 |
Proposals for the Block Edge Mask for the Band 40.5 - 43.5GHz |
In document SE19(00)47 two different block edge masks were proposed as illustrated in figures 1 and 2 below.
Figure 1
Figure 2
Document SE19(00)44 described a method of coordination which had two phases. The first phase was to require each operator to manage a guard band within its block allocation so that there is no requirement for coordination with operators in the same geographical area. The block edge mask that would be used is that shown in figure 1. The block edge mask in this case would follow the limits of EN 301 390. The second phase of coordination would take place after the initial deployment of networks and would involve the operators coordinating transmissions within their guard bands so as to maximize spectral efficiency. This would necessitate an incursion of spectral mask from block A into block B and visa-versa, this is illustrated in Figure 2.
Draft ETSI standard EN 301 390 has two spurious emission masks for point-to-multipoint equipment with fundamental emission above 21.2GHz. One mask is for channel separations between 1MHz and 10MHz, the other is for channel separations greater than 10MHz. The two masks are shown in figures 3 and 4 below.
Figure 3: P-MP equipment for Channel separation 1<CS£ 10 MHz (from draft ETSI EN 301 390)
Figure 4: P-MP equipment for Channel separation CS>10 MHz (from draft ETSI EN 301 390)
The proposed uncoordinated block edge mask uses the spurious emissions mask shown in figure 3 and follows the additional requirement for EN 301 390 for all stations. It is also proposed that the block edge mask values are those transmitted by the antenna and not from reference point C’ shown in the general RF block diagram of figure 1 in draft ETSI EN 301 390. In this proposition the antenna gain used is +15dB. The proposal is shown in figure 5 below.
Figure 5: Block edge mask for non-coordination (assuming minimum CS of 3.5MHz)
The above mask has assumed that the minimum channel separation (CS) is 3.5MHz. The channel centre frequency is –250%CS from the sub-band boundary. If it is felt necessary to accommodate 1MHz channel separation then the non-coordinated block edge mask would be as shown in figure 6 below.
Figure 6: Block edge mask for non-coordination (assuming minimum CS of 1 MHz)
For the above non-coordination block edge mask all channel separations above 1 MHz would be accommodated.
In the frequency range where the –25dBm limit is shown in figures 5 and 6, allowance may need to be given for discrete (CW) spurious emissions. In EN 301 390 the number of discrete (CW) spurious emissions permitted is up to 10 and these are limited to a power level of –15dBm.
The block edge masks proposed are reciprocal about the sub-band boundary where the –25dBm/1 MHz shall extend down to 21.2GHz.
In the second phase of deployment the operators either side of a block boundary would be invited to come to an agreement on placing carriers closer to their boundary edges. This will involve the assessment of two parameters at the block edge:
a) The maximum EIRP transmitted by both operators at the block edge,
b) The size of the channel separation (CS) required by both operators at the block edge.
When the above parameters are known, a block edge mask can be constructed based on the spectral mask that extends into the adjacent block. The point at which the spectrum mask enters the sub-band boundary is proposed to be the 1xCS offset from the channel centre frequency. To obtain the required block edge mask the two spectral masks are examined and the mask that is least intrusive in terms of frequency into the neighbouring block is the mask that is chosen (an alternative proposal may be that the mask with the least overall spectral power intrusion is chosen).. To illustrate how this might be implemented, an example is shown below;
Example:
Operator A wishes to have a carrier of 28MHz CS using 4 state TDMA at the block edge with an EIRP of +45dBm.
Operator B wishes to have a carrier of 7MHz CS using 4 state TDMA at the block edge with an EIRP of +45dBm.
The spectral masks for the above two operators is shown, in figure 7 below, superimposed upon each other. It can be seen that above the –15dBm/1MHz spurious level the 28MHz channel extends further than the 7MHz channel.
Figure 7: Spectral masks for 28MHz and 7MHz 4state TDMA with EIRP of 45dBm
From the above operator requirements the block edge mask would be based on the requirements of operator B. In this proposal the spectral mask of the 7MHz CS enters the block of operator A at 1xCS away from the centre frequency. The mask will be based on absolute spectral density and not relative spectral density.
The proposed mask is shown in figure 8 below.
Figure 8: Block Edge Mask for 7MHz CS with EIRP of 15dBW
For the 40.5 to 43.5GHz MWS band, the idea of a block edge mask has been widely accepted. The details of the block edge masks have not been discussed in any detail. This paper has made detailed proposals to forward the discussion of block edge masks and to hopefully initiate debate of a more detailed nature.
The two-phase approach to the block edge mask allows the regulatory authorities to be impartial when allocating frequency blocks to operators and ensures that during phase two the operators will try to use their narrowest channel widths near block edges to ensure the best spectral efficiency.
Further work will need to be carried out to consider the Terminal Stations (TS) as the above work was based on antennas with 15dBi gain.
|
|
