Analysis of Web-Based IoT through Heterogeneous Networks: Swarm Computing over LoRaWAN

The Internet of Things (IoT) leverages added valued services by the wide spread of connected smart devices. The Swarm Computing paradigm considers a single abstraction layer that connects all kinds of devices globally, from sensors to super computers. In this context, the Low-Power Wide-Area Network (LPWAN) emerges, spreading out connection to the IoT end devices. With the upsides of long-range, low power and low cost, LPWAN presents major limitations regarding data transmission capacity, throughput, supported packet length and quantity per day limitation. This situation makes LPWAN systems with limited interoperability integrate with systems based on REpresentational State Transfer (REST). This work investigates how to connect web-based IoT applications with LPWANs. The analysis was carried out studying the number of packets generated for a use case of REST-based IoT over LPWAN, specifically the Swarm OS over LoRaWAN. The work also presents an analysis of the impact of using promising schemes for lower communication load. We evaluated Constrained Application Protocol (CoAP), Static Context Header Compression (SCHC) and Concise Binary Object Representation (CBOR) to make transmission over the restricted links of LPWANs possible. The attained results show the reduction of 98.18% packet sizes while using SCHC and CBOR compared to HTTP and JSON by sending fewer packets with smaller sizes.

THE DATA REDUCTION USING MATLAB FOR DIFFERENT SPECKLE IMAGES FORM SMALL SURFACES ROUGHNESS

The so-called laser speckles are bright spots and dark spots formed when a coherent ray is incident on a rough surface which scattered randomly in all directions, the interference of these scattered rays form these bright and dark spots (Laser Speckles).In this paper we are concerned with the formation of Objective speckles calculations. Using MATLAB the image can be converted into binary object (0 and 1) as the speckle spots intensities are dark and bright, respectively.To simplify the calculations, two processes (transform and predictive) may be used, and according to the loss of many data for the using of predictive process, the transform process is considered.The calculations are based on the evaluation on small roughness of surfaces in range 0.1 – 1 μm, on the same footing the contrast was considered in the range from zero to one.Fraunhofer diffraction Unfortunately, no calculations in this field had been done from other researchers.