Dear Chairman and SE7 Colleagues,
Motorola Solutions has proposals below to:
ANNEX 2 (On Receiver Performance)
We have been considering if this Annex 2 could be reused as a selectivity analysis as the idea of showing that improved selectivity in the receiver can improve IMD rejection is indeed valid.
However as soon as Annex 2 is moving into modelling, it starts making statements such as "IMD can be neglected if the input to the receiver is less than -40dBm", which is not correct, we had to conclude, that nothing else of this Annex really can be salvaged, other than:
A 2.1 Non-linear Model of the Receiver
In order to model the victim receiver’s behaviour in the presence of blockers, spurious and interferers, whether wideband or narrowband, certain assumptions about its nonlinearity, selectivity and dynamic range have to be made. An obvious way is to derive the receiver’s characteristics from the standards the victim receiver has to comply to. This typically contains: blocking test, intermodulation rejection test, spurious response rejection test and nominal error rate (NER) test.
However, the standards represent minimum performance requirements. Practical designs typically reflect much higher performance, which accounts for tolerances and for the fact that equipment vendors want to give their customers a good experience also in challenging signal scenarios.
- The standard reflects a test with a single blocker whereas in practical usage scenarios there might be multiple ones.
- The standard specifies two interferers, which are placed at certain frequency offsets so that the intermodulation product falls exactly together with the weak wanted signal. In practice however there might be a multitude of interferers and thus more mixing products with larger strength and at higher frequency offsets.
- The spurious response rejection test only allows for 5% exceptions. The receiver design approach must take this into account.
Modern PMR receiver design has to take these factors into account whilst allowing mass market low cost production techniques, and allowing receivers to have high switching bandwidths to minimise the number of models produced, and maximise the flexibility afforded to system implementers.
A 2.2 Trade-off between selectivity and linearity in receiver design
In order to increase robustness of receivers to challenging signal scenarios, one can either go for highest analogue selectivity by very narrow tuneable preselectors as the first stage in receiver or by ensuring that the IP3 of the overall receiver is adequate to meet the demands of type approval specifications and practical system implementation.
Very high selectivity through tuneable preselectors can solve nearly all unforeseeable blocking, spurious and intermodulation problems, however it would imply very high Q filters. These are practical in base stations, but not practical in mobile stations as they would prevent large switching bandwidth and would increase cost and size.
A receiver designer therefore balances out both strategies by implementing moderate selectivity in a tuneable analogue preselector by using resonators with moderate Q factors and by using active devices with moderate IIP3.
To meet the requirements of high switching bandwidth, front end filtering in receivers is limited to ensuring that image frequency rejection using a high intermediate frequency can be achieved. The designer then ensures that the IP3 performance of the entire receiver chain is adequate to meet the non linearity performance needed to meet intermodulation requirements, and other specifications including spurious response. The TETRA standard can be met with a receiver IP3 performance of ‑9.5dBm.