Mohd Ariff Abdullah

A region-to-point indoor localization approach via RSS-magnetic fingerprinting

Indoor localization based on wireless fingerprinting has been an area of interest for many researchers. In this paper, we introduce a method for leveraging RSS and magnetic vectors in order to estimate the location of a smartphone device within an indoor office environment. Prior to determining a location, a fingerprint map has been populated following specific criteria to define a virtual region. We will later reveal that by employing such region-based scheme, the location estimation can become more accurate. The steps that we have used in our indoor localization estimation follow a two-level approach whereby the first step aims at determining the region where the target device resides. The second step narrows down further to determining the exact location of the device based on the magnetic flux reading obtained earlier. However, the second step would be included in our future work. We present our findings based on the location estimation accuracy of our indoor localization method.

Multi hop cluster head multicast protocol

Multicast has been identified as the next generation multimedia transmission protocol for the Internet. With an increasing growth of WiFi enabled mobile devices, WiFi multicasting is rapidly becoming a necessity. Wireless networks do not efficiently support multicast, so there have been proposals to convert multicast to unicast for efficient and reliable delivery. This process is not robust enough to support multiple clients. This paper proposes a cluster head based mechanism to distribute cascaded multicast contents to the wireless clients. The paper further evaluates the performance of a cluster head mechanism against the standard mechanism to deduce robustness related conclusions.