Exhibit 99

CASE REF: BVJA/BV10066/JUL26

CASE REF: BVJA/BV10066/JUL26

Rooms 1101-4, 11/F Harcourt House,
39 Gloucester Road, Wan Chai, Hong Kong
Tel (852) 2529 6878 Fax (852) 2529 6806
E-mail info@romagroup.com
http:// www.romagroup.com

20 August 2026

Astrum Space Inc.

4^th Floor, Harbour Place,

103 South Church Street,

P.O. Box 10240 Grand Cayman KY1-1002,

George Town,

Cayman Island

Case Ref: BVJA/BV10066/JUL26

Dear Sir/Madam,

In accordance with the instructions from Astrum Space Inc. (hereinafter referred to as the “Company”), we have conducted the valuation of 25 MHz L Band Spectrum (hereinafter referred to as “Subject Spectrum Assets”). We are pleased to report that we have made relevant enquiries and obtained other information which we considered relevant for the purpose of providing our valuation as at 31 July 2026 (hereinafter referred to as the “Date of Valuation”).

This report states the purpose of valuation, scope of work, overview of the asset, basis of valuation, investigation, valuation methodology, major assumptions, information reviewed, limiting conditions and remarks, and presents our opinion of value.

This report presents an independent valuation of the Subject Spectrum Assets as of 31 July 2026, undertaken for the specific purpose stated herein. Based on the rigorous application of the Market Approach, benchmarked against comparable primary and secondary market spectrum transactions normalized on a price-per-MHz-POP basis, an adjusted implied valuation multiple range of USD 0.524 to USD 0.873 per MHz-POP has been derived. Upon scaling this baseline multiple range to each target jurisdiction via relative GDP per capita and multiplying by the 25 MHz assigned bandwidth and the respective covered populations, the Fair Value range of the Subject Spectrum Assets is concluded to be USD 2,814,000,000 to USD 4,532,000,000 as of the Date of Valuation. The underlying quantitative analysis and supporting empirical evidence are detailed in the subsequent sections of this report.

This report is prepared solely for the use of the directors and management of the Company. In addition, Roma Appraisals Limited (hereinafter referred to as “Roma Appraisals”) acknowledges that this report shall will be used by the Client in investor presentations, public documents, and regulatory filings, including submissions to the U.S. Securities and Exchange Commission (SEC).

Roma Appraisals assumes no responsibility whatsoever to any person other than the Company in respect of, or arising out of, the contents of this report. If others choose to rely in any way on the contents of this report, they do so entirely at their own risk.

Our valuation conclusion is based on the assumptions stated herein and the information provided by the management of the Company and/or their representative(s) (together referred to as the “Management”).

In preparing this report, we have had discussions with the Management in relation to the licensing status, network deployment plans, technical characteristics, and other relevant information concerning the Subject Spectrum Assets. In arriving at our opinion of value, we have relied on the completeness, accuracy and representation of regulatory, technical, operational, financial and other pertinent data and information of the Subject Spectrum Assets — including but not limited to license grants, buildout and coverage status, encumbrances, lease or sharing arrangements, and population coverage (POPs) — as provided by the Management to a considerable extent.

We have no reason to believe that any material facts have been withheld from us. However, we do not warrant that our investigations have revealed all of the matters which an audit, regulatory review, or more extensive technical examination might disclose. In case of any change in the assumptions stated herein, including but not limited to changes in licensing status, network deployment plans, or comparable transaction data, our opinion of value may vary materially.

The L-band is the IEEE-designated segment of the radio spectrum running from 1 GHz to 2 GHz (wavelengths of roughly 15–30 cm), sitting at the boundary between upper UHF and lower microwave frequencies. It occupies a critical middle ground in propagation physics — better range and building/foliage penetration than higher microwave bands (C, Ku, Ka), but more bandwidth capacity than lower VHF/UHF bands — which is why it has become the workhorse band for satellite navigation, mobile satellite services (MSS), and increasingly terrestrial mobile broadband.

L-band signals exhibit several properties that drive its commercial and technical value:

The 1–2 GHz range is finely segmented by ITU Radio Regulations and national regulators to separate satellite, GNSS, aeronautical, and terrestrial mobile uses and avoid interference:

Historically, L-band capacity in the 1525–1559/1626.5–1660.5 MHz segments has been licensed to Inmarsat and Ligado, with Inmarsat operating Broadband Global Area Network (BGAN) and maritime/aeronautical services under Mobile Satellite Service authority, often paired with Ancillary Terrestrial Component (ATC) rights that let the same spectrum support hybrid satellite-terrestrial networks. This ATC framework is precisely what has made portions of L-band contentious and valuable: Ligado’s proposed terrestrial 5G network in this band drew sustained opposition from GPS receiver manufacturers, DOD, FAA, and NOAA, who argued adjacent-channel interference could degrade GPS civilian signals in the 1559–1610 MHz band — a dispute documented extensively by the U.S. Congressional Research Service.

In Europe, CEPT/ECC has harmonized 1452–1492 MHz specifically for Mobile/Fixed Communications Networks Supplemental Downlink (MFCN SDL), letting LTE-Advanced and later networks aggregate this spectrum onto existing base stations to boost downlink capacity without new site builds.

The Subject Spectrum Assets comprises geostationary satellite spectrum rights held under three complementary International Telecommunication Union (ITU) space network filings — NEASTAR-1 (Germany), ASIABSS (Australia), and NYASIABSS (Papua New Guinea) — all coordinated at the 105° East orbital position and collectively operated under the Astrum Networks group. Each filing is administered through a different sovereign ITU member state, which provides the subject asset with layered regulatory coverage and geographic redundancy across three separate national spectrum-management regimes.

The core frequency asset common to all three filings is the 1,467–1,492 MHz L-band segment (space-to-Earth / downlink), supplemented by additional assigned bands used for feeder links, uplink command paths, and broadcast-satellite service (BSS) delivery, as detailed in Section 3. The filings support satellite broadcasting and related space-station services with associated ground infrastructure licensed separately in Australia.

The 1,467–1,492 MHz L-band segment is common to all three filings and represents the principal downlink capacity of the subject asset. Each filing additionally carries supplementary bands used for uplink command/control and, in the case of NEASTAR-1 and NYASIABSS, wideband feeder-link or C-band capacity.

NEASTAR-1 carries the widest frequency footprint of the three filings (12 total emission/reception bands spanning L-, S-, X-, and Ku-band segments), consistent with a multi-mission satellite payload. ASIABSS is confined to a single emission band (1,467–1,492 MHz) and a single reception band (7,025–7,075 MHz), reflecting its narrower, more mature service definition. NYASIABSS adds a 3,400–4,200 MHz (C-band) emission segment alongside the shared L-band core.

All three filings share the same geostationary orbital position — 105° East — with tight station-keeping tolerances (0.1° east/west longitudinal tolerance for each filing) and inclination excursion limits of 1° (NEASTAR-1) to 5° (ASIABSS, NYASIABSS). This shared orbital slot enables the three filings to function as a coordinated network rather than as independent, competing assignments.

Service areas and beam coverage differ by filing and are defined through named beams in the ITU filings:

Taken together, the beam patterns indicate primary service coverage across the Asia-Pacific region, encompassing East, South, and Southeast Asia, Australia, and the Pacific Islands, with certain beams extending to near-global reach subject to specific country exclusions.

The Subject Spectrum Assets is a geographically layered, multi-jurisdiction satellite spectrum position anchored at the 105° East geostationary orbital slot, built around a common 1,467–1,492 MHz L-band downlink core supplemented by wider auxiliary bands under the NEASTAR-1 (Germany) and NYASIABSS (Papua New Guinea) filings.

The Asia Pacific region excluding China remains the fastest-expanding major economic bloc globally in 2026, though growth is decelerating from a strong 2025 base as US tariff escalation and an energy price shock tied to the Middle East conflict feed through supply chains and consumer prices. This regional backdrop directly frames the demand environment for the subject spectrum asset, whose beam coverage spans Australia, Papua New Guinea, ASEAN, and the wider Pacific Islands.

Forecasts vary modestly by institution, but the direction is consistent: 2026 growth across developing Asia and the Pacific is moderating to roughly 4.9–5.1% under an “early stabilization” scenario for the Middle East conflict, down from 5.5% in 2025, with the Asian Development Bank (ADB) citing higher energy prices, normalizing export growth after 2025’s tariff front-loading, and solid domestic demand in South and Southeast Asia as offsetting factors. Regional inflation is projected to rise to 3.6% in 2026 after easing through 2025. The IMF’s comparable estimate sits at 4.1% for 2026, while the ASEAN+3 Macroeconomic Research Office (AMRO) projects a more moderate 4.0% for both 2026 and 2027 as higher US tariffs weigh on external demand, with headline inflation still contained at 1.4%.

ASEAN remains the region’s principal growth engine but with material dispersion across members. The Mastercard Economics Institute puts Indonesia (5.0%) and the Philippines (5.6%) among the fastest-growing 2026 economies, while Singapore (2.2%) and Thailand (1.8%) are normalizing toward trend. UOB revised its 2026 ASEAN GDP forecast upward to 4.0% (from 3.2%) on sustained AI-related tailwinds, while DBS and Maybank flag inflation as the key 2026 “wild card,” projecting ASEAN-6 headline inflation climbing to 2.2–2.7% from a multi-year low of 1.6% in 2025, with the largest upward revisions in Thailand, the Philippines, and Indonesia. Currency dynamics remain broadly supportive: the Singapore dollar has stayed resilient on MAS pre-emptive tightening, and ING expects most Asian central banks to avoid breaching inflation targets even as some (Korea, Indonesia, the Philippines) continue cutting rates.

Trade policy is the dominant downside risk. US tariffs on Chinese goods exceeding 40% in 2026 have triggered a “great reshuffle” of Southeast Asian supply chains, with Vietnam, Malaysia, Thailand, and Indonesia gaining FDI and manufacturing relocation but facing transshipment scrutiny and an estimated 5–10% cost inflation in affected sectors; one economy anticipates GDP growth reduced by 0.76 percentage points directly from tariff effects. AMRO similarly attributes its growth moderation to “higher US tariffs weighing on external demand.”

Japan’s growth remains structurally weak, with real GDP projected at just 0.9–1.0% in 2026 (up from an estimated 1.2% in 2025), while core inflation excluding fresh food and energy stays persistently above the Bank of Japan’s 2% target, keeping a 25bp rate hike in play. South Korea presents a markedly stronger and improving picture: the government sharply revised its 2026 growth outlook up to 3.0% (from an initial 2.0% estimate) on an “unprecedented semiconductor export boom,” with nominal GDP expected to rise 12.3%, though this comes with inflation revised up to 2.6% and a weaker won environment.

Australia — home to the ASIABSS filing’s ground segment and the ASEAN/East Timor broadcast licence held by AsiaSpace Limited — is experiencing a below-trend recovery. The OECD projects 1.9% GDP growth in 2026 with headline inflation pushed above 4% by an energy price spike, while Deloitte Access Economics has cut its 2026-27 forecast to 1.3%, describing the longest run of sub-2% growth since the early 1990s. Telecommunications and information media is a relative bright spot, with gross value added growth of 3.1–4.1% forecast for FY2025-26 — a favourable signal for continued satellite broadcast and earth-station demand.

Papua New Guinea (PNG), sponsor of the NYASIABSS filing, is projected to grow 3.8–4.6% in 2026, moderating from 5.5–5.6% in 2025 as LNG output plateaus and non-resource sectors (agriculture, services) take over as the primary growth driver, with inflation rising toward 4.8%. Across the wider Pacific Island Countries within ASIABSS’s near-global beam footprint, growth is decelerating more sharply — to 2.8–3.4% in 2026 from over 4% in 2025 — driven by rising energy and shipping costs and softening tourism, with the World Bank flagging persistent structural constraints.

We have appraised the Subject Spectrum Assets on the basis of fair value. According to International Financial Reporting Standard 13 “Fair Value Measurement” (“IFRS 13”), fair value is defined as “the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date”.

In applying this basis of value, fair value is determined from the perspective of a hypothetical market participant, rather than from Astrum Networks’ own entity-specific perspective and accordingly excludes any premium or discount attributable to synergies, strategic considerations, or operational advantages that may be unique to the current holder of the Subject Spectrum Assets, such as their integration with Company’s existing satellite fleet or its multi-jurisdictional filing structure. This is consistent with the requirement under IFRS 13 that fair value be measured using the assumptions that market participants would use when pricing the asset, assuming those participants act in their economic best interest.

The fair value of the Subject Spectrum Assets, comprising the geostationary satellite spectrum rights held under the NEASTAR-1, ASIABSS, and NYASIABSS ITU filings at the 105° East orbital position, has therefore been assessed by reference to observable market evidence, where available, including comparable spectrum transactions, supplemented by appropriate valuation techniques where such market evidence is limited or requires adjustment for differences in frequency band, licensing structure, regulatory maturity, and coverage between the Subject Spectrum Assets and the comparables identified.

Our investigation comprised discussions with members of Management regarding the development, operations, and other information relevant to the Subject Spectrum Assets held by the Company, including their regulatory status, technical deployment, and commercial utilization. We also made such further inquiries and obtained such additional information as we considered necessary for the purposes of this valuation.

We have held discussions with Management concerning the development, operations, and other pertinent aspects of the Company’s spectrum holdings, and have consulted other sources of financial and business information available to us. In forming our opinion of value, we have relied, to a considerable extent, on the completeness, accuracy, and representations of the operational, financial, licensing, and other relevant data and information provided to us by Management.

The valuation of the spectrum assets requires consideration of all pertinent factors that may influence the operation of the Business Enterprise and its capacity to generate future investment returns from the deployment of such spectrum. The factors we have considered include, but are not necessarily limited to, the following:

In arriving at our opinion of value of the Subject Spectrum Assets, we have considered, reviewed and relied upon the following key information provided by the instructing party and from the public domain:

We have assumed and relied upon the accuracy and completeness of the information reviewed for the purposes of this valuation. In addition, we have relied upon the statements, information, opinions, and representations provided to us by the Company, including those concerning the regulatory standing and technical specifications of the Subject Spectrum Assets.

We have discussed these details with Management and have assumed the accuracy of the information provided, relying on such information to a considerable extent in arriving at our opinion of value. Where relevant, we have also sought clarification from Management on spectrum-specific matters such as license conditions, renewal prospects, and any restrictions on transfer or use.

We also conducted independent research using various sources, including government statistical releases, spectrum regulatory publications, and other industry data, to verify the reasonableness and fairness of the information provided. Based on this review, we believe the information relied upon is reasonable and reliable.

Our opinion is based upon existing economic, market, financial, regulatory, and other conditions that can be evaluated as at the date of this report, and we assume no responsibility to update or revise our opinion based on events or circumstances occurring after that date. In reaching our opinion, we have made assumptions with respect to such economic, market, financial, regulatory, and other conditions and matters, many of which are beyond our control or the control of any party involved in this valuation exercise, including future spectrum policy, licensing decisions, and market demand for spectrum capacity.

There are three generally accepted approaches to obtaining the fair value of the Subject Spectrum Assets, namely the Market-Based Approach, Income-Based Approach and Cost-Based Approach. Each of these approaches is appropriate in one or more circumstances, and sometimes, two or more approaches may be used together to cross-check the reasonableness of the valuation conclusion. Whether to adopt a particular approach will be determined by the most commonly adopted practice in valuing spectrum assets of a similar nature, as well as the availability and reliability of relevant data.

The Market-Based Approach values a spectrum asset by reference to prices at which comparable spectrum, such as spectrum of a similar frequency band, bandwidth, or geographic coverage, has changed hands in arm’s length transactions, whether through spectrum auctions, secondary market trades, or M&A transactions involving spectrum-holding entities. The underlying theory of this approach is that one would not pay more for spectrum than one would have to pay for an equally desirable alternative block of spectrum.

The transactions employed in analysing indications of value need to be concluded on an arm’s length basis, assuming that the buyers and sellers are well informed and have no special motivations or compulsions to buy or to sell. There are two methods commonly used to implement this approach for spectrum assets.

The first method relies on benchmarking against comparable spectrum auction results and secondary market transactions, normalized on a per-MHz-per-population (MHz-pop) basis to account for differences in bandwidth and coverage between the subject spectrum and the comparable transactions. Assets which have an established and liquid secondary market may be appraised by this method. However, this method has not been adopted because insufficient comparable auction or secondary market transaction data are available for spectrum of the same or similar frequency band and technical characteristics as the Subject Spectrum Assets. Therefore, there is insufficient relevant comparable transaction data to form a reliable basis for our opinion of value.

The second method derives valuation multiples, such as enterprise value per MHz-pop, from the market prices and financial data of listed telecommunications companies holding similar spectrum and applies these multiples to the Subject Spectrum Assets after appropriate adjustments for differences in spectrum quality, propagation characteristics, and licensed usage rights.

The Income-Based Approach focuses on the economic benefits attributable to the income-producing capability of the spectrum asset, in particular the incremental revenue and cost savings enabled by the spectrum’s deployment in the operator’s network. The underlying theory of this approach is that the value of the spectrum asset can be measured by the present worth of the economic benefits to be received over its useful or licensed life. Based on this valuation principle, the Income-Based Approach estimates the future economic benefits attributable to the spectrum and discounts them to their present values using a discount rate appropriate for the risks associated with realizing those benefits.

A commonly applied variant of this approach for spectrum assets is the Greenfield Method, under which a hypothetical, newly established network is modelled as if built from inception using only the subject spectrum, with the resulting projected cash flows discounted to present value to isolate the value of the spectrum from the value of the existing going concern and other assets of the business. Alternatively, the Relief-from-Royalty Method may be applied, whereby the value of the spectrum is estimated as the present value of the hypothetical royalty payments that the operator would otherwise need to pay a third party to license equivalent spectrum rights, net of tax, over the spectrum’s useful life.

Alternatively, the present value of the economic benefits can be calculated by capitalizing the benefits to be received in the next period at an appropriate capitalization rate, subject to the assumption that the spectrum-enabled business will continue to generate stable economic benefits and growth.

The Cost-Based Approach is based on the general concept that the value of a spectrum asset may be estimated by reference to the cost that would be required to obtain a licence for spectrum of equivalent utility, whether through replacement cost or reproduction cost. The assumption of this approach is that a prudent investor would not pay more for the Subject Spectrum Assets than the cost of acquiring or replicating spectrum rights of comparable frequency band, bandwidth, and technical capability, adjusted for physical, functional, and economic obsolescence where applicable. Under the Cost-Based Approach, the fair value of the spectrum asset refers to the current cost of obtaining a substitute licence with equivalent economic utility as at the measurement date, determined by reasonable valuation approaches having regard to the nature of the spectrum right, including its regulatory conditions, remaining term, and renewal prospects.

The Asset-Based Approach is based on the general concept that the earning power of a business entity is derived primarily from its existing assets. The assumption of this approach is that when each of the elements of working capital, tangible and intangible assets is individually valued, their sum represents the value of a business entity and equals to the value of its invested capital (“equity and long-term debt”). Under the Asset-Based Approach, the fair value of equity of a business entity/group refers to the fair values of various assets and liabilities on the statement of financial position of the business entity/group as at the measurement date, in which the fair value of each asset and liability was determined by reasonable valuation approaches based on its nature.

In this valuation exercise, the Market-Based Approach, implemented through the comparable transactions’ method, was ultimately adopted as the primary approach in arriving at our opinion of value, for the reasons set out below.

The Income-Based Approach, while conceptually applicable, was less suitable as the primary method given the layered, multi-jurisdictional nature of the Subject Spectrum Assets. The three underlying ITU filings, NEASTAR-1, ASIABSS, and NYASIABSS, sit at different regulatory maturity stages (comprising the fully registered, MIFR-confirmed 60-year ASIABSS status, alongside the NEASTAR-1 and NYASIABSS filings that have successfully achieved their bring-into-use milestones and are currently awaiting final MIFR registration), which introduces complex operational variables into any Greenfield or discounted cash flow projection tied to a single, unified revenue stream. Reliably forecasting incremental satellite broadcast and earth-station revenue attributable to each filing separately, net of the shared orbital infrastructure at 105°E, would require assumptions disproportionate to the reliability achievable through direct market benchmarking.

The Cost-Based Approach was similarly considered less appropriate, as the replacement cost of securing an equivalent geostationary orbital slot and ITU coordination position does not necessarily correlate with the commercial value the marketplaces on de-risked, revenue-generating spectrum, particularly given ASIABSS’s demonstrated operating history via AsiaSpace Limited’s ACMA-licensed earth stations and international broadcasting service into ASEAN markets and East Timor.

The following assumptions considered, having significant sensitivity effects in this valuation, have been evaluated and validated in arriving at our assessed values.

General Assumptions

We have adopted certain specific assumptions in our valuation, and the major ones are as follows:

The Subject Asset Spectrum is valued under the market approach, using observed transaction pricing for comparable spectrum as the primary input. No competitive auction or arm’s-length secondary-market transaction has occurred for Supplemental Downlink (SDL) or L-band spectrum anywhere in Asia Pacific between 2020 and 2026, and no direct market for the 1467–1492 MHz band exists in any of the 24 markets in the Coverage Region. Consequently, this valuation relies on a single, well-documented primary transaction — the EchoStar–AT&T spectrum sale — as the base comparable, adjusted through a series of quantified factors to bridge from the observed US low/mid-band transaction price to an indicated value for L-band SDL spectrum in each target market.

This reliance on unobservable, judgment-based adjustment inputs (rather than multiple directly observable market transactions) places this valuation within Level 3 of the fair value hierarchy under IFRS 13 and IVS 2025. Section 3.7 sets out the key judgments and their sensitivity.

The EchoStar–AT&T transaction was selected over four other spectrum transactions identified in the transaction research (the SpaceX–EchoStar S-band/AWS-4 sale, the AST SpaceMobile–Ligado L-band MSS agreement, the T-Mobile–UScellular acquisition, and the Egypt NTRA multi-operator agreement) because it is the largest, most recent (closed July 2026), and most transparent all-cash transaction with publicly disclosed spectrum quantities and covered population — providing the cleanest observable $/MHz-POP data point. The other transactions are referenced as corroborating, non-priced context rather than as base comparables, given their different band physics, deal structures, or regulatory regimes.

The EchoStar–AT&T spectrum sale (announced August 2025, closed July 2026) transferred 3.45 GHz and 600 MHz licenses — an average of ~50 MHz of low-band and mid-band spectrum across more than 400 US markets — for $23.0 billion in an all-cash transaction. Dividing consideration by bandwidth and covered population produces an unadjusted base price of $1.346 per MHz-POP, benchmarked against a US comparable GDP per capita of $90,027.

The EchoStar–AT&T sale is judged the strongest available benchmark for this valuation — ahead of the other transactions identified in the transaction research — for six reasons:

The base price is bridged to a value applicable to 1467–1492 MHz SDL spectrum through six multiplicative adjustment factors, reducing the unadjusted $1.346/MHz-POP to $0.524/MHz-POP before country-level GDP scaling (Lower Bound basis; $0.873/MHz-POP on the Upper Bound basis, which removes the regulatory harmonization discount for markets that have already identified the band for IMT use and the regulatory incumbent (DAB) conflict discount).

The +58.1% propagation adjustment is derived from a power-law curve (price = A × f⁻ⁿ) fitted through the EchoStar–AT&T deal’s own two revealed price points — 600 MHz low-band and 3.45 GHz mid-band — weighted by their respective share of the deal’s ~50 MHz average (20 MHz / 40% low-band, 30 MHz / 60% mid-band). The curve is then used to interpolate the implied price at 1,480 MHz, the midpoint of the target band.

Limitation: this is a two-point curve fit derived from a single transaction. It captures that deal’s own coverage-versus-capacity price spread rather than a universal frequency-value relationship and does not isolate propagation effects from other confounding factors such as regulatory bandwidth-chunk size or demand segmentation. 1,480 MHz falls between the two anchor points (600–3,450 MHz), so this is an interpolation rather than an extrapolation.

The adjusted price per MHz-POP is scaled to each target market by the ratio of local GDP per capita to the US comparable GDP per capita, then multiplied by the 25 MHz assigned bandwidth and the population covered:

Indicated Fair Value = Adjusted Price per MHz-POP × (Target GDP per Capita ÷ Comparable GDP per Capita) × 25 MHz × Population Covered

The Lower Bound reflects the full adjustment cascade, including the −25% regulatory harmonization discount and the −20% regulatory incumbent (DAB) conflict discount. The Upper Bound removes both discounts, assuming an optimized scenario where regulations align properly and incumbent conflicts are fully resolved. This is consistent with the subset of markets identified in supporting research as having already identified or cleared the 1452–1492 MHz band for IMT use (including Japan, Australia, Malaysia, Singapore, Thailand, Vietnam, Myanmar, Bangladesh, and Pakistan). Markets shown with a single point estimate reflect that already-cleared status rather than a Lower–Upper range.

Due to the uncertainties in economic situation, there is no guarantee that the expected financial performance will materialize.  Any changes in global political, economic and social conditions, laws, regulations and policies may have significant impacts on the projections of the future income of the Target Company.  None of these changes can be foreseen with certainty.

The concurrent loosening of monetary policies by the central banks in many developed and developing countries pose a significant risk of inflation, which will erode the profitability of the Target Company.

The performance of the Target Company may be better or worse than the expectation, and the resulting earnings and cash flows can be very different from our estimates.  The possibility of severe operational incidence, whether it is exogenous or endogenous, cannot be precluded.

Any changes in the technological developments and advancements may have significant impacts on the projections of the future income of the Target Company.  To remain competitive in the industry, the Target Company may be required to make substantial capital expenditures to keep up with technological changes.

The revenue of the Target Company is heavily dependent on a single business segment.  There is no guarantee that this business relationship can be maintained and that the products of the Target Company will remain popular.

As part of our analysis, we have reviewed the financial and technical information, such as ITU space network filing summaries, ACMA ground licensing records, and other pertinent data concerning the Subject Spectrum Assets, provided to us by the Company. We have assumed the accuracy of, and have relied on, such information to a considerable extent in arriving at our opinion of value.

We confirm that we have made relevant searches and enquiries, including review of ITU Radiocommunication Bureau publications, coordination records, and comparable spectrum transaction data, and have obtained such further information as is considered necessary for the purpose of this valuation.

The conclusion of value is based on generally accepted valuation procedures and practices that rely substantially on the use of numerous assumptions and the consideration of many uncertainties, not all of which can be easily quantified or ascertained, particularly given the advancing regulatory registration status and successfully completed bring-into-use milestones across the NEASTAR-1, ASIABSS, and NYASIABSS filings underlying the Subject Spectrum Assets. Further, while the assumptions and consideration of such matters are regarded by us to be reasonable, they are inherently subject to significant business, economic, regulatory, and competitive uncertainties and contingencies, many of which are beyond the control of the Company, and Roma Appraisals.

Unless otherwise stated, all monetary amounts stated in this valuation report are in UNITED STATES DOLLAR (USD).

We hereby confirm that we have neither present nor prospective interests in the Company, their associate companies, subsidiaries, the sponsoring administrations of the relevant ITU filings, or the values reported herein.

Based on the investigation stated above and the valuation method employed, the aggregate indicated Fair Value of the Subject Asset Spectrum across all 24 markets in the Coverage Region, as of the Date of Valuation, is USD2,814,000,000 to USD4,532,000,000.

Yours faithfully,

For and on behalf of

Roma Appraisals Limited

YU

Appendix I Limiting Conditions

The valuation reflects facts and conditions existing at the Date of Valuation. Subsequent events or circumstances, including any change in the coordination status, final MIFR publication steps, or registration status of the NEASTAR-1, ASIABSS, or NYASIABSS filings, have not been considered, and we are not required to update our report for such events and conditions.

We would particularly point out that our valuation was based on the ITU space network filing summaries, ACMA licensing records, and other information regarding the company background and business nature of the Business Enterprise and the Subject Spectrum Assets provided to us.

To the best of our knowledge, all data set forth in this report are assumed to be reasonable and accurately determined. The data, opinions, or estimates identified as being furnished by others, including the FCC auction records, M&A transaction data, and secondary-market spectrum comparables referenced in this valuation, have been gathered from reliable sources; yet, no guarantee is made nor liability assumed for their accuracy.

We have relied on the historical and/or prospective information provided by the Management and other third parties, including ITU SpaceExplorer filing exports and ACMA Register of Radiocommunications Licences data, to a considerable extent in arriving at our opinion of value. The information has not been audited or compiled by us. We are not in a position to verify the accuracy of all information provided to us, including the confirmed status of coordination, notification, and bring-into-use records reported across the three filings.

However, we have had no reason to doubt the truth and accuracy of the information provided to us and to doubt that any material facts have been omitted from the information provided. No responsibility for the technical, regulatory, or financial information relating to the Subject Spectrum Assets that has not been provided to us is accepted.

We assumed that the Management is competent and performs its duties under the company regulation, and that the operating agencies named on the respective ITU filings, Astrum Networks, AsiaSpace Ltd., and the Post and Telecommunication Board of Papua New Guinea, discharge their responsibilities in respect of the associated space network filings and ground licences in a responsible manner, unless otherwise stated in this report. Also, ownership of the Company and the Subject Spectrum Assets was assumed to be in responsible hands, unless otherwise stated.

We have not investigated the title to, or any legal liabilities attaching to, the Subject Spectrum Assets, including any encumbrances, disputes, or unresolved coordination matters that may not have been disclosed to us, and have assumed no responsibility for the title to the spectrum rights appraised.

Our conclusion of the fair value was derived from generally accepted valuation procedures and practices that rely substantially on the use of various assumptions and the consideration of many uncertainties, not all of which can be easily quantified or ascertained, particularly given the multi-jurisdictional, differing-maturity nature of the three underlying ITU filings. The conclusion and various estimates may not be separated into parts, and/or used out of the context presented herein, and/or used together with any other valuation or study.

We assume no responsibility whatsoever to any person other than the directors and the Management in respect of, or arising out of, the content of this report. If others choose to rely in any way on the contents of this report, they do so entirely at their own risk.

The working papers and models for this valuation, including the spectrum transaction comparable set and MHz-POP benchmarking analysis, are being kept in our files and would be available for further reference. We would be available to support our valuation if required. The title of this report shall not pass to the Company until all professional fees have been paid in full.