IoT Patent Strategy

Divided infringement is the most significant unique patent challenge for IoT systems, because modern IoT architectures distribute functionality across multiple parties: THE DIVIDED INFRINGEMENT PROBLEM: patent infringement requires that a single party perform all steps of a method claim (for method claims) or use/make a system with all elements (for system claims); IoT systems typically involve: (1) the device (sensor; microcontroller; firmware); (2) a gateway or edge computing node; (3) a cloud service or platform; (4) an end user or operator; if a patented method claim has steps performed by the device AND steps performed by the cloud service AND steps performed by the user, no single party performs all steps; result: no direct infringement by any single party even if all steps are collectively performed; THE AKAMAI STANDARD: Akamai Technologies, Inc. v. Limelight Networks, Inc. (Fed. Cir. 2015, en banc): a party is liable for direct infringement of a method claim when that party: (1) performs all the steps of the claimed method; OR (2) directs and controls a third party's performance of the remaining steps; OR (3) forms a joint enterprise with a third party for performance of the remaining steps; DIRECTION AND CONTROL TEST: a party 'directs or controls' another when: the other party acts under the first party's instructions or directions; the other party performs the steps in a way specified or required by the first party; the first party conditions participation in an activity or receipt of a benefit on performance of the steps; JOINT ENTERPRISE: requires: an agreement (express or implied); a common purpose; a community of pecuniary interest; equal right to voice direction; this is typically harder to establish than direction/control; STRATEGIES TO ADDRESS DIVIDED INFRINGEMENT IN IOT: (1) SYSTEM CLAIMS (most effective): a system claim can be infringed by a single party even when different components are at different locations; if the cloud company provides the complete IoT system (device + platform + network), they infringe the system claim; covers the entire IoT system in a single claim; (2) CRM CLAIMS: a CRM (non-transitory computer-readable medium) stores instructions across the entire system; one party who controls the software infringes the CRM claim even if the software runs on distributed hardware; (3) DEVICE-ONLY METHOD CLAIMS: limit method claims to steps performed entirely on the device (or entirely on the cloud platform); avoid claiming cross-actor steps in a single method claim; (4) PROVIDER-FOCUSED CLAIMS: claim the method from the perspective of the service provider who directs and controls all steps.

IoT devices nearly universally implement one or more communication standards (5G; 4G LTE; Wi-Fi; Bluetooth; Zigbee; Thread; LoRaWAN), all of which are covered by patents that may be standards-essential: WHAT ARE STANDARDS-ESSENTIAL PATENTS (SEPs): a patent is standards-essential (SEP) if it is technically necessary to implement a standard — there is no technically viable alternative to the patented feature for a compliant implementation; standards development organizations (SDOs like 3GPP; IEEE; IETF; ETSI; Bluetooth SIG) require members who contribute to a standard to declare any patents they believe may be SEPs; contributors who declare SEPs commit to license them on FRAND (fair, reasonable, and non-discriminatory) terms; 5G SEPs AND IoT: 5G is the primary connectivity standard for industrial IoT, smart cities, connected vehicles, and cellular IoT; MAJOR 5G SEP HOLDERS: Huawei (largest declared 5G SEP holder); Qualcomm; Nokia; Ericsson; Samsung; ZTE; InterDigital; FRAND RATE FOR 5G TOTAL ROYALTY STACK: varies but industry-wide discussions estimate total 5G royalty stacks of 3-8% or more for 5G-connected devices; individual SEP holder FRAND rate demands vary widely; RECENT SEP LITIGATION: UK Supreme Court (Unwired Planet 2020): UK courts can set global FRAND rates; Nokia v. Daimler (German courts): automotive SEP licensing fights; Ericsson v. Samsung; Qualcomm v. Apple; Wi-Fi SEPs AND IoT: Wi-Fi 6 (802.11ax) and Wi-Fi 7 (802.11be) are used extensively in smart home; industrial IoT; and commercial IoT; Wi-Fi SEP holders include Qualcomm; Broadcom; InterDigital; Intel; Motorola/Google; Wi-Fi Alliance licensing through patent pools (Via Licensing; Sisvel); BLUETOOTH SEPs: Bluetooth 5.x dominates short-range IoT (wearables; smart home; healthcare); Bluetooth SIG membership and royalty structure; LORAWAN AND LPWAN: Semtech holds key LoRa physical layer patents; LoRa Alliance licensing; PRACTICAL IMPACT ON IoT PRODUCTS: an IoT device manufacturer must account for SEP royalties for each standard the device implements; multiple standards on a single device = stacking royalties; small IoT devices at low price points (<$10) may have significant margins consumed by SEP royalties; negotiating direct licenses vs. relying on component supplier pass-through licenses; COMPONENT-LEVEL vs. DEVICE-LEVEL LICENSING: some SEP holders (particularly Qualcomm) insist on device-level licensing (royalty on device sale price); others accept component-level licensing (royalty on chipset); the distinction can be worth billions in a large IoT deployment.

Beyond the connectivity standards landscape, IoT innovation in hardware design, firmware architecture, and sensor systems generates commercially valuable patents: SENSOR AND SENSING SYSTEM PATENTS: MEMS SENSOR PATENTS: MEMS (Micro-Electro-Mechanical Systems) accelerometers; gyroscopes; pressure sensors; microphones are fundamental IoT components; Bosch; STMicro; InvenSense/TDK; Knowles dominate MEMS sensor IP; sensor fusion: combining accelerometer + gyroscope + magnetometer data using specific algorithms (Kalman filter variants; quaternion-based orientation); edge AI for sensor classification (vibration-based machine fault detection; acoustic anomaly detection); POWER MANAGEMENT AND BATTERY LIFE: energy harvesting patents: ambient light; thermal gradient; RF energy harvesting; ultra-low-power MCU operation and sleep/wake protocols; specific power gating architectures; sub-threshold circuit design; SENSOR-TO-CLOUD PIPELINE PATENTS: specific edge preprocessing to reduce data transmission (specific compression; specific feature extraction); specific OTA update mechanisms for constrained devices (differential updates; atomic update mechanisms); specific time synchronization protocols for IoT sensor networks; SECURE IoT PATENTS: specific secure boot and attestation mechanisms for IoT; hardware security module (HSM) integration in embedded systems; specific device identity provisioning (X.509 certificate management; DTLS for IoT); specific side-channel attack mitigation in constrained IoT implementations; MESH NETWORKING PATENTS: Thread protocol IP (Nordic Semiconductor; Google/Nest; Silicon Labs); Zigbee IP (Silicon Labs; NXP); specific self-healing mesh network algorithms; specific distributed routing for battery-constrained mesh networks; INDUSTRIAL IoT (IIoT) SPECIFIC PATENTS: OPC UA (industrial communication standard) implementation patents; specific real-time data synchronization for industrial control; specific time-sensitive networking (TSN) for industrial Ethernet; specific failure detection and redundancy for IIoT; PATENT VALUATION CONSIDERATIONS: the commercial value of an IoT patent depends on: how broadly it reads on actual products (vs. design-arounds); whether it covers a standard implementation or a proprietary innovation; the size of the addressable market; the difficulty of designing around it; IoT hardware patents (specific MEMS structure; specific RF circuit topology) tend to have cleaner infringement analysis than IoT software patents (which face Alice challenges).

Building an effective IoT patent portfolio while managing the complex FTO landscape is a critical strategic challenge for IoT startups: BUILDING THE IoT PATENT PORTFOLIO: START WITH WHAT IS GENUINELY NOVEL: most IoT startups have genuine innovation in: the algorithm (edge AI; sensor fusion; specific control logic); the hardware design (novel circuit architecture; power management); the system architecture (specific device-cloud interaction; specific network topology); the user experience (specific interaction model; specific data visualization); START EARLY: file provisional patent applications before product launch, beta tests, or public demonstrations; the provisional captures priority date at lower cost; file non-provisional within 12 months; CLAIM STRUCTURE FOR IoT PORTFOLIO: (1) System claims covering the complete IoT system (device + gateway/cloud + software), assigned to the product entity; (2) Device-level method claims (steps performed only on the device), for hardware sales scenarios; (3) Platform-level method claims (steps performed only on the cloud/service), for SaaS scenarios; (4) CRM claims (software on a single medium), for software licensing scenarios; FREEDOM-TO-OPERATE FOR IoT PRODUCTS: LAYER 1 — STANDARDS ROYALTIES: identify every communication standard the device implements (5G; Wi-Fi; Bluetooth; Zigbee; Thread; LoRaWAN; etc.); obtain legal opinion on whether component supplier licenses cover device; determine whether direct FRAND licenses are needed; budget for royalty stacks; LAYER 2 — COMPONENT IP: RF chipset IP (Qualcomm; Broadcom; MediaTek); sensor IP (Bosch; STMicro; InvenSense); MCU IP (ARM; MIPS; RISC-V for CPU; various peripheral IP); LAYER 3 — OPERATING SYSTEM AND MIDDLEWARE: RTOS licenses (FreeRTOS Apache 2.0; Zephyr Apache 2.0; ThreadX now open source; RTEMS GPLv2); IoT cloud SDK licenses (AWS IoT; Azure IoT Hub; Google Cloud IoT terms); LAYER 4 — APPLICATION SOFTWARE: compression codec IP; security protocol IP (wolfSSL; mbedTLS); database IP; DEFENSIVE CONSIDERATIONS: join LOT Network to protect against NPE patent assertions from pool members; join Open Invention Network (OIN) for Linux-adjacent IP protection; build cross-licensing relationships with complementary technology companies; consider defensive publication for obvious design decisions that don't merit patenting.