Wearable digital olfactory experience device, system and method

Abstract

The invention discloses a wearable digital olfaction experience device, system and method. The core purpose of the wearable digital olfaction experience device, system and method is to determine that olfaction is a brand-new and high-fidelity digital information medium. The device is positioned as a purely "olfactory display" that includes a nose wearing structure, an olfactory stimulus generating unit, a wireless communication module, and a processing circuit. The core of the olfactory stimulus control device is that the processing circuit adopts pure open-loop control logic and is configured to receive timestamps and parameters defined in an external digital control instruction sequence and drive the olfactory stimulus generation unit to perform high-fidelity micro-dose release strictly according to the timestamps and parameters. An accurate and synchronous olfactory dimension is provided for immersive experience such as VR / AR, a brand-new and private information expression mode is opened up, and a complete technical link from content creation to sharing is constructed around the device.

Description

Technical Field

[0001] This invention relates to the fields of human-computer interaction, intelligent wearable devices, and immersive computing technology. Specifically, this invention provides a wearable digital olfactory experience device, system, and method. The device is a wearable nasal device capable of precisely and privately converting abstract, time-series digital information into a realistic olfactory experience. The method and system are built around this device, covering the complete technical chain from the generation, editing, and sharing of olfactory content to the final experience, aiming to establish olfaction as a new, high-fidelity digital information medium, providing a new dimension and paradigm for information transmission, sensory experience enhancement, and user state guidance. This invention particularly focuses on a purely open-loop control architecture, which achieves precise arrangement and enrichment of the user's sensory experience through feedforward guidance rather than closed-loop feedback, thus fundamentally different from various biofeedback devices in the existing technical field. Background Technology

[0002] Since the birth of computers, the way humans interact with the digital world has undergone several major paradigm shifts. From early command-line character interfaces to graphical user interfaces (GUIs), and now to the mobile computing era defined by smartphones and multi-touch screens, the sensory channels we rely on for information interaction remain essentially highly concentrated on sight and hearing. Whether browsing web pages, watching videos, or receiving notifications, the presentation and reception of information are almost entirely dominated by "seeing" and "hearing." Although vibration motors provide limited tactile feedback, their information bandwidth is extremely low, and their expressive capabilities are patterned, far from being a rich channel for information transmission.

[0003] However, this interaction paradigm, centered entirely on audiovisual perception, is facing increasingly severe challenges in today's information-explosive growth. The most pressing issues are the dilemmas of "information overload" and "attentional intrusion." Our visual and auditory channels are constantly bombarded with notifications, pop-ups, and alerts. These information delivery methods are inherently "public" and "intrusive," forcibly extracting our cognitive resources, leading to a sharp increase in "cognitive load" and widespread "digital fatigue." For example, a loud alert during a meeting can interrupt everyone's train of thought; a news notification popping up while reading can instantly disrupt a flow state. Therefore, there is an urgent need in this field for a new, quieter, more private digital information delivery method that does not occupy core audiovisual cognitive channels.

[0004] Secondly, with the rapid development of immersive computing technologies such as Virtual Reality (VR) and Augmented Reality (AR), we are striving to build a "metaverse" that infinitely approximates physical reality. While today's head-mounted displays can provide a high degree of realism in terms of audiovisual experience, a truly immersive virtual world must encompass a complete sensory dimension. Smell, as the sense most deeply connected to memory and emotion, remains a crucial "final piece" in building the ultimate immersive experience. Existing VR olfactory solutions, such as bulky external modules or environmental generators, suffer from significant shortcomings in portability, privacy, response speed, and flexibility in scent switching. Their fundamental bottleneck lies in the inability to achieve precise control over olfactory information in both the temporal and spatial domains.

[0005] Furthermore, smell plays a vital role in daily life, evoking memories, creating atmosphere, and forming the core of "flavor." However, this rich, profound, and emotionally valuable sensory dimension remains largely unexplored territory in the digital technology landscape. The root cause lies in the lack of a miniaturized, consumer-grade hardware device capable of precisely, controllably, privately, and imperceptibly converting digital signals into a genuine olfactory experience. Without such an "olfactory display," the digital world can never truly "smell."

[0006] In summary, existing technologies have significant gaps: a lack of non-audiovisual, private, and low-intrusion information channels; a lack of miniaturized, highly responsive personalized olfactory devices to complement immersive experiences; and a lack of a complete technological ecosystem to support the creation, sharing, and experience of olfactory content. This invention is proposed against this backdrop.

[0007] Furthermore, almost all current devices attempting to combine olfaction with health monitoring employ a "closed-loop feedback control" logic. This involves monitoring users' physiological indicators through biosensors and automatically releasing fragrance when thresholds are reached. This technological approach has profound flaws: First, it carries the risk of pseudoscience and low reliability. Consumer-grade sensors have limited accuracy, and the algorithms for interpreting physiological indicators and psychological states are still immature, easily leading to erroneous judgments and disrupting the core user experience. Second, it fundamentally infringes on user intent and creative freedom. In scenarios such as VR movies or games, the user's physiological responses are an integral part of the experience; automatic "intervention" essentially "pollutes" the creator's narrative rhythm and emotional arc. Third, it raises serious privacy and security concerns. Continuously collecting personal physiological data raises serious questions about data ownership, usage boundaries, and the risk of misuse. Therefore, how to completely break free from the technological constraints of closed-loop feedback is a core technical challenge that urgently needs to be addressed in this field.

[0008] Furthermore, existing technologies completely ignore the "creative" and "ecological" dimensions of olfactory content. All known solutions are limited to providing a few manufacturer-preset scents, leaving users with no choice but to passively accept them. There is a lack of a standardized, user-friendly toolchain to support a broad range of users and content creators in creating, editing, and sharing rich "olfactory works." This lack of a content ecosystem keeps digital olfactory technology at the level of a "functional accessory," preventing it from developing into a new, expressive digital medium. Summary of the Invention

[0009] The main objective of this invention is to overcome the aforementioned defects and deficiencies in the prior art and to provide a wearable digital olfactory experience device, system, and method. This invention aims to open up the possibility of olfaction as a completely new human-computer interaction channel and transform olfaction from a passive, environmental sensory input into an active, information-based, precisely programmable, and controllable digital medium.

[0010] To achieve the above objectives, a first aspect of the present invention provides a wearable nasal olfactory experience device. The core technical feature of this device is that it is functionally and strictly positioned as an "olfactory display" focused on executing external commands. Specifically, the device includes a main structure that can be stably worn in the user's nasal region (e.g., inside the nose, nasal septum, nasal alae, and bridge of the nose), internally integrating an odor generation unit, a wireless communication module, and a processing circuit as the control core. Unlike existing technologies, a key and essential limitation of the processing circuit of the present invention is that its decision-making logic is completely open-loop. It is configured to receive a sequence of digital control commands containing precise time-series information from an external control device (such as a smartphone, PC, or VR headset) via the wireless communication module. This "open-loop control logic" is a core feature with strict technical limitations, explicitly pointing to a unidirectional, deterministic information execution flow: external command -> device -> odor output. Its essential characteristics are reflected in: the uniqueness of the decision source, with all parameters released by the sense of smell originating solely from external instructions; the fidelity of execution, as the device is a "high-fidelity actuator" that unconditionally reproduces instructions; and the unidirectionality of information flow, as the device, during its decision-making process, does not collect, analyze, or utilize any unconscious physiological parameters (such as heart rate, skin conductance, and electroencephalography) signals used for closed-loop feedback. It is important to emphasize that this invention does not exclude the integration of sensors for various non-physiological feedback functions, such as sensors for device status monitoring (e.g., temperature, battery level), environmental perception and compensation (e.g., respiratory airflow, humidity, noise), or conscious user input and interaction (e.g., touch, button press, posture control), because the signals collected by these sensors do not originate from the user's unconscious physiological state. The fundamental difference lies in the fact that any fine-tuning based on these signals ultimately aims to more perfectly reproduce the original instruction's target (e.g., timestamp, intensity), rather than altering the target itself. These fine-tunings are essentially optimizations of the execution path to overcome physical environmental disturbances. Therefore, this logic, at the strategic level, is always an "open-loop" execution faithful to the intent of the instruction, which is completely consistent with the core spirit of claim 1. The experience guidance achieved by this invention is a novel paradigm of "feedforward guidance." Olfactory information, as part of the preset content, works in synergy with audiovisual information to proactively bring users into the emotional state desired by the creator, thereby respecting the creative intent, protecting user privacy, and avoiding pseudoscience traps.

[0011] A second aspect of the invention provides a method for controlling a wearable olfactory experience. This method outlines a complete process from content generation to experience: First, on the user interface of an external control device, the user actively selects an olfactory formula containing multi-step timing information, or the formula is automatically generated by an application. Next, the external control device converts this high-level formula into a sequence of digital control instructions containing a series of precisely timestamped, timed release events that the device can directly execute. Subsequently, this instruction sequence is transmitted wirelessly to the device worn by the user. Finally, the device executes one or more timed micro-dose releases strictly according to the timing defined by the instruction sequence, perfectly synchronizing the digital timeline with the real-world olfactory experience.

[0012] A third aspect of the present invention provides a wearable digital olfactory experience system. In its most basic embodiment, the system includes at least: a) a wearable nasal olfactory experience device according to the first aspect of the invention; and b) an external control device (e.g., a smartphone) running an application configured to execute the control method described in the second aspect of the invention. In a more preferred embodiment, the system further includes a server. The server is configured to provide a user-generated content (UGC) community platform that not only provides a visual "digital perfumery" editing interface allowing users to create their own olfactory recipes, but also allows users to upload, store, and share these olfactory creations, and allows other community members to easily download and "one-click replicate" them. In this way, the system defined by the present invention fully covers the entire chain from cloud content ecosystem to terminal control logic to personal wearable execution.

[0013] Compared with the prior art, the present invention has the following beneficial effects:

[0014] It opens up a brand-new non-intrusive information channel: through the transmission of private olfactory information, it effectively alleviates attention fatigue and cognitive load in the digital age.

[0015] It completes the sensory landscape of immersive experiences: providing a crucial olfactory dimension for VR / AR in a lightweight and imperceptible way, enhancing the realism and emotional depth of the virtual world.

[0016] An open ecosystem of olfactory content has been built: by providing a complete software and hardware system from content creation (UGC) to sharing and reproduction, the foundation has been laid for the birth and prosperity of "olfactory narrative," a brand-new digital art form.

[0017] Therefore, this invention has great technical value and broad market prospects in the fields of human-computer interaction, wearable devices, digital entertainment, and assistive technology. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0019] Figure 1 This is a schematic diagram of the structure of an intranasal intelligent olfactory capsule according to an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the structure of a C-shaped open annular (nasal septum type) device according to another embodiment of the present invention.

[0021] Figure 3 This is a flowchart of the active olfactory clearance (AOC) system based on cyclodextrin described in this invention.

[0022] Figure 4 This is the overall system architecture and information flow diagram of the wearable olfactory experience system of the present invention.

[0023] Figure 5 This is an example diagram of the visual editing interface of the "Digital Perfume Stand" in the user terminal application described in this invention.

[0024] Figure 6 This is a schematic diagram of the composition and data flow of the feeding olfactory enhancement system based on an external motion sensor as described in this invention.

[0025] Figure 7 This is a schematic diagram of the process and interface for personalized olfactory experience calibration as described in this invention.

[0026] Figure 8 This is a schematic diagram of the consumable identification and management system in the device described in this invention.

[0027] Figure 9 This is a schematic diagram of the workflow of the "one-click replication" function in the user-generated content (UGC) ecosystem according to an embodiment of the present invention. Detailed Implementation

[0028] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0029] Example 1: Structure and Design of Intranasal Intelligent Olfactory Capsule

[0030] a) Overall structure and "intranasal ring" wearing method

[0031] The main structure of the device is defined as an "open ring structure," which is a broad geometric concept designed to encompass all O-shaped, C-shaped, U-shaped, or non-complete ring shapes that can be worn on the user's nose area in a non-invasive, gentle, and elastic manner. Figure 1 The asymmetrical O-shaped ring shown, thicker at the top and thinner at the bottom, is a highly optimized manifestation of this design. This design allows it to be placed comfortably and stably inside the user's nasal vestibule. The core objective of its design is to allow this micro-ring to be placed laterally or semi-laterally into one side of the nasal vestibule, and to form a stable three-point or multi-point support structure by relying on its own slight elastic tension.

[0032] Ergonomic adaptation: The curvature, channel size, and overall dimensions of this ring structure are determined by parametric modeling and topology optimization after data analysis of the three-dimensional anatomical morphology of the nasal vestibule, resulting in several common sizes. Figure 1 This is just a shape example; the actual shape may need further customization and adjustment based on individual differences.

[0033] Main body 101: This is the larger and more fully formed end of the ring structure. Figure 1 The upper part (of the nasal cavity) is designed to house the device's core electronic components. When worn, this part is designed to rest stably against the upper part of the nasal vestibule.

[0034] Elastic arm 102: This is the thinner end of the ring structure. Figure 1 (Both sides of the middle). It smoothly transitions and connects with the main body 101, together forming an open ring. Its slender design and flexible material selection allow it to produce slight elastic deformation when worn, providing a stable yet extremely gentle elastic tension.

[0035] Surface Engineering and Materials Science: The device employs a two-component injection molding process. Internally, it features a semi-rigid framework (e.g., made of medical-grade polycarbonate PC), while the entire exterior is completely encapsulated in medical-grade liquid silicone (LSR) with extremely low Shore A5-15 hardness. The mold cavity for the outer silicone layer undergoes optical-grade mirror polishing, resulting in a final product with a surface roughness below Ra 0.1 micrometers, achieving ultra-low friction and biocompatible contact with the nasal mucosa.

[0036] Size and weight: Its size is extremely compressed. For example, the maximum unfolded size of the ring does not exceed 15 mm, the diameter at the thickest point does not exceed 6 mm, and the total weight is controlled to less than 4 grams, ensuring that it can be stably accommodated in the nasal vestibule without causing discomfort.

[0037] b) "Dual-channel airflow system":

[0038] 1. Breathing ventilation channel (macro / normally open / passive): In order to fundamentally solve the "blockage" problem that any nasal in-situ device may face, the open ring structure of this embodiment itself constitutes a natural and huge breathing ventilation channel.

[0039] Structure and Function: This channel is the central open area of ​​a C-shaped or U-shaped ring. It does not rely on additional openings, but rather utilizes the inherent openness of the structure itself to provide a low-resistance, passive air passage for the user's macroscopic physiological breathing. Its sole design goal is to minimize breathing resistance to achieve "comfort" and "unawareness" when worn.

[0040] Physical state: Its physical state is always open, allowing bidirectional, continuous or intermittent airflow, perfectly simulating the natural breathing state when not wearing any clothing.

[0041] Fluid dynamics optimization: The cross-sectional shape of the ring (e.g., designed as a streamlined ellipse) and its overall form both require precise computational fluid dynamics (CFD) simulation optimization. The simulation model models the pressure drop and flow velocity distribution at different breathing rates (from 6 L / min at rest to 18 L / min during light activity) within a three-dimensional nasal cavity containing the wearable device. The optimization goal is to ensure that the average increase in breathing resistance due to the wearable device is kept below 15%, thus ensuring that the user does not feel the device's impact on breathing while at rest, talking, or even during light exercise.

[0042] 2. Odor Injection Channels (Microscopic / Normally Closed / Active): Completely separated from the macroscopic main respiratory channel, the main body 101 of the device integrates a set of physically isolated, parallel microscopic odor injection channels. Crucially, each independent channel is connected to a dedicated micro-jet outlet, which together form the outlet array 103. This "one odor, one dedicated channel, one independent outlet" architecture fundamentally eliminates cross-contamination (i.e., crosstalk) between different odors. Each microscopic injection channel in the system is a completely independent microfluidic system controlled by its corresponding built-in microvalve (such as the thermo-actuated microvalve corresponding to the "flexible microheater array" 106). The sole function of each channel is as an active actuator, unidirectionally ejecting a micro-dose of olfactory material (gas / aerosol) from the corresponding micro-chamber of the "multi-odor sac" 105 in a controlled jet form from its dedicated outlet.

[0043] A key design feature of this invention is its highly controlled, active "jet" delivery method. In a preferred design, each channel is itself designed as a miniature curved conduit with a specific curvature. This design efficiently redirects the original, vertically upward pressure pulse generated by the microheater within the corresponding odor sac into an airflow directed toward the nasal cavity at an optimal jet angle.

[0044] This active "jetting" method uses actuators (such as microheaters) to propel minute amounts of olfactory liquid at high speed in the form of microdroplets, giving them initial momentum. Ultimately, this microdroplet, carrying a high concentration of olfactory information, rapidly evaporates in the air of the nasal vestibule—its temperature rapidly dropping to room temperature due to heat absorption during evaporation, ensuring absolute safety—forming a highly concentrated cluster of olfactory information in space and time. This allows it to actively traverse the critical distance between itself and the olfactory epithelial cells the instant the command is issued (within milliseconds). This mechanism fundamentally guarantees precise control over the timestamps of olfactory events and is the physical basis for constructing high-fidelity, fast-paced dynamic olfactory narratives.

[0045] The reason for choosing this active, directional spraying method, rather than other seemingly simpler reagent release methods (such as passive evaporation, heating evaporation, or ultrasonic atomization), is based on the need for high-fidelity transmission of olfactory information:

[0046] In contrast to passive evaporation / diffusion: this method is completely uncontrollable, cannot be "turned on" or "turned off," and therefore cannot be synchronized with the time of digital content. It can only produce a continuous, stable background odor, and cannot carry dynamically changing information.

[0047] Compared to conventional heating evaporation, this method suffers from insurmountable thermal inertia. Whether heating to "turn on" the odor or cooling to "turn off" it, there is a severe delay on the order of seconds. This makes it completely unsuitable for the millisecond-level refresh rate requirements of modern digital media, and thus unable to achieve clear and rapid odor switching.

[0048] In contrast to ultrasonic atomization: although atomization can produce tiny droplets, the resulting "aerosol cloud" lacks sufficient initial momentum and directionality. These aerosols remain suspended near the nozzle, and whether they are ultimately perceived by the user still highly depends on the user's unpredictable, passive inhalation behavior. This reintroduces significant uncertainties in time and dosage, fundamentally violating the core principle of "precise control" of this invention.

[0049] Therefore, the active "spraying" method used in this invention is the only technical approach that can simultaneously meet all of the following stringent conditions:

[0050] Timing precision: By imparting initial momentum to molecules, it eliminates the dependence on passive breathing by the user and achieves millisecond-level command response.

[0051] Dosage repeatability: Each spray is a discrete, quantified event, ensuring a stable output of olfactory intensity.

[0052] Spatial targeting: The design of micro-curved channels allows for the precise delivery of a limited number of olfactory molecules to the most efficient olfactory sensing areas.

[0053] In conclusion, the choice of "jetting" was not an arbitrary engineering decision, but a fundamental technological decision necessary to transform olfaction into a high-bit-rate, high-fidelity digital information medium.

[0054] c) Internal component layout and selective odor release mechanism

[0055] 1. Overview of Internal Component Layout: The components inside the device are stacked in layers to maximize the use of the irregular internal space.

[0056] Energy and control core: An extremely thin, irregularly shaped rigid-flex PCBA 104 and a miniature arc battery 107 together constitute the electronic and energy core of the device, which is placed in the elastic arm of the ring structure.

[0057] Olfactory stimulation generation unit: This is a pre-assembled, highly integrated functional unit that occupies the upper half of the annular structure (main body 101). The module is precisely laminated from the lower "multi-odor sac" 105 and the upper "flexible microheater array" 106.

[0058] 2. The "multi-odor sac" 105 is a physical carrier for storing multiple olfactory information, and the following engineering solutions can be used:

[0059] Multilayer composite film material structure: The odor capsule wall is not a single material, but a high-performance multilayer composite film, manufactured through co-extrusion or composite processes. A typical layered structure may include:

[0060] Inner layer (contact layer): Made of highly chemically inert materials, such as modified polyethylene (PE) or polypropylene (PP). This layer must be compatible with various complex fragrance and flavor formulations, and must not undergo any leaching, adsorption, or chemical reaction to ensure the purity of the scent.

[0061] Intermediate layer (barrier layer): This is crucial for ensuring the long-term stability of consumables. Materials with extremely low gas permeability are used, such as ethylene-vinyl alcohol copolymer (EVOH) or metallized polyester (e.g., aluminized PET). EVOH has excellent barrier properties against most small organic molecules (fragrance molecules) and oxygen, effectively preventing odor leakage, volatilization, and oxidative deterioration, ensuring the stable performance of consumables throughout their shelf life.

[0062] Outer layer (structural and welded layer): Materials with good mechanical strength and heat-sealing properties, such as nylon (PA) or polyester (PET), are selected. This layer provides the necessary toughness and puncture resistance to the odor sac and allows for precise shaping and sealing via hot pressing or ultrasonic welding.

[0063] Fabrication process of microcavities: Multiple isolated microcavities can be fabricated on a multilayer composite film using the following precision processes:

[0064] Hot-press welding: Two composite films are hot-pressed together using a heated mold with finely textured linear raised lines. The raised parts of the mold fuse the films together, forming a sealed "weld" that defines multiple independent, arc-shaped micro-cavities between the films. The width and temperature of the weld need to be precisely controlled to ensure seal strength and avoid damage to the barrier layer.

[0065] Laser welding: Using a high-precision laser beam, a composite thin film is scanned along a preset path. The laser energy is absorbed by the material, achieving localized, non-contact melting and welding. Laser welding can achieve finer and more complex weld paths, making it suitable for manufacturing irregularly shaped cavities.

[0066] Microfluidic Independent Channels and Outlet Array: Each microchamber has an extremely fine outlet that connects directly to a completely independent microfluidic channel. These channels are arranged in parallel, each extending from its corresponding odor microchamber to a dedicated micro-jet nozzle located in the outlet array 103. This design physically eliminates any form of channel convergence, thereby fundamentally eliminating the risk of odor crosstalk caused by "dead volume." The design of each independent channel has been optimized using microfluidic simulations to minimize its internal flow resistance and surface area, thereby minimizing odor molecule residue after a single jet.

[0067] 3. The “flexible micro heater array” 106 is the core actuator that converts digital electrical signals into precise thermal pulses, and is the key to achieving high-fidelity olfactory rendering.

[0068] Substrate and Structure: Array 106 is fabricated on an extremely thin (e.g., 25 μm) flexible polyimide (PI) substrate. PI material is chosen as an ideal substrate due to its excellent high-temperature resistance (withstanding instantaneous high temperatures >300°C), superior electrical insulation, and good flexibility.

[0069] Fabrication and materials of micro heating elements: Each individual micro heating element on the array can be fabricated on a PI substrate using the following mature microfabrication processes:

[0070] Thick film technology: A special conductive heating paste is printed on a predetermined position on a PI substrate through a high-precision screen printing process, and then cured at high temperature to form a resistance heating element.

[0071] Heating materials: Ruthenium-based resistive paste or carbon nanotube conductive ink can be used. Ruthenium-based paste is a mature thick-film resistive material with a stable temperature coefficient of resistance (TCR) and high reliability. Carbon nanotube ink has a faster electrothermal response and higher energy efficiency.

[0072] Wire material: The wires connecting the heating element are printed with silver paste, which has better conductivity.

[0073] Thin-film technology: For applications requiring higher precision and consistency, thin-film processes from the semiconductor industry can be employed. A metal or alloy thin film (such as platinum (Pt), tungsten (W), or nickel-chromium alloy (NiCr)) is deposited on a PI substrate using magnetron sputtering or vacuum evaporation. Then, photolithography and etching processes are used to precisely define the serpentine or linear geometry of each micro-heating element. Thin-film technology can produce heaters that are smaller, have better resistance consistency, and faster response times.

[0074] Electrical Connections and Packaging: Each heating element on the array has an independent lead, which converges at the edge of the array to form a connection area (typically a gold finger or pad). This connection area is thermo-bonded with anisotropic conductive adhesive (ACF) or connected to the drive circuitry on the main PCBA 104 via a flexible ribbon cable connector. Finally, the entire heater array (except for the working surface in contact with the odor sac) is encapsulated and protected with an insulating cover film to prevent short circuits and electrochemical corrosion.

[0075] 4. Overall Workflow

[0076] When the System-on-Chip (SoC), the control core located on PCBA 104, receives a digital instruction, its firmware generates a high-precision pulse width modulation (PWM) signal based on the intensity parameters defined in the instruction. This signal is amplified by the drive circuit to form a high-energy electrical pulse lasting only a few microseconds, which is then precisely applied to the selected heating element in the "flexible microheater array" 106.

[0077] Due to the extremely low heat capacity of the micro-heating element, its temperature follows Joule's law, increasing by more than 10 kJ / m² within 1-3 microseconds. 8 The rate of heat surges to over 300°C in K / s. This intense heat pulse penetrates the tightly fitted, micron-thin composite membrane wall through an efficient heat conduction path, instantly heating the extremely thin layer of olfactory liquid in the corresponding microcavity that is in contact with its inner wall to a metastable "superheated" state.

[0078] This extreme overheating state immediately triggered explosive nucleo-boiling, in which a tiny steam bubble formed and expanded rapidly in less than a microsecond. The explosive expansion of this bubble instantly generated a local high-pressure pulse of up to tens of atmospheres within the nanoliter space confined by the microchamber.

[0079] This powerful pressure wave is the microscopic actuation source of the entire system. Acting on the remaining liquid in the chamber, it forces a fine jet of liquid to be violently ejected at speeds up to 15 m / s from its dedicated micro-nozzle (one of the outlet array 103) according to the principles of fluid dynamics. This high-speed liquid jet breaks apart during its flight due to surface tension, forming one or more highly uniform microdroplets at the nanoliter level.

[0080] The electrical pulse stops the instant the droplet ejection is complete, the heater cools rapidly, and the vapor bubbles in the chamber collapse, preparing for the next ejection cycle. Finally, this microdroplet carrying a high concentration of olfactory information rapidly evaporates in the air of the nasal vestibule, forming a highly concentrated cluster of olfactory information that is captured by the user's breathing airflow, thus completing a precise, clear, and truly high-fidelity digital olfactory experience.

[0081] d) Precision drive of microheaters and physical principles of micro-dose injection

[0082] The core physical principle of the "flexible microheater array" described in this invention, which enables precise micro-dose jetting, is highly similar to the mature and widely used thermal inkjet printing technology. It does not simply heat the liquid to boiling; instead, it applies a precisely controlled, instantaneous high-energy electrical pulse to heat an extremely thin layer of liquid within the microcavity to a "metastable superheated" state far exceeding its boiling point within microseconds (μs), thereby triggering an explosive phase transition (nuclear boiling). The pressure of the generated vapor bubbles then propels the microdroplets out at high speed.

[0083] 1. The role of PWM signals – constructing instantaneous high-energy pulses:

[0084] In this context, the application of PWM (Pulse Width Modulation) is not in the traditional sense of continuously adjusting average power (such as adjusting LED brightness), but refers to its core logic—precisely controlling the total energy injected into the microheater by precisely controlling the width (duration) of a single electrical pulse.

[0085] Pulse width: This is the most critical control parameter. It directly determines the total injected energy (E∝V). 2 *t / R). The pulse width is typically in the range of 1 to 10 microseconds. If the pulse width is too short, the energy is insufficient to bring the liquid to a metastable superheated state; if the pulse width is too long, it may lead to energy dissipation, reduced efficiency, or even "film boiling," which in turn hinders effective jetting.

[0086] Pulse amplitude: The voltage of the pulse determines the heating rate or power (P = V). 2 / R). Higher voltage means a faster rate of temperature rise. To achieve "metastable superheating" rather than slow, conventional boiling, the rate of temperature rise must be extremely high, typically requiring 10. 8 On the order of K / s.

[0087] Pulse frequency: The frequency determines the repetition rate of the spray, which is the "olfactory refresh rate" in this invention or the pulse density used to simulate continuous odors. Its upper limit is limited by the "refill time" of the microchamber, that is, the time it takes for the vapor bubble to collapse and for new liquid to replenish it through capillary action after the last spray, which is usually several hundred microseconds.

[0088] 2. Metastable superheating and vapor bubble pressure:

[0089] When the aforementioned high-energy pulse is applied to the microheater, the temperature of the liquid film adhering to its surface instantly surges to approximately 300-400°C. Due to the extremely short heating time (on the order of microseconds), the liquid does not have time to form conventional bubble nuclei, thus entering a metastable superheated state. This state is extremely unstable and can instantly form a vaporization nucleus at some point on the heater surface or on impurities inside the liquid, triggering explosive boiling.

[0090] This process creates a microbubble with pressures reaching tens of atmospheres. The rapid expansion of this bubble acts like a micropiston, propelling a portion of the liquid within the microchamber out of the microjet nozzle at high speed (typically 5-20 m / s), forming one or more microdroplets with volumes ranging from picoliters (pL) to nanoliters (nL).

[0091] 3. Functional relationships – empirical calibration rather than simple formulas:

[0092] The specific functional relationship between PWM parameters (pulse width, voltage) and the final results (droplet volume, velocity, vapor bubble pressure) is an extremely complex multivariable nonlinear problem involving thermodynamics, fluid mechanics, and materials science. It does not have a universal analytical solution, but rather is highly dependent on:

[0093] Liquid properties: Different fragrance formulations have different viscosity, surface tension, boiling point, specific heat capacity, etc.

[0094] Hardware geometry: the size and resistance of the microheater, the volume of the microchamber, and the shape and diameter of the nozzle.

[0095] Material properties: thermal conductivity of heater material, substrate material, and cover layer, etc.

[0096] Therefore, in practical engineering implementation, this functional relationship is obtained through experimental calibration. For each odor consumable, a series of optimal drive waveform parameters are calibrated in the laboratory and stored in the device firmware's lookup table (LUT). When the upper-layer application issues a command such as "intensity 0.7", the firmware queries the LUT for that odor and selects the corresponding pulse width and voltage parameters to execute the spray. This method ensures consistent and repeatable spraying results for different odors.

[0097] Those skilled in the art should understand that the above-mentioned physical principles of thermally actuated injection, the key control parameters (pulse width, amplitude) of PWM pulse as the core method, and the given typical numerical ranges (such as pulse width of 1-10 microseconds, amplitude of 10) are not the core principles of thermally actuated injection. 8 The temperature rise rate (K / s) and the engineering implementation method using experimental calibration and lookup table (LUT) together constitute a clear and complete technical implementation framework.

[0098] Specifically, those skilled in the art can, based on the device structure disclosed in this invention and using the microsecond-level pulse width and high-energy pulse as the starting point for debugging, systematically test the effect of different driving waveforms on the spraying effect of specific olfactory liquids (considering their unique viscosity and surface tension) through conventional experimental methods (e.g., using high-speed cameras or droplet analyzers), thereby establishing a dedicated, optimized drive parameter lookup table (LUT) for each consumable.

[0099] This debugging process is standard engineering practice in the development of thermal inkjet printheads or similar microfluidic actuators and does not constitute an excessive experimental burden. Therefore, the disclosure of this invention is sufficient to enable those skilled in the art to stably achieve the precise and repeatable micro-dose olfactory information release required by this invention under different hardware geometries and liquid formulations, and to perform subsequent system maintenance and performance tuning.

[0100] Example 2: Design of a C-shaped open-ring olfactory experience device

[0101] This embodiment aims to describe an optional physical form of a wearable olfactory experience device. Unlike the "intranasal" design of Embodiment 1, this embodiment adopts an "external nasal surround" wearing method (including septal nasal type, wing-nose type, etc.), which aims to provide users with different wearing experience options and provide an extended platform for integrating more complex functions (such as multiple scents, longer battery life, stronger wireless communication, etc.).

[0102] The most fundamental structural difference between this embodiment and Embodiment 1 is that, as a device worn outside the nostrils and surrounding the nasal septum, it lacks a "main breathing ventilation channel" that runs through the entire device body for physiological respiration. Therefore, to avoid physical obstruction of normal airflow at the user's nostrils, the device in this embodiment must be extremely thin in its structural design. However, regarding the core multi-odor release mechanism, this embodiment adopts the exact same architectural design and actuation principle as Embodiment 1, based on "one odor, one dedicated channel, and one independent outlet," and a fixed micro-heater array.

[0103] a) Wearing method and slim ergonomic design

[0104] Please see Figure 2 This figure is a schematic diagram of the structure and function of an external nasal septum device. In terms of overall structure, the device of this embodiment is designed as a delicate C-shaped or U-shaped open ring structure, which is specifically designed to be "hung" or "clamped" on the user's nasal septum (columella) in a non-invasive manner (the same can be applied to the alar type).

[0105] Wearing method: The C-shaped opening of the device allows it to be easily slipped onto the columella from below, hanging stably by its own slight clamping force or gravity. Its two arms extend upwards, fitting snugly against both sides of the nasal septum, with the ends positioned directly below or slightly inwards from the nostril entrance.

[0106] Shape Design: Since the two arms of the device are located precisely at the entrance of the nasal airflow passage, their thickness directly affects the smoothness of breathing. To avoid any perceptible airflow obstruction or "blockage," the cross-sectional thickness of the two arm tips is strictly controlled to a very small range, for example, no more than 4 mm. The entire device's shape has been optimized for fluid dynamics to create a smooth contour that minimizes disturbance to nasal airflow. To ensure a "feel-free" wearing experience and freedom of speech and facial expression, the total vertical height from the lowest point of the device to its two arm tips has been optimized to a length not exceeding 12 mm and an overall maximum thickness not exceeding 10 mm.

[0107] Materials and Aesthetics: Utilizing the same dual-injection molding process of "rigid inside, soft outside," it features a rigid biocompatible polymer inner skeleton and an ultra-soft, skin-friendly silicone outer layer, balancing durability and wearing comfort. Its minimalist and elegant ring shape makes it resemble a cutting-edge technological accessory when worn, lowering the psychological barrier to wearing it in public. If you find the nose-external style unattractive for everyday wear, you can wear it over a face mask.

[0108] b) Modular partitioning layout of internal components

[0109] 1. Olfactory stimulus generation unit (located at the ends of both arms of the C ring):

[0110] At the two upper arms of the C-shaped ring, a completely independent olfactory stimulation generation unit is integrated. Each module is precision-laminated from a lower multi-odor sac 105 and an upper flexible microheater array 106. This dual-arm layout allows the device to carry twice the number of odor types, or to provide different stereoscopic olfactory experiences for the left and right nostrils.

[0111] At the very top of each olfactory stimulation generation unit is an outlet array 103. The positions of these micro-nozzles are precisely designed to face the inhalation airflow path of the user's nostrils to ensure that the ejected olfactory information clusters can be efficiently captured.

[0112] A key innovative design feature is that the olfactory stimulation generation unit is pluggable and replaceable. As shown in the figure, each module is electrically connected and physically secured to the main power and control module, which contains the micro-arc battery 107 and PCBA 104, via a sliding rail fixing device (or other similar mechanical or magnetic structure).

[0113] This modular design cleverly defines the entire olfactory stimulation generation unit (including the odor consumable itself and the sophisticated microheater actuator) as a single replaceable consumable unit. This offers a significant advantage: when a user wishes to experience a completely new set of scents, there's no need for cumbersome internal odor capsule replacements; instead, the entire olfactory stimulation generation unit can be directly replaced. If a user owns multiple olfactory stimulation generation units pre-installed with different themes (scents), it enables a variety of scent combinations and pairings, greatly expanding the device's application scenarios and enriching the user experience.

[0114] 2. Energy module (located in the vertical arm of ring C):

[0115] Within the vertical arm connecting the bottom PCBA and the top olfactory engine, two slender, miniature arc-shaped batteries 107 are housed. This distributed energy layout not only makes full use of the internal space of the arm but also helps to achieve the weight balance of the entire device.

[0116] 3. Control core module (located in the bottom arc area of ​​ring C):

[0117] The device's "brain"—PCBA 104—is located in the bottom arc of the C-shaped ring. PCBA 104 integrates all the core computing and communication units, including the SoC (System-on-a-Chip), wireless communication module, and expansion sections for connecting additional sensors. For example, a capacitive touch sensor can be added to perform advanced gestures such as "single," "double," "long press," or "swipe," allowing users to actively command preset functions, such as temporarily enhancing olfactory intensity or activating standby mode. It is electrically connected to the batteries 107 on both sides and the olfactory stimulation generation unit at the top via flexible circuitry integrated within the ring-shaped frame.

[0118] c) Equivalent implementation of core technology principles

[0119] Despite the differences in physical form and component layout, this embodiment fully replicates the advanced technical principles detailed in Embodiment 1 in its core multi-odor release mechanism.

[0120] 1. The "one odor, one channel, one outlet" architecture: Each multi-odor sac 105 located in both arms is also internally divided into multiple independent microchambers. Each microchamber is connected to a dedicated micro-jet port in the outlet array 103 via a dedicated, physically isolated microfluidic channel. This design also physically eliminates any form of channel convergence, thereby fundamentally eliminating the risk of odor crosstalk caused by "dead volume" and ensuring the highest purity of olfactory information.

[0121] 2. Realization of "Multiple Odor Sacs" and "Flexible Microheater Array":

[0122] The multi-odor sac 105 is manufactured using the same multilayer composite thin film material (including an EVOH barrier layer) and precision thermoforming or laser welding processes detailed in Example 1. The flexible microheater array 106 is also manufactured using mature microfabrication processes such as thick-film technology (screen printing) or thin-film technology (sputtering and photolithography) on a flexible PI substrate. Each microheating element is precisely aligned with a microchamber of the underlying multi-odor sac 105.

[0123] 3. Overall workflow of microsecond-level actuation and nano-level jetting: The overall workflow of this embodiment is completely consistent with that of Embodiment 1. When the SoC on PCBA 104 receives a digital instruction, its firmware generates a high-precision PWM signal based on the intensity parameters. This signal is amplified by the drive circuit to form a high-energy electrical pulse lasting only a few microseconds, which is applied to the selected element of the microheater array 106.

[0124] The component temperature exceeds 10 8 The rate of K / s increases sharply, triggering explosive nucleo-boiling through thermal conduction on the inner wall of the liquid in the corresponding micro-chamber of the lower multi-odor sac 105. The resulting local high-pressure pulse ejects a nanoliter-level liquid jet at a speed exceeding 10 m / s from its dedicated micro-nozzle located in the outlet array 103, forming a highly concentrated olfactory information cluster in space and time.

[0125] In summary, the C-ring design of this embodiment not only provides a different wearing aesthetic and interaction method, but also demonstrates a significant platform advantage over the "intranasal capsule" of Embodiment 1 in terms of core functionality. Firstly, regarding multi-odor capability, the two independent olfactory stimulation generation units integrated at the ends of the arms are replaceable, directly doubling the types of odors the device can carry, or enabling "three-dimensional olfactory" stimulation of different odors for each nostril, greatly enriching the layering and immersion of the olfactory experience. Secondly, in terms of battery life, despite pursuing an extremely thin arm, the distributed structure allows the device to house two independent micro-batteries. Its total battery capacity potential far exceeds the extremely compressed single-capsule design in Embodiment 1 due to volume constraints, thus significantly extending the usage time on a single charge, better meeting the needs of high-intensity, long-duration applications.

[0126] Example 3: Active olfactory clearance (AOC) based on a cyclodextrin molecule capture system

[0127] This embodiment aims to illustrate a highly optimized system and method for achieving Active Olfactory Cleansing (AOC) functionality. This functionality, by pre-filling a specially formulated molecular trapping agent into designated microcavities of the "multi-odor sac" 105 described in Embodiments 1 and 2, endows the device with unprecedented olfactory "refreshing" capabilities without increasing the volume of additional hardware. This AOC mechanism aims to overcome the inherent olfactory residue and cross-contamination problems of traditional olfactory devices, thereby greatly improving the "temporal resolution" and "signal-to-noise ratio" of the device when processing rapid, continuous, and variable olfactory information, making it a true high-fidelity "olfactory display" capable of carrying complex narratives. Unlike traditional solutions such as simple physical purging or solvent dilution, this embodiment creatively proposes a complete system solution integrating advanced chemical principles, precision physical engineering, and intelligent software algorithms. Please refer to... Figure 3 This is a flowchart of the operation of the AOC system based on cyclodextrin.

[0128] a) The core chemical principle of AOC: Targeted molecular capture based on cyclodextrin

[0129] The core chemical principle of the AOC mechanism in this embodiment is based on the unique properties of a macrocyclic oligosaccharide molecule called cyclodextrin.

[0130] 1. Molecular Structure and Properties: Cyclodextrins are cyclic molecules formed by multiple glucose units linked by α-1,4-glycosidic bonds. Their three-dimensional morphology resembles a hollow, truncated cone, often figuratively described as a "molecular basket" or "donut." A key feature of their molecular structure lies in the difference in hydrophilicity between their inner and outer surfaces: their outer wall is composed of hydrophilic hydroxyl groups, making them readily soluble in water; while their inner cavity is composed of hydrophobic CH bonds and glycosidic oxygen bridges, forming a nonpolar, lipophilic microenvironment.

[0131] 2. Molecular Capture Mechanism: The vast majority of odor substances that trigger human olfaction (such as esters, aldehydes, and alcohols in floral and fruity aromas) are essentially volatile, hydrophobic small organic molecules. When these hydrophobic odor molecules (guests) encounter cyclodextrin molecules (hosts) in an aqueous environment, due to hydrophobic interactions (hydrophobic effect), they spontaneously and preferentially penetrate the hydrophobic cavity of the cyclodextrin and are "encapsulated," forming a thermodynamically more stable, non-covalent "host-guest inclusion complex." Once the odor molecule is encapsulated in the "cage" of cyclodextrin, it loses its volatility and can no longer bind to olfactory receptor proteins on the nasal mucosa. This thoroughly and efficiently "neutralizes" or "shields" residual odor signals at the molecular level.

[0132] 3. Material Selection and Safety: In this embodiment, chemically modified cyclodextrin derivatives, such as hydroxypropyl-β-cyclodextrin (HPβCD), are preferably used. HPβCD not only has excellent water solubility but also possesses a wealth of publicly available safety data, making it a safe pharmaceutical excipient for oral, injectable, and even nasal spray medications (typically used to increase the solubility and stability of poorly soluble active pharmaceutical ingredients). Choosing HPβCD as the core component of the scavenger ensures that this AOC mechanism has extremely high, scientifically validated biosafety, which is crucial for a product that needs to be in prolonged contact with the nasal mucosa. Furthermore, HPβCD itself is a completely odorless and tasteless macromolecule, and will not introduce any secondary odor interference.

[0133] b) Development of a system-level solution: Overcoming the challenges of water-based carriers

[0134] While cyclodextrin aqueous solutions offer unparalleled advantages in terms of chemical principles and safety, their water-based carriers present challenges such as slow evaporation, a moist feeling in the nasal cavity, and residue, which must be addressed through systems engineering approaches. This embodiment constructs a complete and intelligent solution through a close integration of physical engineering, dosage science, and software algorithms.

[0135] 1. Physical engineering solution: Achieve rapid removal through "flash evaporation".

[0136] To ensure that the aqueous carrier evaporates almost completely before contacting the nasal mucosa, leaving only "dry," activated cyclodextrin microparticles for "capture," the method employed is ultra-microdropletization: This solution cleverly utilizes the thermal jetting technology based on a microheater array, already present in Examples 1 and 2. By specifically optimizing the electrical pulse parameters driving this technology (such as pulse width, frequency, and energy), the volume and diameter of the sprayed cyclodextrin aqueous solution droplets can be controlled within an extremely small range. The evaporation rate of the droplets is directly proportional to the ratio of their surface area to volume. By controlling the droplet diameter to below 10 micrometers, their enormous specific surface area allows the droplets to rapidly undergo "flash evaporation" within tens of milliseconds after leaving the nozzle, through collision and friction with the air. The final result is that the device no longer sprays "wet" droplets, but rather a cloud of "dry mist" carrying highly active cyclodextrin microparticles at near-room temperature. This aerosol can diffuse efficiently in the nasal vestibule in gaseous form, thoroughly mixing with the residual odor molecule cloud and completing molecular capture. Since most of the moisture evaporates during the flight, almost no "liquid water" will be deposited on the nasal mucosa, thus completely avoiding the problems of "dampness" and large-scale solvent residue.

[0137] 2. Intelligent Dosing Control Solution: Precise Dosing Based on "Chemical Measurement"

[0138] To avoid "overspraying," only the minimum amount of scavenger is sprayed, just enough to neutralize most of the residual molecules, minimizing the potential impact of "idle" cyclodextrin molecules on subsequent odors. A smart dosing algorithm can be used: an olfactory residue estimation model can be integrated into the device firmware. This model tracks the history of olfactory events in real time. When an AOC command is needed, the algorithm assesses the residue, estimating the approximate molar number of odor molecules that may remain in the nasal cavity from a pre-set database, based on what odor was just released (different molecules have different adsorption properties), its intensity, and its duration. Based on the inclusion ratio of cyclodextrin to typical odor molecules (typically 1:1), the minimum effective dose (stoichiometric equivalent) of cyclodextrin required to capture these residual molecules is calculated. The calculated dose is then converted into specific PWM parameters to drive the spray system, and a precise spray is executed.

[0139] 3. Software Control and Optimization Scheme: Introducing "Post-Clearing Delay" and "Intensity Compensation"

[0140] Through software scheduling logic, the system proactively avoids and counteracts any possible trace physical residue effects, ensuring a seamless transition in the olfactory experience. Forced refresh delay: After the AOC clearance command is executed, the system scheduler can be configured to forcibly insert a very short "silent period," such as 200-500 milliseconds. During this silent period, no new olfactory events are allowed, specifically to allow the cyclodextrin particles carrying inclusion complexes to further diffuse and dissipate with normal respiratory airflow. Subsequent odor intensity compensation: If a new odor must be emitted immediately after the clearance command (without a silent period), the system can automatically provide a small, dynamic pre-compensation for its initial intensity (e.g., increasing the command intensity by 10% in the first 100 milliseconds). This enhanced "first puff" of odor has enough molecules to "saturate" any "idle" cyclodextrin molecules that might remain in the olfactory pathway, ensuring that enough odor molecules still reach the olfactory receptors, thus achieving a smooth and seamless transition in olfactory perception.

[0141] Example 4: System Overall Architecture and Information Flow

[0142] This embodiment aims to illustrate the overall architecture of the wearable olfactory experience system of the present invention from a macroscopic, global perspective, and how information flows between the various components of the system. A clear system architecture is key to understanding how the present invention transforms abstract digital information into a realistic olfactory experience. Please refer to... Figure 4 This diagram illustrates the overall system architecture and information flow of the wearable olfactory experience system of this invention. The system mainly consists of four logical layers:

[0143] a) Level 1: Content and Event Source Layer 401

[0144] The topmost content and event source layer, 401, is the "starting point" or "trigger" for all olfactory experiences—the source of the need for olfactory release. This layer is highly open and extensible, encompassing a wide variety of events that can be digitally captured or defined. It primarily includes:

[0145] 1. Digital Media Sources: This is the core of immersive entertainment applications. For example, movies, VR / AR games, etc., have "olfactory track" files embedded within them or olfactory event commands generated in real time by the game engine, which constitute a major source of information for this layer.

[0146] 2. Operating system and application services: such as the calendar, clock, or any authorized third-party app on a smartphone. The olfactory notification requests they generate based on preset rules or user actions also belong to the event source layer 401.

[0147] 3. User-generated content (UGC): The "olfactory recipes" created and saved by users through the "digital perfumery" are an important user-driven content source in this layer.

[0148] 4. External physical world sensors: such as motion sensors attached to tableware, whose collected physical motion data are defined as trigger events in event source layer 401.

[0149] b) Layer Two: User Terminals and Control Core Layer 402

[0150] Information flows from the event source layer 401 to the user terminal and control core layer 402 below. For example... Figure 4 As shown, this layer is the "brain" and "command center" of the entire system, typically a dedicated application (APP) running on the user's smartphone, PC, or VR headset. Its core responsibility is to receive, parse, and process various raw information from the upstream event source layer 401, and convert it into low-level instructions that downstream wearable devices can understand and execute. The main functional modules include:

[0151] 1. Information Reception and Parsing Engine: Responsible for listening to and receiving various inputs from Level 1 401.

[0152] 2. Olfactory Synchronization and Timing Engine: Responsible for accurately aligning olfactory events with the timestamps of external media sources.

[0153] 3. Logic Judgment and Strategy Selection Engine: Responsible for executing the preset rules of context-based notification applications.

[0154] 4. Instruction Compilation and Serialization Module: This is the core of this layer, responsible for "translating" high-level olfactory requirements into low-level digital control instruction sequences with precise timestamps and physical parameters.

[0155] 5. User Interface (UI): A graphical interface that provides users with the means to create, set up, and interact with content.

[0156] c) Layer 3: Wireless Communication Link Layer 403

[0157] The compiled sequence of digital control instructions is transmitted from the control core layer 402 via the wireless communication link layer 403. For example... Figure 4 As shown, this layer acts as the "nerve" connecting the "brain" and the "limbs." This layer is responsible for reliably and with low latency transmitting instruction sequences to the wearable device. Bluetooth Low Energy (BLE) technology is preferred. The communication link 403 must employ a strong encryption mechanism to ensure the confidentiality and integrity of the instructions.

[0158] d) Layer 4: Wearable Devices and Execution Layer 404

[0159] The instructions ultimately reach the wearable device and execution layer 404 at the bottom layer. This layer is the final "executor" of the olfactory experience, namely the wearable device described in this invention (such as in Embodiment 1 or 2). Its role in the system architecture is strictly limited to a "faithful execution unit," without making complex decisions. Its firmware's main functions include:

[0160] 1. Command Reception and Parsing Module: The device firmware receives data packets from the link layer 403 through the BLE module and parses out the specific execution commands from them.

[0161] 2. High-precision timer actuator: The device utilizes its internal hardware timer, which is unaffected by timing jitter in the communication link 403, and strictly follows the time parameters defined in the instruction to drive the odor generation unit to release micro-dose at precise moments.

[0162] 3. Status feedback (optional): Device 404 can send simple status feedback to the upstream user terminal 402 to form a complete status monitoring closed loop.

[0163] A typical example of an information flow path: Taking the scenario of "watching a bonfire party in a VR movie" as an example, let's track the information flow... Figure 4 The complete flow path in the architecture shown:

[0164] [Level 1 401] When the VR movie player plays to the timestamp 01:25:000, it reads an olfactory event from its encapsulated olfactory track: {"time":"01:25:000","aroma":"wood_smoke","intensity":0.7}.

[0165] [Level 2 402] The olfactory synchronization engine running in the VR headset receives the event. It calculates that the communication latency is approximately 80 milliseconds. Therefore, when the master clock reaches 01:24:920, it compiles the event into a low-level instruction: "Switch to 'wood smoke' scent; execute a series of jet pulses with 70% intensity parameters within a 400ms time window."

[0166] [Layer 3 403] This instruction is sent from the VR headset (part of the user terminal 402) to the wearable device worn by the user (execution layer 404) via the BLE link 403.

[0167] [Level 4 404] Upon receiving the instruction, device 404 immediately parses and activates its internal high-precision hardware timer. At the precise moment (when the external world master clock reaches 01:25:000), it drives its micro-dose release unit to release a realistic wood smoke odor.

[0168] [User Experience] While seeing the roaring campfire in virtual reality, users also smell the smoke and fire released by device 404 in perfect synchronization, greatly enhancing their immersion and realism.

[0169] Through this clear, layered system architecture 400, the present invention constructs a powerful, flexible, and scalable platform that can accurately transform digital content and events from an endless source of content and events 401 into a rich, colorful, and personalized olfactory experience through intelligent processing of the control core layer 402 and reliable transmission of the wireless link layer 403, and finally by the execution layer 404.

[0170] Example 5: Application of "Digital Perfume Mixing Table" on User Terminal

[0171] This embodiment describes a "digital perfumery table" that runs on a user terminal device (typically a smartphone or tablet). This is a visual editing interface that transforms the complex process of creating olfactory recipes into an intuitive and creative user experience, enabling non-professional users to easily create and arrange olfactory narratives, much like editing audio and video.

[0172] Please see Figure 5 This image shows a typical user interface example of a "digital perfumer." Its design draws on the mature interaction paradigms of modern digital audio workstations (DAWs) and non-linear video editing software, and incorporates deep innovation and optimization for the unique characteristics of the sense of smell.

[0173] a) Interface layout and core components, such as Figure 5 As shown, the "digital perfumery station" mainly consists of the following core components:

[0174] 1. Multitrack Timeline 501:

[0175] Occupying the core area of ​​the interface is a horizontally unfolding timeline. The timeline scale can be freely scaled by the user, with millisecond-level precision. Below, multiple parallel "olfactory tracks" 502 are provided. The multi-track design allows users to arrange and overlay multiple scents to create complex layers. In an advanced view mode, the timeline 501 can also display a dedicated "AOC control track," allowing professional users to manually drag in a "cleansing event" marker at a specific moment in the timeline for precise manual control over the timing of olfactory cleansing, which is crucial for creating fast-paced, high-contrast olfactory narratives.

[0176] 2. Scent Material Library 503:

[0177] Located on the left side of the interface is the user's personal scent library. This library automatically synchronizes and displays all available scents currently owned by the user and installed in their olfactory devices. Each scent is presented as a "scent block" with an intuitive icon and name.

[0178] 3. Creation Process:

[0179] The creation process is very intuitive. Users simply drag and drop the desired scent elements from the resource library 503 onto the olfactory track 502 below. An "olfactory event clip" 504 is then created. Users can drag the edges of the clip to adjust its duration or freely drag its position on the timeline.

[0180] b) Depth parametric editing of olfactory event fragments

[0181] Double-clicking or long-pressing the olfactory event fragment 504 will bring up a detailed parameter editing panel 505.

[0182] 1. Strength Envelope Editor:

[0183] This editor allows users to create arbitrarily complex intensity variation patterns by adding and dragging "keyframe" control points. To simulate a more natural olfactory experience, the editor also offers a "random modulation" function. Users can overlay an adjustable, weak random noise signal onto the intensity curve to simulate the natural fluctuations in odor intensity caused by airflow disturbances in the real world, such as simulating the scent of flowers swaying in the wind or the varying intensity of smoke from a burning campfire.

[0184] 2. Templates and Presets:

[0185] To lower the barrier to entry for creative work, the editing panel 505 provides an "Envelope Preset Library." Users can directly select preset intensity curve templates such as "campfire," "waves," "breathing," and "heartbeat," quickly apply them to clips, and then make personalized fine-tuning.

[0186] c) Enhanced creation and preview functions

[0187] 1. Previewing and managing scent materials:

[0188] Each scent block in the scent library 503 has a "test smell button" 506. When the user clicks this button, the device will immediately spray the scent once with a standardized intensity and duration to help the user make a quick decision.

[0189] 2. Multimedia synchronization and advanced real-time preview:

[0190] The "Digital Perfumery" maintains a real-time Bluetooth connection with the user's wearable device throughout the entire process. In addition to basic previewing by dragging the play pointer, it supports more advanced preview modes. Users can select one or more olfactory event segments 504 and enable "Loop Playback" mode to repeatedly and attentively experience the effect of the segment while fine-tuning parameters. Furthermore, for multi-track projects, "Solo" and "Mute" buttons are provided, allowing users to temporarily listen to only one track or mute another during preview, facilitating precise mixing and balancing of complex olfactory layers.

[0191] 3. Integration of AOC (Active Olfactory Cleaning) function:

[0192] To ensure a "what you hear is what you get" real-time preview experience, an "Auto AOC" switch 507 is provided on the interface. When this switch is turned on, the system will automatically trigger an AOC clearing command before playing a new scent when the playback pointer jumps or when switching between two segments with significantly different scents, ensuring the accuracy of the preview.

[0193] 4. Undo / Redo function:

[0194] The main toolbar of the interface features a standard undo / redo button 508. The application supports multi-step undo and redo, ensuring a safe and smooth user creation process.

[0195] d) Recipe Packaging and Sharing

[0196] Once the user is satisfied with their work, they can click the "Save" button. The app will encapsulate all olfactory events on timeline 501 (including manually placed AOC events), envelope data (including keyframes and modulation parameters), and user-added metadata (title, description, tags) into a structured, standardized Olfactory Scripting Language (OSL) file. This file not only records "when to release what scent" but also precisely defines "how to release it," ensuring that other users can obtain an olfactory experience almost identical to the creator's when "one-click replicating." Subsequently, the user can share this complete OSL file to the UGC community platform of this invention with one click.

[0197] Example 6: Olfactory Synchronization Method with External Digital Media Sources

[0198] This embodiment illustrates the method and core technical details of accurately synchronizing olfactory experience with external digital media sources (such as movies, VR / AR applications, and music) in this invention.

[0199] a) "Scented script" language and creation tools on the content creation side

[0200] To make olfactory synchronization widely applicable and highly expressive, this invention envisions a standardized "olfactory scripting" language (which can be a markup language based on Extensible Markup Language (XML) or JavaScript Object Notation (JSON).

[0201] 1. Core elements of the Olfaction Scripting Language (OSL):

[0202] OSL is not a programming language, but a descriptive language designed to allow content creators to intuitively and precisely define a complete, dynamic timeline of olfactory experiences. An OSL file (i.e., an "olfactory track") consists of a series of "olfactory event" nodes. Each event node contains at least the following attributes:

[0203] Timestamp: Defines the precise time when the event occurred, in the format HH:MM:SS.mmm. This serves as the baseline for synchronization with audiovisual content.

[0204] Odor Identifier (aromaID): A string used to uniquely identify the odor to be released. For example, "coffee_arabica_roast". This ID needs to correspond to the odor ID defined in the electronic identifier on the odor component (as described in the embodiments below).

[0205] Event Type: Defines the basic behavior of the event. Major types include, but are not limited to: Pulse: Performs a single, instantaneous pulse release; Sequence: Performs a series of preset pulses with different parameters; Curve: Performs continuous (simulated by high-frequency pulses) or variable-speed pulse releases along a user-defined intensity curve.

[0206] The parameter set (params) is an object containing specific parameters for the event. Its content varies depending on the event type. For the PULSE event, the parameters can be: {"intensity":0.8,"duration":0.2}, representing the pulse intensity (mapped to pulse width or voltage) and duration (in seconds), respectively. For the SEQUENCE event, the parameters can be: {"pulses":[{"delay":0.0,"intensity":0.9},{"delay":0.5,"intensity":0.5},{"delay":1.0,"intensity":0.2}]}, defining a sequence of three pulses, each with its own delay relative to the event timestamp and its own intensity. This can be used to simulate the rapid decay or echo of an odor. For the CURVE event, the parameters can be: {"duration":10.0,"curvePoints":[{"time":0.0,"intensity":0.0},{"time":2.0,"intensity":1.0},{"time":8.0,"intensity":0.8},{"time":10.0,"intensity":0.0}],"mode":"continuous"}, defining an event lasting 10 seconds whose intensity smoothly varies along a Bézier curve defined by four keyframes. The mode can be specified as continuous (high-frequency pulses simulating continuous release) or pulsed (pulsed pulse frequency varying according to the slope of the curve).

[0207] 2. Visual creation tools: In order to enable creators without technical backgrounds (such as film directors, perfumers, and musicians) to easily create olfactory scripts, this invention conceives of a professional creation software, namely the "digital perfumery table" described in Example 5.

[0208] b) The algorithm of the user terminal's "adaptive olfactory synchronization engine"

[0209] This is the core processing module that runs on the user's mobile phone or VR headset, responsible for translating OSL scripts into precise control over the device. Its "adaptive" nature is reflected in its ability to dynamically compensate for communication latency and individual device differences.

[0210] 1. Initialization and Device Characteristic Investigation:

[0211] When the synchronization engine first connects to a wearable device, it performs a "handshake and probe" process to understand the device's "properties":

[0212] Query capabilities: The engine reads a series of "device information characteristic values" from the device's general attribute configuration file to determine the functions supported by the device. For example, what is the maximum pulse intensity level it supports? Does it support multiple odor switching? Which aromaIDs are included in the currently installed odor components?

[0213] Dynamic latency measurement: The engine sends a series of request-response latency tests (a write request requiring the device to immediately send a read response) with precise timestamps, and calculates the average round-trip time (RTT) from command to device response under the current network environment by measuring multiple round-trip times. Half of this RTT value can be used as an estimate of the initial one-way communication latency_base.

[0214] 2. OSL script parsing and event queue scheduling:

[0215] When media playback begins, the synchronization engine parses the loaded OSL file, sorts all sniffer events by their timestamps in ascending order, and places them into a priority queue called the "future event queue." The engine continuously queries the media player's master clock T_current at a very high frequency (e.g., every 10 milliseconds). In each query cycle, it checks the head event of the "future event queue," Event_next.

[0216] 3. Prediction and scheduling logic:

[0217] This is the core algorithm for achieving precise synchronization. The engine's scheduling trigger condition is not T_current greater than or equal to Event_next.timestamp, but rather T_current greater than or equal to (Event_next.timestamp - T_predictive_buffer). Here, T_predictive_buffer is a dynamically calculated value, calculated as: T_predictive_buffer = latency_base + T_execution_delay + T_safety_margin. Where: latency_base is the initially measured or continuously updated communication latency; T_execution_delay is the device's expected execution latency (internal processing and mechanical delays from receiving the instruction to physically releasing the signal), which can be obtained during device probing; and T_safety_margin is a small safety margin (e.g., 20-50 milliseconds) to handle network jitter.

[0218] 4. Instruction compilation and dynamic adaptation:

[0219] When an Event_next event meets the scheduling triggering condition, it is retrieved from the "future event queue" and enters the instruction compilation stage:

[0220] Translation from high-level parameters to low-level instructions: The compiler module translates descriptive OSL events into a sequence of binary low-level instructions that the device firmware can directly understand, based on the event type and parameters. For example, an event of type 'CURVE' will be translated into a long series of tiny pulse instructions executed at an extremely high frequency (e.g., once every 50 milliseconds), with the intensity parameter of each pulse calculated based on the precise Y value of the curve at that moment.

[0221] Dynamic intensity adaptation: If the synchronization engine learns from device feedback (or long-term statistics) that the release efficiency of the currently connected specific device is too high or too low (due to individual differences), it can perform a dynamic, device-specific correction on the intensity values ​​defined in the OSL script at compile time. For example, for a device with "strong jetting power," it might multiply all intensity values ​​by a factor of 0.8.

[0222] 5. Execution and Feedback Loop:

[0223] The compiled instruction sequence is immediately sent out via BLE. Simultaneously, the synchronization engine records the instruction's sending timestamp, T_sent.

[0224] Execution Acknowledgment and Delay Update: On the device side, when a critical instruction (such as the start of a sequence) is physically executed, the device can immediately send back a very short "execution acknowledgment" packet containing the instruction's unique ID. When the synchronization engine receives this acknowledgment packet, it can calculate the actual total latency of this operation, T_real_latency = T_now - T_sent. This T_real_latency is fed into a Kalman filter or a simple moving average filter to continuously and dynamically update and optimize the prediction of future communication latency, thereby making T_predictive_buffer increasingly accurate.

[0225] Through this integrated solution that combines standardized creation, dynamic latency compensation, device characteristic adaptation, and closed-loop feedback optimization, this invention ensures that regardless of the network environment or playback device, the olfactory effect experienced by the user can achieve "sensory-level" precise synchronization with the content they watch or hear, thus providing an unbreakable technical guarantee for a truly immersive experience.

[0226] Example 7: Personalized Olfactory Experience Calibration and Intelligent Management System

[0227] This embodiment aims to elaborate in detail on a technical solution for addressing differences in olfactory sensitivity among individual users and providing intelligent and secure experience management, corresponding to... Figure 7 The process and interface are shown. This solution integrates an "olfactory sensitivity calibration" module and an "olfactory experience intelligent management engine" into the user terminal application (APP). This engine aims to prevent users from experiencing olfactory fatigue due to prolonged, high-intensity exposure to a single odor and can automatically apply safety policies based on user scenarios. Crucially, all mechanisms described in this embodiment are based on user-defined rules and logic algorithms executed by an external controller (APP), completely avoiding the collection and feedback of real-time, unconscious physiological parameters from the user. Therefore, it perfectly maintains the technical purity of the core open-loop control and feedforward guidance of this invention.

[0228] a) The necessity and technical principles of calibration and management

[0229] Scientific research shows that there are significant individual differences in human olfactory perception. For the same odor pulse emitted by a device at the exact same physical concentration, a user with a highly sensitive sense of smell may find it too stimulating, while a user with a less sensitive sense of smell may find it too faint. Furthermore, even for the same user, the perception threshold for different chemical categories of odor molecules (such as esters in floral scents and sesquiterpenes in woody scents) can vary significantly; this phenomenon is known as a subclinical manifestation of "specific olfactory agnosia."

[0230] Therefore, the technical principle of this embodiment is extended to two levels:

[0231] Multi-dimensional personalized calibration: Through a guided calibration process based on psychophysical methods, not only is a mapping relationship established between the objective physical intensity output by the device and the subjective psychological intensity of the user for a benchmark odor, but it also allows for differentiated calibration for different odor categories, ultimately generating a user-exclusive, multi-dimensional "olfactory calibration profile".

[0232] Intelligent Experience Management: Through a rule-based, forward-looking software engine, olfactory commands to be executed are preprocessed to ensure the effectiveness of the long-term experience and the safety and comfort in specific scenarios.

[0233] b) System Composition and Workflow

[0234] The calibration and management functions are primarily implemented by functional modules in an application (APP) running on the user's terminal device and interact with the wearable olfactory device.

[0235] 1. Guided calibration process: Users can start the "olfactory sensitivity calibration" wizard in the "Personal Settings" of the APP.

[0236] Step 1: Basic Calibration (Based on Reference Odor):

[0237] The calibration wizard first guides the user through a benchmark odor that is chemically stable, universally applicable, and unlikely to cause olfactory fatigue (e.g., a weak concentration of n-butanol or a neutral sandalwood) by using a chemically stable, universally applicable odor. The algorithm, as described earlier, uses a series of "release-feedback-record" loops to determine the user's baseline perceptual threshold P_threshold and a global intensity mapping function f_global(x). This function is typically an S-shaped curve (such as the sigmoid function), which better simulates the non-linear perceptual characteristics of human senses to stimulus intensity (Weber-Fechner law).

[0238] Step 2: Multi-odor difference calibration: After completing the basic calibration, the system will provide an optional advanced calibration process.

[0239] Scent Classification: The system classifies scents in the consumables library into several major categories based on their main chemical components or fragrance family, such as "floral", "fruity", "woody", and "spicy".

[0240] Differentiated Fine-tuning: The system selects a representative scent from each category and releases it once at the "clear and comfortable" intensity just calibrated by the user. The interface then prompts the user and provides a simple adjustment slider. For example, when releasing the "rose" scent, if the user feels its intensity is significantly higher than the previous "sandalwood," they might adjust the slider to "-20%." The system records this differentiated calibration coefficient for the "floral" category as C_floral = 0.8. This process is repeated for each scent category, ultimately generating a multi-dimensional calibration profile, such as: {C_floral: 0.8, C_fruity: 1.0, C_woody: 1.1, ...}.

[0241] 2. Application and manual fine-tuning of calibration coefficients:

[0242] In the instruction compilation stage described in Example 6, for any olfactory instruction from the content source with an intensity of I_original and belonging to Category_Y, its final intensity I_final sent to the device is calculated in layers as follows:

[0243] I_intermediate = f_global(I_original) (Applying global mapping curve)

[0244] I_final = I_intermediate * C_Y (the difference coefficient applied to this odor category)

[0245] Furthermore, in the app's "Advanced Settings," the system provides a manual adjustment interface for professional users, allowing them to directly fine-tune or override all parameters generated by automatic calibration in a graphical or numerical manner. Users can manually drag keyframes on the global mapping curve f_global(x) to intuitively change its shape; they can precisely input the calibration coefficients for each odor category; and they can even adjust the attenuation constant k and exposure threshold in the fatigue mitigation model, which will be described below. This allows users to fully customize the response characteristics of the entire system to their personal preferences, giving them complete freedom in fine-tuning.

[0246] 3. Olfactory Experience Intelligent Management Engine: In addition to the calibration function, the APP also integrates a management engine that works continuously in the background.

[0247] Module 1: Olfactory Fatigue Relief Module:

[0248] This module dynamically tracks a user's exposure to a specific odor over a period of time by maintaining an "exposure" counter with a weighted time decay algorithm for each odor aromaID. As described earlier, the algorithm automatically reduces the release intensity of an odor when its exposure exceeds a user-defined (or system default) moderate or high threshold. For example, if a user is in a game scenario with a "coffee" background smell for an extended period, this module will automatically and smoothly reduce the intensity of all subsequent "coffee" odor commands by 20%-40% to slow down the user's olfactory adaptation to the coffee aroma, ensuring that the user's olfactory system can still respond effectively when a critical, clearly perceived "coffee" incident occurs.

[0249] Module 2: Context-Aware Security Profiles

[0250] This module allows users to set a set of rules, including odor whitelists / blacklists and global intensity limits, for special scenarios such as "sleep mode". When the mode is activated (manually, by timer, or in conjunction with the operating system), the engine automatically filters all incoming olfactory commands, allowing only mild olfactory stimuli that comply with safety rules to be executed.

[0251] c) Beneficial effects of this embodiment

[0252] Achieving a truly personalized experience: Through a multi-dimensional calibration mechanism, the system not only adapts to the user's overall olfactory sensitivity, but also takes into account the differences in their perception of different odor categories, elevating the personalized experience to an unprecedented level.

[0253] Giving users complete control: By providing a full range of manual fine-tuning options, this solution offers intelligent convenience while fully respecting users' personal preferences and autonomy.

[0254] Adhering to the core invention concept: the entire calibration and management process is a user-initiated setting or a forward-looking algorithm based on objective rules. The result is an optimized expression of the "feedforward" signal, which is fundamentally different from closed-loop biofeedback that aims to form a real-time response.

[0255] Enhancing long-term effectiveness and safety: Through intelligent fatigue relief and situational safety management, this invention ensures the long-term usability of the product and the comfort and safety of users in any scenario.

[0256] Example 8: A method for enhancing the sense of smell during eating based on an external motion sensor

[0257] This embodiment will illustrate the specific engineering implementation details of the olfactory flavor enhancement method, especially its core, machine learning-based real-time eating action recognition algorithm. We will describe the entire technical solution from data acquisition and model training to real-time inference deployment, thereby demonstrating the powerful capabilities of this invention in a specific application scenario.

[0258] a) System hardware composition and data flow

[0259] Please see Figure 8 This diagram illustrates the system composition and data flow upon which the method of this embodiment relies. The system mainly consists of the following parts:

[0260] 1. Wearable olfactory device: The device as described in Example 1 or 2, pre-loaded with an odor component containing the aroma of a target food (e.g., "strawberry", "chocolate").

[0261] 2. Miniature wireless motion sensor (which we call a “sensor module”): This is a core, new external device.

[0262] Hardware Configuration: The sensing module integrates a high-performance, low-power six-axis inertial measurement unit (IMU) that can synchronously output three-axis acceleration and three-axis angular velocity data at a configurable frequency (e.g., 50-100Hz). It also includes a Bluetooth-enabled SOC (for data transmission) and a miniature rechargeable battery.

[0263] Physical form: Its outer shell is designed as a tiny, waterproof, flat module that can be firmly and non-invasively attached to the handles of various cutlery (spoons, forks, chopsticks) via a replaceable, food-grade, flexible silicone clamp.

[0264] 3. User terminal device (smartphone): running the core "flavor enhancement engine" application (APP).

[0265] 4. Data Stream: The sensing module continuously sends the six-axis IMU data stream it acquires to the mobile app in real time via BLE. The engine in the app processes and analyzes the data stream, and once it identifies a key eating action, it immediately generates an olfactory command, which is sent to the olfactory device for execution via another BLE link.

[0266] b) Construction of a real-time eating action recognition algorithm (complete machine learning process)

[0267] This is the core technology of the method in this embodiment, which enables the system to "understand" the user's eating actions.

[0268] Phase 1: Construction of a large-scale, high-quality training dataset

[0269] Data collection: A large number of subjects of different ages, genders, dominant hands, and eating habits were recruited. They were provided with various tableware equipped with sensing modules and asked to eat various types of food (solid, liquid, and paste) naturally while being recorded by a camera.

[0270] Synchronization and Annotation: Using specialized software tools, the high frame rate video stream and the synchronously recorded IMU data stream are aligned. Then, trained annotators manually and frame-by-frame add precise action labels to the IMU data stream on the video. The action class set should include at least the following: Idle: cutlery resting on the table or in hand; Lifting: lifting cutlery from the table; Scooping / Forking: the action of scooping or forking food from a container; Transporting: the action of transporting cutlery with food from a container to the mouth; Mouth-Entry: (key action) the precise moment the cutlery enters the lips; Mouth-Exit: the moment the cutlery leaves the lips; Returning: the action of placing empty cutlery back into the container or on the table.

[0271] Data augmentation: To enhance the model's generalization ability, a series of data augmentation operations can be performed on the collected IMU data, such as adding random noise, random rotation (simulating different holding postures), and scaling on the time axis (simulating different eating speeds). Ultimately, we will obtain a large-scale dataset containing tens of thousands of eating action segments with precise time-series labels, which is the foundation for training a high-precision model.

[0272] 2. Phase Two: Data Preprocessing and Feature Engineering

[0273] The raw IMU data is a high-dimensional time series, and directly inputting it into the model does not yield good results. It needs to be processed and representative features extracted.

[0274] Coordinate system alignment: Since the sensor module may be fixed to the tableware at any angle, an algorithm is needed to align its coordinate system to a unified coordinate system that is relevant to the tableware itself. This can be achieved by requiring the user to perform a simple calibration action (such as placing the tableware horizontally on the table for a few seconds) before each use.

[0275] Data windowing: The continuous IMU data stream is divided into fixed-length, overlapping sliding windows (e.g., each window is 2 seconds long and in 0.5-second increments).

[0276] Feature Extraction: For the data within each window, a series of statistically and physically meaningful features are calculated to comprehensively describe the motion patterns over that time period. Extracted features may include: Time-domain features: mean, variance, standard deviation, maximum, minimum, root mean square (RMS), zero-crossing rate, and integral values ​​(representing displacement and angle changes) for each of the six axes. Frequency-domain features: Fast Fourier Transform (FFT) is performed on the signal for each axis to extract its energy proportion, dominant frequency, and spectral entropy in different frequency bands (e.g., 0-5Hz, 5-10Hz, etc.). Cross-axis correlation features: covariance or correlation coefficient between signals from different axes. Finally, each sliding window is transformed into a high-dimensional (e.g., hundreds of dimensions) feature vector.

[0277] 3. Phase Three: Selection and Training of Machine Learning Models

[0278] Model Selection: Considering the inherent temporal dependencies of eating actions (e.g., Mouth entry almost always follows transiting), a model capable of effectively processing time-series data is the optimal choice. This embodiment preferably employs a recurrent neural network (RNN) based on Long Short-Term Memory (LSTM). LSTM networks, through their internal "gating" mechanism, can effectively learn and remember long-term dependencies in time series data, making them highly suitable for recognizing human activities. The model structure can be a deep network containing 2-3 LSTM layers, ultimately connected to a Softmax classification layer.

[0279] Model Training: All feature vectors extracted in the previous step, along with their corresponding labels, are input into the LSTM network for supervised learning training. Standard machine learning procedures are used, including dividing the dataset into training, validation, and test sets, and employing backpropagation and an optimizer to minimize the cross-entropy loss function. Monitoring on the validation set prevents overfitting and helps find the optimal hyperparameters.

[0280] Model quantization and optimization: After training, in order to enable the model to perform real-time inference efficiently on resource-constrained smartphones, it is necessary to perform optimization operations such as model quantization (e.g., converting 32-bit floating-point weights to 8-bit integers) and pruning, so as to greatly reduce the model size and computational complexity with almost no loss of accuracy.

[0281] 4. Phase Four: Deployment of Real-Time Inference and Triggering Logic

[0282] The trained and optimized model is deployed to the "flavor enhancement engine" of the mobile app.

[0283] Workflow Description: The app receives IMU data from the sensor module in real time and performs data windowing and feature extraction using the exact same parameters as during training. The generated feature vectors are fed into the deployed LSTM model. The model outputs a probability distribution vector for each input window in real time, representing the probability that the window belongs to each action category. To avoid abrupt user experience due to misclassification of a single window, the engine uses a simple state machine or a Hidden Markov Model (HMM) to smooth and post-process the classification results of multiple consecutive windows to obtain a more stable and robust final action recognition sequence.

[0284] 5. This is the "final step" in transforming the action recognition results into a perfect user experience that conforms to the principles of multi-sensory science. The engine's triggering logic is not a simple "recognition-execution" process, but a sophisticated timing control system based on key events and consisting of two phases.

[0285] Phase 1: Synchronous release of lead aromas

[0286] The engine pays particular attention to recognizing the action of "transporting"—the process of a user moving utensils containing food from a container to their mouth. Triggering timing: Once the system confidently recognizes the start of the "transporting" action and the trajectory of the utensils clearly points towards the user's face, the engine immediately triggers a low-intensity, short-duration olfactory release. Design purpose: This stage of the design aims to simulate and enhance the aroma of food that the user naturally smells through the prenasal passages before putting the food in their mouth.

[0287] Simulating a natural experience: The first aroma we smell is usually the most direct and purest "fragrance" of food. For example, when a piece of strawberry cake is brought to our mouths, we first smell the pure aroma of strawberries. This release of the "lead aroma" is like the "overture" to the culinary experience; it arouses the user's anticipation and lays the sensory groundwork for the upcoming oral experience. Its aroma characteristics are designed to be closer to the simple top notes of food.

[0288] Phase Two: Delayed Blending of Core Flavors

[0289] After the first phase, the engine continues to track actions and focuses on the most critical event, the "MOUTH_ENTRY".

[0290] Triggering timing: Once the system identifies the "MOUTH_ENTRY" event with high confidence, it will not immediately trigger olfactory release again. Instead, it will start a "flavor blending delay" timer that can be fine-tuned by the user in the app according to personal preferences.

[0291] The length of this delay is a key parameter, with an optimal range of 500 to 1500 milliseconds (depending on different chewing speeds at different ages or individual differences).

[0292] Design Insight and Purpose: Only after this initial delay has elapsed will the engine generate and send a second (typically stronger and more complex) olfactory release command. This deliberate, configurable delay is the core insight of this invention in user experience design. It aims to maximize the synchronization and integration of externally supplied aromas with the aromas released during chewing—when the food matrix is ​​broken down, enzymatically decomposed, and heated—and which then ascend through the posterior nasal cavity, all within the olfactory cortex of the brain.

[0293] Simulating complex flavors: The retronasal olfaction produced after chewing is the core of "flavor," a blend of taste, touch, and more complex aroma molecules. The second release of aroma is designed to complement and enhance this complex middle and base notes. Achieving "enhancement" rather than "interference"—this precise point of synchronization—is key to producing the natural feeling of "flavor enhanced," rather than the disturbing sensation of "I smell two unrelated scents."

[0294] Closing Logic and Technical Boundary Definition: After recognizing the "MOUTH_EXIT" event, the engine can intelligently decide how to handle the odor's fading. For example, it can send an "AOC" command (if the device supports it) or simply let the odor dissipate naturally to ensure that the olfactory environment remains clean between meals, avoiding an unpleasant olfactory experience.

[0295] Through this two-stage triggering logic, the present invention no longer simply simulates the sense of smell, but deconstructs a complete eating action into two key sensory nodes, "prenasal smell" and "postnasal smell", and enhances and arranges them separately and precisely, thereby constructing a highly flexible and powerful flavor enhancement framework.

[0296] Those skilled in the art will understand that one of the main challenges in achieving the best experience stems from the precise flavor tuning for different foods, especially the accurate setting of the aroma characteristics, intensity ratio, and timing of "lead aromas" and "core flavors." For example, the dynamic processes of naturally releasing aromas in liquids (such as soup) and solids (such as steak) are completely different, requiring different sets of parameters. However, this tuning of flavor parameters and content creation for specific foods falls under the application layer development problem based on the platform and methodology provided by this invention. The core inventiveness of this invention lies in providing the "platform" and "method" themselves, which have two-stage triggering logic and configurable timing parameters necessary to achieve such fine-tuning, rather than focusing on the final tuning result of any particular food flavor.

[0297] Through the above-mentioned deep implementation of the entire chain from data acquisition to deployment, this invention transforms a simple motion sensor into an "eye" that can accurately interpret subtle human behaviors. Based on this, an unprecedented intelligent system is built that can proactively enhance the human eating and drinking experience in real time and in a personalized way.

[0298] Example 9: Consumables Identification and Management System

[0299] a) Physical identification and authentication of modular devices (such as the olfactory stimulus generation unit in Example 2)

[0300] For the design of the olfactory stimulus generation unit as described in Embodiment 2, which is detachable and replaceable, this system additionally integrates a physical recognition and authentication mechanism to support a more flexible and richer consumable ecosystem:

[0301] Identification and Content Matching: Each replaceable olfactory stimulus generation unit integrates a non-volatile memory chip (e.g., an NFC tag). This chip pre-stores key data such as the module's unique identifier (UID), detailed information about the included scent combinations (e.g., a "forest-themed pack" containing scent channel mappings for pine, moss, and post-rain soil), production batch number, and expiration date. When a user installs a new olfactory stimulus generation unit onto the main device via a sliding rail mounting device, the main control chip reads the chip information through pre-reserved electrical contacts or near-field communication.

[0302] Genuine Product Authentication and Security: To ensure a high-quality and secure user experience and prevent the use of unauthorized third-party consumables, the storage chip also includes an encryption key. When recognizing a new module, the device performs an encrypted challenge-response authentication process. Only genuine modules that pass authentication will be accepted by the device for their scent configuration information and allowed to operate. If authentication fails or expired consumables are detected, the device will refuse activation and issue a clear notification to the user via the application ("Unrecognized consumables" or "Expired consumables, please replace").

[0303] b) Intelligent estimation and management of olfactory liquid balance

[0304] Since directly measuring the liquid balance in a sealed, opaque micro odor sac is technically extremely challenging, this embodiment employs an intelligent estimation algorithm based on "digital consumption counting".

[0305] 1. Quantitative model of dose consumption: Before the device leaves the factory, the single spray dose (in nanoliters) of each odor liquid under different release parameters (PWM duty cycle, pulse duration, etc.) has been precisely calibrated in a laboratory environment and formed a multi-dimensional "dose-parameter" lookup table or mathematical model, which is pre-burned into the device firmware.

[0306] 2. Real-time Consumption Accumulation Algorithm: The device firmware SoC has a non-volatile memory area for storing a counter for "cumulative dose consumption". Whenever the device executes an olfactory release command, the SoC queries its internal model based on the specific parameters used in this command, calculates the theoretical dose consumed in this operation, and accumulates it into the counter.

[0307] 3. Remaining Capacity Estimation and Feedback: The device calculates the remaining capacity percentage in real time based on the formula "(Initial Total Capacity - Cumulative Consumed Dosage) / Initial Total Capacity", and can report it to the user's terminal APP via BLE as needed. When the remaining capacity is lower than a preset threshold (e.g., 10%), the device will proactively send a "Consumables are about to run out" notification to the APP.

[0308] c) Battery health and power management

[0309] In addition to standard battery percentage monitoring, this system also includes a more advanced battery health (SoH) management function.

[0310] 1. Precise power metering: It adopts an advanced "power meter" IC or algorithm, and accurately calculates the remaining power by integrating the inflow and outflow of battery current in real time using the "coulomb counting method", avoiding the problem of inaccurate power display caused by voltage fluctuations.

[0311] 2. Battery State of Health (SoH) Monitoring: The system estimates SoH (the degree of capacity degradation of the battery relative to its new state) by tracking and recording the number of complete charge-discharge cycles and the actual maximum capacity measured in each cycle. When the SoH falls below a threshold (e.g., 80%), the app can alert the user that the battery is aging, allowing for reasonable range expectation management.

[0312] 3. Charging optimization and protection: The system also includes intelligent charging management logic, such as slowing down the charging speed (trickle charging) after charging to 80% to extend the battery's cycle life.

[0313] d) Precision temperature monitoring and safety management

[0314] Since the core actuation mechanism of this invention involves instantaneous heating, accurate temperature monitoring is crucial to ensuring the safety and consistent performance of the device.

[0315] 1. Multi-point temperature monitoring: The device integrates multiple temperature sensors. These include an internal temperature sensor integrated within the SoC chip, and one or more external NTC thermistors strategically placed near the micro-heater array and the battery. Thanks to the miniaturization of modern electronic components, these external NTC thermistors can be packaged in sizes as small as 0.6mm x 0.3mm or even smaller, making their volume negligible. Therefore, their integration does not affect the device's original compact design.

[0316] 2. Core Functions and Security Management Logic:

[0317] Actuation performance calibration: The processing circuit dynamically compensates the PWM parameters of the release command based on the real-time temperature measured by the NTC thermistor to ensure a consistent olfactory experience under different ambient temperatures.

[0318] Overheat protection: An absolutely safe upper temperature limit (e.g., 40°C) is set in the firmware. If any temperature sensor detects that the device temperature is close to this upper limit, the system will automatically enter a temporary "cooling" mode, temporarily refusing to execute new olfactory release commands until the temperature returns to a safe range.

[0319] Charge / discharge protection: Temperature sensors are also used to monitor the battery temperature during charging and discharging, preventing charging and discharging at excessively high or low temperatures, thereby ensuring battery safety and extending its lifespan.

[0320] e) Supporting charging and storage system

[0321] To provide users with convenient energy replenishment and secure daily storage, this invention also includes a matching charging and storage system specifically designed for olfactory devices. The design concept of this system is highly consistent with the charging cases of modern true wireless Bluetooth earphones.

[0322] Its design features a compact and portable case with a magnetic flip-top. The case contains one or more recessed compartments precisely matching the shape of the olfactory device, and uses positioning magnets to automatically attach and align the device. Depending on the device's form, at least two different configurations are available: The first configuration is designed for the integrated capsule-shaped device as described in Example 1. The case contains multiple independent recessed compartments precisely matching the device's shape, allowing users to simultaneously store and charge multiple backup devices to meet all-day or multi-scenario usage needs. The second configuration is tailored for the modular device as described in Example 2. The case contains a main compartment for storing and charging a single C-shaped ring-shaped main unit, and around it, multiple dedicated secondary compartments for securely storing and displaying replaceable olfactory stimulation generation units. This allows users to carry a complete set of different themed olfactory combinations and switch them at any time. This embodiment preferably uses a contact charging solution. The bottom of the charging chamber features two or more gold-plated, spring-loaded charging probes that correspond to pre-installed, waterproof metal contacts on the device. Once the device is drawn into the chamber, the physical circuit is activated, and charging begins automatically. The power chamber itself has a built-in high-capacity battery that can be charged via a USB-C interface and can provide multiple full charges for the olfactory device. An LED status indicator light on the outside of the power chamber clearly displays the battery level and charging status.

[0323] Example 10: Community Platform Supporting User-Generated Content (UGC)

[0324] This embodiment will describe in great detail the server system used to support the personal olfactory experience in this invention, especially the implementation details of its user-generated content (UGC) community platform, which is one of its core functions. The platform is designed to transform the wearable device of this invention from a passive content consumption tool into an active and creative social medium, thereby building a content ecosystem with strong network effects and user stickiness.

[0325] a) Strategic goals and core values ​​of UGC community platforms

[0326] From the outset, we established three core strategic goals for the UGC community platform:

[0327] 1. Unlimited content supply: Breaking the limitations of official pre-made content in terms of quantity, style, and application scenarios, by empowering millions of users to create, providing the entire ecosystem with a continuous stream of diverse and trend-setting content.

[0328] 2. Deepen user engagement and sense of belonging: Through creation, sharing, interaction, and co-creation, transform users from mere consumers into co-builders of the ecosystem. This sense of participation and the social recognition gained through creation are key to cultivating a highly loyal core user group.

[0329] 3. Driving Data-Driven Product Evolution: UGC communities are a "gold mine" for understanding users' real needs and preferences. Through big data analysis of popular community content, creation trends, and user feedback, the most direct and valuable data support can be provided for the R&D of official scent consumables, the iteration of hardware functions, and the formulation of business strategies.

[0330] b) Backend technical implementation of the UGC community platform

[0331] The server system adopts a modern, scalable cloud-native architecture, and the specific microservice modules are implemented as follows:

[0332] 1. User Account and Authentication Service: Employing OAuth2.0 and OpenIDConnect protocols, it provides secure identity authentication and authorization. It supports registration via mobile phone number / email, as well as unified login from third-party platforms. The user profile database stores not only basic information but also user preference tags, creation history, and more.

[0333] 2. Content Storage and Management Services: OSL recipe files (typically tens of KB JSON / XML files) and unstructured data such as cover images and preview videos are stored in an infinitely scalable cloud object storage service (such as Alibaba Cloud OSS or AWS S3), and configured with CDN for global accelerated distribution. Recipe metadata is stored in a high-performance document database (such as MongoDB), with composite indexes built for key fields such as creator ID, tags, and publication time to achieve millisecond-level complex queries.

[0334] 3. Social Graph and Interaction Services: Using graph databases (such as Neo4j) to model the follow / fan relationships between users enables efficient multi-degree social relationship queries (e.g., "people followed by people I follow"). Frequently written interaction data such as comments and likes can be cached (e.g., Redis) and message queues (e.g., Kafka) to smooth out peak traffic and asynchronously write to the main database, ensuring system responsiveness under high concurrency. Notification services are delivered in real-time via WebSocket or mobile push services (APNS / FCM).

[0335] 4. Recommendation and Discovery Engine: This is a complex data processing system consisting of offline and real-time components. Offline component: Runs batch processing tasks (such as Spark) late every night to calculate collaborative filtering models, user profiles, and content feature vectors based on user historical behavior. Real-time component: When a user opens the app, it retrieves the latest trending lists in real time and combines them with the offline-calculated models and the user's current context (such as time and location) to dynamically generate a personalized recommendation list.

[0336] c) End-to-end technical process for "one-click replication"

[0337] Please see Figure 9 The diagram illustrates a flowchart of the "one-click replication" function, a bridge connecting community content and personal experience. In the integrated or disposable device form factor of this invention, this process is cleverly designed to intelligently match the user's existing device inventory.

[0338] Its backend implementation process is as follows:

[0339] Step 1 (User Request): User B, on their app, clicks the "One-Click Replica" button for an olfactory recipe (e.g., "Cyberpunk Rainy Night") shared by User A. The app immediately sends a POST API request to the server's API gateway, carrying User B's JWT authentication token in the request header.

[0340] Step 2 (Server-side verification and response): The server verifies the token, queries the recipe metadata, and generates a time-limited, signed OSL file download URL, which is then returned to user B's APP.

[0341] Step 3 (Intelligent matching of user terminals): This is the core logic for adapting to the form factor of integrated devices.

[0342] Requirement: User B's app downloads and parses the OSL file, extracting the set of all required aroma IDs for the recipe. For example, {"rain", "neon", "old leather"}.

[0343] Inventory Check: The app will check the "My Devices Library" linked to the user's account. This library records all olfactory stimulation generating devices that the user currently owns and that are in an active state. For each device, the app knows what its internally fixed or integrated odor combination is (this information is reported and stored via Bluetooth when the device is first paired).

[0344] Building the Available Aromas: The app iterates through the user's "My Devices" library, aggregating all scent IDs from all their devices into a large set of available scent IDs. For example, a user might have a "Nature" themed device with {"Floral", "Rain", "Green Grass"} and a "Study" themed device with {"Coffee", "Old Leather", "Tobacco"}. Their available aromas would then be {"Floral", "Rain", "Green Grass", "Coffee", "Old Leather", "Tobacco"}.

[0345] Step 4 (Intelligent Matching Judgment and Guidance Logic): The APP calculates locally whether requiredAromas (requirements: {"rain", "neon lights", "old leather"}) is a subset of availableAromas (possessions: {"floral scent", "rain", "green grass", "coffee", "old leather", "tobacco"}).

[0346] Scenario 1: Partial match or no match

[0347] The app calculates the difference between missingAromas and requiredAromas. In this example, missingAromas is {"neon lights"}.

[0348] Then, the app will pop up a carefully designed onboarding screen, clearly informing the user: "To perfectly experience 'Cyberpunk Rainy Night,' you are missing the key scent of 'Neon Lights.'"

[0349] The interface will provide one or more smart purchase recommendations at the bottom, such as: "Purchase the 'Cyberpunk' themed olfactory stimulation generator (with 7 built-in scents including 'neon' and 'metal')". After clicking, users can be directly redirected to the built-in e-commerce module to complete the purchase of the new device.

[0350] Scenario 2: Complete match

[0351] If requiredAromas is a subset of availableAromas, the "One-Click Replica" button on the app interface will change to the "Start Experience" button.

[0352] The control core of this invention possesses an advanced collaborative scheduling engine capable of managing and executing complex olfactory recipes that exceed the capabilities of a single olfactory module, achieving a "three-dimensional" or "layered" olfactory experience. This engine seamlessly adapts to different hardware configurations: for multiple independent devices (as in Embodiment 1), the engine decomposes the unified recipe into independent instructions and distributes them to the corresponding devices based on their device IDs, enabling synchronous execution across devices. For a single device with multiple channels (as in Embodiment 2), the engine generates a single instruction sequence containing internal channel addresses (e.g., left / right arm), which is then routed and executed internally by the device.

[0353] Through the powerful creative capabilities granted by the aforementioned "digital perfumery platform," and the discovery, interaction, and seamless experience loop provided by the community platform, this invention democratizes the olfactory creative rights that originally belonged to a very small number of professional perfumers, granting them to every ordinary user. This is not merely an addition of functions, but the construction of an olfactory healing culture and content ecosystem that can self-reproduce, self-evolve, and possess infinite vitality.

[0354] Example 11: Feasibility and Key Parameter Analysis of Equipment Engineering

[0355] The embodiments aim to comprehensively demonstrate the engineering feasibility of the two core device forms described in this invention through the most detailed quantitative analysis based on rigorous geometric modeling and real-world engineering constraints, and to supplement the statements on respiratory synchronization and odor safety.

[0356] a) Part One: Intranasal Intelligent Olfactory Capsule (Example 1)

[0357] 1. Analysis of volume, weight, and multi-odor capacity

[0358] The goal of this device is to achieve all functions while maintaining a maximum unfolded size of no more than 15 mm, a maximum diameter of no more than 6 mm, and a total weight of less than 4 grams. The core engineering challenge stems from the large central channel designed to ensure unobstructed breathing. Based on a rigorous geometric model (a semi-annular structure with a centerline diameter of 15 mm and a cross-sectional diameter of 4.5 mm), and through precise calculations using the Paps-Gourdin theorem, the net usable volume truly available to house all components is limited to 400 mm². 3 .

[0359] To accommodate all the necessary systems within this limited space, a realistic trade-off must be made regarding the device's multi-scent capabilities. Therefore, the number of individual scent chambers in the intranasal capsule will be limited to two to four. This engineering constraint, in turn, translates into a clever product positioning strategy: defining the intranasal capsule as a series of "themed" devices (e.g., a "focused alertness" theme including mint, citrus, and rosemary scents) rather than a "one-size-fits-all" device.

[0360] Based on this, the volume and weight of each component are broken down as follows: The energy module (10mAh micro battery) occupies approximately 120mm². 3 (30%), weighing approximately 0.4g; the olfactory release system (containing 2-4 chambers, total capacity 0.05mL) occupies approximately 140mm². 3 (35%), weighing approximately 0.2g; the control and communication module (PCBA) occupies approximately 40mm². 3 (10%), weighing approximately 0.3g; the remaining 100mm 3 (25%) will be used for structural, encapsulation, and assembly space, weighing approximately 1.6g. In summary, the total volume of all components exactly fills 400mm². 3 The net usable space is limited, under demanding but engineering-feasible conditions. The total weight is estimated at approximately 2.5 grams, well below the target of 4 grams, making it entirely feasible.

[0361] 2. Analysis of Battery Life and Battery Capacity

[0362] This analysis is based on the 10mAh (3.7V) battery selected after the above trade-offs, with a total energy storage of 133.2 joules. The device's power consumption can be broken down into two main parts. The first is static power consumption, which is the consumption of the device in deep sleep mode. At this time, the current is approximately 5μA, and the daily static energy consumption is calculated to be approximately 1.43 joules.

[0363] The second is dynamic power consumption, which is the instantaneous energy consumption of performing a single olfactory event. This process lasts approximately 30 milliseconds, and its total energy consumption is the sum of the energy consumption of each stage. A typical event includes: 1) SoC wake-up and BLE reception stage (approximately 5mA, 5ms); 2) Heating control execution stage (approximately 20ms), where the total current is the sum of the microheater operating current (~100mA) and the current required for the SoC to maintain its active mode (~2mA); 3) Closure and sleep entry stage (approximately 5mA, 5ms). According to the formula E=Σ(V*I*t), after precise calculation, the total energy consumption of a single olfactory event is approximately 0.0069 joules.

[0364] Based on this model and by introducing a charging and storage system as a holistic solution, we arrive at the following conclusions regarding battery life:

[0365] Device battery life: For heavy gamers (4000 events per day), after accounting for 30% wear and tear, the actual battery life is approximately 3.3 days. For light notification users (50 notifications per day), the actual battery life is approximately 53 days.

[0366] Total system battery life: A typical charging case (e.g., 400mAh) can provide the device with more than 30 full charges. Therefore, for heavy gamers, the total system battery life will be extended to more than 3 months.

[0367] The conclusion is that the device's battery life is a reasonable trade-off designed for a single immersive experience. As a complete system including the charging case, its overall battery life is extremely powerful and entirely feasible.

[0368] 3. Analysis of Heat and Safety

[0369] The core heat source of the device is the instantaneous pulse of the microheater, with a single event generating only 0.0066 joules of heat. This is an extremely small energy value, and the tiny amount of heat generated is quickly carried away by the abundant capillary blood flow under the nasal mucosa (a highly efficient "living cooling system"). Therefore, the actual temperature rise of the device's outer casing is almost undetectable (below 0.1°C), making it very safe with no significant risk from temperature rise.

[0370] 4. Detailed analysis of consumable usage time

[0371] This analysis is based on an odorant sac volume compressed to 0.05 mL. To ensure the scientific rigor of the analysis, we first need to determine a typical single-dose value that is both clearly perceptible and conforms to the "micro-dose" principle of this invention.

[0372] Scientific Basis Notes: Calculation from injection volume to perceived concentration

[0373] Whether a nose can detect an odor depends on whether the concentration of odor molecules reaching the olfactory epithelium exceeds the substance's "olfactory threshold." Taking limonene (citrus scent), a substance with a moderate olfactory threshold, as an example, its olfactory threshold is widely reported in authoritative industrial hygiene and olfactory science literature at the ppm (parts per million) level, with typical values ​​of 1 ppm or 1000 ppb. Calculations show that when a device sprays 1 nanoliter (nL) of a limonene solution containing a high concentration (e.g., 50%-100%), after complete evaporation in a typical inhaled airflow (e.g., 100 mL), the resulting average molar concentration can stably reach or exceed the 1000 ppb threshold. Therefore, using 1 nanoliter (nL) as a typical, average single-spray consumption value is a scientifically sound and practically feasible choice.

[0374] Total usable uses and usage time calculation: Based on an average consumption of 1 nL per application, a total capacity of 0.05 mL (i.e., 50,000 nL) can support 50,000 applications. Converted to usage time under different scenarios:

[0375] Heavy gamer (4000 times per day): approximately 12.5 days.

[0376] Moderate entertainment user (1000 times per day): Approximately 50 days.

[0377] Light notifications to users (50 times per day): Approximately 1000 days

[0378] Feasibility conclusion: Completely feasible. Even for the most intensive users, a single device can provide nearly two weeks of continuous use, a cycle that reasonably matches the device's charging cycle. The physical lifespan of consumables far exceeds the electronic lifespan (determined by battery cycle life) and chemical shelf life (the recommended shelf life of fragrance formulations, typically 12-18 months).

[0379] b) Part Two: Analysis of the C-shaped Open Ring Device (Example 2)

[0380] The C-ring device (Example 2) demonstrates overwhelming engineering advantages and broader application prospects while providing the same core functions as an intranasal capsule. Its distributed "external nasal surround" structure allows its net usable volume to easily reach 800 mm². 3 This is twice the size of an intranasal capsule. This significant space margin is the fundamental source of all its core advantages.

[0381] First, the C-ring device makes no compromises in its multi-odor capability. Its larger olfactory stimulation generation unit can easily accommodate a complex linear odor capsule divided into 8, 10, or even 12 independent chambers. This allows it to transcend the limitations of a "themed" approach and become a "professional" or "all-in-one" device capable of perfectly executing, mixing, and layering multiple complex olfactory formulations. Users don't need to replace the device itself to experience different types of olfactory content; a single C-ring device can meet all their needs in various scenarios such as entertainment, work, and relaxation, greatly enhancing the product's value and convenience.

[0382] Secondly, the C-ring device represents a significant leap forward in battery life, a core user experience metric. Its larger size allows for easy integration of custom battery packs with a total capacity of 40-60mAh in both arms. For example, with a 50mAh battery (666 joules of total energy), even under the same heavy gaming conditions as an intranasal capsule (consuming 28.3J daily), the device's actual battery life can reach (666J / 28.3J / day)*70%≈16.5 days, exceeding two weeks. In moderate usage scenarios, it easily achieves the ideal "monthly charging" experience. This fundamental improvement in battery life enables it to handle longer periods of continuous use, such as sleep aids during long flights and all-day ambient lighting effects.

[0383] Finally, the C-ring also demonstrates significant advantages in engineering implementation and functional expansion. Its modular partitioned layout (such as...) Figure 2As shown, this significantly reduces component integration density and manufacturing difficulty, which is beneficial for improving product yield and long-term reliability. In addition, the detachable olfactory stimulation generation unit also increases the possibilities of odor combination. The extra space also allows for the integration of more powerful antennas to optimize communication distance, or the addition of non-physiological sensors (such as adding small buttons) to achieve richer interaction methods and environmental adaptation functions, without worrying about encroaching on the space of core functions (battery, olfactory system).

[0384] c) Final Summary

[0385] Intranasal capsules (Example 1): While technically feasible, they represent a highly compromised product designed for ultimate invisibility and privacy. Their biggest drawbacks are limited battery life and multiple scent capabilities, making them more suitable as entry-level devices for light use in specific scenarios.

[0386] C-ring device (Example 2): While providing the same core functions, it is significantly superior to intranasal capsules in terms of engineering feasibility, battery life, multi-odor capability, and functional expandability, representing a more mature, powerful, and user-friendly full-function device.

[0387] d) Fine-tuning of release timing based on respiratory synchronization

[0388] In a preferred embodiment of the invention, particularly in a device configuration with greater space margin as described in Embodiment 2, sensors for monitoring the user's respiratory airflow can be further integrated. Such sensors may include, but are not limited to, microelectromechanical systems (MEMS) thermistor or piezoresistive airflow sensors, miniature microphones for analyzing respiratory acoustics, or strain sensors for monitoring minute deformations of the nose.

[0389] Those skilled in the art will understand that breathing behavior has a unique triple nature: it includes unconscious, rhythmic physiological activity, can be consciously and actively regulated by the user (such as deep breathing, breath-holding, or "inhaling" actions to perceive odors more clearly), and also constitutes the physical airflow environment in which this device operates. Therefore, in the context of this invention, breathing signals are not strictly classified as purely "unconscious physiological parameters" as explicitly excluded in claim 1, but are closer to a physical signal that can be actively interacted with and used for environmental compensation.

[0390] The core objective of this embodiment is to achieve higher-fidelity feedforward guidance, which can be extended to an enhanced interactive experience. Whether this function needs to be integrated depends on the performance parameters of the device's jetting mechanism. If the olfactory stimulus generation unit can generate a sufficiently high jetting velocity, allowing the olfactory information cluster to actively and rapidly travel a distance to reach the olfactory epithelial cells, the influence of respiratory airflow is relatively small, and this respiratory synchronization function can be omitted to simplify the design. Conversely, if the jetting velocity is difficult to achieve the expected speed in engineering, the physical transmission process of the olfactory information cluster is more easily disturbed by the user's respiratory airflow.

[0391] In this scenario, the processing circuitry can be configured to establish a dynamic delivery delay model while executing the external digital control command sequence. This model comprehensively considers both the device's fixed execution delay (from receiving the command to the completion of ejection) and the variable environmental delay (i.e., the impact of the user's respiratory cycle on molecular delivery time) acquired in real time by the respiratory sensor. To this end, the processing circuitry uses the acquired respiratory cycle signal (e.g., predicting the start point and duration of the next inspiratory phase) to predictively fine-tune the release event timestamp in the command. Specifically, the system actively counteracts the blowing effect caused by exhalation and utilizes the acceleration effect generated by inhalation by dynamically advancing or delaying its actual ejection time. The ultimate goal of this timing adjustment is to compensate for the total delay in the entire physical delivery path, thereby ensuring that the actual arrival time of the olfactory molecules at the user's olfactory epithelial cells is as consistent as possible with the target timestamp defined in the original digital control command sequence.

[0392] In more advanced applications, the system can also analyze breathing patterns and intensity. For example, a short, forceful inhalation can be interpreted as a user's "request for enhancement," and the system can temporarily increase the injection dose based on the original intensity of the script. This fine-tuning and interaction does not aim to change the content of the instruction itself (such as the type or intensity of the odor), but rather to more perfectly and unbiasedly reproduce and enhance the intent carried by the instruction. Therefore, this does not constitute a closed-loop biofeedback mechanism to change the user's state, but rather an optimization of the performance of a "high-fidelity actuator," falling within the scope of the present invention's "environmental compensation" or "user-initiated interaction," and is entirely within the protection spirit and scope of this invention.

[0393] e) Personalized content interaction and execution at the system level

[0394] Those skilled in the art will understand that the "open-loop feedforward" olfactory experience device disclosed in this invention constitutes the core execution terminal of the entire digital olfactory ecosystem. In another optional, system-level implementation, a personalized content recommendation and scheduling engine can be constructed to work in conjunction with this device.

[0395] This engine can be deployed on cloud servers or user terminal devices, and its core function is to act as an intelligent "content scheduler." Its purpose is not to interfere with the device's faithful execution of specific olfactory command sequences, but rather to provide users with a more personalized experience at a higher level of content selection. For example, the system can be authorized to integrate user status information from other sources, one important option being generalized physiological data from various wearable sensing devices.

[0396] In this advanced application architecture, the entire workflow is strictly divided into two functionally independent and time-separated stages:

[0397] Phase 1: Content Generation and Selection. In this phase, the content recommendation engine analyzes the user's state data over a specific time period before playback begins. Based on this analysis, the engine generates or selects a completely new, complete, and static olfactory content script. This script itself is structurally identical to any other pre-made script, containing a series of release events with precise timestamps.

[0398] Phase Two: Content Execution and Playback. Once the content script is finalized, it is sent to the olfactory device of this invention as an unalterable, static digital file. From this moment on, all dynamic connections with the user's real-time state are severed. The device of this invention will act as a pure, high-fidelity "open-loop" executor, strictly, unconditionally, and faithfully reproducing every instruction in this script file, without relying on any real-time external feedback signals during its execution decision-making process.

[0399] This separation architecture of "content-layer intelligence" and "execution-layer fidelity" ensures that the core technical features of the device of the present invention remain consistent, serving merely as a final presentation device for the analysis and decision-making results.

[0400] f) Safety considerations and technical boundary statement regarding odor substances

[0401] The core innovation of this invention lies in providing a set of devices, systems, and methods for accurately generating, transmitting, and executing time-series olfactory experiences. Its technical focus is on the conversion process from "information" to "experience." However, the medium carrying this experience is a chemical odorant, and its safety is crucial for the eventual commercialization of the product. To clarify the technical boundaries of this patent, the following statements are hereby made:

[0402] 1. Raw Material Compliance Declaration: In the envisioned commercial application of this invention, all base odor substances used to fill odor components must comply with the relevant laws, regulations, and industry standards of their respective sales regions, such as the usage standards of the International Fragrance Association (IFRA) and national standards. It must be ensured that the substances used are safe for human use and do not pose any toxicity, sensitization, or other potential health risks under their specified usage and at micro-dosage.

[0403] 2. Formulation Stability and Evaluation Requirements: Considering the extremely low probability of unintended chemical reactions occurring in various odor substances during storage or release (such as when heated), any officially provided or certified "olfactory formulation" must undergo rigorous chemical stability testing, toxicological evaluation, and human safety testing before being recommended or sold, in order to eliminate any potential risks.

[0404] 3. Clear Definition of the Technical Solution and Implementation Responsibilities of this Invention: This patent application, as a disclosure of a technical solution, focuses its protection scope and inventive core strictly on the "platform" itself—namely, the mechanical and electronic structure of the hardware, the open-loop control logic of the software, the communication protocol of the system, and the method for constructing the UGC content ecosystem. The aforementioned requirements regarding the safety of chemical substances are regulatory compliance and quality control responsibilities that product developers, manufacturers, and content providers must fulfill when transforming this technical solution into a consumer-facing commercial product. These safety safeguards are prerequisites for applying the technology of this invention, but they do not themselves constitute part of the technical solution of this invention. The logical relationship here can be compared to a patent for a high-definition display screen: this patent protects the structure and driving algorithm of the display panel, but does not include or guarantee the legality or safety of any specific video content played. Similarly, the purpose of this invention is to provide a reliable, high-precision tool for "playing" olfactory information, and the safety of the "content" (i.e., specific odor substances) being "played" should be guaranteed by the content provider (manufacturer or user) in practical application and in accordance with relevant laws and regulations.

[0405] In summary, this invention, by proposing a wearable digital olfactory experience device, system, and method, pioneers a completely new dimension of digital information transmission and sensory experience. From the ultra-miniaturized intranasal and extranasal hardware design, to the intelligent algorithms and system architecture that achieve high-fidelity experiences, and further to the construction of a sustainable content ecosystem, a series of embodiments of this invention collectively demonstrate a complete, feasible, and highly promising technological platform.

[0406] The foregoing, with reference to the accompanying drawings and specific embodiments, has described in detail one or more implementations of the present invention. However, the present invention is not limited to the details disclosed in the above embodiments and drawings. For those skilled in the art, various modifications, combinations, and substitutions can be made without departing from the spirit and scope of the present invention, and all such changes should be considered to fall within the protection scope of the present invention. The protection scope of the present invention should be determined by the scope of the claims.

Claims

1. A wearable digital olfactory experience device, characterized in that, The device comprises: a body structure configured to be wearable on the nasal region of a user; an olfactory stimulus generation unit disposed within the body structure for storing and releasing an odor-producing composition; a wireless communication module; and a processing circuit configured to: a) receive, via the wireless communication module, a digital control instruction sequence from an external control device, the sequence comprising one or more timed release events with precise time stamps; and b) drive the olfactory stimulus generation unit to perform one or more micro-dose releases in an open-loop control logic based solely on the timing and parameters defined in the digital control instruction sequence, wherein the open-loop control logic does not collect or utilize any signals outputted by biosensors for collecting real-time, unconscious physiological parameters of the user to form a closed-loop feedback control during its decision-making process. The body structure is selected from one of:

2. The apparatus of claim 1, wherein, an intranasal structure, such as a ring or capsule structure configured to be placed non-invasively in the vestibule of the nasal cavity of a user; or an external wearable structure to be worn on the external region of the nose of a user, such as the nasal bridge, nasal septum or ala. The olfactory stimulus generation unit comprises an electrically controlled micro-ejection mechanism configured to eject the odor-producing composition as micro-droplets at high speed in a single release event.

3. The apparatus of claim 1, wherein, The electrically controlled micro-ejection mechanism comprises a micro-heater array configured to trigger explosive nucleate boiling of the odor-producing composition by applying a transient high-energy electrical pulse to the odor-producing composition, thereby ejecting the micro-droplets at high speed.

4. The apparatus of claim 3, wherein, The olfactory stimulus generation unit further comprises an olfactory scavenger configured to be released upon receiving a specific instruction to actively remove other olfactory stimuli remaining in the nasal cavity; wherein in a preferred embodiment, the olfactory scavenger is a formulation based on the principle of molecular capture, such as an aqueous cyclodextrin solution.

5. The apparatus of claim 1, wherein, The olfactory stimulus generation unit comprises a replaceable or integrated odor container; the device is further configured to be able to determine identity authentication information associated with the odor container; and the processing circuit is further configured to determine whether to respond to the digital control instruction based on the verification result after verifying the identity authentication information.

6. The apparatus of claim 1, wherein, The method comprises:

7. A wearable olfactory experience method, characterized in that, a) generating or obtaining, on an external control device, a digital control instruction sequence encoding a timed release protocol, the protocol comprising one or more timed release events with precise time stamps; b) transmitting the digital control instruction sequence to the device according to claim 1 via wireless communication; and c) driving the olfactory stimulus generation unit of the device to perform one or more micro-dose releases in an open-loop control manner strictly according to the timing and parameters defined in the instruction sequence without collecting and analyzing real-time, unconscious physiological parameters of the user to form a closed-loop feedback. The digital control instruction sequence is generated based on real-time synchronization signals received from an external digital media source, including a virtual reality application, an augmented reality application, a video player or a music player.

8. The method of claim 7, wherein, ​ 9. The method of claim 7, wherein, The digital control instruction sequence is generated in real-time based on data collected from an external motion sensor attached to a utensil or food container to synchronously release food-related aromas during a user performing a specific eating action.

10. A wearable digital olfactory experience system, characterized in that, The system comprises: a) a wearable digital olfactory experience device as claimed in any of claims 1 to 6; and b) an external control device configured to: i) generate or obtain a digital control instruction sequence encoding a timed release protocol; and ii) transmit the digital control instruction sequence to the wearable digital olfactory experience device via wireless communication to drive its execution of the release.

11. The system of claim 10, wherein, The system further comprises a server configured to provide a user-generated content (UGC) community platform allowing users to upload, share and download digital control instruction sequences corresponding to olfactory recipes.

12. The system of claim 10, wherein, An application running on the external control device provides a visual editing interface allowing users to create or edit olfactory recipes containing multi-step timing information, the interface comprising a multi-track timeline and a library of scent materials.

Images