DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-4, 11, and 18-20 are rejected under 35 U.S.C. § 102(a)(1) as being anticipated by Kwon (US 2014/0037018 A1, henceforth referred to as "Kwon", cited by the applicant)
PNG
media_image1.png
193
500
media_image1.png
Greyscale
Fig. 1 of Kwon annotated by the examiner for ease of reference.
Regarding claim 1, Kwon discloses an OFDM base station receiver 100 (see FIG. 1; §0021, §0035, §0038–§0039, §0050) comprising processor-based hardware configured to execute programmatic operations for distortion compensation. Specifically, Kwon's distortion compensation unit 120 and OFDM receiving unit 130 are implemented as digital signal processing components within the base station, necessarily comprising at least one processor and associated memory storing executable program logic. See §0038 (describing the OFDM receiver 100 as including an RF receiving unit 110, distortion compensation unit 120, and OFDM receiving unit 130 which collectively operate on received OFDM signals in a programmatic, sequential fashion). The preamble apparatus limitations are therefore fully met.
Kwon also discloses selecting a compensation function (i.e., a distortion model) from a set of multiple such functions. Specifically, Kwon describes a distortion compensation unit 120 that "include a plurality of preset compensation functions for respective pieces of data among a plurality of pieces of characteristic identification information data" and that selects the compensation function "corresponding to a respective piece of data of the characteristic identification information data included in the OFDM signal." See §0017, §0038–§0041, §0048–§0049.
Kwon further describes a codebook containing "a plurality of compensation functions respectively corresponding to the plurality of pieces of characteristic identification information data" (§0048), from which the appropriate compensation function is called upon detecting the characteristic identification information in the received signal (§0049). This codebook of multiple compensation functions — each tailored to a specific type of power amplifier distortion characteristic — constitutes "a set of power amplifier distortion models" within the meaning of Claim 1. The act of calling the appropriate function from the codebook based on the detected identification information constitutes "selecting" that distortion model from the set. See also Claim 4 of Kwon; §0044, §0058–§0059 (describing type A, type B, and type C distortion characteristics each having a corresponding compensation function).
Kwon also teaches compensation functions that are preset (i.e., pre-trained/pre-determined before use) and configured specifically to compensate power amplifier distortion. See §0017 ("plurality of preset compensation functions"), §0038–§0041, §0048.
A "pre-trained machine learning model" in the context of the present application is a model that has been trained prior to deployment to perform a compensation function — i.e., it has been fitted or optimized, using data, to produce an inverse or corrective output relative to a known distortion characteristic. Kwon's preset compensation functions are precisely this: they are derived from the distortion characteristic function f(x) of the power amplifier and are analytically constructed as the inverse function f'(x) = ax/f(x) (see §0019, §0025, §0045–§0047), meaning they are pre-computed/pre-trained based on known power amplifier distortion data and stored in the codebook before runtime operation.
Kwon's compensation functions satisfy the functional description of a "pre-trained machine learning model" because: (1) they are derived/fitted from knowledge of the power amplifier's distortion characteristics prior to deployment (pre-training phase); (2) they are stored in a codebook and retrieved at runtime (inference phase); and (3) they are "configured to compensate power amplifier distortion" — which is the express purpose described throughout Kwon (see §0016–§0017, §0071).
Claim 1 does not further specify how the machine learning model is trained, what training data or algorithm is used, or any architecture of the model. Without such further specificity, the claim scope encompasses any pre-trained model configured to compensate PA distortion — a functional description that Kwon's pre-derived, pre-stored inverse compensation functions satisfy.
Kwon further discloses the OFDM base station receiver 100 receiving RF OFDM signals from an OFDM transmitter (terminal device 10). See §0021 ("The OFDM receiver may be included in an OFDM base station receiving an RF OFDM signal from the OFDM transmitter"); §0035–§0036; FIG. 1 (showing signal path from OFDM Terminal 10 → ANT1 → RF Receiving Unit 110 of OFDM Base Station 100). The OFDM terminal 10, comprising an OFDM transmitting unit 11, power amplifier 12, and RF transmitting unit 13 (§0036; FIG. 1), constitutes "a terminal device" within the plain meaning of Claim 1. The RF OFDM signal carrying data transmitted from the OFDM terminal to the OFDM base station constitutes an "uplink data transmission."
Kwon then teaches that after selecting/calling the appropriate compensation function, the distortion compensation unit 120 compensates for distortions of the received OFDM signal using that function. See §0017 ("compensating for distortions of the OFDM signal from the RF receiving unit by using a compensation function corresponding to a respective piece of data of the characteristic identification information data"); §0038–§0041; §0045 (compensation function set as inverse function of distortion characteristic, applied by multiplying it against the received signal); §0052–§0056; FIG. 2 (step S400: "Compensate for Distortion of OFDM Signal Based on Called Compensation Function").
Specifically, Kwon teaches that the compensation is performed "based on" the selected distortion compensation function (the distortion model), which is the exact relationship recited in Claim 1. See §0056 ("compensating for the distortions of the received OFDM signal by using the called compensation function").
wherein, per claims 2 and 18, Kwon expressly discloses that the distortion compensation unit 120 selects its compensation function based on "characteristic identification information" included in the OFDM signal, which directly identifies the specific power amplifier distortion type of the terminal's power amplifier. See Kwon §0017, §0037, §0041, §0043–§0044, §0048–§0049.
Specifically, Kwon discloses at §0037 that the OFDM transmitting unit 11 "includes one distortion characteristic function corresponding to distortion characteristics of the power amplifier adopted for a current OFDM transmitter among a plurality of distortion characteristic functions in the OFDM signal so as to identify the distortion characteristics of the power amplifier." This characteristic identification information embedded in the transmitted signal constitutes an "identifier of a power amplifier distortion model associated with the power amplifier" and/or an "identifier of a power amplifier of the terminal device" within the meaning of Claim 2.
Kwon further discloses at §0044 and §0059 that the power amplifiers of the OFDM transmitter may be classified into type A, type B, and type C distortion characteristics, each having a corresponding characteristic identification code — constituting distinct PA identifiers from which the appropriate compensation function is selected.
With respect to "operating conditions," the characteristic identification information of Kwon encodes the distortion type of the power amplifier, which in practice encompasses the operating conditions of the PA (e.g., frequency band, bias point, supply voltage) that determine its distortion characteristics. A person of ordinary skill in the art would recognize that the type A/B/C classification in Kwon represents operating-condition-dependent distortion profiles. The claim requires selection based on "at least one of" the listed criteria; Kwon discloses selection based on a PA identifier/distortion model identifier, which is expressly listed.
Wherein per claims 3 and 19, Kwon discloses an OFDM system in which the OFDM terminal 10 transmits an OFDM signal that has been processed by its power amplifier 12 and is therefore distorted by the power amplifier's characteristic distortion function f(x). See Kwon §0036–§0037; FIG. 1 (signal path S11 → Power Amplifier 12 → S12 → RF Transmitting Unit 13 → over the air → RF Receiving Unit 110 of Base Station 100).
Critically, Kwon §0037 discloses that the OFDM transmitting unit 11 includes a known, pre-determined distortion characteristic function in the OFDM signal — meaning the transmitted signal is a pre-defined OFDM waveform that, after passing through the power amplifier 12, becomes distorted by the PA distortion. The base station 100 receives this distorted version of the pre-defined OFDM signal (S12, FIG. 3). This constitutes "one or more first reference signals" that "comprise a pre-defined signal distorted by the power amplifier distortion." See FIG. 3 (waveform S11 = original pre-defined OFDM signal; S12 = the same signal distorted by PA 12; S32 = signal after distortion compensation at base station).
The term "reference signal" in the context of the present application means a known signal used by the receiver to identify and compensate for distortion — precisely what the characteristic-identification-information-bearing OFDM signal of Kwon represents. Kwon's OFDM signal serves both as a data carrier and as a reference for distortion identification. See §0037, §0043, §0049 (base station detects the characteristic identification information from the received signal to identify distortion and select the compensation function).
wherein per claims 4 and 20, Kwon expressly discloses that the base station selects (calls) the compensation function based on the characteristic identification information detected from the received OFDM signal. See Kwon §0049 ("when the characteristic identification information included in the OFDM signal is detected, the compensation functions corresponding to the characteristic identification information may be called from the code-book"); §0041, §0054–§0055; FIG. 2 step S300 ("Call Compensation Function Corresponding to Characteristic Identification Information").
Since the received OFDM signal constitutes "the first reference signals" as established in the Claim 3 analysis above, and since the distortion model (compensation function) is selected based on information detected from that signal.
Claim 11 is the method counterpart of apparatus Claim 1. The element-by-element analysis and mapping to Kwon's disclosure is identical in substance to the Claim 1 rejection. Kwon discloses each step of this method through the operation of its OFDM base station receiver 100, as mapped below.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 5-10, 21-23 are rejected under 35 U.S.C. 103 as unpatentable over Zhiming et. al. (CN 107819476 B, a machine translation is relied upon for current rejection) in view of Norio.
Regarding claims 5 and 21, Kwon discloses a base station that (a) receives distorted OFDM signals (the first reference signals); (b) detects the characteristic identification information to identify the PA distortion; and (c) selects the appropriate compensation function. See Kwon §0038–§0041, §0048–§0049, §0054–§0056.
Kwon's distortion compensation unit 120 already performs an implicit evaluation of the distortion model when it detects the characteristic identification information and confirms which compensation function to apply. The step of evaluating the model's performance based on the received reference signal and then updating/refining the compensation function based on such evaluation is a routine and well-known design choice in adaptive signal processing. A POSITA would have been motivated to add model re-training/adaptation functionality to Kwon's system for the following reasons:
(1) Adaptive or iterative refinement of signal compensation functions based on received signals was a standard technique in the wireless communications field at the priority date, as it directly improves compensation accuracy over time in the face of time-varying channel conditions and PA drift.
(2) Kwon's system already collects the necessary data (received distorted reference signals) to perform such evaluation and adaptation — making the addition of a re-training step a straightforward and obvious implementation choice with a predictable result: improved compensation accuracy.
(3) The concept of updating or fine-tuning a pre-trained model using newly received signals is a basic and well-understood operation in machine learning and adaptive filtering, requiring no more than ordinary skill to implement.
There would have been a reasonable expectation of success in implementing re-training because Kwon already provides the infrastructure (the codebook model, the received signal data, and the compensation evaluation path) needed for such adaptation. The modification amounts to combining known elements using known methods to yield a predictable result—paradigmatic obviousness.
Regarding claims 6 and 22, Kwon discloses a base station that compensates for PA distortion introduced by the terminal's power amplifier (Kwon §0016–§0017, §0038–§0041). The purpose of this receiver-side distortion compensation is expressly stated: to allow the power amplifier to operate more efficiently by relieving the terminal from needing to employ PAPR reduction techniques that constrain the PA's operating range. See Kwon §0071 ("expanding the operating area of the power amplifier").
Given this express purpose, a POSITA would immediately recognize that once the base station confirms it can compensate for PA distortion, the logical and obvious next step is to signal this capability back to the terminal so that the terminal can relax its power control constraints — specifically, reduce its power backoff (the margin the PA must maintain below its peak power to avoid distortion) and/or reduce maximum power reduction (MPR), thereby operating at higher efficiency. This feedback mechanism is a standard element of adaptive power control protocols in wireless systems (e.g., LTE/5G uplink power control), and implementing it in the context of Kwon's distortion compensation system would have been an obvious design choice with a well-understood and predictable benefit: increased uplink transmits power and improved link budget.
The motivation is explicit in Kwon itself — the system is designed to "expand the operating area of the power amplifier" (§0071) and to reduce PAPR-related constraints (§0006–§0008, §0010). Communicating this operational expansion to the terminal via a signaling message is the natural and obvious implementation of that stated goal.
Further per claims 7 and 23, as established in the Claim 3 analysis above, Kwon's base station already receives signals from the terminal that contain characteristic identification information and are distorted by the PA. Kwon further teaches a codebook-based compensation architecture in which different compensation functions correspond to different PA characteristics (§0048–§0049).
The additional step of receiving a second (subsequent) set of reference signals and using them to update or adjust the distortion model is a well-known adaptive compensation technique. A POSITA would have been motivated to extend Kwon's system with ongoing model adjustment using subsequently received reference signals for the following reasons: (1) PA distortion characteristics can vary with operating conditions (temperature, bias voltage, frequency, aging), necessitating periodic model updates; (2) Kwon's architecture already provides the model selection and compensation infrastructure needed to implement model adjustments; and (3) using reference signals to update compensation functions is a routine technique in adaptive equalizers and channel estimators in the wireless communications field. The combination involves no more than the application of known techniques to a known system for a predictable improvement in compensation accuracy.
Also, per claim 8, Kwon discloses the terminal-side (OFDM terminal 10) transmitting characteristic identification information in the OFDM signal to enable the base station to select the correct compensation function. See Kwon §0037 ("the OFDM transmitting unit 11 includes one distortion characteristic function corresponding to distortion characteristics of the power amplifier adopted for a current OFDM transmitter among a plurality of distortion characteristic functions in the OFDM signal so as to identify the distortion characteristics of the power amplifier"). This characteristic identification information is precisely "an identifier of a power amplifier distortion model associated with the power amplifier" of Claim 8, and its transmission to the base station constitutes transmitting a first indication that enables the base station to perform its distortion compensation — functionally equivalent to indicating support for distortion compensation at the base station.
With respect to the "machine learning based" qualifier: as established in the Claim 1 rejection analysis, Kwon's preset compensation functions read on "pre-trained machine learning model" under the broad functional scope of Claim 1. Accordingly, transmitting the PA identifier to enable Kwon's distortion compensation scheme is transmitting a first indication supporting that scheme.
With respect to the "second indication indicating to operate the power amplifier according to a reduced power backoff and/or reduced maximum power reduction": as analyzed under Claim 6 above, Kwon expressly motivates the concept of reducing PA operating constraints (§0071) as a direct consequence of its distortion compensation system. A POSITA would have found it obvious for the base station to transmit a corresponding signal to the terminal confirming this relaxed operation, as such feedback signaling is a standard component of uplink power control in wireless communication standards (e.g., 3GPP LTE/NR power control mechanisms). The two-way nature of the capability/permission exchange (terminal signals PA identity → base station signals relaxed operation back to terminal) represents the obvious and natural system-level implementation of Kwon's distortion compensation design goal.
Regarding claim 9, Kwon discloses the OFDM terminal 10 transmitting OFDM signals to the base station 100 via its power amplifier 12. See Kwon §0036; FIG. 1. The standard operation of a scheduled wireless system — in which the base station issues an uplink grant and the terminal transmits uplink data in response — was a fundamental and universally known feature of cellular wireless systems (e.g., LTE PUSCH scheduling) at the priority date. Combining this standard uplink scheduling mechanism with Kwon's disclosure that the terminal operates at relaxed power constraints (reduced power backoff / MPR) following capability exchange would have been entirely routine for a POSITA and would produce the predictable result of an uplink transmission occurring under the relaxed power conditions already established by Claims 6 and 8. No inventive step beyond ordinary skill is required.
Regarding claim 10, Kwon discloses that the distortion compensation unit identifies distortion characteristics of "various power amplifiers" (§0020, §0026, §0048) and that these distortion characteristics (type A, B, C) are intrinsic properties of each PA model. The physical operating conditions that govern PA distortion characteristics — namely frequency band, temperature, power supply voltage, and bias voltage — are fundamental and well-established PA behavioral parameters in RF engineering. These parameters directly determine the shape of a PA's distortion characteristic function f(x), as any RF engineer would know.
Kwon's classification of PAs by distortion type (A/B/C) encompasses, at minimum, the operating-condition-dependent distortion profiles. The specific enumeration of frequency band, temperature, supply voltage, and bias voltage in Claim 10 is an obvious and non-inventive specification of the well-known physical parameters that characterize PA operating conditions. Including these parameters in the identifier transmitted by the terminal is a straightforward implementation of Kwon's characteristic identification information concept using known PA characterization parameters. A POSITA would have immediately recognized these as the relevant operating conditions for PA distortion identification and would have found their inclusion obvious.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HAFIZUR RAHMAN whose telephone number is (571)270-0659. The examiner can normally be reached M-F: 10-6.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Andrea Lindgren Baltzell can be reached on (571) 272-1769. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
/HAFIZUR RAHMAN/Primary Examiner, Art Unit 2843.