PTAB

IPR2018-00614

EchoStar Corporation v. Realtime Data LLC

Log In

1. Case Identification

2. Patent Overview

  • Title: Data Compression Systems and Methods
  • Brief Description: The ’728 patent describes techniques for adaptive data compression. The system analyzes data within a data block to identify its parameters or attributes, a process which excludes relying solely on an external descriptor. If attributes are identified, the system performs "content dependent data compression" using a tailored algorithm; if not, it uses a single default data compression encoder.

3. Grounds for Unpatentability

Ground 1: Obviousness over Franaszek, Hsu, and Sebastian - Claim 25 is obvious over Franaszek in view of Hsu and Sebastian under 35 U.S.C. §103.

  • Prior Art Relied Upon: Franaszek (Patent 5,870,036), Hsu (a 1995 article, "Automatic Synthesis of Compression Techniques for Heterogeneous Files"), and Sebastian (Patent 6,253,264).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Franaszek taught a base system that selects compression algorithms based on a data type field attached to a data block (i.e., a descriptor). When the data type is known, a preselected list of methods is used; when unknown, a default list is used. Petitioner contended Hsu taught the core novelty of the ’728 patent: analyzing the data within the block itself, using statistical sampling and calculating "redundancy metrics" to determine both the data type and its compressibility. This internal analysis was used to select the optimal compression algorithm from a database, directly mapping to the claim limitation of analyzing internal data attributes while excluding reliance solely on a descriptor. Finally, Sebastian was argued to teach using a single, generic encoder when a specific format-dependent encoder is unavailable, which corresponds to the ’728 patent’s fallback to a "single data compression encoder."
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine these references to create a more robust, efficient, and versatile compression system. A POSITA would be motivated to integrate Hsu's internal analysis into Franaszek's framework to improve compression for heterogeneous files and to handle cases where Franaszek's external descriptor is missing or inadequate. This combination improves the accuracy of selecting a compression algorithm. Furthermore, a POSITA would be motivated to substitute Sebastian's single generic encoder for Franaszek's process of testing a default list of encoders, thereby reducing processing time and improving overall system speed when a data type is unknown.
    • Expectation of Success: Petitioner asserted a high expectation of success. The combination involved applying known techniques (Hsu’s internal analysis, Sebastian’s generic encoder) to a known system (Franaszek) to achieve the predictable results of improved compression accuracy and speed. The references share common goals of optimizing space savings and efficiency and rely on similar underlying compression technologies (e.g., Lempel-Ziv, Huffman), making their integration a matter of routine engineering.

Ground 2: Obviousness over Franaszek, Hsu, Sebastian, and Kawashima - Claim 25 is obvious over Franaszek in view of Hsu, Sebastian, and Kawashima under §103.

  • Prior Art Relied Upon: Franaszek (Patent 5,870,036), Hsu (a 1995 article), Sebastian (Patent 6,253,264), and Kawashima (Patent 5,805,932).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground built upon the Franaszek/Hsu/Sebastian combination from Ground 1 and added the teachings of Kawashima. Petitioner argued that Kawashima specifically addressed the problem of "negative compression" or "wasteful data expansion," where a compression attempt results in an output file larger than the input. Kawashima taught comparing an "actual compression ratio" to a preset threshold. If the data expanded (i.e., failed to meet the threshold), the system would output the original, pre-compression data instead of the larger, compressed version. This mapped directly to the claimed function of determining whether to output a block in received or compressed form.
    • Motivation to Combine: A POSITA, having already devised the system in Ground 1, would be motivated to incorporate Kawashima's teachings to further refine the system's efficiency. The primary motivation was to improve space savings and system performance by explicitly preventing the output of negatively compressed data, a well-known inefficiency in the art. This avoids wasting storage space and subsequent processing time on data that did not benefit from compression.
    • Expectation of Success: Petitioner claimed a high expectation of success because adding a threshold check to prevent data expansion is a simple, logical, and predictable improvement. The modification addresses a known problem in a way that is fully compatible with the goals and methods of the other combined references.

4. Key Claim Construction Positions

  • "data block": Petitioner proposed that this term should be given its broadest reasonable interpretation to mean "a single unit of data, which may range in size from individual bits through complete files or collection of multiple files." This broad construction supports applying the prior art's teachings to various data granularities.
  • "data block in [a/the] received form": Petitioner argued this term means a "data block that was not further compressed." This construction was critical to their argument, as it allowed them to map prior art references that teach outputting the original, uncompressed data (e.g., when compression is inefficient) to this claim limitation.

5. Key Technical Contentions (Beyond Claim Construction)

  • The petition's central technical contention was that the ’728 patent’s allegedly novel feature—analyzing data within a block rather than relying on an external descriptor—was not novel at all. Petitioner's expert argued that Hsu explicitly taught this concept by using statistical sampling of the block's content and calculating various "redundancy metrics" (e.g., degree of variation in character frequency, average run length) to devise a "compression plan." This method of looking at the inherent statistical properties of the data was presented as a direct and obvious alternative to Franaszek’s descriptor-based approach.

6. Relief Requested

  • Petitioner requested the institution of an inter partes review and the cancellation of claim 25 of Patent 9,054,728 as unpatentable.