On June 2, 2004, Elandia/Vivato submitted a supplemental proposal that contained information regarding network design and pricing that was not part of the original document. The original document consisted of three published papers that described Vivato technology and its implementation by Elandia. On Friday May 28, 2004, Doug Welch of Elandia and Shanuj Sarin of Vivato visited Fullerton for the purpose of conducting a preliminary site survey in preparation for a formal proposal. After considerable discussion, the Technology Working Group determined that it could not accept the supplemental proposal because the original one that had been submitted by the May 12, 2004, deadline was too incomplete to be considered a formal proposal. It was determined by the TWG members that accepting the new proposal would require that the RFP process be reopened to all, and the Working Group was not willing to do that.
Members of the TWG discussed what are the proper procedures to follow during the RFP process. Several members of the TWG followed significantly different procurement procedures in the corporate world where there is considerably more flexibility in evaluating equipment and selecting vendors.
Joseph Hsieh began a presentation of Wireless Hotspot's network proposal at 10:00 a.m. He was assisted by four persons representing Firetide, the wireless equipment manufacturer, and Vernier, who would provide the control system for the wireless network. Wireless Hotspot was founded in 2002. It is a privately funded company based in Los Angeles and incubated by the FAME Renaissance Center. It is currently engaged in the deployment of a small outdoor mesh in Culver City. This deployment appears to be the first outdoor mesh using Firetide equipment. Most of Wireless Hotspots experience has been in the creation of hotspots for businesses and serving as a wireless Internet service provider.
Wireless Hotspot would deploy 11-15 Firetide 1000R outdoor mesh routers along the streets in downtown Fullerton. The routers would be served by 23-30 YDI directional flat-panel access points. The system would be monitored by a Vernier control management system that would provide client registration and authentication, monitor usage, support network portals, permit the creation of VPNs, restrict bandwidth, control usage and support other functions. Because the radios are used only for backhaul, the addition of many directional access points significantly adds to the cost of the system. The high power quoted for these access points and the large number of them raises questions about possible self-interference and degradation of system performance. Firetide 1000S indoor routers can be used to extend the WiFi signal indoors.
Shekhar Deshpande began his presentation of SkyRiver's network proposal shortly after 11:00 a.m. He was accompanied by several SkyRiver engineers. SkyRiver Communications currently operates the largest fixed broadband wireless networks in Southern California including the cities of San Diego, Ontario, and Palm Springs, as well as the San Fernando Valley, serving the small-to-medium business market. SkyRiver is a privately held company which has been developing and implementing WiFi solutions since 2000, but it has not had much experience deploying WiFi mesh networks.
Although SkyRiver uses Tropos equipment, it brands the network radios with the SkyRiver name and discusses the technology as though it was developed by SkyRiver when, in fact, it was developed by Tropos. Even much of the material in the written proposal was prepared by Tropos because it appears verbatim in the CDCE proposal. Consequently, it will be possible to compare directly the SkyRiver system design with that proposed by CDCE Wireless which uses the exact same radios. SkyRiver proposes deploying 24 radios/access points over the 24-block area. In comparison, CDCE believes that 9 radios/access points will adequately cover the same area. It is disconcerting to see two proposals with such great variation in design parameters when the same equipment is being deployed in both. Too many radios are likely to lead to self-interference in areas of significant overlap and increase the number of hops between nodes before reaching a gateway to the Internet. This is likely to degrade system performance while greatly increasing the cost of the system. Similarly, too few radios are likely to result in gaps in coverage. The estimated cost of the SkyRiver deployment is over twice the cost of the CDCE system that was designed with input from Tropos engineers.