Kordia™ Group Limited

Nordia™ Group Limited Submission to the Ministry of Economic Development, Kordia™ Submission on Discussion Paper - Radio Frequency Auction: 2.3 GHz and 2.5 GHz Bands

Introduction

Summary of points

Process

Question 1: Considerations Relating to Lot Design

Question 2: Proposed Lot Design

Question 3: Maori Interests

Question 4: Eligibility to Bid

Question 5: Duration of Acquisition Limits

Question 6: Use or Lose Provisions

Question 7: Settlement

Question 8: Managed Spectrum Parks: Proposed Allocation Procedures and Usage Rules

Question 9: Transition Plan for the 2.5 GHz Band

Annex 1 - Comparison between: two separate MSP blocks, and one aggregated MSP block

Annex 2 - Case against asymmetrical guard bands - Extract from Kordia™’s27 April 2007 submission on Radio Frequency Auction No.9: Overview

Introduction

1. Kordia™ Group Limited (Kordia™) thanks the Ministry of Economic Development (Ministry) for its invitation to provide comments on the issues raised in thdocument:
   
   

   Radio Frequency Auction: 2.3 GHz and 2.5 GHz Bands, Discussion Paper

   
   published in August 2007 by the Radio Spectrum Policy & Planning Group, Energy and Communications Branch of the Ministry (Discussion Paper).
2. Kordia™’s contact people for the purpose of this submission for commercial and policy matters and technical matters respectively are:
   

Susie Stone

Group Corporate Affairs Manager

Kordia™Group

PO Box 2495

AUCKLAND

Telephone: (09) 916 6513

Fax: (09) 916 6446

Email: susie.stone@kordia.co.nz

Ian Goodwin

Strategic Technology Manager

Kordia™Limited

PO Box 98

WELLINGTON

Telephone: (04) 914 8032

Fax: (04) 914 8094

Email: ian.goodwin@kordia.co.nz

Summary of points

• The Ministry must follow process, and involve industry in a review of the needs for the 2.5 GHz band, prior to making a decision to auction it.

• The Ministry should defer any auctioning of the 2.5 – 2.69 GHz band until it has consulted with the industry on the proposed uses of the 2.5GHz spectrum.

• Guard bands should be implicit in the rightholder’s responsibility under the Radiocommunications Act, not explicit in the form of denied bands. Guard bands are only one of many “tools” available to facilitate technical co-ordination.

• An exception to implicit guard bands is required adjacent to the 2.690 – 2.70 GHz Radio Astronomy band where an explicit guard band is recommended to protect the Radio Astronomy Service.

• There should be no more than one Managed Spectrum Park (MSP) in each band to minimise the number of inefficient boundaries between spectrum blocks.

• The preferred location of the MSP in the 2.3 GHz band is at the upper end of the band, adjacent to the ISM band.

• The preferred location of the MSP in the 2.5 GHz band is at the lower end of the band, adjacent to the ISM band.

• Users of MSPs should be advised by the Ministry, as Band Manager, of the requirement to co-ordinate with neighbouring spectrum users, in addition to coordination among themselves within their park.

• MSPs should be administered under three methods:

1. Rural areas - GUSL (i.e. no restricted entry)

2. Provincial centres - Private shared MSPs with limited-entry

3. Six main centres - Private shared MSPs, restricted to the nationwide BWA Band Managers

• Existing O-band users should be permitted to use the MSP for TV OB linking.

  • The Option B lot design is not preferred because it could result in inefficient spectrum usability.

  • All lots (including MSPs) should include AFELs to enable operation up to the edge of the block when practical.

  • Lot design should not be distorted to cater for CMAR.

  • Access to MSPs should be limited to Broadband Wireless Access (BWA) and, when congestion occurs in the OX band, for TV OB linking.

  • CMAR should not be permitted in the Managed Spectrum Parks.

  • Existing CMAR licences created in the last few years must not be given incumbent spectrum licences.

  • Simultaneous Ascending Auction (SAA) is preferred.

Process

Industry review of the best use of the 2.5 GHz band

3. Kordia™ is concerned with the lack of industry consultation over the best use of the 2.5GHz band. In the past the Ministry has consulted with the industry on proposed new spectrum uses and auctions. Kordia™ is particularly supportive of the use of discussion papers as a means to gather views and for the industry to be informed of other participants’ viewpoints.

4. To date, the only industry involvement in a review of the 2.5 GHz band has been under the guise of a BWA discussion paper, and a 2.3 GHz Auction Overview. The Ministry has not yet undertaken an industry discussion of the alternative uses for the “2.5 GHz band”, including an open analysis of the demands if any for the alternative uses. Kordia™ considers the Ministry should follow its past process and involve the industry in a considered review of the needs and uses of the 2.5 GHz band, prior to any decision to auction it.

5. Kordia™ is concerned that there is a requirement to clear and sell the band for BWA, when there has not yet been any industry involvement to determine the relative requirements of the three contending uses for the 2.5 – 2.69 GHz band:
   
   

   • TV OB and ENG linking - Existing use
   • IMT-2000 - Identified by WRC-2000
   • BWA - Potential application

    

6. The Ministry should therefore defer any auctioning of the 2.5 – 2.69 GHz band until after undertaking proper consultation with the industry.

7. At paragraph 11 of the Discussion Paper the Ministry notes the ITU World Radio Conference (WRC-2007) which is being held in November of this year. The upcoming WRC-2007 will discuss possible convergence of IMT-2000 and BWA services such as WiMAX, suggesting that after that time, the international framework should be resolved, and the development of technologies and services in other countries can be expected.

8. Kordia™ supports the decision to hold the auction after WRC-2007. Given that the Ministry is attending WRC 2007, and that the outcome of the WRC 2007 will be known by mid November, Kordia™ requests that the Ministry update the industry promptly as to the outcomes of the WRC with regard to the 2.3GHz and the 2.5 – 2.69 GHz bands.

9. Kordia™ notes that the timing of the proposed auction is very close to the end of WRC 2007 and Kordia™ has concerns about whether industry and the Ministry will in fact have sufficient time to properly consider the implications of any decisions made at WRC 2007 prior to the auction being held in December.

Question 1: Considerations Relating to Lot Design

Do you agree with the considerations discussed in this chapter concerning lot design? Please provide any comments under the following headings:

a. Expected technical developments and likely demand.

b. Size and location of the managed spectrum park or parks, including:

• whether two MSPs should be provided for in the 2.5 GHz band in order to allow for FDD (including CMAR) uses.

c. Acquisition limits.

d. Technical issues such as:

• accommodating TDD and FDD technologies;
• guard-bands;
• other.

e. Preferences regarding auction type.

f. Other.

Q 1.a Expected technical developments and likely demand

10. The WiMAX Forum has defined a more limited subset of the highly flexible IEEE 802.16 Wireless MAN standard, as the technology for a common industry standard for fixed, nomadic and mobile broadband access. Initial implementations of WiMAX are TDD based. However FTT based products are expected eventually. With Intel and other major industry players strongly backing WiMAX, it is hoped that the inevitable inclusion of WiMAX chips in most new PCs, will lead to a crucial lowering of the cost of user terminals for WiMAX, both PC based and stand-alone WiMAX CPE terminals.
11. New Zealand’s lagging position in OECD rankings for broadband service uptake is in part a result of the relative high price of competitive RF based broadband delivery. This relative high price is largely the result of the present high cost of suitable wireless CPE devices, and the sensitivity of the BWA business case to the cost of CPE terminals for licensed BWA bands.
12. As WiMAX is a widely supported open standards based technology, like WiFi, it is likely to accelerate the introduction of cost-competitive BWA terminals into the marketplace.

Q 1.b Size and location of Managed Spectrum Parks

MSP Size

13. The size of MSPs should be comparable to the TDD block size for nationwide lots.
    Hence 35 MHz is an appropriate size, as shown further below in Kordia™’s suggested
    Option C version 2.
    

MSP Number and Location

14. In the 2.3 GHz band, Kordia™ supports the Ministry’s proposed location of a single MSP at the upper end, where it is adjacent to the unlicensed ISM band.
15. Regarding the 2.5 GHz band, Kordia™ believes the Ministry’s proposed alternative options require some revision to optimise the MSP location.
16. Industry indications are that initially, WiMAX will be available for TDD, and that FDD may eventually follow. Providing for FDD operation in MSPs may unnecessarily complicate the band plan, and would reduce the amount of net useable spectrum. This is because of the greater number of unnecessary spectrum boundaries that would be created if the MSP spectrum was to be divided into two separated blocks. Hence in the 2.5 GHz band, there should be only one Managed Spectrum Park. See
    annex 1.
17. In the 2.5 GHz band, the location of the single MSP could potentially be at either the low, mid or upper end of the band, however each location has different advantages and disadvantages:
    
    

    Low

    The low end has the benefit of locating the MSP next to the unlicensed ISM band. This location would have only one boundary with private band managers and would
    thus lessen the amount of necessary co-ordination. Given the ad hoc nature of BWA operations in the ISM bands, situating the MSP adjacent to the ISM band
    would have obvious synergies, compared with locating a private managed band adjacent to the ISM band.
    
    

    Mid

    The middle of the band would be useful if the main Management Right lots are configured for FDD.
    
    

    Upper

    This location would have only one boundary with private band managers and would thus lessen the amount of necessary co-ordination.
    
18. If a TDD optimised plan such as one similar to Option 1 were to be adopted, thenlocating the MSP at the low end would seem most efficient.
19. If a plan to accommodate both TDD and FDD is adopted, such as one similar to Option 3, then a MID location for the MSP would have some merit, where it could
    occupy the mid-band-gap. However, considering the need to cater for some TDD, in such a plan, it would be better to locate nationwide TDD block(s) in the mid band,
    and keep the MSP at the low end of the band, where it would have a better operational interface with the ISM than would a private managed band.
    
20. Hence Kordia™ prefers Version 2 of Option-C as illustrated further below, with the MSP at the low end of the 2.5 GHz band.

No CMAR in the MSPs

21. The Discussion Paper asks whether MSPs should be designed to accommodate CMAR in the 2.5 GHz band.
22. The answer to that question is no.
23. In the 1.5 GHz “L-band”, there are almost 2000 CMAR links licensed to Telecom throughout rural New Zealand, and a further 185 links licensed to other organisations. L-band has the benefit of good propagation where microwave paths are not always ideal, and economically priced CMAR equipment is available. Although there are a large number of links in L-band, the geographical isolation of their rural locations enables good spectrum re-use. Alternative spectrum is not necessary for CMAR.
    
24. Telecom has owned eight of the 8 MHz blocks in the 2.3 GHz Management Rights for 17 years. Until recently they have been unused. The only licences registered in the 2.3 – 2.396 MHz band are for 17 CMAR links in the rural area inland from Kuranui near Waverly. These did not exist prior to April last year. Before that date L-band links carried the CMAR traffic over these 17 links. With the remaining 1900 links operating throughout the rest of rural New Zealand in L-band, Kordia™ does not accept that the very recent registration of these few 2.3 GHz links by Telecom is because the spectrum is required for CMAR. Rather it appears to be a late attempt to use the vacant 2.3 GHz spectrum to now justify renewal of those previously unused management rights.

Q 1.c Acquisition limits

25. Kordia™ supports the use of acquisition limits for the combined 2.3 and 2.5 GHz Auction. We agree that 35 MHz is appropriate for a DTT based spectrum block, and that 20 MHz plus 20 MHz is appropriate for a pair of FDD based spectrum blocks.
26. See Kordia™’s response to Question 8.d below, regarding the exclusion of metropolitan MSP spectrum when determining compliance with the acquisition limit.
27. Kordia™ recommends that the acquisition limits should apply only to the spectrum bands included in the auction.

Q 1.d Technical issues: accommodating TDD and FDD; guard-bands; etc

More about technical compatibility, guard bands and Managed Spectrum Parks

28. The Discussion Paper does not adequately address the need for co-ordination between operators using the MSPs and neighbouring band managers.
29. Paragraphs 48, 50 and 117 of the Discussion Paper touch on the need for MSP operators to co-ordinate among themselves, and although paragraph 50 refers to the inclusion of a 5 MHz guard band, this is not mentioned again in the Discussion Paper, and does not appear in the alternative spectrum plans.
30. Kordia™ is concerned that in the Discussion Paper, the emphasis of co-ordination for MSP operators, is co-ordination among themselves. We see co-ordination between users of adjacent bands as equally important. The practicality of co-ordinating between the manager of a “normal” band, and the multiple users of an MSP band is likely to be difficult, and time consuming.
31. The licence conditions for MSP licences should address the need for co-ordination, by requiring users of the MSP licence to not only “enter into cooperative process to protect the quality of service for the various users of the park” as suggested in the discussion paper, but should require MSP users to also jointly or individually enter into co-ordination negotiations with neighbouring band managers to ensure technical compatibility and to avoid harmful interference – in either direction.

Q 1.e Preferences regarding auction type

32. Kordia™ submits that its preference is for a Simultaneous Ascending Auction (SAA).
33. Kordia™ also recommends that if an open outcry auction process is to be used, then each lot should be auctioned separately, and bidding should be available to all bidders, except bidders who have already reached their maximum allowed spectrum.

Question 2: Proposed Lot Design

Do you prefer Option A, B or C for the lot design? Please explain why. If you prefer Option C is your preference for C(i) or C(ii)?

Why?

If there is an alternative option you prefer please specify it and explain why it would be

preferable.

Single MSP - at the lower end of the 2.5 GHz band

34. In our answer to question 1.b, we have recommended that any MSP spectrum in a band should be:
    
    
    1. aggregated into one block in order to minimise the amount of necessary co-ordination with neighbouring right holders, and
    2. to be situated adjacent to the ISM band. (At the upper end of the 2.3 GHz band, and at the lower end of the 2.5 GHz band).

35. None of the three options in the discussion paper quite provide for these requirements.
    
36. Kordia™ does not support Option-B because any bidder left with non-adjacent lots at the close of the auction would have significantly reduced amount of available
    spectrum after any necessary guard band allowances are taken into account. Option-B is likely to result in inefficient spectrum usability.
    

37. With respect to the remaining options for the lot design, Kordia™’s preference and alternatives are shown below. Further explanation is set out in an annexure to these submissions.
    

Lot Design - No explicit guard bands

38. An important part of lot design is the need for technical co-ordination and compatibility with licences created in neighbouring management rights. The Discussion Paper emphasises the need for guard bands to manage co-ordination. Kordia™’s experience is that there are a number of methods that can provide for or contribute to a combined solution for compatibility. These include:
    
    • antenna directivity
    • polarisation discrimination
    • geographical separation
    • co-location of base stations
    • co-operative channel planning, and
    • isochronous operation of neighbouring TDD services, etc.

39. Kordia™ recommends that the need to leave unoccupied guard bands should not be an explicit condition of a management right.
    
40. Kordia™ recommends that the Auction Rules should include a clear reminder to prospective Management Right owners of their responsibility to ensure the technical compatibility of any licences they register. They need to be compatible with other licences, whether registered or planned, as required by the Radiocommunications Act 1989 and the international Radio Regulations (Article 3.3).
41. In responding to the Ministry’s previous discussion paper (27 April 2007), Kordia™ expressed concern about the proposal to require guard bands on one side of a
    boundary (paragraphs 44 – 46 of our submission.) Consequently Kordia™ is concerned to see reference to “the need for a 5 MHz guard band” in various parts of
    the Discussion Paper. The language of paragraph 55 is particularly worrying. It includes such references to explicit guard bands as: “the burden of co-ordination, possibly within certain limits, would be borne by the guard band owner.”
    
42. Kordia™ submits that the responsibility to create any necessary guard bands in particular geographic areas would generally be the responsibility of both of the
    neighbouring band managers, and such guard bands should generally be symmetrical about the Management Right boundary.

No explicit asymmetrical guard bands

43. Kordia™ agrees with the sentiment in Option-A which states that:
    
    Bidders would be required to manage interference issue within the lots they acquire. They could do this by providing for guardbands within the lots they acquire and/or by agreement with neighbouring rights holders.
    
44. Kordia™ disagrees with the proposal in Option-C that:
    
    “Guardbands of 5 MHz would be specified on the upper bound of each management right…”
    
    
45. Kordia™ refers the Ministry to paragraphs 43-46 of its previous submission of 27 April 2007 on the Auction No.9 Overview, as copied in Annex 2 to this submission.
46. Kordia™ submits that the responsibility to create any necessary guard bands in particular geographic areas would generally be the responsibility of both of the neighbouring band managers

Explicit Guard band below 2.690 GHz

47. Although paragraph 24 of the Discussion paper mentions the need for a guard band to protect the Radio Astronomy allocation in the band 2690 – 2700 MHz, there is no other reference to this need in the remainder of the Discussion Paper. Its implementation is not something that should be left to the MSP users themselves. This guard band, as an exception, should be part of the frequency plan.
48. Looking ahead to New Zealand’s likely involvement in the Square Kilometre Array, it would also be helpful to Radio Astronomy to include a caution on the conditions of the Management Right below 2690 MHz, to pay special attention to out-of-band emissions falling above 2690 MHz, to ensure these are as low as possible, and below
    the -50 dBW/MHz eirp power floor implicit in the Act. (Note that Radio Astronomy works with very small “Jansky” pfd units which are -200 dBW/MHz/m2 and hence -
    50 dBW/MHz eirp emission levels are comparatively very large.)

Adjacent Frequency Emission Limits for Management rights

49. In the above discussion under “lot design – no explicit guard bands” Kordia™ identifies a number of technical methods for achieving technical compatibility between services operating in adjacent management rights. These techniques will often enable one or other operator to use channels right up to the MR boundary. To facilitate this efficient use of spectrum, Kordia™ recommends that:
    
    

    The Ministry should incorporate practical AFEL profiles on all lots, and on MSP blocks, sufficient to accommodate WiMAX transmissions in edge channels. (Note however the need to protect the Radio Astronomy service at 2690 MHz.)

Question 3: Maori Interests

a. What provision should be made regarding allocation of spectrum in the 2.3 GHz and 2.5 GHz bands for use by Maori (i.e. Maori service providers)?

b. What provision should be made regarding allocation of spectrum in the 2.3 GHz and 2.5 GHz bands for use for Maori (i.e. Maori as service consumers, or in the interests of Maori language and culture)?

c. What terms and conditions should apply to this spectrum?

Q 3.a Allocation of spectrum

50. Kordia™ believes that the auction should be commercially neutral to all potential users of the spectrum, in the same way that the Ministry has a philosophy of technology neutrality.

Q 3.b Maori as service consumers, or in the interests of Maori language and culture

51. Kordia™ recognises the need to encourage Maori language and culture and believes that where radiocommunications systems such as BWA can help facilitate this, any government or other support should be directed at the services provided by such radiocommunications systems, as opposed to diverting the spectrum resource itself.

Q 3.c Terms and conditions

52. Terms and conditions such as eligibility to enter the auction should be impartial to ethnicity or race.

Question 4: Eligibility to Bid

53. Kordia™ agrees that there should be no restrictions on the eligibility of parties to bid for lots. However the requirements of the Commerce Act 1986 still need to be met, following the completion of the auction.
    
54. Kordia™ notes that the “use or lose” conditions being proposed for the spectrum lots are intended to deter speculators. 
55. Under question 8, Kordia™ has recommendations concerning eligibility to use MSP spectrum.

Question 5: Duration of Acquisition Limits

a. Do you consider the expiry of acquisition limits 1 year before the (December) use or lose date to be satisfactory?

b. Do you prefer a different acquisition limit date? Please explain reasons for your view?

c. Do you have any other comments on the time for acquisition limits to expire?

Q 5.a Expiry of acquisition limits

56. Kordia™ supports the proposal in paragraph 91 of the Discussion Paper to set an expiry date for the acquisition limit, one year prior to the “use or lose” date. This would allow a band manager who has been unable to meet the “use” conditions to trade out or sell out of their commercial dilemma.

Q 5.b Do you prefer a different acquisition limit date?

57. No.

Q 5.c Do you have any other comments on the time for

acquisition limits to expire?

58. No.

Question 6: Use or Lose Provisions

Do you prefer a date of December 2012, 2014, or 2016, for applying the use or lose test? If not what alternative date would you prefer to implement the Cabinet decisions on use or lose? Please explain.

Do you prefer Option A or Option B or some other option regarding the test for ‘use’? If you do not agree with the proposals, what ‘use or lose’ provisions do you propose? Should financial consequences result from not using rights, either in addition or as an alternative to loss?

How should the amount of any financial consequence be calculated?

General Concerns

59. Kordia™ repeats its concern that the Auction Rules are not the appropriate legal mechanism for introducing a “use or lose” regime.

60. Kordia™ notes that the Australian position provides greater administrative certainty. In Australia, only spectrum licences not management rights are issued. Under Division 6 of Part 3.2 of Chapter 3 of the Radiocommunications Act 1992 the Australian Communications and Media Authority has express statutory powers to compulsorily resume spectrum licences and pay compensation Kordia™ urges the Ministry to refrain from compulsorily resuming management rights in the absence of similar legislative mechanisms.

Q 6. December 2012, 2014, or 2016, “use or lose” test?

61. Kordia™ prefers the date of December 2016 for the use or lose test.
    
62. December 2016 will give band managers a reasonable time to develop a viable business and to meet the use or lose criteria, following the expected availability of WiMAX or other suitable technologies at cost effective prices.

Q 6. Option A or Option B for the “use” test?

63.  Kordia™ prefers Option-A for the “use” test.

Q 6. Financial consequences of not using rights?

64. Kordia™ strongly recommends that no further financial consequences should be imposed on band managers who fail to meet the “use” test, other than their obligation to either:
    
    
    a) comply with the “use” criteria within a reasonable time, or
    b) sell their management right on the open market.
    
65. The requirement to sell should allow the manager reasonable opportunity to sell at a fair market price.
66. No further financial consequences should be imposed, as it is sufficient penalty in itself for a band manager to have invested in the management rights, and in the infrastructure and other business costs to develop a BWA service, that then failed to achieve the necessary “use” threshold, and for the manager to then to have to sell the management right.
67. In a case of flagrant anti-competitive behaviour, penalties should be imposed under currently available legislation, such as the Commerce Act 1986.

Question 7: Settlement

Do you agree with the proposed settlement terms (30 days following completion of the auction)?

If not, what other factors do you see as relevant?

30-days settlement terms

68. Kordia™ accepts that normal commercial 30-days settlement terms are appropriate for the spectrum auction.

Question 8: Managed Spectrum Parks: Proposed Allocation Procedures and Usage Rules

a) Do you agree with the suggested eligibility criteria for access to a MSP?

b) Do you agree with the suggested core technical and usage requirement?

c) Of the three options outlined above, which is your preferred method for implementing a MSP? Why?

d) Are there better alternatives or variations on these implementation options?

e) What incentives for gaming arise under the various options, and what measures, if any, could be taken to minimise such incentives?

f) What fees or resource charges should be levied (particularly for your preferred option)?

Q 8.a. Eligibility criteria for access to a MSP

69. Kordia™, with the exception of the special case when there is congestion in the OX band, is against special provision being made to allow non-BWA services to have access to an MSP. In particular, Kordia™ objects strongly to CMAR being permitted to use the MSPs.
70. Please refer to Kordia™’s response above to Question 1.b regarding “denying CMAR access to MSP bands”.

Q 8.b Core technical and usage requirement

71. Kordia™ agrees in principle, but notes a significant omission from the technical coordination proposals discussed in paragraph 117 of the Discussion Paper.
72. In addition to a requirement to employ interference management techniques, and to co-ordinate among themselves, MSP users must also be required to co-ordinate with licence existing and planned holders in neighbouring spectrum. If the Crown as band manager for the MSPs fails to pass on that requirement to MSP users, then the Crown itself or the Crown’s approved radio engineer who certifies licences within an MSP, may be responsible for the consequences of any harmful interference between systems in the MSP and in neighbouring bands.

Q 8.c Preferred method for implementing a MSP

73. See Kordia™’s response to Question 1.b above on the preferred location of the MSP within each band, and our response to Question 8d below.
    

Option-Two preferred

74. Kordia™ considers that Option-Two is the preferred method for administering MSP spectrum in provincial areas, (with the proviso as stated elsewhere in this submission, that CMAR should not be permitted.)
    
75. Kordia™ agrees with Option-Two’s provision, inter alia, for:

• A maximum number of users – to avoid excessive interference and congestion.
• Area based licences - such as TLA based areas.
  
• Clearly documented technical usage rules.
• Registration of base station transmit sites in the publicly accessible Radio Register.
• 2-yearly statutory declaration of use.
• 6-yearly renewal, 2-years prior to expiry.
• An application fee to cover MED’s engineering costs for determining permitted number of users.

Application fee

76. Kordia™ considers it only fair to charge successful applicants for access to a provincial or metropolitan MSP. An application fee could be charged to all applicants provided that it was refundable to unsuccessful applicants.

Resource fee

77. Kordia™ supports the principle of the Ministry charging all successful applicants a resource fee. The level of fee for access to each area MSP should be based on the mean auction price of nationwide management rights, scaled pro rata: per population within the area, per the bandwidth of the MSP, and per the number of service providers permitted in the particular area.

Q 8.d A better alternative implementation option

78. The MSP concept is a relatively new and undeveloped method for efficiently administering the spectrum. It is expected to be useful in situations midway between those for which licensed and unlicensed spectrum management are appropriate.
79. Kordia™ believes that the demand for spectrum and the other factors mentioned above vary significantly in different areas: rural, provincial and metropolitan.

Rural areas - GUSL

80. In rural areas, demand for spectrum is light, and the prospect for unmanageable interference and congestion are low. Accordingly we believe in rural areas, the “MSP” spectrum would best be administered under a General User Spectrum Licence (GUSL).
81. Use of the GUSL spectrum in rural areas should be restricted to allow only BWA service providers who are not owners of nationwide lots in the 2.3 and 2.5 GHz bands.
82. The conditions for the use of each rural GUSL should specifically exclude use for CMAR systems and contain conditions with regard to the potential use for live vide contribution links.

Provincial areas – MSP

83. In provincial areas, demand for spectrum to satisfy local enterprise BWA services, is likely to be higher than in rural areas, but is likely to be well matched to providing the necessary spectrum access through a shared licensed regime, such as the MSP concept. Kordia™ therefore supports the use of MSP licensing for provincial areas. Provincial areas would include all cities and towns outside the six main metropolitan cities, Auckland, Hamilton, Tauranga, Wellington, Christchurch and Dunedin.
84. In addition conditions should apply to the MSP with regard to the potential use for live video contribution links.

Metropolitan areas – MSP access for nationwide BWA operators only

85. MSP spectrum in the metropolitan areas should only be accessible to nationwide BWA service providers in order to alleviate their spectrum congestion in the main centres.
86. Limiting access to the MSP spectrum for local enterprise BWA operators, to provincial and rural areas only, would also avoid users of “cheap” MSP spectrum from undermining in metropolitan markets, the commercial viability of the “nationwide” service providers who have paid significantly more for their management rights.
87. In metropolitan areas, traffic demands on BWA services will be higher than in provincial and rural areas. Nationwide service providers who may find that their spectrum is adequate for the traffic demands of rural and provincial areas, will find their spectrum much more congested in metropolitan areas. That congestion would be best matched by providing for the metropolitan overflow into the MSP spectrum. Naturally “nationwide” operators wishing to access that spectrum would have to pay for their MSP area licence on the same basis that the resource fee if calculated for provincial MSPs in the remainder of the country.
88. Access to metropolitan MSP spectrum by “nationwide” operators, should not be included when determining their compliance with the acquisition limit.
89. In addition conditions should apply to the MSP with regard to the potential use for live vide contribution links.

Q 8.e Incentives for gaming and counter measures

90. Under Option-Two, Kordia™ sees no obvious gaming opportunities, given the requirement to co-ordinate with the spectrum users.

Q 8.f Fees and resource charges

91. Application fees and resource fees are discussed in our response to Question 8.c above.

Question 9: Transition Plan for the 2.5 GHz Band

Do you have any comments on the proposed transition plan for existing licences in the 2.5 GHz band?

TV OB Operation to be permitted in the MSPs

92. If the Ministry proceeds with the 2.5 GHz auction prior to involving the industry in a review specific to the 2.5 – 2.69 GHz band (O-band), then the existing O band licensees should be permitted to operate within the MSPs in that band.
93. In providing for the operation of TV OB linking in the MSPs, the Ministry should ensure that the bandwidth and AFELs can accommodate the link emissions. There is not time to provide the necessary AFEL profiles with this submission. We invite the Ministry to liaise with the TV OB industry to determine these parameters.
94. Major events such as the 2011 Rugby World cup highlight the desperate shortage of TV OB linking spectrum in New Zealand. Providing access to suitable spectrum for this event should not be restricted to use until 2011, as demands for TV OB linking will continue for other events in the future.
95. Kordia™ proposes that all users in the OX-band be limited to digital equipment. Furthermore, Kordia™ proposes that the Ministry should reserve the right to limit the bandwidth individual links use in the OX-band should congestion arise.

Portable Use

96. Kordia™ proposes a modified approach to the license cancellation to mitigate OX band congestion and provide flexibility for the 2011 Rugby World Cup.
97. Although equipment is available in the OX-band and V-band as the Ministry states, the situation is more complicated than outlined due to economies of scale and logistics issues. These issues mean an operator needs to choose one band or another. This is best explained by explaining how Kordia™ will approach the situation.
98. Kordia™ provides existing services by connecting portable point to point radios through many fixed dishes and feeder infrastructure. Some, but not all, installed infrastructure is suitable for OX-band so a programme of upgrading to OX-band will be required. In addition Kordia™ provides portable point to point links over long distances which V-band is unsuitable for. Kordia™ will therefore have 90% of its equipment in the OX-band.
99. Other operators have existing equipment and infrastructure in the O-band, some of which will be upgradeable to OX-band, but not V-band. They will more than likely to the same approach as Kordia™ and have the majority of their equipment in the OXband.
100. We therefore see congestion developing in the OX-band especially with events such as the 2011 Rugby World Cup.
101. Kordia™ has experience with the use of 2.4GHz ISM band links and has found that the reliability of ISM band links is unpredictable. This is not acceptable for providing live video contribution links for broadcasters.
102. Kordia™ proposes that the band license be only partially cancelled in December 2008. Kordia™ proposes instead that the band be made available to portable point to point links in the situation where OX band congestion occurs. The basis for use would be:

103. This approach provides a sensible mitigation in the event of OX-band congestion, and would provide potential for flexibility for the 2011 Rugby World Cup and would maximise the use of spectrum. equipment in the O-band.

Camera Mounted Equipment

104. Kordia™ considers that the Ministry’s proposed date of December 2010 is too early and proposes an alternative date of December 2011 to ensure the Rugby World Cup coverage can be achieved with as much frequency flexibility as possible.
     
105. Kordia™ notes that it is possible the rugby authorities will require sidelines for the bigger matches to be free of cables, as was the case with the 2005 Lions tour. This will necessitate the use of high definition wireless camera links. In addition to the local broadcasters, there will also be international broadcasters wanting wireless camera links for interviews. Kordia™ is therefore concerned that restricting use to the OX band frequencies will lead to interference.
106. Kordia™ also observes that mobile WiMax services may appear before the TVOB services may moved out of the O-band. Mobile WiMax devices operating in this band could appear in sports stadiums in an uncontrolled way so there is the strong possibility that interference caused by consumer devices may make the O-band frequencies unusable. Testing will be required to see if camera mounted equipment links will operate satisfactorily in the presence of consumer equipment. Kordia™ again notes that the lack of a formal public review of the requirements for the uses of the 2.5GHz spectrum, including its current uses, may result in some significant issues being missed by the Ministry.
107. Kordia™ also notes that 2.4GHz ISM band is unlikely to be of use for high quality broadcast wireless camera links due to interference within sports stadiums but may be workable in television studios.
     
108. Kordia™ also proposes that the Ministry reserves the right to impose bandwidth use limitations on individual links. This would provide another way for mitigation of congestion in the OX-band.

Helicopter Based Use

109. Helicopter links are the most problematic issue technically because frequency reuse is unlikely to be possible across a large area.
     
110. Use of the 2.4 GHz ISM band is not considered a feasible option for the use of helicopter links because interference issues will not provide a sufficiently robust signal for broadcast quality video contribution.
111. Kordia™ would use the MSP for helicopter links until other users started causing quality issues.
     
112. The 2011 Rugby World Cup will see requirements for portable point to point links, camera links and helicopter links. We envisage all requirements will be for high definition services.
113. Kordia™ is concerned that OX band congestion will mean that not all requirements can be met in that band.
     
114. Kordia™ proposes that use of the 35MHz 2.5GHz MSP for WiMax be delayed as much as possible. One method to achieve this is for all other MSP spectrum to be used first. Ideally the delay would extend until December 2011.

Annex 1 - Comparison between: two separate MSP blocks, and one aggregated MSP block

1. All three Lot design options in the Ministry’s Discussion Paper have two MSP blocks in the 2.5 GHz band. These options all have a total of three boundaries between the MSP block and neighbouring management rights. Option C illustrates this in the figure below.

2. Kordia™’s version 2 of Option C, shown below aggregates the MSP spectrum into a single block at the low end of the band, adjacent to the ISM band at 2500 MHz. Consequently, the MSP block has only one boundary with neighbouring management rights. This is important because technical co-ordination between a single band manager and multiple users of an MSP is likely to be difficult, due to the complexity of interactions between one band manager, and multiple operators in their shared band.

3. This lot design provides a guard band for radio astronomy above 2690 MHz, and has a mix of FDD and TDD blocks sufficient for four “nationwide management right” operators.

Annex 2 - Case against asymmetrical guard bands - Extract from Kordia™’s27 April 2007 submission on Radio Frequency Auction No.9: Overview

1. We also note the proposal for 5MHz guard band on the lower boundary when necessary from within their 30MHz. We understand why this method of requiring the guard band inside the lower boundary avoids management rights out of step with the underlying 5MHz raster, however we suggest that further study be conducted into the supposed raster constraint, in order to investigate a more balanced guard band philosophy.

2. Kordia™ cautions that the proposed one-sided responsibility for providing the guard band might lead to serious tensions at the boundaries between adjacent operators.

3. To illustrate, at a boundary where potential interference is expected, a lower band manager might argue for the upper band manager to use their edge spectrum as a guard band, and cancel their licences in the area, while the upper band manager may see other technical measures that the lower band manager or both managers might implement as more economic and preferable.

4. The WiMAX Forum™ Mobile System Profile (Release 1.0 Approved Specification – November 2006) in Tables 5 and 128 indicates that 250 kHz channel step size is recommended for the 2.3 GHz band. This negates a need for the asymmetrical guard bands proposed in the Overview.