FCI Technology Investigations: L band Compatibility Criteria and Interference Scenarios Study

Transcription

1 FCI Technology Investigations: L band Compatibility Criteria and Interference Scenarios Study Deliverable C4: Draft compatibility criteria for GSM/UMTS900 Edition Number 1.0 Edition Date 24/08/2009 Status Final COOPERATIVE NETWORK DESIGN

2 Document information Document title Compatibility criteria and test specification for GSM/UMTS900 Author Simon Dunkley Produced by Helios 29 Hercules Way Aerospace Boulevard - AeroPark Farnborough Hampshire GU14 6UU UK Produced for Eurocontrol Helios contact John Micallef Tel: Fax: john.micallef@askhelios.com Produced under contract C Version 1.0 Date of release 24/08/2009 Document reference P1031C4 P1031C4v10 HELIOS 2 of 39

3 Contents 1 Introduction General About this document GSM UMTS Compatibility criteria Testing plan A Worst-case interference scenario B Relevant specifications C Operators adjacent to 960MHz band edge D LDACS candidates summary E UMTS BER measurements (courtesy of Agilent Technologies) Loopback Bit Error Ratio Measurement procedure Loopback BER Measurement Parameters Loopback BER Measurement Results Input Signal Requirements F GSM reference sensitivity measurements (courtesy of Agilent Technologies).. 37 Measuring the Bit Error Ratio P1031C4v10 HELIOS 3 of 39

4 1 Introduction 1.1 General Recognising that there is insufficient spectrum in the standard VHF band to support future aeronautical communications needs, two options for an L-Band Digital Aeronautical Communications System (L-DACS) have been identified by the European and US ICAO ACP members under the joint development activity known as the Future Communications Study (Action Plan 17). The first option for L-DACS is a frequency division duplex (FDD) configuration utilizing OFDM modulation techniques. The second LDACS option is a time division duplex (TDD) configuration utilising a binary (GMSK) modulation scheme One of the key questions with respect to these candidate L-DACS technologies which needs to be addressed is that of its compatibility with other, existing L-Band systems. Not only must the candidate systems be able to operate effectively whilst in the presence of interference from other systems, but they must also cause the minimum possible interference to the legacy systems. These compatibility analyses are required in order to assess the feasibility of using the competing L-DACS systems both in a ground, and in particular in an airborne environment As the specific details of the L-DACS implementation options, including the Air Interface, are currently being specified and validated, this document considers only the compatibility of the legacy systems i.e. the interference path from L-DACS TO the legacy systems. The development of the criteria in the reverse case (to L- DACS) will be carried out at a later stage. The following diagram outlines the process currently foreseen by EUROCONTROL, showing the steps to be undertaken to complete the L-DACS selection. Development of LDACS1/2 TX prototype Testing and LDACS Specifications Development of Evaluation Selection RX prototype Interference Scenarios, Criteria and Testing Plan Figure 1 Overall evaluation process currently foreseen This document is a deliverable of the study covering the grey box in Figure 1 addressing the interference criteria, scenarios and testing plan. The overall study addresses two aspects of the current systems. The first one considers the current systems as victims and aims to define the appropriate spectrum compatibility criteria with a new system. The second one considers the current systems as P1031C4v10 HELIOS 4 of 39

5 interferers and aims to define the appropriate interference scenarios to be used when evaluating the impact of the current systems to a new system For the first part, there are 5 deliverables covering DME, UAT, SSR, GSM/UMTS and GNSS (C1, C2, C3, C4 and C5). For the second part, there is one deliverable consolidating the interference scenarios for all the previously considered systems and JTDS/MIDS in addition. There is also a combined deliverable (C6/S6) covering both the criteria and scenarios for the RSBN system. Finally there is one deliverable C7 providing an analysis of the potential usage of the suppression bus by a new system. 1.2 About this document This document is deliverable C4 of the Spectrum Compatibility criteria and Interference Scenarios for existing systems operating in the L band study produced by Helios for Eurocontrol under Contract C as contribution to the Future Communication Study (FCS) activities, and in support of the work to realise one of the recommendations of the FCS to develop an L-band data link The development of the L-band data link is identified in the development activities for the SESAR Implementation Package 3 (IP3) in the post 2020 timeframe. Therefore, the outcome of this deliverable will be used as input to the SESAR JU development activities under WP This document sets out the compatibility criteria which would allow the proposed LDACS system to coexist with both GSM and (future) UMTS900 systems operating in the band up to 915MHz The restrictions to be placed on the LDACS systems are derived by considering the minimum performance of GSM/UMTS equipment (as set out in relevant specifications) in the presence of interferers in adjacent bands. The proscribed level of interference in these specifications defines the allowed transmissions that may be produced by the LDACS system under typical operating conditions. This leads to two Tx masks with which the LDACS systems must comply in order to be compatible with the GSM/UMTS systems under typical deployment conditions A testing plan is then defined which goes on to describe: - how LDACS prototypes can be tested for their conformance with these compatibility criteria (ie the Tx masks); - a practical test to confirm that the operation of the GSM/UMTS systems are not unreasonably degraded in the presence of the LDACS system (under extreme conditions) The criteria identified in this document are based on the current expected operating conditions of the proposed L-DACS systems, and on the published protection and susceptibility criteria for the technology alongside which L-DACS must operate. As such, issues such as frequency separation and emission masks for L-DACS are based on current expectations of the system parameters and final test validation will be necessary in order to confirm the implementation and test setup details herein documented. This paper is thus intended to provide the groundwork for the prototyping and compatibility trials bearing in mind that the compatibility criteria identified so far address interference in one direction where the legacy system is the victim The remainder of this document is as follows: P1031C4v10 HELIOS 5 of 39

6 - section 2 describes the GSM900 system; - section 3 describes the UMTS900 system; - section 4 defines the compatibility criteria with which the LDACS system should comply; - section 5 sets out practical tests to ensure that the LDACS system complies with these criteria, and that the two systems, under extreme conditions, are able to operate with minimal performance degradation. P1031C4v10 HELIOS 6 of 39

7 2 GSM System overview GSM, based on a TDMA (Time-Division Multiple Access) and FDD (Frequency-Domain Duplex) duplexing scheme, is the most successful of the 2nd generation digital mobile standards globally, and is practically ubiquitous (in various forms) around the world. Development of the 2nd generation GSM technology has been significant since its introduction in the 1990s, and includes air interface developments such as the introduction of EDGE (Enhanced Data Rates for Global Evolution), offering data rates of 384kbps. The components of the system are illustrated in figure 1. Mobile Base Station Base Station Controller Network Figure 1: GSM system components Mobile Stations (MSs) are classified according to the maximum power that they can transmit1, and range (in the 900MHz band) from Class 2 (8W or +39dBm) to Class 5 (0.8W, +29dBm). Base stations (BSs) range from +20 to +43dBm (100mW to 20W) per carrier. BS antennas are typically configured as omni-directional or tri-sectored, with a maximum gain of +18dBi2 (leading to a maximum EIRP of +61dBm 1.25kW). Carriers are modulated to occupy 200kHz channels, with cell re-use patterns of typically four, and the TDMA structure makes eight timeslots available. Frequency Division Duplex (FDD) is used with uplink (UL) and downlink channels (DL) separated by between 45MHz and 95MHz, depending on the band. The modulation used on the GSM air interface varies according to enhancements that may have been carried out, but is either GMSK or 3/8-shifted 8-PSK (EDGE). Cell planning is a complex process, balancing cost, quality of service and resistance to new base station sites to achieve as spectrally efficient systems as possible. In rural 1 Note: the GSM specifications have been modified to allow low-power MSs to conform to a less stringent specification, leaving them more vulnerable to external interference and blocking. They are referred to as Small MSs. 2 Some operators, such as Proximus in Belgium, use 6-sector antennas with up to 25 dbi gain, especially in very dense urban areas. P1031C4v10 HELIOS 7 of 39

8 areas, cell sizes are large and interference from others low; in urban areas, the system typically operates under interference-limited conditions, where the thermal noise floor is, in effect, raised by up to 10dB. The 900 and 1800MHz bands were reserved by the European Commission3 during the 1980s to be used exclusively for 2G GSM technology. The GSM system is primarily rolled out in these bands4, typical assignments at 900MHz being illustrated in the figure below. The spectrum immediately adjacent to the ARNS band is used for downlinks (base station transmit), the first uplink spectrum being 45MHz lower. Figure 2: GSM900 spectrum Mobile telephone use is currently forbidden on most aircraft, however, in April 2008, the Commission ruled that the use of mobile devices in midair is to be allowed by means of on-board picocells (GSM on-board (GSMOB)) in the 1800MHz band linked back to the fixed infrastructure via satellite. Safe GSMOB operation is achieved by restricting use to above 3000m, constraining the power that the on-board BSs and MSs can transmit (and coordinating the channels used with nearby aircraft) and using an on-board Network Control Unit (NCU) transmitting a low-power signal in the DL band of all mobile technologies, including CDMA-450, GSM-900 and UMTS2000 as well as GSM-1800 to mask any signals that individual MSs might detect from ground-based BSs and thus attempt to ensure that MS only transmit on the frequencies selected for use on the aircraft concerned. 3 GSM Directive (87/372/EEC) MHz in countries such as the USA. P1031C4v10 HELIOS 8 of 39

9 3 UMTS System overview The UMTS air-interface, UTRAN (UMTS Terrestrial Radio Access Network), is based on wideband CDMA protocols, offers still higher data rates, and evolutions of the specification, HSDPA/HSUPA (High-Speed Downlink/Uplink Packet Access), offer payload rates of up to 14.4 Mbps. Current HSDPA deployments support down-link speeds of 1.8, 3.6, 7.2 and 14.4 Mbps/s. Further speed increases are available with HSDPA+, which provides speeds of up to 42 Mbps. UMTS systems have primarily been rolled out in the 2.0 GHz band5 and in an FDD configuration, although many operators have also been assigned associated TDD (Time-Domain Duplex) spectrum. Most, however, are only recently considering exploitation of this resource. The development of 3G, however, has seen traffic increasingly transferring to this new technology operating in the 2.0 GHz band, relaxing the pressure on the lower bands. With time, operators are likely to want to convert some of their 900/1800 MHz spectrum for use by new technologies (as has already happened in a limited trial in Finland), and the principle of liberalisation was embraced by the European Commission in The components of the UMTS system are, for the purposed of this document, similar to those of the GSM system (see figure 1-1), with an MS being replaced by a UE, a BS by a Node B, and a BSC by a Radio Network Controller (RNC). The RNC has considerably more functionality than its GSM counterpart, coordinating soft handovers between cells (ie with no disruption to the radio bearer) and having the ability for more than one Node B to communicate to a single UE, thereby offering macro-diversity 6. UMTS is a wideband CDMA (Code-Division Multiple Access) technology, carriers being modulated to occupy 5MHz channels. The technology has initially been implemented in the 2.1GHz band, using FDD blocks separated by 90MHz, but spectrum is also available for Time Domain Duplex (TDD) mode operation, which is only now beginning to be exploited. UMTS900 would not use TDD mode and would conform to the existing 45MHz duplex spacing. CDMA operates by modulating the data stream using QPSK modulation, and then spreading the resultant signal across the 5MHz channel using a high-rate (typically 1.28Mbps or 3.84Mbps) spreading code7 to rapidly switch the sense of the carrier. The power flux associated with individual signals falls to extremely low levels, often below the thermal noise floor, and means that lower powers can be used for any one link, typically +20dBm for MSs. UMTS BS antennas are often similarly tri-sectored, but the nature of the CDMA technology is that reuse factors of one can be achieved, boosting the spectral efficiency of the system. Complex simulations are used to plan CDMA systems where factors such as interference from within a cell and neighbouring cells, and changes to the size 5 Terrestrial FDD allocations of MHz paired with MHz; TDD of MHz & MHz; FDD satellite allocations of paired with MHz 6 Improving the capacity of the network by allowing links between Node Bs and UEs to be achieved with lower power. 7 The spreading codes are generated by signal processing technology, are unique for each link and are used at the receiver to retrieve the signal. This is the basis of the multiple access mechanism. P1031C4v10 HELIOS 9 of 39

10 cells under high load conditions, or cell breathing, influence the way in which the system is deployed and managed. Furthermore, the power for each link needs to be carefully controlled in order to maximise the capacity of each cell. P1031C4v10 HELIOS 10 of 39

11 4 Compatibility criteria The co-ordination of GSM services with services in bands adjacent to the 900MHz band (eg ARNS and broadcast services) has been carried out on a case-by-case basis, leading to the finalisation of the relevant ETSI specifications. For example, when the GSM standard was under development in the 1980 s, its co-existence with DME services was investigated and restrictions placed on the transmission mask of the GSM system accordingly. GSM s resilience to interference from neighbouring systems was designed on the assumption that appropriate guard bands would be in place (eg 1.5MHz to DME ground stations), and specified in the form of adjacent channel interference performance (both from a system s own units and neighbouring systems8); resistance to spurious emissions; and front-end blocking performance, the latter likely to be the limiting factor in coexistence with an LDACS systems. The in- and out-of-band9 blocking performance of GSM MSs is shown in figure 3: this represents the level at which an interferer10 would begin to degrade system performance if individual links are noise-limited and operating at the limit of their sensitivity. Uplink (BS to MS) operations are removed from the 960MHz boundary by 45MHz and so are unlikely to be degraded by an LDACS systems, except, possibly, if BSs are collocated. Figure 3: GSM in- and out-of-band blocking immunity small/other MS 8 Neighbouring GSM systems, both spectrally and geographically are typically achieved through biparty co-ordination agreements. 9 The band effectively overstepping the 960MHz threshold by 20MHz according to the ETSI specifications, in an attempt to protect GSM services. 10 The specifications define an interferer as being a CW signal injected into a system at varying offsets from the required central carrier s frequency. P1031C4v10 HELIOS 11 of 39

12 Similarly, the blocking performance 11 of UMTS MS receivers are illustrated in figure 4, which, although the band extends only 15MHz (to 975MHz) has an more conservative specification ie -30dBm across most of the L band. Figure 4: UMTS in- and out-of-band blocking immunity 4.1 Spectral comparison A combination of the two profiles (and raising the UMTS profile by 7dB - see footnote) and depiction of the relative position of the two candidate LDACS systems is illustrated in figure 5. This defines the maximum power flux from interfering systems that should allow UMTS/GSM systems to operate unperturbed as noise-limited/self-interferencelimited systems, respectively. Deployed mobile telephony equipment typically performs much better than the minimum criteria set out in ETSI specifications, but operators typically make use of this headroom by deploying high densities of equipment. The operators would be reluctant, therefore, to relax these stringent conditions (although Ofcom in the UK recently allowed the use of low power operations in channels adjacent to operational systems), 11 Using a similar CW injection method. An additional specification for out-of-band narrow-band systems (appropriate for the candidate LDACS systems), defines performance expected in the presence of a narrow-band system operating at an offset of 300kHz from the 960MHz boundary, but anticipates a 7dB higher operating margin ie the wanted signal being 7dB stronger than the cases where a CW signal is injected. Given that the nature of CDMA technology means that systems are almost always operating in an interference-limited mode, it seems reasonable to raise the blocking thresholds defined for CW signals by a corresponding 7dB in defining the system s resilience to external narrow band interferers, as shown in figure 5. P1031C4v10 HELIOS 12 of 39

13 and are likely to argue that any new system should offer no more interference than existing ARNS systems. Figure 5: Spectral comparison: resilience of GSM and UMTS to CW (solid line) and narrow band interferers (dashed line); and candidate LDACS systems Two transmission masks 12 can be defined therefore which respect these limits, and are illustrated below. Each defines the maximum signal that can be received at the input port of the MS receiver into a 1MHz and 30kHz bandwidth, respectively (figure 6 and 713). Note: This profile is dominated by the blocking performance of UMTS MS receivers. 12 The mask extends to MHz, but it is assumed that systems operating higher up the L band will modify this mask accordingly 13 The thresholds for measurements into 30kHz are raised by 8dB, reflecting the power density that would be expected for a carrier bandwidth of 200kHz, the worst case for the current LDACS proposals P1031C4v10 HELIOS 13 of 39

14 Figure 6: LDACS Tx mask (into 1MHz) Figure 7: LDACS Tx mask (into 30kHz) P1031C4v10 HELIOS 14 of 39

15 5 Testing plan The strategy for testing the compatibility of the candidate LDACS systems with GSM/UMTS900 is approached in two ways. 5.1 Conformance with Tx mask The production LDACS system should conform to the minimum criteria set out in the form of the Tx masks in figures 6 and 7. The output of the LDACS system should be measured using standard power measurement equipment with measurement bandwidths of 30kHz and 1MHz. An attenuator of XdB should be inserted to replicate the lowest attenuation expected between the LDACS output and the GSM input. The centre of the measurements should be as set out in the table below. Measurement centre (MHz) Maximum signal into 1MHz (dbm) to in 1 MHz steps to in 1 MHz steps to in 1 MHz steps to in 1 MHz steps -8 Table 1: LDACS conformance Tx mask into 1MHz Measurement centre (MHz) Maximum signal into 30kHz (dbm) to in 30 khz steps to in 30 khz steps to in 30 khz steps to 12749,985 in 30 khz steps -15 Table 2: LDACS conformance Tx mask into 30kHz Failure to comply with the Conformance with Tx mask test would indicate that the LDACS equipment is not conforming to the restrictions highlighted above and may disrupt the operation of MSs operating at the top of the mobile band. 5.2 System compatibility P1031C4v10 HELIOS 15 of 39

16 The LDACS systems should be operated in close proximity to a selection of GSM/UMTS900 systems to confirm compatibility14. Each MS should be set up and tested for in- and out-of-band resilience to interfering signals as set out in the relevant specifications15 and described below GSM The BER (see Annex B) shall not exceed when a useful signal modulated with the relevant supported modulation (GMSK or 3π/8-shifted 8-PSK16), at frequency MHz, 3 db above the reference sensitivity level (Small MS -102dBm; Other MS -104dBm) and the LDACS signal at a frequency (f) (the lowest in the L band proposed to be used by the system) at a signal level as determined by table 3. The traffic passing over the LDACS link shall be equivalent of the maximum data throughput achievable. LDACS carrier centre (f) f+bw/2 < LDACS signal (dbm) -43 dbm < f+bw/2 < dbm < f+bw/2 < dbm 980 < f+bw/2-0 dbm BW = Carrier bandwidth of the LDACS system Table 3: LDACS interfering signal strength at victim receiver (GSM) input For each unit tested, the reference sensitivity should be demonstrated to be achieved in the absence of the LDACS signal UMTS 14 Conformance with the Tx mask (constructed from the in- and out-of-band blocking criteria) may, although conservative, not be sufficient to ensure that spurious emissions do not upset the operation of the mobile systems. 15 ETSI TS for GSM and TS for UMTS 16 Used by EDGE P1031C4v10 HELIOS 16 of 39

17 The BER (see Annex B) shall not exceed for the parameters specified in table 4 and the LDACS signal at a frequency (f) (the lowest in the L band proposed to be used by the system) at a signal level as determined by table Parameter Value Units DPCH_Ec Refsens dB or -104 db/3.84mhz Îor REFÎor + 10dB or db/3.84mhz UE transmitted mean power 20 (for power class 3 and 3bis) 18 (for power class 4) dbm Table 4: Set up conditions for UMTS BER tests LDACS carrier centre (MHz) f+bw/2 < LDACS signal (dbm) -49 dbm < f+bw/2 < dbm 1020 < f+bw/2 < dbm 1044 < f+bw/2-8 dbm BW = Carrier bandwidth of the LDACS system Table 5: LDACS interfering signal strength at victim receiver (UMTS) input For each unit tested, the required BER should be demonstrated to be achieved in the absence of the LDACS signal. Failure to pass the System Compatibility tests would indicate that, although the LDACS equipment is conforming to these restrictions, its presence is disrupting the operation of MSs operating at the top of the mobile band and further investigations should be carried out as to the exact cause. 17 Using the tests specified in section (Minimum requirement narrow band blocking), TS Defined in table 7-2 (Test parameters for reference sensitivity), TS P1031C4v10 HELIOS 17 of 39

18 P1031C4v10 HELIOS 18 of 39

19 A Worst-case interference scenario The scenarios that are likely to most threaten the operation of GSM networks are illustrated below. LDACS UE to terrestrial mobiles The LDACS transceiver on aircraft are likely to be downward facing, but omni-directional, transmitting at between +36dBm and in the MHz band (for LDACS2, assuming the current 1.5MHz DME guard band is respected). In urban areas, including those surrounding around airports, external GSM signal levels received by mobile telephones is likely to be no less than -84dBm19. Aircraft approaching airports on a 3 degree approach path will be at altitudes as indicated by the table below (assuming flat terrain). Distance from airport (km) Altitude (m) Path loss to LDACS base station (db) Maximum LDACS Tx power* (dbm) Path loss vertically (db) Signal received ground (dbm) on * to achieve an SNR of 10dB (for the LDACS system) plus a safety margin of 10dB. Assumes dynamic power control. maximum allowed at 961.5MHz, -48dBm (UMTS); -42dBm (GSM) Table 6: Worst case interference scenario: LDACS UE to terrestrial mobiles LDACS BS to terrestrial mobiles Assuming the LDACS base station is located at an airport, but at least 100m from terminal buildings, and the LDACS2 base station will transmit at +35dBm (assuming a single carrier). The antennas are arranged to provide at least 10diB isolation to MSs at ground level. The table below indicates the signal received by GSM MSs at varying distance from the base station. 19 Minimum reference sensitivity (-102dBm) plus the required C/I (+9dB) plus the typical noise floor uplift (+10dB) P1031C4v10 HELIOS 19 of 39

20 Distance from LDACS BS (m) Path loss to LDACS base station (db) Signal received by victim (dbm) maximum allowed at 961.5MHz, -48dBm (UMTS); -42dBm (GSM) Table 7: Worst case interference scenario: LDACS BS to terrestrial mobiles P1031C4v10 HELIOS 20 of 39

21 P1031C4v10 HELIOS 21 of 39

22 B Relevant specifications The following material has been referenced from ETSI TS V ( ): Digital cellular telecommunications system (Phase 2+); Radio Transmission and Reception; (3GPP TS version Release 1999). Reference sensitivity level (GSM900) Small MS -102dBm Other MS -104dBm Blocking characteristics,with wanted signal of level +3dB above reference sensitivity, to maintain BER 0.001, in presence of CW signal: MS ( ) In-band 600kHz f-f0 < 800kHz -38dBm, -43dBm (for small MS) 800kHz f-f0 < 1600kHz -33dBm, -43dBm (for small MS) 1600kHz f-f0 < 3000kHz -23dBm, -33dBm (for small MS) 3000kHz f-f0-23dbm, -23dBm (for small MS) Out-of-band (>980MHz) 0dBm Spurious response - Not specified Receiver inter-modulation characteristics defined, but less onerous than blocking criteria. The following has been derived from ETSI TS V8.3.0 ( ): Universal Mobile Telecommunications System (UMTS); User Equipment (UE) radio transmission and reception (FDD) (3GPP TS version Release 8) Reference sensitivity level (UMTS) P1031C4v10 HELIOS 22 of 39

23 Band Unit DPCH_Ec <REFSENS> <REFÎor> VIII dbm/3.84 MHz Blocking characteristics In-band (defined as in UE receive band or 15MHz above ie MHz), to maintain BER 0.001, in presence of CW signal f-f0 = 10MHz f-f0 > 15MHz -56dBm -44dBm Out-of-band (15MHz above UE receive band ie >975MHz), to maintain BER 0.001, in presence of CW signal Frequency range 1 ( MHz) -44dBm Frequency range 2 ( MHz) -30dBm Frequency range 3 ( MHz) -15dBm Narrow-band blocking: to maintain BER 0.001, in presence of GMSK interfering signal as defined in TS , with Ior = Ref Ior+10dB, 2.8MHz from channel centre, -56dBm Spurious response Receivers are allowed to fail out-of-band blocking tests at 24 points in the range MHz, but must not fail for a power of -44dBm. Receiver inter-modulation characteristics Not specified P1031C4v10 HELIOS 23 of 39

24 P1031C4v10 HELIOS 24 of 39

25 C Operators adjacent to 960MHz band edge Country Operator Separation from band edge (khz) Austria Orange Austria 400 Belgium Base 100 Czech Republic Vodafone Czech Republic 200 Denmark Sonofon 1200 Finland Radiolinja Origo Oy 100 France SFR 100 Germany T- Mobile Deutschland GmbH 100 Greece Panafon SA 0 Hungary Vodafone 300 Ireland O2 300 Italy Wind Telecomunicazioni 200 Portugal TMN 1100 Romania Orange Romania 0 Spain Vodafone 100 Sweden Telenor 1100 Switzerland TDC Switzerland 100 Netherlands Vodafone 1100 Turkey Turkcell 1000 United Kingdom O2 (UK) Ltd 100 Table 8: Users of spectrum adjacent to the L band P1031C4v10 HELIOS 25 of 39

26 P1031C4v10 HELIOS 26 of 39

27 D LDACS candidates summary System Duplex Forward link (MHz) Reverse (MHz) link EIRP (dbm) LDACS2 (LDACS2) B-AMCX (LDACS1) TDD FDD /48 Table 9: Candidate LDACS systems summary P1031C4v10 HELIOS 27 of 39

28 P1031C4v10 HELIOS 28 of 39

29 E UMTS BER measurements (courtesy of Agilent Technologies) Background on the measurements Loopback Bit Error Ratio Measurement procedure 3GPP TS , F defines bit error ratio (BER) as follows: "The Bit Error Ratio is defined as the ratio of the bits wrongly received to all data bits sent. The bits are the information bits above the convolutional/turbo decoder." When the test set measures loopback bit error ratio, it sends a known data pattern (see DL DTCH Data ) on the downlink dedicated traffic channel (DTCH) to a UE that is configured in loopback mode 1 or 2. (Although the measurement will operate with UE Loopback Type set to 2, all TS tests utilizing the BER measurement require loopback mode 1). The UE decodes the data and retransmits it on the uplink DTCH. The test set analyzes the uplink data to see how closely it matches the data bits originally sent on the downlink. The test set compares the downlink and uplink data one transport block at a time and reports the Bit Error Ratio. To decrease test time, you may set Confidence State to On and specify the BER Requirement, which allows the test set to stop testing when an early pass or early fail event occurs, and to report a Confidence Test Result (see 3GPP TS F.6 and 3GPP TS ). 3GPP TS Figure F Because the test set must compare the UE's uplink data with the corresponding downlink data, the measurement must be able to synchronize to the UE's signal (align received uplink transport blocks with their corresponding downlink transport blocks. In doing so, the test set determines the UE's Loopback Delay ). See Input Signal Requirements. When testing the UE's receiver performance, it is assumed that the uplink is error free. Any errors occurring on the uplink invalidate the receiver test. The loopback BER measurement returns information about uplink CRC Errors to assist in ensuring that the uplink is error free (to perform the loopback BER measurement, you must set Uplink DTCH RMC CRC Presence to Present so that an uplink CRC is generated). To perform the loopback BER measurement while in Active Cell Operating Mode, you must be on an RB test mode call, with Channel Type set to a symmetrical RMC. Loopback BER can not be measured for Asymmetrical RMCs. The trigger source for this measurement is always protocol-based (see Trigger Source Description ). UE Control During FDD Test Mode Operation The FDD Test Operating Mode does not provide radio bearer setup control of the UE for test mode operation. You must configure the UE in loopback mode 1 or 2 and synchronize it to the test set's downlink signal in order to make this measurement in FDD test operating mode. P1031C4v10 HELIOS 29 of 39

30 The uplink transport format must include both the DTCH and the DCCH; however, only the DTCH information is used for the loopback bit error ratio measurement. Loopback BER Measurement Parameters Number of Bits to Test When Confidence State is Off, you may set the number of data bits to test during the measurement using the Number of Bits to Test setting. However, the actual number of bits tested is always the largest multiple of the number of bits needed to fill a DTCH transport block for the transmitted format. For example, for the 12.2 kbps reference measurement channel (RMC) the transport block is 244 bits. If you set the number of bits to test to 1000, the actual number of bits tested will be 1220 (1000 bits/244 bits per block = 4.1 blocks, so 5 blocks are required to test at least 1000 bits. 5 blocks x 244 bits per block = 1220 bits). When Confidence State is On, you may set the maximum number of bits to test during the measurement using the Number of Bits to Test setting. However, the actual number of bits tested will be less than this number if an early pass or early fail event occurs (see Confidence Test Result ). If a Confidence Test Result of Max Bits Tested is returned, then the actual number of bits tested is the largest multiple of the number of bits needed to fill a DTCH transport block for the transmitted format, as explained above. GPIB command: SETup:WBERror:COUNt Trigger Arm - see Trigger Arm (Single or Continuous) Description. Measurement Timeout - see Measurement Timeouts. Confidence State To decrease test time, you may set Confidence State to On and specify BER Requirement, which allows the test set to stop testing when an early pass or early fail event occurs, and to report a Confidence Test Result (see 3GPP TS F.6 and 3GPP TS ). From the front panel, Confidence State can be set by pressing the ON or OFF key while Confidence Level is selected, or by simply entering a valid value for Confidence Level. GPIB commands: SETup:WBERror:CONFidence[:SLEVel], SETup:WBERror:CONFidence:STATe Confidence Level 3GPP TS F.6 defines: F = probability of a wrong decision (1 - F) = probability of a correct decision The probability (risk) to fail a good DUT shall be </= F according to the following definition: A DUT is failed, accepting a probability of </= F that the DUT is still better than the specified error ratio (Test requirement (see BER Requirement )). P1031C4v10 HELIOS 30 of 39

31 The probability to pass a bad DUT shall be </= F according to the following definition: A DUT is passed, accepting a probability of </= F that the DUT is still worse than M times the specified error ratio (see BER Requirement ). (M>1 is the Bad DUT Factor ). The Confidence Level parameter is equivalent to (1 - F). 3GPP TS table F specifies an F of 0.2% ( Confidence Level of 99.8% ) for all Receiver and Performance Characteristics tests in 3GPP TS sections 6 and 7 (except for the fail limit for 6.5 Blocking Characteristics). Confidence Level is thus fixed to 99.8%. Similarly, Bad DUT Factor is fixed to 1.5. GPIB commands: SETup:WBERror:CONFidence[:SLEVel], SETup:WBERror:CONFidence:LEVel BER Requirement 3GPP TS F.6 defines TR = Test Requirement. As testing progresses, the bit error ratio trajectory converges to a value (see 3GPP TS Figure F ). The BER value that the UE is converging to shall not exceed the Test Requirement. The BER Requirement parameter is equivalent to Test Requirement expressed as a percentage. 3GPP TS table F specifies a Test Requirement of between and 0.1 (0.1% and 10% BER Requirement ) for the Receiver and Performance Characteristics tests in 3GPP TS sections 6 and 7. This setting is only available when Confidence State is On. GPIB command: SETup:WBERror[:RATio]:REQuirement Bad DUT Factor 3GPP TS F.6 defines M = bad DUT factor. A Bad DUT Factor of greater than 1 shifts the early pass limit curve up (the early pass limit curve is defined by 3GPP TS F equation (2). See also 3GPP TS Figure F ) This allows an early pass event to occur after fewer bits have been tested for a given number of errors, which results in a higher bit error ratio (see the early pass equation for NL( ne ) in 3GPP TS F.6.1.9). As M is increased, a greater number of bad DUTs are passed. 3GPP TS table F specifies a Bad DUT Factor of 1.5 for all Receiver and Performance Characteristics tests in 3GPP TS sections 6 and 7. Thus, Bad DUT Factor is fixed to 1.5. This setting is only available when Confidence State is On. GPIB command: SETup:WBERror:CONFidence:DUTFactor Uplink DTCH RMC CRC Presence UE Loopback Type Loopback BER Measurement Results Integrity Indicator Bit Error Ratio The ratio of incorrectly received bits to the Bits Tested, expressed as a percentage (%). P1031C4v10 HELIOS 31 of 39

32 GPIB command: FETCh:WBERror<[:ALL] :INTegrity :RATio :COUNt :BITS>? Confidence Test Result The Confidence Test Result is only available when Confidence State is On. o o Pass : Few enough bit errors were detected for a large enough number of Bits Tested that the test set is able to predict (with the specified Confidence Level ) that the bit error ratio trajectory is converging to a value below the BER Requirement. An early pass event is determined according to 3GPP TS equation (2). Fail : Enough bit errors were detected for the number of Bits Tested that the test set is able to predict (with the specified Confidence Level ) that the bit error ratio trajectory is converging to a value above the BER Requirement. An early fail event is determined according to 3GPP TS equation (1). o Max Bits Tested : After testing the number of bits specified by Number of Bits to Test, the test set is unable to predict (with the specified Confidence Level ) whether the bit error ratio trajectory is converging to a value above or below the BER Requirement. GPIB command: FETCh:WBERror:RATio:FAIL? Bit Error Count The number of incorrectly received bits. GPIB command: FETCh:WBERror<[:ALL] :INTegrity :RATio :COUNt :BITS>? Bits Tested Number of bits tested to determine the Bit Error Count and Bit Error Ratio results. GPIB command: FETCh:WBERror<[:ALL] :INTegrity :RATio :COUNt :BITS>? Missing Blocks The UE sends a TFCI for each uplink DTCH transport block. When operating in Active Cell Operating Mode, the test set monitors the uplink TFCI, and can report uplink missing blocks. If the TFCI indicates that there is no data block sent for that TTI, the test set reports a missing block. When a missing block occurs, Bit Error Count and the number of Bits Tested are not incremented. The test set returns a Questionable MS-to-Cell Data Integrity Indicator. When in FDD Test Operating Mode, the test set does not monitor the uplink TFCI; rather, it assumes that a data block is sent in every TTI. However, when a missing block occurs, the test set detects a CRC error. Thus, the data within the block is not analyzed, and the number of Bits Tested is not incremented. GPIB command: FETCh:WBERror:MBLocks? CRC Errors To determine if an uplink CRC error has occurred, the test set calculates the CRC for the data sent in the uplink DTCH transport block, and compares the result to the CRC sent by the UE for that transport block. When a CRC error occurs, the data within the block is not analyzed, and the number of Bits Tested is not incremented. If a CRC P1031C4v10 HELIOS 32 of 39

33 error occurs, the test set returns a Questionable MS-to-Cell Data Integrity Indicator. GPIB command: FETCh:WBERror:CRCerrors? Loopback Delay 3GPP TS section specifies that the uplink frame transmission take place 1024 chips after reception of the corresponding downlink frame. In reality, there is often a larger delay before the UE is able to transmit the uplink frame. 3GPP TS , specifies this delay between received downlink radio frames and their corresponding uplink radio frames (produced from the received data) as the loopback delay. Loopback delay is measured at the UE antenna connector and is specified in units of radio frames. While the UE test loop is closed and the radio bearer configuration is not changed, the UE is required to maintain a fixed loopback delay, which must not exceed the number of radio frames corresponding to 10 times the TTI (Transmission Timing Interval) of the actual transport channel configuration. For example, if TTI is 10 ms, the UE is required to maintain a fixed loopback delay not to exceed 10 radio frames (the test set can actually tolerate a loopback delay of up to 50 TTI). To calculate loopback delay, the test set correlates the data in the uplink radio frame to the data sent in the downlink radio frame. Thus, the data sent by the test set must be different for each radio frame ( DL DTCH Data must not be set to All Zeros or All Ones ). GPIB command: FETCh:WBERror:LDELay? Input Signal Requirements The expected power into the test set must be set in order to maintain the link with the UE. This is done using the UE Target Power setting or by using manual receiver power control and specifying the expected power level. See Receiver Control. Frequency: 800 to 1000 MHz, 1700 to 1990 MHz Input level: -61 dbm to +28 dbm The test set can synchronize to the UE's signal only if the bit error ratio is less than 10% on a block-by-block basis for three consecutive blocks, and the loopback delay is less than 50 transmission time intervals of the DTCH. See Synchronization requirements. Carrying out the measurements The test equipment used is the Agilent E6703A W-CDMA lab application for the 8960 Series 10 (E5515C) wireless communications test set. P1031C4v10 HELIOS 33 of 39

34 Figure 8: Agilent 8960 wireless communications test set The test equipment is set up as shown below. The attenuator is set to a level equivalent to the worst case expected ie such that the signal received at the GSM transceiver conforms with the masks set out in figures 6 and 7. The traffic passing over the LDACS link shall be equivalent of the maximum data throughput achievable. The second LDACS unit maybe required in order for the system to work correctly. RF in 8960 Wireless Communications Test set RF out GSM/UMTS transceiver Combiner Attenuator Splitter LDACS system (Tx) LDACS system (Rx) Figure 9: Experimental setup The test equipment used is the Agilent E6703A W-CDMA lab application for the 8960 Series 10 (E5515C) wireless communications test set. P1031C4v10 HELIOS 34 of 39

35 Measuring Loopback Bit Error Ratio The following is a description of the test procedures. 1. Press the Measurement selection key. 2. Select the Loopback Bit Error Ratio measurement. 3. Select Bit Error Setup ( F1 ). Set Number of Bits to Test to Set Trigger Arm to Single. Set Confidence Level to 99.80% (press the ON key). Set BER Requirement to 0.10%. Set Uplink DTCH RMC CRC Presence to Present and UE Loopback Type to Type 1. Select Close Menu ( F6 ). 4. Establish an RB Test Mode connection between the UE and test set (see How Do I Set Up a Call and Make a Connection? ). 5. Set UE Target Power to +21 dbm (for a Power Class 4 UE). 6. Set Cell Power ( F7 ) to -106 dbm (for Operating Band I). 7. SET THE LDACS POWER APPROPRIATELY 8. Press the START SINGLE key to begin the measurement. 9. If the Confidence Test Result is Max Bits Tested, start the measurement again. You may need to increase the Number of Bits to Test to enable a Pass or Fail result. 10. Adjust Cell Power ( F7 ), AWGN Power ( F4 ), Soft Handover ( F3 ) and Cell 2 Power ( F2 ) as desired. An example measurement result is shown below: P1031C4v10 HELIOS 35 of 39

36 P1031C4v10 HELIOS 36 of 39

37 F GSM reference sensitivity measurements (courtesy of Agilent Technologies) Background on the measurements Bit Error (BER) Measurement Procedure During BER measurements, the test set generates a downlink TCH with pseudorandom binary sequence, PRBS-15, data at a known power level. The mobile station receives the data, loops it back to its transmitter, and returns the data to the test set. The test set compares data sent to data received, and BER is calculated. SETup subsystem commands are sent to the test set to specify the time taken to close its loopback path, whether to open or close a loop during downlink signaling operations (for example, channel assignment), the number of bits to test, measurement type, loopback delay, measurement units, trigger arm, and measurement timeout values. When a call is established on a TCH, the loopback type, corresponding to one of the BER measurement types, must be sent to the mobile station. The test set closes the loopback automatically and re-opens it when the measurement is closed (that is, when INITiate:BERRor is OFF). You must set the measurement type from one of the available measurement types, (see SETup:BERRor[:TYPE] ). If you query a residual result when a nonresidual measurement is initiated, the test set returns 9.91 E+37 (NAN). The Measurement type must be set before initiating a BER measurement. See BER Measurement Types The loop must be closed before a BER test can start. Using the close loop signalling delay time feature allows time for the loop to close. See SETup:BERRor:CLSDelay[:STIMe] for more details. Each mobile station may have a different time delay between receiving a frame and re-sending it on the uplink. By default, the test set is configured to automatically determine the amount of delay needed. This is determined automatically when the test set has, for two frames, correctly received 80% of the downlink bits back on the uplink. The test set can be queried for the loopback delay value. If necessary, you can manually set the loopback delay (see SETup:BERRor:LDControl:AUTO ). In case the test set is not able to correlate the data it transmits on the downlink with the data it receives on the uplink, a Measurement Timeout value should be set. If a timeout is not set and the test set is unable to correlate, the measurement appears to "hang". The BER measurement trigger source is always set to immediate. The BER measurement does not offer multi-measurement results. See Statistical Measurement Results P1031C4v10 HELIOS 37 of 39

38 BER, FBER, and DAUDIO (uplink speech level) measurements are mutually exclusive measurements. Whichever of these measurements is activated last forces the others to stop. BER Measurement Types Type IA, Type IB, and Type II for both residual and non-residual refer to the different class types describing GSM speech coding bits. Residual: Residual Type IA (50 bits per speech frame) Residual Type IB (132 bits per speech frame) Residual Type II (78 bits per speech frame) Residual Data When one of these residual measurement types is selected, the test set sends the mobile station a message instructing it to close loopback A. When the mobile station is in loopback A, it erases frames received with CRC (cyclic redundancy check) errors and replaces all the bits in those frames with zeros. The BER measurement is then able to return the frame erasure count or ratio results as well as the results of the bitwise comparison of the un-erased frames looped back from the mobile station. Non-residual: Type IA (50 bits per speech frame) Type IB (132 bits per speech frame) Type II (78 bits per speech frame) Data When you select one of these non-residual measurement types, the test set sends the mobile station a message instructing it to close loopback B. A BER measurement with CRC's (cyclic redundancy check) returns the CRC count or ratio results. The mobile station does not indicate if any frames in the downlink were erased. Residual (D): Residual Type IA (D) (50 bits per speech frame) Residual Type IB (D) (132 bits per speech frame) Residual Type II (D) (78 bits per speech frame) Residual Data (D) When one of these residual (D) measurement types is selected, the test set sends the mobile station a message instructing it to close loopback D. The test set works almost the same as the correspondent Residual measurement type with one exception: the FER display during the measurement is actually Frame Error Rate plus Unreliable Frames Rate. P1031C4v10 HELIOS 38 of 39

39 BER Measurement Results The results of a BER measurement can be displayed in two ways, (number of errors counted) or (the ratio bad bits (errors) to total bits counted). If you are using the test set manually, select either Count or % from the Measurement Units field. If you are using the test set remotely, these results are available using the FETCh command, see FETCh:BERRor:COUNt[:BITS]? or FETCh:BERRor:RATio[:BITS]?. Alternatively the FETCh:BERRor[:ALL]? or FETCh:BERRor:FULL? can also be used to return the results. Type A Residual Measurement Results Integrity Indicator Bit Error Ratio Bits Tested Bit Error Count Frame Error Rate Frame Erasure Count Type B Non-Residual Measurement Results Integrity Indicator Bit Error Ratio Bits Tested Bit Error Count CRC Ratio CRC Count Type D Residual Measurement Results Integrity Indicator Bit Error Ratio Bits Tested Bit Error Count Frame Error Rate (which is actually Frame Error Rate plus Unreliable Frames Rate) Frame Erasure Count Using the GSM Bit Error Measurement for Circuit Switched Data (CSD) Bit Error Testing The GSM bit error measurement is also useful for CSD bit error testing. For more information on GSM CSD, see GSM Circuit Switched Data (CSD). You must start a CSD call before initiating the GSM BER measurement. P1031C4v10 HELIOS 39 of 39

40 Measurement Types for CSD Bit Error Since all the data in the CSD data payload is equally important and convolutionally encoded with same protection, the test set allows you to choose two new measurement types to ensure appropriate results are returned: Residual Data and Data. In addition, the frame-by-frame Fast Bit Error measurement is also applicable to CSD bit error testing. Operating Considerations GSM Bit Error Measurement is available when the operating mode is set to Active Cell, GSM BCH+TCH, GPRS BCH+PDTCH+TCH and EGPRS BCH+TCH+PDTCH. Carrying out the measurements The test equipment used is the Agilent E6568C W-CDMA/GSM/GPRS/EGPRS Wireless Test Manager for the 8960 Series 10 (E5515C) wireless communications test set. Figure 10: Agilent 8960 wireless communications test set The test setup used is equivalent to that used for the UMTS measurements (figure 9). P1031C4v10 HELIOS 40 of 39

41 Measuring the Bit Error Ratio This section describes the test procedure. This procedure assumes that the BCH parameter for Cell Band is set up correctly for your mobile station and that the appropriate RF IN/OUT Amplitude Offsets have been set to offset loss between the mobile station and the test set. To make bit error measurements your mobile must be fitted with a Test SIM card and have ETSI Type B connection capability. 1. Press the CALL SETUP key. 2. Press the More key (bottom right) to obtain the Call Parms (3 of 3) screen. 3. Press the Multislot Config ( F7 ) key and select 1 down, 1 up. 4. Press the Data Conn Type ( F4 ) key and select ETSI Type B. 5. Connect the mobile. 6. Turn the mobile on and wait for Attached in the Active Cell: field. For mobiles that don't perform GPRS attach automatically, set the mobile to data mode. 7. Press the Start Data Connection ( F3 ) key and watch for the Active Cell field changing to Transferring. 8. Press the Measurement selection key. 9. Select the Bit Error measurement. 10. Press the Bit Error Setup ( F1 ) key. 11. Set the measurement parameters as needed for your measurement situation, including: Measurement Timeout = 13.0 s A typical bit error ratio measurement result is shown above. P1031C4v10 HELIOS 41 of 39