6.3 Downlink Beamforming
6.4.3 Discussion
To determine which beamforming scheme should be used in TDD/CDMA systems is a complicated tradeoff issue between performance improvements and implementation costs.
0 1 2 3 4 5 6 0.80
0.82 0.84 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1
Number of neighboring cells generating the opposite direction interference Probability(γ > γth)
Scheme IV
Scheme II
Scheme III
Solid Line : γth = 8 dB Dashed Line : γth = 7 dB
Scheme I
Figure 6.9: Performance comparison of four beamforming schemes with different numbers of cells generating the opposite direction interference, where an antenna array with nine elements is deployed at base stations.
Scheme IV, using the uplink MVDR beamforming and downlink transmit beam-steering, can effectively suppress the opposite direction interference, thereby providing greater flex-ibility in delivering asymmetric traffic services. In Scheme IV every TDD/CDMA cell can independently designate traffic patterns for either uplink modes or downlink modes in every time slot according to its own rate of traffic asymmetry. On the other hand, using a simpler beam-steering method in both the uplink reception and downlink transmission, Scheme III provides satisfactory performance only when the number of cells generating the opposite direction interference is not large. Thus, it is suggested to combine Scheme III with other sectorization or channel assignment techniques to reduce the number of cells generating the opposite direction interference.
Scheme II is another effective technique to reduce the impact of the opposite direction interference. Recall that Scheme II utilizes the MVDR beamforming only in the uplink.
Note that the performance of Scheme II is better than Scheme III but worse than Scheme IV. As remarked earlier, the extra cost of implementing downlink transmit beam-steering may not be very high. If so, Scheme IV will be a better choice than Scheme II provided that the MVDR beamformer has already been adopted in the uplink. As for Scheme I, it is shown that only using beam-steering in the uplink can not provide acceptable performance.
Although we concentrate on the uplink performance of TDD/CDMA systems, antenna beamforming can also be exploited to improve the downlink performance. For example, by taking advantage of the reciprocity of TDD channels, downlink transmit beamforming from neighboring base stations can lower the effective interfering power to the mobile station in the home cell. Furthermore, when the mobile station is employed with a small number of array sizes [108, 47], the downlink performance can be further enhanced with the beamforming techniques similar to the Schemes III and IV.
6.5 Chapter Summary
This chapter have investigated the effect of four antenna beamforming schemes on rejecting the opposite direction interference in the TDD/CDMA system. From the network view to exploit the multiple antennas at the base stations of adjacent cells, we propose a simultaneous downlink transmit beamforming and uplink receive beamforming scheme to alleviate the negative impact of the opposite direction interference with a low implementation cost. Based on the numerical results in this chapter, we can make the following summary:
• Schemes IV, which adopts the MVDR beamformer in the uplink and the beam-steering in the downlink, can effectively suppress the strong opposite direction interference of TDD/CDMA systems, thereby allowing every cell to provide asymmetric traffic services with different rates of traffic asymmetry.
• Scheme III, which adopts the beam-steering method in both the downlink transmission and uplink reception, can provide satisfactory performance when the number of cells generating the opposite direction interference is not large. When combined with other
sectorization or channel assignment techniques, Scheme III can be a very effective mechanism to overcome the opposite direction interference in the TDD/CDMA system with lower implementation costs.
• If only the uplink beamforming is considered, the MVDR beamformer (Scheme II) instead of the conventional beamforming method (Scheme I) should be adopted since the conventional beam-steering can not effectively suppress the opposite direction in-terference.
As a result, even with the severe impact of the opposite direction interference, a feasible and economical beamforming mechanism (e.g. Scheme III) can be found, which can enable the TDD/CDMA system to deliver asymmetric traffic services within the entire service area with greater flexibility.
Chapter 7
Concluding Remarks
7.1 Dissertation Summary
The story is coming to an epilogue. In the dissertation, it has been seen that the mar-riage of the multiplexing-based antenna technique with the multiuser scheduling system is finally proved to a wonderful ending due to their complementary diversity-multiplexing char-acteristics. Another marriage of the diversity-based antenna technique with the multiuser scheduling system, however, may lead to a pitfall due to the intrinsic conflict – one prefers fluctuations while the other makes tranquility. In the third marriage, a happy ending follows when the tough opposite direction interference in the TDD/CDMA systems is tamed by an appropriate match of beamforming techniques.
To sum up, through the lens of the network perspective to exploit the interaction between the multiple antenna technique and its underlying communication system, this dissertation has developed effective strategies to apply the MIMO technique for the multiuser scheduling and TDD/CDMA systems. The dissertation includes the following research topics:
1. The fading mitigation based antenna technique (or diversity-based antenna technique) for the multiuser scheduling system.
2. The throughput enhancement based antenna technique (or multiplexing-based antenna technique) for the multiuser scheduling system.
3. The throughput enhancement based antenna technique for the multiuser scheduling
system with the zero-forcing receiver.
4. The interference suppression based antenna technique (or beamforming technique) for the TDD/CDMA system.
Contributions from this research are listed as follows.
1. Developed an analytical framework to investigate the interaction of antenna diversity and multiuser scheduling. Through a unified capacity analysis, the chaotic interplay among fading characteristics, multiuser diversity and various diversity-based antenna schemes is unraveled.
2. Suggested an innovative strategy to replenish the diversity-deficient spatial multiplex-ing MIMO system with multiuser diversity. With only scalar feedback, the proposed SWNSF scheduling can significantly improve the cell coverage and system capacity of the multiuser MIMO system.
3. Leveraged the cross-layer cooperation between the simple zero-forcing receiver and multiuser scheduling to achieve the full theoretical capacity of the MIMO system.
Also, a proof is given to illustrate the asymptotic optimality of the zero-forcing in a scheduling environment with many users.
4. Invented a cross-cell cooperation among the multiple antennas of adjacent base sta-tions. The exploitation of such a cooperation along with the TDD channel reciprocity enables an effective and economical beamforming solution to resolve the opposite di-rection interference issue for the TDD/CDMA system.
The following summaries the results from the above contributions.