The exponential rise in wireless transmission has gained widespread attention to meet major mobile communication purposes including, Internet of Things (IoT) and Machine-to-Machine (M2M) communications. Mobile Ad-hoc Network (MANET) has emerged as one of the most viable solution to meet aforesaid mobile communication purposes; however, being decentralized and infrastructure-less network it undergoes adversaries including dynamic topology and security breaches due to malicious node attachment during routing. On the contrary, data security and allied Quality-of-Service (QoS) provision are inevitable in NextGen communication systems. Unlike routing-based security measures, Medium Access Control (MAC) based approaches are found more effective for MANETs. However, most of the classical MAC designs either address QoS or security as standalone objective. Unlike existing MAC solutions, in this paper a state-of-art novel Huffman Coding and Multi-Generation Mixing (MGM) assisted random linear network model-based MAC design (HM2-MAC) is proposed for MANET. Realizing the robustness of the network coding algorithms towards reliable, secure and error-free multicast transmission, we designed HM2-MAC in such manner that Huffman coding helps securing the source data packets, while MGM concept enables reduction in redundant packets to make overall communication resource efficient and secure. Unlike redundant packet-per-generation, MGM concept helps in reducing redundant transmission and hence achieves resource efficiency. Since, in HM2-MAC model the coefficient matrix used to encode the data is known only to the sink, no intermediate node can decode it or can pollute it. It makes multicast transmission more secure over MANET. Additionally, the use of iterative buffer flush technique preserves resources or buffer to accommodate more data for transmission and hence higher throughput. Noticeably, error sensitive packetization and MGM control strengthens our proposed model to retain optimal performance. HM2-MAC has been applied as a sub-layer of native IEEE 802.11 MAC and hence retains backward compatibility towards real-world implementation. MATLAB based simulation revealed that our proposed HM2-MAC protocol achieved almost 99.6% throughput even under varying link-loss patterns, which reveals its robustness to ensure QoS delivery with secure data transmission in MANET.

No abstract

No abstract

The face of human may be a muddled visual dimension model and is therefore extremely difficult to create a computing model for the cognitive basic process. The paper displays a system for perceiving the human face smitten by image-based highlights. The technique proposed is available in 2 phases. In an image using Viola-Jones calculation, the main preparation distinguishes the human face. Using a combination of Principle Component Analysis and Artificial Neural Network, the distinguished face within the image is perceived at the next stage contrasting the execution of the proposed strategy with existing ways. The proposed strategy recognizes greater accuracy in the acknowledgement.