Project topics and materials
Get Complete Project Material File(s) Now! »
Spectrum Availability and Constraints
For multi-mode operation, the choice of frequency of operation is an important deliberation for hardware design, more so from a network system management point-of-view. From a spectrum bandwidth and geographical standpoint there exist only a few bands which can allow the co-existence of GSM/EDGE, WCDMA/HSPA and LTE. The Table 3.1 gives an indication of the FDD frequency bands available for different standards, with Band Category 2, and Multi Standard Radio Bands 2/II, 3/III, 5/IV and 8/VIII referring to those currently amenable to multi-mode operation. In order to simplify the problem at hand, the focus of this work shall be limited European (ETSI) FDD bands while not being oblivious to limitations posed by respective bands in other geographies (e.g. in North America). Within Europe, there exist two bands EGSM900 / UTRA-VIII (925 − 960MHz) and the DCS1800 / UTRA-III (1805 − 1880MHz), which can support GSM, UMTS, and LTE. A typical telecommunication network license in most countries is issued for one or more standards. A growing trend is to re-use or ’re-farm’ spectrum or bands currently allocated to one technology such as GSM/EDGE, to deploy more spectrally efficient technologies such as WCDMA/HSPA and LTE/LTE-A. This is because lower frequencies allow better propagation conditions, especially when large coverage distances and depth i.e. availability of services indoors, or coverage area, are key. This takes advantage of the existing infrastructure and high penetration of GSM/EDGE technologies while significantly reducing the network deployment costs. Network planning and performance management issues still remain to be tackled but the discussion of these issues is beyond the scope of this work. Some part of the benefit from lower frequency operation, however, is lost, since antenna gains tend to get smaller at lower frequencies. To maintain the antenna gain at lower frequency would require a physically larger antenna which is not always feasible at base station sites and in small terminals. As a side note, only by around 2015 will the Digital Dividend band (470 − 862MHz) be fully available, providing bandwidths similar to existing bands (≈ 100MHz for FDD and TDD). It is however unlikely to be globally harmonized.
Contrarily, for more congested or high density urban scenarios, known as ’hot spots’, higher frequencies become interesting. This is because they are less than the lower bands, they offer larger bandwidths and consequently greater potential for higher data-rates. This would make the DCS1800 / UTRA-III an ideal candidate for throughput evolution. However, being closer to the popular 2100MHz UMTS band, it would offer only a minor cost benefit to shift, providing an obvious tradeoff. Returning to spectrum re-farming, the GSM specific 900MHz and 1800MHz spectrums in Europe have been re-farmed over the last few years to also support WCDMA networks [17, 18]. There are already several Base Station products that support this development, although most of them need to be factory reconfigured when switching bands. Software reconfiguration is still a growing trend. Interestingly, all of the LTE (E-UTRA) bands are common with the UTRA bands, implying that LTE and WCDMA can be deployed together, however, with the caveat that not all WCDMA (UTRA) bands support full bandwidth (20MHz) operation of LTE, being limited either by the spectrum availability or the bandwidth allocated to the operator (which typically varies from band to band and is different for each country). This would also apply to the GSM (GERAN) bands that also can support LTE [19, 20].
Spectrum Reuse or Re-farming
The issue of recovering spectrum from incumbent, but less efficient bands to re-use them for more efficient technologies has been introduced in Section 3.2. Extending WCDMA/HSPA and LTE operation to GSM bands does not imply immediate cessation of GSM services 7 but is actually intended to harmonize the transition from GSM to the more spectrally efficient technologies over a period of time. With advanced power control, cooperation and resource management strategies, UMTS and LTE standards (which are also backwards compatible with GSM/EDGE) are well suited to co-exist with the older technology in the same band. The requirements for the coexistence (coordinated and uncoordinated operation) between UMTS and GSM networks in these bands have partially been covered in their respective 3GPP standards (although not for same band of operation) and to a certain extent also in [41]. Recent releases of the GSM/EDGE standards and their evolutions are also focused on harmonizing performance and coexistence requirements with the newer standards. Because of the inter-carrier distortions, the RF design engineer would need to pay particular attention to the frequency separation between the similar and multi-mode carriers and also to the relative carrier power levels which will depend on network-level settings and hardware limitations. Examples of this are discussed in Chapter 4.
Multi-Operator resource sharing
With base stations offering multi-band solutions (not limited to region specific bands, as is most likely the case) spectrum allocation no longer remains the primary constraint for design. Manufacturers such as Nokia Siemens Networks propose different degree of sharing resources [42] such as
• Roaming-Based-Sharing: an operator can share another operator’s RAN indirectly via the core networks controlling the RAN.
• RAN-Sharing: operators have dedicated carriers but share network elements up to and including the radio network controller (RNC), or possibly only the base stations.
• Site-Sharing: shared equipment room, transmission equipment, power supply, roof top resources such as towers, poles and outdoor wire channels and other auxiliary facilities such as monitoring equipment and air conditioning.
Table of contents :
1 Introduction (en fran¸cais)
1.1 Motivation
1.2 M´ethodologie
1.3 Contributions
2 Introduction
2.1 Motivation
2.2 Methodology
2.3 Contributions of this work
3 Background
3.1 Identifying a Use-Case Scenario
3.2 Spectrum Availability and Constraints
3.3 Architectural and Topological Considerations
3.4 Co-existence Issues
4 Transmitter – System Level Analysis
4.1 Architecture of a contemporary Base Station radio transmitter
4.2 Synthesizing Link-Level Performance Requirements from System-Level Re- quirements
4.3 Spectrum Emissions Requirements
4.4 Summary
5 Multi-Mode Transmitter Performance Budgeting
5.1 Cascade Analysis for Cumulative Errors
5.2 Other Errors
5.3 Power Control Issues in a Multi-mode Transmitter
5.4 Multi-mode AQM with Mode-specific Gain Control
5.5 Summary
6 The Design of a Multi-Mode Variable Gain AQM
6.1 NXP-QUBiC4X Technology
6.2 AQM Input Stage
6.3 Mixers
6.4 Quadrature LO Generation
6.5 Common Base Output Stage
6.6 Simulation results
6.7 Time domain waveforms
6.8 S-parameter simulation results
6.9 Parametric sweeps
6.10 Monte Carlo simulations
6.11 Summary
Conclusion
A The Simplified Concept of Pre-Distortion
A.1 Basics of Pre-Distortion
B Basic Polynomial Non-linearity Mechanisms
B.1 Intermodulation Distortion
B.2 Composite Triple Beat
B.3 Cross Modulation
B.4 Intermodulation Enhancement
C Multi-mode DAC Dynamic Range Requirements
D Derivations from Chapter 6
D.1 Derivation RF voltage visible at the IF node of the Mixers
D.2 Input Impedance of the Common-Emitter Stage with Series-Series/Shunt-Shunt Feedback
Bibliography
