TDD/CDMA Systems with Asymmetric Traffic
6.4 The Proposed Cross-slot Interference-based DCA algorithmalgorithm
6.5.2 Effect of Traffic Asymmetry
Figure 6.5 shows the overall outage probability performance of the four schemes in the different traffic asymmetry conditions. The outage probability is defined as P rob{γi <
γt}. Consider the traffic load factor TF = 0.8 and the system parameters in Table I.
As shown in the figure, the increased degree of traffic asymmetry degrades the overall outage performance. One can find that the proposed Scheme IV outperforms other schemes. Furthermore, the performances of Schemes II and IV are better than those
A
Figure 6.4: The cellular system with grouped cells, where cell A has a symmetric load, cell B has more downlink traffic than uplink traffic, and cell B has more uplink traffic than downlink traffic.
Table I : System Parameters
Cell radius R=577 m
Base station antenna hight hb = 15 m Mobile antenna height hm =2 m Shadowing standard deviation between
home base station and mobile σs = 6 dB Shadowing standard deviation between
adjacent base station and mobile σn = 8 dB Shadowing standard deviation between
base station and base station σb = 3 dB Shadowing standard deviation between
mobile and mobile σm = 10 dB
Processing Gain P G = 32
Number of Antennas M = 5
Normalized inner region radius fs = 0.8
SNR 20dB
Max. allocatable power per user 1 w Target downlink SIR γt(d) = -6 dB
Target uplink SIR γt(u) = -9 dB
of Schemes I and III. This is because the impact of base-to-base cross-slot interference is much higher than that of mobile-to-mobile cross-slot interference. Using the MVDR beamformer, the base-to-base cross-slot interference can be effectively alleviated.
Figure 6.6 evaluates the outage probability for the downlink users in the outer region. The outage probability of these users will increase significantly when the degree of traffic asymmetry increases. With the help of the proposed DCA algorithm, the time slots that hace less cross-slot interference are reserved for these users. With Λ = 5, this figure demonstrates that Scheme IV can reduce at least 8 times of outage probability compared to Scheme II, while Scheme III improves the outage probability by 4 times compared to Scheme I.
Figure 6.7 illustrates the effect of traffic asymmetry on the outage performance for all the downlink users in the system. For Scheme I, the outage probability increase from 0.5 % to 4.5 % as degree of traffic asymmetry changes from 0 to 5. By using the cross-slot interference-based algorithm with uplink and downlink beam-steering (Scheme III), the outage probability van be improved to 2.5 % when the degree of traffic asymmetry is equal to 5. By adopting the uplink MVDR beamforming and downlink beam-steering in the cross-slot interference-based DCA (Scheme IV), the outage probability for the downlink users can be further reduced to 0.4 %.
Figure 6.8 illustrates the outage probability against the degree of asymme-try for the four different schemes. From the figure, one can see that the cross-slot interference-based DCA with the uplink and downlink beam-steering (Scheme III) can not significantly improve the uplink outage performance. This phenomenon is different from the downlink case. As seen previously in Fig. 6.7, Scheme III can sig-nificantly improve the downlink outage performance as compared to Schemes I and II. However, in the uplink case, Scheme III only performs slightly better than Scheme I and ever worse than Scheme II. This is because the base-to-base cross-slot inter-ference can effectively reduced by the MVDR beamformer. As shown in the figure,
at the degree of asymmetry equal to 5, the uplink outage probability of Scheme II can be reduce to 3 % as compared with 15% of outage probability of Schemes I and III. By using MVDR beamformer and cross-slot interference based DCA, the uplink outage probability can be reduced to 0.5 % in the same traffic asymmetry condition.
From the above results, it has been shown that the proposed cross-slot interference-based DCA algorithm combining with MVDR beamformer can significantly reduce both the mobile-to-mobile and base-to-bas cross-slot interference. The MVDR beam-former can suppress the base-to-base cross-slot interference significantly. However, the mobile-to-mobile cross-slot interference will cause high outage probability for the mo-biles near the cell boundary, thereby reducing the cell coverage area. The proposed slot interference-based DCA algorithm can avoid the mobile-to-mobile cross-slot interference by reserving good channels for the potential users having mobile-to-mobile cross-slot interference.
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0
2 4 6 8 10 12
The degree of traffic Asymmetry
Outage Probability (%)
Scheme1 Scheme2 Scheme3 Scheme4
Figure 6.5: Effect of traffic asymmetry on the overall outage performance with both downlink and uplink users.
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0
5 10 15
The degree of traffic Asymmetry
Outage Probability (%)
Scheme1 Scheme2 Scheme3 Scheme4
Figure 6.6: Effect of traffic asymmetry and mobile-to-mobile cross-slot interference on the outage performance for the downlink users in the outer region.
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
The degree of traffic Asymmetry
Outage Probability (%)
Scheme1 Scheme2 Scheme3 Scheme4
Figure 6.7: Effect of traffic asymmetry and the mobile-to-mobile cross-slot interfer-ence on the outage performance of all the downlink users in the system.
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0
2 4 6 8 10 12 14 16 18 20
The degree of traffic Asymmetry
Outage Probability (%)
Scheme1 Scheme2 Scheme3 Scheme4
Figure 6.8: Effect of traffic asymmetry and base-to-base cross-slot interference on the outage performance for all the uplink users.
145