Network Strategies says the key conclusions of this study "are likely to be invalid, or at least questionable, due to weaknesses within the cost analysis methodology and assumptions."

According to Network Strategies, "The Analysys modelling does not estimate the costs of a realistic network in the key areas of: voice service and demand to meet universal Service requirements; WiMAX coverage and capacity restrictions; spectrum allocations and availability in Australia; real-life engineering of the coverage of wireless access systems; capacity and coverage of 3G/HSPDA systems."

Analysys claimed that its calculations were high-level in order to provide indicative costs, but this is not a valid defence of its results, according to Network Strategies. "The problems we have identified are sufficiently substantial that they cast significant doubt on even the high level results of the model...We conclude that the Analysys results should not be interpreted as an accurate reflection of the comparative costs of wireless access technologies in Australia."

Two key inputs to the voice component of the modelling are the network capacities needed to carry voice and an estimate of the total demand for voice traffic on the network.

According to Network Strategies, "The network capacity set aside for voice traffic is likely to be significantly lower than what is realistically required, especially if the network must meet universal service quality requirements and carry other typical PSTN signals such as fax, modem and teletype."

Analysys was not able to obtain local call information for Australia, and instead used New Zealand data as a proxy. Network Strategies questions the appropriateness of this "because New Zealand has free residential local calling which may distort the proportion of calls made to local numbers." And it notes that "The Analysys voice traffic estimates may be readily checked against publicly available data for Australia."

Doing this, Network Strategies found that "Telstra traffic alone is more than double the Analysys projections." It notes that the Telstra traffic data would also include dial-up Internet calls as well as voice calls. but that "Total PSTN traffic would be even higher once traffic from the other fixed line carriers was included....We estimate that the actual voice traffic per household and per business location may be up to two times that estimated by Analysys."

The Network Strategies report is available here. {moscomment}