ABSTRACT
Contents 8.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236 8.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 8.3 Performance Measures of Cellular System . . . . . . . . . . . . . . . . . . . . . . . . . 239
8.3.1 Constraints on Network Resources . . . . . . . . . . . . . . . . . . . . . . . . . . 239 8.3.2 Performance Measures of Forward Link . . . . . . . . . . . . . . . . . . . . 240 8.3.3 Performance Measures of Reverse Link . . . . . . . . . . . . . . . . . . . . . 243
8.4 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 8.4.1 Forward Link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
8.4.1.1 Scheduling Problem Discussion . . . . . . . . . . . . . . . . . . . 246 8.4.1.2 Resource Allocation Problem Discussion . . . . . . . . . . 247 8.4.1.3 Formulation on the Forward Link . . . . . . . . . . . . . . . . . 249
8.4.2 Enhancements on Reverse Link 1xEV-DV . . . . . . . . . . . . . . . . . 250 8.4.2.1 Transmission Characteristics . . . . . . . . . . . . . . . . . . . . . . . 251
8.4.2.2 Rate Control Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252 8.4.2.3 System Model Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 8.4.2.4 Formulation on the Reverse Link . . . . . . . . . . . . . . . . . . 253
8.5 Proposed Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 8.5.1 Solutions for the Forward Link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
8.5.1.1 Performance Analysis of B&B, CE, and HEU Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
8.5.1.2 Dynamic Scheduling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 8.5.2 Solutions for the Reverse Link. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
8.5.2.1 Algorithm Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 8.5.2.2 Illustrative Example Results . . . . . . . . . . . . . . . . . . . . . . . . 262
8.6 Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 8.6.1 Recommendations for Future Work . . . . . . . . . . . . . . . . . . . . . . . . . 266
Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
8.1 Overview Third-generation (3G) standards are designed to meet the increasing demand for high-speed packet data transmission. CDMA2000 1xEV-DV is one of the 3G standards that can support the integration of multimedia services on a single 1.25-MHz carrier. In this chapter we develop joint scheduling and resource allocation schemes for both forward and reverse link of CDMA2000 1xEV-DV system. The existing resource management schemes mostly consider the benefits of either service providers or subscribers. On the other hand, our resource management schemes not only benefit the subscribers by optimizing the data rate or (Quality-of-Service) QoS supported, but also satisfy the service providers by maximizing the revenue generated. The schemes also provide flexible trade-offs between the maximization of data rate, QoS, and revenue. The combined scheduling and resource allocation problems are formulated into binary integer linear programming (LP) with multiple constraints. First, we investigated solving the binary integer LP using exact algorithms, which are Branch and Bound (B&B) and Complete Enumeration (CE). The results show that these algorithms are inefficient for this problem, as they require high computational time. We then created heuristic algorithms that consume much shorter computational times although they sometimes compromise the objective function values. These heuristics are more appropriate for our scheduling and resource allocation schemes, which require short computation time for implementation. In the end, we are able to make recommendations to service providers on how to choose the algorithms to properly schedule the transmissions.
