Obviousness

Obviousness Analysis of US Patent 8019332 under 35 U.S.C. § 103

An invention is considered obvious under 35 U.S.C. § 103 if "the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious... to a person having ordinary skill in the art to which the claimed invention pertains." The analysis involves determining the scope and content of the prior art, identifying differences between the claimed invention and the prior art, and evaluating the level of ordinary skill in the pertinent art. A motivation to combine prior art references is also required, which can come from the knowledge of those skilled in the art, from the prior art references themselves, or from the nature of the problem to be solved.

The PTAB's institution decision in IPR2017-00754 found a reasonable likelihood that claims 1-14 and 16-20 of US8019332 are unpatentable as obvious over combinations of prior art, specifically citing 3GPP TS 25.211 V8.0.0 (Dec. 2007) and 3GPP TS 36.321 V8.0.0 (Dec. 2007). Although this IPR was settled, the institution decision suggests that viable obviousness arguments exist.

Prior Art References:

• 3GPP TS 25.211 (various versions, including V6.8.0, V6.9.0, V3.1.1, V3.11.0 released between 2000-2008): This technical specification describes the physical channels and mapping of transport channels onto physical channels in the FDD mode of UTRA (UMTS Terrestrial Radio Access). It defines various channels like the Forward Access Channel (FACH), Paging Channel (PCH), and Random Access Channel (RACH), and discusses physical layer procedures.
• 3GPP TS 36.321 (various versions, including V8.0.0, V8.2.0, V8.9.0, V10.10.0 released between 2007-2013): This specification details the Medium Access Control (MAC) protocol for E-UTRA (Evolved Universal Terrestrial Radio Access), which is part of 3GPP LTE. It covers MAC architecture, services expected from the physical layer, mapping between logical and transport channels, multiplexing, HARQ, scheduling, and priority handling. It mentions the use of RNTI on the PDCCH for Random Access Response messages.
• US8989208B2 ("PDCCH search space design for LTE-A multi-carrier operation"): This patent, though issued later (2015), describes concepts relevant to PDCCH search space design, including UE-specific search spaces, CCE aggregation levels (1, 2, 4, 8), and the derivation of PDCCH starting CCE indices based on UE-specific IDs, available CCEs, and/or CCE aggregation levels. It mentions that the PDCCH starting CCE index may be randomly derived.
• General LTE/PDCCH/CCE knowledge: Several sources (e.g.,) detail the fundamental concepts of PDCCH, Control Channel Elements (CCEs), aggregation levels (1, 2, 4, 8), and search spaces (common and UE-specific) in LTE systems. They explain that PDCCH carries downlink control information (DCI), that CCEs are resource units for control information, and that UEs monitor specific search spaces to find their DCI. The aggregation level refers to the number of CCEs used for a PDCCH, with higher levels being suitable for poor channel conditions. UE-specific search spaces have starting locations that can vary per subframe or UE, often determined by a hash function.

Motivation to Combine Prior Art References:

A person having ordinary skill in the art (POSITA) in mobile communication technologies, particularly in 3GPP LTE, would have been motivated to combine known techniques to improve the efficiency and reduce power consumption in decoding control information. The overarching problem addressed by US8019332 — limiting PDCCH regions to be decoded by each UE to reduce processing and power consumption while minimizing search space overlap — was a recognized challenge in the evolution of mobile communication systems like LTE.

The motivation to combine elements from the identified prior art to arrive at the claimed invention would stem from:

1. Improving resource allocation and control signaling efficiency: The 3GPP specifications (TS 25.211 and TS 36.321) provide the foundational framework for physical and MAC layer operations in UMTS and LTE, including control channel concepts. A POSITA would constantly seek ways to optimize these foundational elements.
2. Addressing the challenge of blind decoding: UEs perform blind decoding because they don't know the exact location or format of their PDCCHs. This requires significant processing. The concept of limiting search spaces was a known approach to mitigate this.
3. Minimizing collisions and maximizing simultaneous UE control: As highlighted in US8019332 itself, if all UEs decode the same limited PDCCH region, the number of simultaneously controllable UEs is restricted. Allocating different, non-overlapping PDCCH decoding regions (search spaces) to different UEs was a clear objective to overcome this limitation.
4. Leveraging established pseudo-random sequence generation techniques: Randomization or pseudo-random number generation for channel allocation or identification was a common practice in wireless systems to avoid systematic collisions and distribute resources.

Obviousness Arguments for Claims 1, 6, and 11:

Claims 1, 6, and 11 of US8019332 all recite substantially similar technical features regarding the method for determining the start position of a PDCCH search space using an iterative formula Yk = (A * Yk-1) mod D and a subsequent modulo 'C' operation, where C = floor(N/L).

Combination: 3GPP TS 36.321 (V8.0.0 or later relevant versions) in view of general knowledge of PDCCH/CCE allocation and pseudo-random sequence generation, as exemplified by US8989208B2.

Analysis:

• Known elements:

  • PDCCH, CCEs, Aggregation Levels, Search Spaces: It was well-known in LTE (as evidenced by general knowledge and specifications predating the priority date of US8019332, such as 3GPP TS 36.321 from December 2007) that PDCCHs carry DCI, are transmitted using CCE aggregations (aggregation levels of 1, 2, 4, 8), and that UEs monitor specific search spaces (common and UE-specific) to find their DCI.
  • UE-specific search space start positions: The concept of varying the starting location of a UE-specific search space for each subframe or UE, often determined by a hash function (which is a form of pseudo-random generation), was also known. US8989208B2, for example, states that the "PDCCH starting CCE index may be randomly derived based on the UE-specific ID, a number of available CCEs on the carrier with the search space, and/or a CCE aggregation level."
  • Modulo operations for resource mapping: Modulo operations are a fundamental tool in resource allocation within fixed-size pools to ensure indices wrap around correctly. The calculation of C = floor(N/L) to determine the number of candidate positions based on total CCEs and aggregation level would be a straightforward engineering design choice for a POSITA optimizing resource utilization within a search space.
  • Iterative pseudo-random number generation: Linear Congruential Generators (LCGs), which follow the form Xn = (a * Xn-1 + b) mod m, were a well-established and computationally efficient method for generating pseudo-random sequences in computer science and communication systems for tasks like scrambling, hopping, or sequence generation. The formula Yk = (A * Yk-1) mod D is a specific instance of an LCG where B=0.
  • Subframe-dependent allocation: The concept of subframes (e.g., in 3GPP LTE, a 10ms radio frame comprises 10 1ms subframes) and subframe-specific resource allocation was inherent to the time-division aspects of LTE. Using a value from a previous subframe (Yk-1) to derive a value for the current subframe (Yk) is a natural way to introduce time-dependent variation and maintain a sequence across subframes, which is a known technique for generating identification-dependent randomization numbers every subframe.

• Motivation for combination:
  A POSITA, seeking to efficiently allocate PDCCH search spaces to multiple UEs in an LTE system to minimize collisions and reduce UE blind decoding complexity (a known problem), would have been motivated to combine these known elements.

  • They would recognize the need for a mechanism to generate unique or pseudo-unique starting positions for UE-specific search spaces across UEs and over time (subframes). The use of a UE ID as an initial seed for a pseudo-random generator, as suggested in US8019332 itself (using UE ID for x for the first subframe), and for deriving Yk-1 for subsequent subframes, would be an obvious choice to achieve UE-specific and time-varying starting positions.
  • The problem of PDCCH decoding regions overlapping for UEs with different CCE aggregation levels (as shown in FIG. 10 of US8019332) provides a clear motivation for generating different randomization numbers for each CCE aggregation level or incorporating the CCE aggregation level into the initial value.
  • Using an efficient and well-understood pseudo-random number generator, such as an LCG (represented by Yk = (A * Yk-1) mod D), would be a predictable choice for generating these dynamic start positions. The modulo 'D' operation (with D chosen to be a prime number larger than the range of the input value for good randomization characteristics, as discussed in US8019332 itself) would be standard practice for such generators.
  • The final modulo 'C' operation, where C = floor(N/L), would be an obvious step to map the output of the pseudo-random generator to a valid starting index within the available CCE aggregations for a given L. This ensures the start position is always within the bounds of the search space for that specific aggregation level.

• Predictable Result: The combination of these elements would yield the predictable result of providing a UE-specific, subframe-dependent, and CCE aggregation level-aware method for determining the start position of a PDCCH search space, thereby reducing blind decoding efforts and minimizing collisions. This is an "obvious solution" to a "known problem".

Therefore, the methods and apparatus described in claims 1, 6, and 11, encompassing the use of an iterative modulo-based pseudo-random sequence for determining the starting position of a UE-specific PDCCH search space, where the number of candidate positions C is derived from the total CCEs and the CCE aggregation level, would have been obvious to a POSITA by the priority date of US8019332. The specific values for A, B, and D (e.g., A=39827, B=0, D=65537) mentioned in dependent claims 2, 7, and 12 would be mere optimizations or design choices within the realm of an obvious mathematical function.