|
Frequency Allocation Plan
for the band 40.5 – 43.5 GHz
|
This note reviews the proposals to accommodate FDD (symmetric and asymmetric) and TDD systems in the 40 GHz band. The proposals show how symmetric and asymmetric FDD systems could be accommodated without wasting large amounts of spectrum. At the same time, it is noted that the current text encourages the use of paired but widely spaced blocks for all systems, including TDD (for which this arrangement may have significant disadvantages).
Although assignments may be of any size, it is noted that many administrations appear to favour large assignments (up to several hundreds of MHz). This increases the probability that a simple interpretation of the existing text will encourage widely split assignments within a single TDD system.
Annex 1 "Guidance and steps towards a plan"
The penultimate bullet point is misleading. Whilst it is true to say that TDD systems could operate using widely spaced blocks, there can be a serious cost disadvantage. The 40 GHz plan is attempting to avoid bias towards or against any particular technology or system architecture. Therefore a more appropriate text could be:
In figure 1 "General concept of paired equal blocks"
The diagram shows two blocks, spaced by 1500 MHz. The size of the blocks is not specified. Subsequent text and diagrams show how these paired blocks could be used for FDD symmetric systems, FDD asymmetric systems and TDD systems (fig. 5).
In the case where there are multiple TDD systems deployed by different operators, the assignments could be in different blocks, which causes no problem. In the case where large frequency assignments are considered (greater than is available or residual in one of the paired blocks) the TDD systems could be forced to operate with a very large frequency gap. This is a serious design issue, with significant potential cost implications.
Given that it is desirable that:-
(a) All technologies receive "equal"
treatment, with potential access to the same amount of spectrum
(b) The plan should be able to adapt to actual demand, irrespective of technology
type (i.e. a situation could develop where most systems are TDD or where
most systems are FDD, but this can not be predicted at the start)
Then a potential way to proceed could be
(1) To make the gap between the paired blocks larger than either of the blocks. A possible starting point could be two nominal 500MHz blocks with a 1000MHz block between them (but, see 4,5 below).
(2) To begin to assign frequencies for FDD systems, starting at the low frequency end of the paired block(s)
(3) To begin to assign frequencies for TDD (and broadcast) systems, starting at the high frequency end of the block between the two paired blocks.
(4) If demand for FDD systems dominates, the paired blocks could be extended from the upper frequency end (ending up with the situation of fig. 2)
(5) If demand for TDD systems dominates, the middle block could be extended downwards into the lower of the FDD paired blocks.
(6) In a situation where demand for different types of system is more balanced, TDD systems could be given assignments first in the block between the paired blocks and only when this is full could be given parts of the paired blocks.
The above procedure will not solve all problems but it does avoid the difficulty of automatically giving disadavantage to one type of technology. In the event that equal demand for systems with different architectures occurred, the total amount of spectrum available for each would be similar. In the unequal case, there can be mechanisms to adjust the plan to approximately follow demand, without imposing serious constraints on system design.
| BFWATG(01)06 |
|
