Obviousness

Obviousness Analysis of US Patent 9,235,259 under 35 U.S.C. § 103

This analysis evaluates whether the claims of US Patent 9,235,259 would have been obvious to a Person Having Ordinary Skill in the Art (PHOSITA) at the time of the invention. Under 35 U.S.C. § 103, an invention is unpatentable if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains.

A PHOSITA in this context would be an engineer or computer scientist with experience in digital signal processing for embedded or mobile systems, particularly with knowledge of audio processing techniques and power management for resource-constrained devices.

The central claims of US 9,235,259 revolve around a two-step "coarse-to-fine" method for detecting an audio "tick." A computationally inexpensive coarse processor identifies a potential tick, which then triggers a more resource-intensive fine processor to confirm the event, often by comparing it to a pre-trained, device-specific acoustic fingerprint. This approach is explicitly designed to conserve processing power and battery life on mobile devices.

Based on the provided prior art, a strong case for obviousness can be constructed by combining several references.

---

Primary Combination of References

A compelling argument for obviousness arises from the combination of EP1978508A1 (Sony) and US20090153342A1 (Sony Ericsson).

• EP1978508A1 (Sony) teaches the core signal processing method claimed in '259. The Sony reference discloses a two-stage method for detecting musical beats:

  1. A coarse step that detects "large instantaneous peaks in the time-series waveform."
  2. A fine step that performs a more detailed analysis on a spectrogram using an FFT (Fast Fourier Transform) to accurately determine the beat position.

  This is, in essence, the "coarse-to-fine" processing architecture claimed in '259. The patentee's attempt to distinguish this by limiting its application to "musical rhythm" is a weak argument. A "tick" from a device tap is simply a singular, aperiodic impulsive sound event, functionally equivalent to a single "beat" from a drum hit described in the Sony reference. A PHOSITA would readily recognize that the same signal processing technique used to identify a beat could be applied to identify a tick.

• US20090153342A1 (Sony Ericsson) provides the clear motivation for this application. It describes the problem domain and context: using a physical "tap" between devices to initiate communication or pairing ("tap-to-pair"). This reference establishes the goal of reliably detecting a tap sound on a mobile device for a specific function.

Motivation to Combine:

A PHOSITA tasked with implementing the "tap-to-pair" functionality described in Sony Ericsson ('342) on a battery-powered mobile device would immediately confront the dual challenges of detection accuracy and power consumption, the very problems the '259 patent claims to solve. When seeking an efficient method for detecting a short, impulsive audio event (the tap), the PHOSITA would look to known signal processing techniques. The coarse-to-fine method disclosed in Sony ('508) presents a direct and known solution for efficiently processing such audio events.

The motivation to combine these references is therefore strong and direct: to apply a known, computationally efficient audio event detection method (from Sony '508) to solve the specific problem of tap detection for device interaction (from Sony Ericsson '342). This combination would have rendered the core method of Claims 1, 5, 20, and 22 obvious, as it would have been a straightforward application of a known technique in an analogous field with a predictable result.

---

Secondary Combination and Device-Specific "Fingerprint"

An alternative argument can be made by combining US20030132950A1 (Surucu) with EP1978508A1 (Sony).

• US20030132950A1 (Surucu) teaches using the specific acoustic signature of a tap on a device's housing as a form of user input. Crucially, it discloses that the system can learn and recognize different types of taps based on their acoustic characteristics. This directly teaches the concept of training a system to recognize a specific acoustic event tied to the device itself.

• EP1978508A1 (Sony), as before, provides the efficient two-stage processing architecture.

Motivation to Combine:

A PHOSITA starting with the tap-input system from Surucu ('950) and seeking to improve its efficiency for a mobile device would be motivated to find a less computationally demanding processing method. The coarse-to-fine method from Sony ('508) provides an obvious path to achieving this efficiency.

Furthermore, this combination renders the "device-specific" and "previously trained" limitations of Claim 1 and others obvious. Surucu teaches training a system on the acoustic characteristics of a tap. A PHOSITA would understand that the acoustic response of a physical tap is inherently "characteristic for the device" due to the device's unique physical construction, materials, and microphone placement. Therefore, combining Surucu's concept of a trainable recognizer with Sony's efficient detection algorithm would lead directly to a system that uses a coarse-to-fine method to compare an incoming tap sound against a "previously trained set of reference properties that are characteristic for the device." There is no inventive step in recognizing that a learned tap signature on a device is, by its nature, specific to that device.

Conclusion

The independent claims of US Patent 9,235,259 appear to be obvious in light of the cited prior art. The core inventive concept—a two-stage, coarse-to-fine processing method to save power—is directly taught by EP1978508A1 for the analogous purpose of detecting audio impulses. The motivation to apply this known method to the problem of "tap-to-pair" is clearly provided by references like US20090153342A1. Finally, the concept of using a trained, device-specific acoustic signature is taught by US20030132950A1. A person having ordinary skill in the art would have been motivated to combine these teachings to arrive at the claimed invention with a reasonable expectation of success.