The relentless global population growth and the ever-increasing food demand pose formidable challenges to the agricultural sector. Farmers grapple with the ongoing challenge of wildlife-induced crop damage and human loss, which not only impedes food production but also exacerbates supply and demand imbalances. However, the rise of TinyML enables Edge AI as a promising avenue for implementing resource-efficient deep learning techniques on low-end edge devices. In this paper, we introduce an innovative solution that harnesses the power of Edge AI using tinyML-based deep learning algorithms in conjunction with the Internet of Things (IoT) for animal intrusion detection and deterrence system. The proposed system is developed to create remotely managed defense system tailored to safeguard vast agricultural expanses. It integrates a laser detection system and an AI-CAM with light weight deep learning algorithms for animal intrusion detection and classification. This system also ensures efficient animal deterrence and real-time monitoring for farmers, enabling them to assess the situation with the assistance of an intelligent rover build using IoT. This work emphasizes on proposing a light-weight deep learning model named EvoNet for animal classification. Results reveal that the proposed model achieves the highest accuracy at 96.7%, surpassing other models presented in this paper. However, for edge devices where compact file sizes are crucial, the model also offers comparable accuracy with file sizes as low as 1.63MB with the help of pruning and quantization techniques. This conceptualized solution has the potential to revolutionize agricultural wildlife management, ushering in a new era of crop protection and economic resilience.
In this letter, we consider IEEE 802.11ah based IoT network with rate adaptation, where all the devices contend for the channel access using group-based restricted access window (RAW) mechanism with default uniform grouping scheme. In such multi-rate networks, as each group contain distinct data rate devices, we have observed drastic degradation of throughput performance due to performance anomaly. To resolve this problem, we propose a novel grouping scheme based on data rates. Analytical and simulation results show that the proposed scheme significantly improves aggregate throughput of RAW mechanism with minimal additional overhead, when compared to default uniform grouping scheme.
IEEE 802.11ah defines amendments to IEEE 802.11 to support the Internet of Things (IoT). IEEE 802.11ah implements restricted access window (RAW) mechanism to reduce the contention and energy consumption in dense IoT networks. The RAW mechanism is a group-based MAC protocol that partitions the devices into various groups and confines the channel access of a group of devices to the restricted time interval known as the RAW slot. However, the standard does not specify, grouping mechanism, duration of RAW slots, and the number of RAW slots while configuring the RAW mechanism. In an IoT network, each device has distinct transmission requirements. Thus, it is necessary to find the optimal number of RAW slots that can maximize the network performance, to group the devices with similar transmission requirements and to assign a RAW slot that adaptively varies with the traffic requirements of the respective group. In this paper, we exploit fuzzy logic to find the optimal number of RAW slots by considering network size, collision probability, and modulation and coding schemes. Further, we propose a traffic-aware adaptive RAW slot allocation (TARA) scheme that uses fuzzy c-means clustering algorithm to group the devices with similar traffic requirements and to assign each group with a RAW slot whose duration adaptively varies with the transmission requirements of the devices. We have also presented a simple yet accurate analytical model to evaluate the performance of the RAW mechanism. Results show that the optimal number of RAW slots found using fuzzy logic significantly enhances the performance of the RAW mechanism in terms of throughput and energy consumption. Further, it is observed that the TARA scheme can effectively meet the traffic requirements of different group of devices when compared to the uniform grouping scheme. Finally, extensive simulations are conducted using ns-3 to validate the analytical results.