Summary
In a sweeping 259-page judgment, the Delhi High Court upheld the validity of Indian Patent No. 240893, owned by Canadian company Communication Components Antenna Inc., for split-sector antenna technology that uses asymmetrical beams to increase subscriber capacity in cellular networks. The court rejected every ground of revocation - novelty, inventive step, insufficiency, and non-patentability raised by German antenna manufacturer Rosenberger Hochfrequenztechnik GmbH and its group companies. Finding infringement established, the court awarded approximately ₹152 crore in damages computed at a 20% royalty on the Defendants' total antenna sales, issued a permanent injunction, and granted a certificate of validity under Section 113 of the Patents Act. The ruling is a landmark in Indian patent law for its damages methodology, its introduction of the "Dartboard Model" to critique scattershot prior art challenges, and its caution against AI-assisted prior art searches designed to overwhelm patent holders.
A patent dispute spanning nearly a decade, three prior infringement suits, proceedings before the US Patent Trial and Appeal Board, and patent offices in Europe and China has culminated in what may be one of the largest patent infringement damages awards ever passed by an Indian court. The Delhi High Court’s 259-page judgment in Communication Components Antenna Inc. v. Rosenberger Hochfrequenztechnik GmbH does not merely resolve a bilateral commercial dispute; it lays down significant principles on damages methodology, the treatment of foreign patent office decisions, and the limits of prior art challenges in Indian patent litigation.
Background
Communication Components Antenna Inc. (“CCA”), a Canadian company specialising in cellular base station products, is the registered owner of Indian Patent No. 240893 (“IN’893”), titled “Asymmetrical Beams for Spectrum Efficiency.” The patent was originally granted on 9th June 2010 to CCA’s predecessor-in-interest, TenXc Wireless Inc., and was assigned to CCA through an asset purchase in 2012. IN’893 covers a method and apparatus for replacing conventional sector antennas with split-sector antennas that emit asymmetrical beams, thereby reducing handover zones and increasing subscriber capacity while maintaining the same critical coverage area.
Rosenberger Hochfrequenztechnik GmbH & Co. KG (“Rosenberger Germany”), a German antenna manufacturer, and its affiliated entities – Prose Technologies India Pvt. Ltd. (formerly Rosenberger Electronic Co. (India) Pvt. Ltd.), Rosenberger Asia Pacific Electronic Co. Ltd. (Beijing), and Prose Technologies (Suzhou) Co. Ltd. (collectively, “Rosenberger”), were accused of manufacturing and selling multi-beam antennas in India that infringed Claims 1, 10, 12, and 13 of IN’893.
CCA had previously enforced IN’893 in three prior suits. In the first suit against Mobi Antenna Technology, the court found infringement and awarded USD 2.6 crore in damages. In the second suit, CommScope Technologies obtained a global licence and settled. In the third suit against Ace Technologies Corp., the court granted an interim injunction directing deposit of 10% of sales, a direction upheld by the Division Bench and the Supreme Court. The present suit was filed in December 2019. Trial was conducted through summary adjudication under the Delhi High Court Patent Rules, with evidence of four witnesses recorded over just two days using live transcription.
Questions Before the Court
-
-
- Whether IN’893 is liable to be revoked on any of the grounds under Section 64(1) of the Patents Act, 1970; including lack of novelty under Section 64(1)(e), lack of inventive step under Section 64(1)(f), insufficiency under Section 64(1)(h), failure of claims to define the invention under Section 64(1)(i), false representation under Section 64(1)(j), and non-patentability under Section 64(1)(k)?
- Whether Rosenberger’s antenna products infringe IN’893?
- Whether CCA is entitled to a permanent injunction and damages, and if so, at what rate?
-
CCA’s Arguments
On Validity:
CCA argued that all existing prior art at the priority date taught the use of symmetrical beams and actively taught away from asymmetry, which was seen as a distortion. The suit patent’s inventive leap was to deliberately introduce asymmetrical beams in a split-sector antenna, reducing handover zones while preserving the same critical coverage area – effectively doubling subscriber capacity without network-wide changes.
On insufficiency, CCA submitted that the specification fully disclosed the invention. Power and phase weightings were expressly provided. Element spacing could be derived using a standard half-wavelength formula, and the single-element pattern followed a known cosine function. CCA highlighted that Rosenberger’s own expert (DW-2) admitted in cross-examination to having generated MATLAB beam patterns using these values but chose not to file them, proving the disclosure was workable.
CCA distinguished the European patent regime from India’s, noting that under Article 84 EPC, claims must be self-contained without reference to the specification, unlike Indian law which permits purposive construction reading claims with the specification. The European application was abandoned for business reasons – the technology was more viable in densely populated markets like India. CCA relied on the Division Bench ruling in Ace Technologies and the single-judge decision in Mobi Antenna as having already upheld the patent’s validity.
On Infringement:
CCA’s expert (PW-1, Mr. Cosgrove) proved infringement by applying the power and phase weightings from the suit patent at operating frequencies stated in Rosenberger’s own product brochures, using standard element spacing and cosine function calculations, and generating beam patterns in MATLAB. When overlaid on Rosenberger’s admitted brochure patterns, they showed near-identical asymmetrical configurations. CCA also produced a comparison document that Rosenberger itself had prepared for Reliance Jio, placing its beam patterns alongside CCA’s to demonstrate they were virtually identical.
On Damages:
CCA claimed lost profits from sales lost entirely to Rosenberger and from price erosion – having to slash prices when competing against infringing products for customers like Reliance Jio and Bharti Airtel. CCA also sought exemplary damages and, in the alternative, royalty-based damages pegged to the confidential CommScope licence agreement.
Rosenberger’s Arguments
On Invalidity – Insufficiency (Primary Ground):
Rosenberger’s central attack was insufficiency under Sections 64(1)(h) and 64(1)(i). The patent failed to disclose how to create an asymmetrical beam that maintains substantially the same critical coverage area. The term “critical coverage area” was coined by CCA, appeared only twice in the entire specification, and was never defined. Without disclosure of element spacing, array arrangement, or the correlation between asymmetry and maintained coverage, a person of average skill could not work the invention.
Rosenberger relied on the EPO’s examination report, which rejected the corresponding European application on similar grounds – finding that 100% identity in coverage areas was physically impossible and that the specification failed to explain how asymmetry was to be induced. CCA had abandoned that application after these objections. Rosenberger further argued that Claims 1 and 10 were impermissibly broad – Claim 1 claimed a monopoly over the theoretical principle of asymmetrical beams, and Claim 10 merely restated it as a product claim without adding any constructional features.
On Invalidity – Lack of Novelty and Inventive Step:
Rosenberger argued that bi-sector antennas, asymmetry in beam patterns, and capacity-enhancing sectorisation were all well-established in the prior art, citing the Butler Matrix design (known since 1982), the Wastberg (Ericsson) patent, and published literature on dual-beam antennas. Rosenberger also pointed to the USPTO’s initial obviousness objection before the patent was allowed with amendments.
On Non-Infringement:
Rosenberger raised several defences. CCA had not conducted a proper claims-versus-product comparison – PW-1 compared beam patterns rather than the physical antenna or its constructional features. The MATLAB overlay was unreliable because the element spacing data was not disclosed in the affidavit.
Most significantly, Rosenberger raised a Gillette defence: its antennas used the prior-art Butler Matrix design, which had no mechanism to induce or optimise asymmetry. Any asymmetry in Rosenberger’s beam patterns was uncontrolled and accidental, arising from noise, distortion, and environmental factors – not deliberately induced. DW-2 categorically stated that Rosenberger’s antennas were designed to radiate symmetric beam patterns only. CCA itself had acknowledged in paragraph 27 of the plaint that “uncontrolled asymmetry” existed in the prior art, and Rosenberger argued its products fell squarely within that category.
On Damages:
Rosenberger contended that CCA lost orders due to its own inferior product quality, not infringement. It claimed to have conducted adequate due diligence before entering India. On quantum, Rosenberger submitted that if held liable, the only appropriate measure would be a royalty at less than 5%, since variation in antenna prices was driven by factors other than the patented technology.
Court’s Observations and Analysis
No Prior Art Document Anticipates IN’893
The court examined each prior art document cited for lack of novelty. Gabriel disclosed only straight sector-for-sector antenna exchange without disclosing a plurality of sub-sector coverage areas, asymmetrical beam patterns as defined in IN’893, or maintenance of critical coverage area. Rhodes, incorporated by reference into Gabriel, contained no concept of critical coverage area. The TenXc Article, while mentioning asymmetric patterns, provided no technical details on how such patterns would be generated, how sectorisation would be achieved, or how coverage area would be maintained; at best, it was a hint of what was to come, not anticipation. Applying the settled principle from Merck Sharp & Dohme v. Glenmark Pharmaceuticals, 2015 (63) PTC 257 (Del), that prior art must disclose subject matter which, if performed, would necessarily infringe the patent, the court held that novelty was established.
The “Dartboard Model” and the Rejection of Hindsight Mosaicing
In a significant analytical contribution, the court introduced what it termed the “Dartboard Model” to describe Rosenberger’s approach to inventive step. The court observed that Rosenberger had cited nine unrelated prior art documents, ranging from lens systems (Herscovici) to weight factor transmitters (Ylitalo-II); cherry-picking the word “asymmetry” from entirely different technological contexts in an exercise of impermissible hindsight bias. The court held that for multiple prior art documents to establish obviousness, they must be inter-linked by the relevant field or the problem being solved. Citing the Calcutta High Court’s decision in Guangdong OPPO v. Controller of Patents and the Delhi High Court’s own ruling in Sterlite Technologies v. HFCL, the court rejected the mosaic as disconnected. Notably, the court cautioned that the increased use of AI tools and Large Language Models makes it easy to cite several unrelated prior arts to challenge inventive step, and courts must exercise deeper scrutiny to ensure that the threshold of inventive step is not set so high as to stifle human innovation.
Sufficiency Stands Despite Foreign Rejections
The court rejected the insufficiency challenge, holding that the terms “critical coverage area,” “asymmetry,” and “substantially equivalent” were clearly understandable to a person skilled in telecommunications. Non-disclosure of element spacing did not invalidate the patent because, as the Plaintiff’s expert established, element spacing is derived once the operating frequency is fixed. It is frequency-dependent and does not need separate disclosure. The court distinguished the EPO and Chinese decisions, noting that both proceeded on the basis of the priority document rather than the complete specification as granted in India, and that the statutory requirements under Indian law (Section 10 of the Patents Act) differ from Article 83 of the EPC and Article 21 of China’s implementing rules. The court emphasised that claims must be construed in the Indian context, relying on the Supreme Court’s ruling in Bishwanath Prasad v. Hindustan Metal Industries.
Infringement Established Through Beam Pattern Comparison
On infringement, the court found that Rosenberger’s defence collapsed at trial. DW-2, the Defendants’ technical witness, admitted that no simulation was conducted on MATLAB or any other software using the suit patent’s disclosed power and phase weightings to generate the Defendants’ beam patterns. He was unable to deny that the beam patterns generated by CCA’s expert from the suit patent’s disclosures were identical to Rosenberger’s product brochure patterns. The Defendants failed to produce their actual antennas or any independent beam pattern data. Their claim that any asymmetry was “unintentional” was contradicted by Rosenberger’s own marketing document, which compared its beam patterns against CCA’s and claimed slight superiority, thereby proving that the parties were direct competitors selling functionally identical antennas.
Damages at 20% Royalty
Rejecting CCA’s claim for lost profits as the sole measure, the court held that the appropriate method was royalty-based damages. After examining the confidential CommScope licence agreement in sealed cover, the court determined that Rosenberger, as a wilful infringer rather than a good-faith licensee, should pay at the upper end of the royalty range. The court fixed the royalty at 20% of sales, applied it to total disclosed sales of USD 60.9 million and INR 184.7 crore, and arrived at total damages of approximately ₹152.32 crore (inclusive of the USD component converted at the decree-date exchange rate of ₹94.65 per dollar). Interest at 7% per annum was ordered if payment is not made by 30th June 2026.
Findings
In view of the observations and the arguments presented by both the parties, the Delhi High Court held that:
-
-
- The Defendants failed to establish any ground for revocation of IN’893. The suit patent is novel, involves an inventive step, is sufficiently disclosed under Section 10 of the Patents Act, 1970, and does not fall within any exclusion. The counterclaim for revocation was dismissed.
- The Plaintiff established that Rosenberger’s multi-beam antennas infringe Claims 1, 10, 12, and 13 of IN’893 by producing asymmetrical beam patterns that maintain substantially the same critical coverage area as the replaced sector antenna.
- A decree of permanent injunction was granted restraining Rosenberger and its group entities from manufacturing, selling, or offering for sale eleven identified antenna models and any other models infringing IN’893.
- Damages of ₹152,32,36,783.90 were awarded, computed at a 20% royalty on the Defendants’ total disclosed sales, payable by 30th June 2026, failing which simple interest at 7% per annum applies.
- For any additional sales not yet disclosed in the Defendants’ affidavits, updated sales figures must be filed within four weeks, with damages calculated at the same 20% royalty rate.
- A certificate of validity was granted in respect of IN’893 under Section 113 of the Patents Act, 1970.
- The Plaintiff was awarded complete actual costs, to be determined by the Taxation Officer.
- The Registry was directed to communicate the judgment to the Controller General of Patents, Designs and Trademarks.
-
Case Citation: Communication Components Antenna Inc. v. Rosenberger Hochfrequenztechnik GmbH & Co. KG & Ors., CS(COMM) 653/2019 and CC(COMM) 22/2022, Delhi High Court, decided on 30th March 2026. Available at https://indiankanoon.org/doc/195388033/
Authored by Gaurav Mishra, Patent Attorney, BananaIP Counsels