11th ECC TG3 MEETING
Copenhagen, 12-16 Sept 2005 / TG3#11_..
Date issued:
Source: UK
Status: For Consideration
Subject: PSD limits for UWB below 2GHz
Password protected: yes / no / x
Summary
This paper considers some factors that should be taken into account by ECC TG3 in defining PSD limits for UWB below its operating band, and especially below 2GHz.
Proposals
The PSD limits for UWB below 3.1GHz should be based on impact analysis down to at least a frequency of 876MHz.
It would not be useful to apply the ITU definitions relating to out-of-band and spurious domains to UWB. However, it should be noted that these definitions are, in any case, not applicable to UWB in a regulatory context.

1  Introduction

The studies of UWB in ECC TG3 have rightly focused on the proposed operating band for UWB (3.1 – 10.6GHz) and the bands immediately surrounding it. However, ECC TG3 now needs to define an emission mask for UWB, which will need to include PSD limits for a much wider frequency range.

This paper considers the appropriate basis to derive limits for UWB below 3.1GHz, and especially below 2GHz.

At present, there are a number of different technologies for UWB, which use fundamentally different modulation processes. The UK strongly supports technology neutrality for the CEPT provisions for UWB. It would therefore not be appropriate to base these provisions on the characteristics of individual UWB technologies.

1.1  UWB technologies

The current UWB technologies can be broadly divided into three categories, according to the modulation process:

1)  Modulation of a carrier

e.g. MB-OFDM

2)  Pulse sequence incorporating frequency shaping

e.g. DS-CDMA

3)  pulse sequence without frequency shaping

e.g. PPM

These modulation processes of these three categories have inherently different spectral properties, and rely to different degrees on separate filtering to achieve out-of-band PSD requirements.

It should be noted that, for pulse based UWB without frequency shaping, the modulation process inherently generates a higher PSD at lower frequencies. It is only the subsequent filtering of this signal that produces a transmitted UWB signal centred on the frequency range 3.1-10.6GHz.

2  Discussion

2.1  FCC Rules and their relevance to CEPT

The FCC rules apply limits based on impact of UWB on other radio services down to a frequency of 960MHz. This frequency is not based on the characteristics of UWB, but is a boundary in the USA frequency allocation table – the FCC decided that the services above 960MHz required protection, but not those below 960MHz.

CEPT has recognised the need to protect UMTS from interference from UWB. ECC TG3 and Administrations have conducted extensive studies on UMTS in the 2GHz band, which have concluded that a UWB PSD of -85dBm/MHz is necessary to protect UMTS.

There have been far fewer studies of the GSM 900 band. However, the considerations and scenarios for interference are identical to UMTS at 2GHz. From a simple scaling for the differences in frequency and in the system parameters for GSM and WCDMA, the protection requirement for the GSM 900 band can be shown to be lower than for WCDMA at 2GHz for an equivalent degree of protection (around -90dBm/MHz). The bottom frequency of the GSM 900 band (including R-GSM) is 876MHz.

As this PSD is lower than the FCC limit at 960MHz, it is appropriate for the CEPT limit to be based on impact on radiocommunication services at frequencies below 960MHz, at least down to 876MHz.

2.2  Spurious and unwanted domain[1]

This section discusses whether it is appropriate to apply the ITU definitions for spurious and unwanted domain to UWB. The annex to this document gives a background discussion on spurious emissions.

Spurious and unwanted domains are defined as follows in the Radio Regulations:

1.146A out-of-band domain (of an emission):The frequency range, immediately outside the necessary bandwidth but excluding the spurious domain, in which out-of-band emissions generally predominate. Out-of-band emissions, defined based on their source, occur in the out-of-band domain and, to a lesser extent, in the spurious domain. Spurious emissions likewise may occur in the out-of-band domain as well as in the spurious domain.(WRC-03)

1.146B spurious domain (of an emission):The frequency range beyond the out-of-band domain in which spurious emissions generally predominate.(WRC-03)

1.144 out-of-band emission:Emission on a frequency or frequencies immediately outside the necessary bandwidth which results from the modulation process, but excluding spurious emissions.

It is clear that these definitions relate to the characteristics of a specific technology.

For a pulse-based UWB technology, the repetition rate of the pulses is a fundamental part of the modulation process. By this definition, the out-of-band domain would therefore extend down to at least the pulse repetition frequency.

ITU Recommendation SM.329-10 recommends that the following definition of spurious domain should generally apply:

2.3 According to the principles stated in Appendix 3 to the RR, the spurious domain generally consists of frequencies separated from the centre frequency of the emission by 250% or more of the necessary bandwidth of the emission.

ITU Recommendation SM.329-10 provides a definition of the boundary for wideband signals. If this definition was applicable to a UWB signal, the following frequencies would result for the lower boundary between the unwanted and spurious domains:

SM.329-10
MB-OFDM without frequency hopping / 2.1GHz
MB-OFDM with frequency hopping / 100MHz
Generic UWB signal occupying the band 3.1-5GHz / 0Hz
Generic UWB signal occupying the band 3.1-10.6GHz / 0Hz

It is therefore clear that, even if the unwanted and spurious domains were applicable to UWB, it would not be useful to attempt to define them.

3  Conclusions

1)  It is appropriate for the PSD mask for UWB devices to be based on protection of radio services at least down to a frequency of 876MHz, and quite possibly lower.

2)  It would not be useful to apply the ITU definitions of out-of-band and spurious domains to UWB. However, it should be noted that these definitions are in any case not applicable to UWB in a regulatory context.

Annex: Background discussion on spurious emissions

Spurious emissions were first defined in an ITU Recommendation in 1948. At that time, transmitters generally used crystal oscillators and multipliers for frequency generation and valve power amplifiers with tuned outputs. This transmitter architecture is only capable of generating spurious emissions at a small number of frequencies (adjacent multiples of the crystal oscillator, and harmonics). The definition of spurious emissions therefore implicitly assumed that there is a low probability of equipment generating a spurious emission on a particular frequency.

Since that time, technology advances in transmitters have resulted in more ways in which spurious emissions can be generated, such as synthesisers for frequency generation and broadband power amplifiers. However, the definition of spurious emissions has remained largely unchanged.

Spurious emissions[2] are distinguished from out-of-band emissions[3], which fall immediately outside the transmission channel and result from the modulation process. There is a significant probability that a transmitter will generate emissions on a particular frequency in the out-of-band domain, but a low probability that emissions will be generated on a particular frequency in the spurious domain.

Developments in radio technology and increase in use of the spectrum are beginning to place strains on the provisions relating to spurious emissions (especially the boundary between the spurious and out-of-band domains). However, they are still relevant, provided that they are not applied in the out-of-band domain.

The R&TTE Directive[4] establishes a regulatory framework in the EU for, inter alia, the placing on the market and putting into service of radio equipment. One route to demonstrate compliance with the directive is through conformity with harmonised standards. If a product conforms to an applicable harmonised standard, there is a presumption of compliance with the relevant essential requirements. Article 3.2 requires that “radio equipment shall be so constructed … so as to avoid harmful interference”. Harmonised standards for radio equipment are developed by ETSI. The spurious emissions limits in these standards are generally based on the limits in CEPT Recommendation 74/01 (except where there is a specific justification for different limits being applied); these are in turn based on ITU Recommendations. The emissions limits in the out-of-band domain are specific to each standard.

4/5

[1] Note: For convenience, the discussion in this section uses terminology generally used in ITU to describe emissions. However, this does not imply that it is appropriate to apply these terms to UWB in a regulatory context.

[2] Radio Regulations Article 1.145 Spurious emission: emission on a frequency or frequencies which are outside the necessary bandwidth and the level of which may be reduced without affecting the corresponding transmission of information. Spurious emissions include harmonic emissions, parasitic emissions, intermodulation products and frequency conversion products, but exclude out-of-band emissions.

[3] Radio Regulations Article 1.144 out-of-band emission*: Emission on a frequency or frequencies immediately outside the necessary bandwidth which results from the modulation process, but excluding spurious emissions.

[4] Directive 1999/5/EC of the European Parliament and of the Council of 9 March 1999 on radio equipment and telecommunications terminal equipment and the mutual recognition of their conformity.