As the MAS software source with 14 years of experience in the field, the company can offer a few guidelines for MAS implementation. A technical discussion of specific algorithms is beyond the scope of this article, but here are some general principles:

• Do your homework thoroughly: Many tools for network or economics analysis and performance simulation from single-antenna domains (e.g., interference averaging) yield misleading results when applied to MAS-enabled gear. Getting MAS analysis right is more complicated, but essential.

• Think integrated: The highest performance for the least marginal unit cost is obtained by integrating MAS into client and infrastructure designs from the outset, not adding them on after the fact.

• Consider network performance, not just the link: MAS modes that achieve useful results for an individual link (e.g., the baseline form of STC MIMO in WiMAX) can fail in a multicell, multi-user context. Network-level analysis of fully loaded systems is essential.

• Use multiple approaches: Use the right tool for the job. Operator requirements and subscriber behaviors vary from one market to another and from one moment to another. Different MAS architectures have unique strengths and weakness in different applications — there is no single “best” approach. It is better to include all approaches in the system (possible, since this is software) and let environmental conditions dictate which one is used.

This brings us to anticipate dynamic, seamless use of all approaches. It has been shown in the PHS implementation, where eight different MAS algorithms are selected on the fly for optimized performance — on a frame-by-frame and user-by-user basis — that MAS architectures can be dynamic systems. Many levels of radio system control (beyond individual cells to the network level, for example) can be incorporated into this sort of self-organizing optimization process.

• NTT DoCoMo Inc. Annual Operating Data (http://www.nttdocomo.co.jp/english/corporate/investor_relations/business/fiscal_e.html).

• Unstrung, “CDMA Reports on B'band,” May 8, 2006.

• Cingular Wireless press release, “Cingular Wireless Reports Strong Second Quarter 2006 Results,” July 20, 2006.

• Unstrung, ibid.

• Personal Broadband Australia and Unwired.

• Based on latest U.S.-market wireless voice ARPU and MOU data from Cingular and T-Mobile; consumer packages for wired broadband Internet access from AT&T (ADSL) and Comcast (cable modem); typical broadband Internet usage figures from Earthlink; consumer cable TV packages from Comcast; and the latest A.C. Nielsen figures for average amount of television watched per day per individual in the U.S.

• Visant Strategies, “Intelligent Antennas,” May 2006.

Steven Glapa is vice president of marketing at ArrayComm LLC, San Jose, Calif. Glapa holds an MBA from Columbia Business School and undergraduate degrees in physics from Carleton College and in computer science and mechanical engineering from George Washington University.

Previous 1 2 3 4

RSS    Save to Del.icio.us  Digg This