GSM-IoT Smart Pet Feeders

• GSM-IoT smart pet feeders are autonomous devices that leverage GSM SMS protocols for remote feeding control and real-time food-level monitoring.
• The design integrates an Arduino-based control system with ultrasonic sensing and servo-actuated dispensing to achieve ±2.67% portion accuracy and efficient power management.
• Its low-cost, modular architecture and network independence enable scalable deployment in resource-constrained or geographically diverse environments.

GSM-IoT smart pet feeders are autonomous, remotely-operated devices that use cellular GSM networks for communication and control of pet feeding tasks. These systems enable pet owners to monitor and command feeding events from virtually any location with cellular coverage, obviating the dependence on local internet infrastructure. Designed with scalability, cost-efficiency, and robustness as primary objectives, the GSM-IoT pet feeder platform integrates microcontroller-based control, GSM SMS-based remote operation, precise electromechanical dispensing, real-time food-level sensing, and a mobile application interface. A salient implementation is detailed in "Design and Development of a Low-Cost Scalable GSM-IoT Smart Pet Feeder with a Remote Mobile Application" (Nishat et al., 13 Jan 2026), which sets rigorous benchmarks in reliability, precision, autonomy, and cost.

1. System Architecture and Hardware Design

The GSM-IoT pet feeder architecture comprises distinct software and hardware subsystems (Fig. 1 (Nishat et al., 13 Jan 2026)), with the Arduino Uno (ATmega328P) serving as the central controller. The key hardware modules are:

Component	Function	Key Specs
Arduino Uno	Control logic, scheduling	16 MHz, 32 KB
SIM800L	GSM/SMS communication	850–1900 MHz
HC-SR04	Ultrasonic food-level sensing	2–400 cm, +5 V
SG90 Servo	Dispensing gate actuation	180°, 2.5kg-cm
Buck Converter	Voltage regulation	12V→5V @ 1.6A
LiPo Battery	Standalone power	12 V, 1600 mAh

The wiring topology (Fig. 2) is optimized for low electrical noise and reliable operation. The Arduino digital outputs control the ultrasonic Trig/Echo, the PWM line for the servo, and serial connections (with level shifting) to the SIM800L module. The system employs a step-down buck converter to supply a regulated 5 V rail for all major components. Power management strategies include maintaining the servo in a rest state between activations, infrequent polling by the ultrasonic sensor, and invoking SIM800L deep-sleep modes to minimize energy draw (Fig. 12). Baseline consumption is ~125 mA (Arduino), with up to 2 A bursts during GSM transmissions, ~500 mA on servo actuation, and ~15 mA when sensing.

2. Communication Protocols and Remote Operation

Remote control leverages standard SMS protocols for total internet-independence. Command formats adhere to <[PIN](https://www.emergentmind.com/topics/tree-like-pairwise-interaction-network-pin)> <CMD> where $\text{CMD} \in \{\text{FEED, STATUS, RESET}\}$. SMS command parsing is governed by the embedded software (Algorithm 1):

1. Poll for incoming SMS: $G.\text{available}() \rightarrow$ $rawSMS = G.\text{readMessage}()$
2. Parse sender number and command string.
3. Authenticate sender and validate PIN.
4. If authorized, execute corresponding control actions; else, error notification via SMS response.

The dedicated mobile application, built with MIT App Inventor, facilitates user SMS entry via a graphical workflow (Figs. 4a, 4b). Users input the SIM number, PIN, and command, triggering Android's SMS API for transmission and updating the UI in response to status messages received from the feeder.

3. Sensing and Portion-Control Mechanisms

Food-level quantification utilizes time-of-flight ranging with the HC-SR04 ultrasonic sensor:

$d = \frac{v_{\text{sound}} t}{2},$

where $v_{\text{sound}} \approx 343\,\text{m/s}$ and $t$ is the measured interval. If the measured distance $d$ exceeds a preset $d_{\text{threshold}}$, a low-food alert is triggered through SMS.

Dispensing volume per feeding is controlled by the angular displacement of the SG90 servo, itself proportional to the PWM signal:

$$\theta = \frac{\text{PWM}_\text{signal}}{255} \times 180^\circ.$$

Empirical calibration at discrete servo positions $\theta_i$ yields the linear model

$V(\theta) = a\,\theta + b$

with $b \approx 0$ found in prototype trials and $a$ chosen such that a full feed cycle delivers $V_{\text{target}}$. Achieved dispensing repeatability is within $\pm 2.67\%$ variance.

4. Performance, Validation, and Reliability

System performance is validated across key operational metrics:

• SMS command success rate: 98% across 100 trials in multiple locations/providers (Fig. 10b), with end-to-end response latencies of 8–12 s (Fig. 10a).
• Portion control accuracy: For 30 target $\sim$50 g feedings, trial variances remain at $\pm 2.67\%$ (Fig. 11), computed as

$$\text{Variance}(\%) = \frac{|V_{\text{meas}} - V_{\text{target}}|}{V_{\text{target}}} \times 100$$

• Reliability: In continuous runs over 30 days at three scheduled feedings/day, the system incurred no missed feeds or SMS failures, demonstrating robust autonomous operation.
• Battery autonomy exceeds 72 hours under normal duty cycles.

5. Scalability, Cost Structure, and Modularity

The bill of materials totals approximately \$35 (2500 BDT), with major contributors:

Item	Cost (USD/BDT)
Arduino Uno	\$4 / 380
SIM800L	\$8 / 350
Servo	\$4 / 150
Ultrasonic	\$2 / 100
Others	\$17 / 1520

Physical modularity is achieved via a PVC chassis with slide-in component mounts and standardized bus connections. This supports "daisy-chaining" additional feeders or integrating diverse sensors (weight/load cells, temperature, etc.). Firmware parameters distinguish multiple feeders through unique PINs or prefix codes, facilitating deployment in multi-pet or large-scale environments.

6. Network Independence and Comparative Evaluation

GSM-IoT feeder systems distinguish themselves by full functionality in the absence of local internet via SMS-based control. Compared to alternatives:

• GSM/SMS: Ubiquitous coverage without dependence on Wi-Fi, secured by PIN and sender authentication. Limitations include instantaneous current draw and 8–12 s SMS latency.
• Wi-Fi: Allows high data rates and low per-packet power, but depends on local router/internet and is vulnerable to outages. Costs for Wi-Fi/AI smart feeders commonly exceed \$100.
• LoRaWAN (not implemented): Theoretically provides long range and ultra-low power at the expense of reliance on private/public gateways and increased protocol complexity.

The comparative table (Table V (Nishat et al., 13 Jan 2026)) documents that the GSM feeder achieves the lowest cost, full scheduling, real-time alerts, and true offline operation, as opposed to mechanical feeders (no remote features) or advanced Wi-Fi/AI feeders (higher cost, internet reliance).

Current trade-offs include the power/sporadic latency intrinsic to GSM; for ultra-remote or extremely low-power deployments, solar augmentation or LoRaWAN adoption are suggested. System extensibility toward RFID or camera-based pet identification, solar charging, weight-based volumetric correction, and GPRS-driven OTA firmware updates are identified as future work.

In sum, the GSM-IoT pet feeder achieves a distinctive position by combining professional-grade reliability (98% SMS success, ±2.67% portion variance), autonomy, scalable modularity, and true network independence at sub-\$40 total cost, tailored for resource-constrained, geographically diverse deployment environments (Nishat et al., 13 Jan 2026).