As people have become more aware of their actions and how they affect their surroundings, they have realized the dire state of the environment. As a result, the recycling movement has gained momentum as a measure to save it. Contemporarily, the recycling industry has not seen a major shift and the problems that existed decades earlier persist. Trash classification is at the core of these problems, because if you can’t classify it, obviously you can’t recycle it. Manual classification often leads to misclassification as humans classify or judge things based on their experiences, knowledge, and not just absolute facts. Additionally, if the waste to be sorted is toxic, being in direct contact may be physically harmful to the people involved. Until a solution is found for this problem, the recycling industry won’t be on par with the rise in recycling culture. Thus, this is the problem we have set out to solve; this paper proposes a Deep Learning model based on EfficientNet Architecture that can classify different kinds of trash with an accuracy high enough to make it a viable solution for the industry while using a comparatively lower number of parameters than existing methods. We achieved an accuracy of 98% on the TrashNet dataset, the standard data for Trash classification, and outperformed all existing models. Additionally, as no large dataset with a varied set of trash images was present, we created a new dataset of 8135 images by combining various datasets and standardized them, achieving a classification accuracy of 92.87% with EfficientNet B3.

Internet of things (IoT) devices are resource-constrained devices with limitations such as low computation power, low communication capabilities, low bandwidths, high latency and short lived power. Therefore, securing the communication among these devices is a key challenge for various sensitive applications. However, conventional encryption and decryption algorithms, known as ciphers, cannot be implemented because of their inherent complexities of implementation and power requirements. One of the promising options available is to implement light-weight ciphers for these resource constrained devices. Moreover, the choice of lightweight encryption tool has great dependency on the type of IoT devices being used in an application. In this paper, a lightweight cellular automata (CA) based cipher, named as Lightweight CA Cipher (LCC) has been proposed for IoT applications. In the proposed method, encryption is done at the perception layer, where the sensor nodes are deployed and decryption is done at the network layer where gateway devices are installed. Experimental results show that the proposed method is efficient than some of the existing ciphers like DES, 3DES when randomness, execution time and implementation simplicity are considered as prime requirements. This cipher passes the randomness tests as prescribed by National Institute of Standards and Technology (NIST), and it also passes all the DIEHARD tests and it establishes the security feature of LCC. Though it is specially designed for resource-constrained environments, it can be scaled up for a large number of sensor nodes.

Cellular Automata (CAs) is one of the most engrossing field for encrypting sensor data applied in Wireless Sensor Networks (WSN). In wireless communications, requirement of security and privacy of information is must. Therefore, transmission of sensor data over wireless communication channel plays a crucial role using cryptography techniques. The usage of cryptography characteristics of cellular automata are still not much explored in WSN. Hence in this paper, we present a symmetric key cryptography technique of block cipher using cellular automata (CAs) rules applied to sensor data in WSN. This cryptography technique uses non- complemented cellular automata rules and hybrid CA rule vector to form a group cellular automata that would be used to encrypt and decrypt sensor data. Proposed methodology has been implemented in C to validate the results.