Obviousness

The obviousness of US patent 10292138 under 35 U.S.C. § 103 can be analyzed by combining the teachings of explicit and implicit prior art references mentioned within the patent itself. A person having ordinary skill in the art (POSA) in wireless communication systems would be motivated to combine these references to address known challenges in Quality of Service (QoS) differentiation and resource management.

The key aspects of claims 1-14 of US10292138 involve a User Equipment (UE) that:

1. Receives parameters for radio bearers (RBs).
2. Determines and transmits separate buffer occupancies for individual RBs to a network.
3. Receives a single allocation of uplink resources from the network.
4. Selects data from RBs for transmission using an iterative process involving a first iteration (selecting a subset of RBs based on received parameters) and a second iteration (selecting data based on buffered data for respective RBs).
5. Transmits the selected data.

Here, we consider the combination of the following prior art references:

• U.S. Pat. No. 6,845,100 (referred to as '100 patent): This patent is explicitly discussed in the background and summary of US10292138. It describes a system with two schedulers, a packet scheduler and a QoS scheduler, where the QoS scheduler prioritizes packets based on their assigned radio bearer. However, it is noted to prioritize only within a user's allocation.
• Hui Zhang, "Service discipline for guaranteed performance service in packet-switching networks" (1995): This is cited as an introduction to schedulers. It would teach general principles of packet scheduling, Quality of Service (QoS), and various service disciplines, including potentially weighted fair queuing or other mechanisms for allocating resources based on priority or weights.
• PCT publication WO 03/049320 (referred to as '320 publication): This publication is by the same applicant as US10292138 and is referenced in the description for specific steps within the iterative algorithm for limiting the number of queues. It describes how the process may operate normally when a queue limitation is not exceeded, and how the RAAUq'' (resource allocation units) may be modified. This indicates that the '320 publication teaches iterative resource allocation and queue management principles.
• General UMTS/3GPP standards (e.g., 3GPP TS 25.401, 3GPP TS 23.060): These are referenced for the typical architecture and operation of a UMTS network, including concepts like Radio Network Controllers (RNCs), Node-Bs, User Equipment (UEs), radio bearers (RBs), and Packet Data Protocol (PDP) contexts.

Obviousness Analysis

A POSA, aiming to enhance wireless communication systems to provide more flexible and efficient QoS differentiation for IP data flows, particularly across multiple users as opposed to merely within a single user's allocation, would have been motivated to combine the teachings of these references.

Motivation for Combining References:

1. Overcoming Limitations of '100 Patent (Cross-User Prioritization): The '100 patent teaches QoS prioritization based on radio bearers but is limited to a single user's allocation. A POSA would recognize the significant advantage of extending this to prioritize specific services (e.g., HTTP over FTP) across multiple users, a feature explicitly stated as a goal of the present invention. This motivation would drive the need for more granular information and control.
2. Granular Buffer Occupancy Reporting (UE to Network): To enable the network to perform cross-user prioritization on a per-radio bearer basis, it is evident that the network requires knowledge of the data buffered for each individual radio bearer at the UE. Thus, a POSA would be motivated to modify the UE (e.g., through its signal processor) to identify and transmit separate buffer occupancy for each radio bearer to the network. This is a logical extension of existing buffer status reporting mechanisms in wireless systems.
3. UE-Side Scheduling for Single Uplink Allocation: In an uplink scenario where the network provides a single physical allocation of resources to the UE, the UE must intelligently divide this allocation among its active radio bearers according to their QoS requirements. The principles of schedulers (Zhang) would suggest using weighted allocation. Therefore, a POSA would be motivated to signal network-defined weight values (S_tier parameters) to the UE. This allows the UE to mirror the network's scheduling logic, splitting the single allocation proportionally among its radio bearers and ensuring consistent QoS prioritization, rather than relying on a simpler, absolute priority scheme. The patent describes this as a "UE mirror scheduler".
4. Iterative Selection for Limiting Queues: As acknowledged by US10292138, a system with many services and users could lead to a large number of small resource allocations, increasing signaling overhead. To mitigate this, a POSA would be motivated to implement a mechanism to limit the number of queues/services allocated resources at any given time.
   • The '320 publication, being from the same applicant and explicitly referenced for parts of the iterative queue limiting algorithm (e.g., for RAAUq calculation and modification), would provide a POSA with a strong teaching or suggestion for employing iterative processes in resource allocation and queue management within a wireless network context.
   • Combining this with general knowledge of scheduling (Zhang), which often involves iterative refinement to balance different criteria (e.g., fairness, throughput, latency, number of active connections), a POSA would find it obvious to apply an iterative approach for selecting data.
   • Specifically, the "first iteration" of selecting a subset of RBs based on received parameters (S_tier via W'q and RAAUq') and ordering serves as an initial filtering step to manage the number of active queues. The "second iteration" of refining this selection based on actual buffered data (Nq' influencing W''q and RAAUq'') and incorporating running_RAAU_delta from previous allocations, ensures long-term fairness and responsiveness to actual data availability, even with short-term limitations. This iterative refinement is a common and obvious design choice for optimizing resource allocation under constraints.

Therefore, a POSA, motivated by the need for enhanced QoS differentiation across users and efficient resource management, would find it obvious to combine the teachings of the '100 patent (for RB-based QoS), general scheduling knowledge (Zhang), and iterative resource allocation techniques from the '320 publication (by the same applicant and referenced for specific algorithmic steps) to arrive at the claimed invention. The resulting system would allow granular buffer occupancy reporting, network-controlled but UE-executed weighted scheduling for uplink transmissions, and an iterative method for limiting active queues to optimize signaling and resource utilization.