on the size of inner region under the frequency-selective fading channel in FFR-based TDD-OFDMA systems.
0 5 10 15 20 25 30 0.87
0.88 0.89 0.9 0.91 0.92 0.93 0.94 0.95 0.96
Number of Subcarriers
Success Probability
CTS=2 CTS=4 CTS=6
Figure 4.7: Effects of frequency diversity gain under the frequency-selective fading channel in FFR-based TDD-OFDMA systems with inner region radius equals to 500 m.
inner region radius is determined, more subchannel units are allocated in inner region. The success probability is higher than system is unused inner region, because the subchannels be assigned to MS located in inner region will have higher received signal power than MS located in outer region. Nevertheless, if inner region size is too large, the serious inter-cell interference will degrade SINR levels and decrease success probability. For the various traffic environment, we can observe that the CTS=6 that will generate serious BS-to-BS cross-slot interference. Hence, the system must reduce the inner radius to 400 meters to guarantee the success probability at the cost of sacrificing spectrum efficiency. For CTS=-6, the system can increase the inner radius up to 900 meters to increase spectrum efficiency because of MS-to-MS cross-slot interference just have minimal degradation for receive SINR. Therefore, we can adaptively design the inner radius for various CTS values to reduce the effects of cross-slot interference and maintain success probability in the FFR-based TDD-OFDMA systems.
Frequency-Selective Fading Channel
The numerical result of the frequency-selective fading channel with 18 subcarriers are grouped as a subchannel is shown in Fig. 4.6. We can observe the analytical approach al-most matches simulation result and little error may be caused by limitation simulation time and number of arrival MSs. Comparison results of Fig. 4.5 and Fig. 4.6, we can realize there will has some advantages to improve success probability in the frequency-selective fading. It almost improve 0.03 in the frequency-selective fading channel than in flat-fading channel because of the frequency diversity gain. Fig. 4.7 shows the success probability versus the number of subcarriers are grouped as a subchannel for cell 1 receivers receive serious inter-cell interference with CTS=[2,4,6] and inner region radius design as 500 m.
More than one subcarriers are grouped as a subchannel will get frequency diversity gain.
However, it will approximate a maximum value no matter how many subcarriers are
in-creased for grouping a subchannel. In the two-cell scenario, if the number of subcarriers is almost larger than 5, the outage performance will not be better.
Now we consider the worst case of CTS=7 and traffic load ρ = mµλ = 1 to do a simulation to observe the probability of generating cross-slot interference for randomly and regularly assignment approach. We can using (4.38), (4.42), (4.45), (4.47) to analyze the probability of generating each inter-cell interference. As we know, the probability of [C1, C2] = (U, U ) and [C1, C2] = (D, U ) is zero because CTS=7 and observed cell 1 is symmetric traffic. As simulation shown, Fig. 4.10 illustrates the randomly assignment will has higher cross probability (the case of [C1, C2] = (U, D)) than the regular assignment shown in Figs. 4.8, 4.9. Hence, using the regular method to assign resource to MS can limit the probability of generating cross-slot interference especially when traffic load is low. From simulation result of Figs. 4.5, 4.6, 4.8 and 4.9, we can realize the analytical approach can help us to design FFR factors and calculate probability of generating each inter-cell interference as more accuracy.