Abdel-Mehsen Ahmad

3D gesture-based control system using processing open source software

The following paper discusses the design and implementation of a 3D gesture-based control system that can substitute any complex human-machine interface with a simpler and more familiar one. This system that can be used by any user disregarding his/her background (age, education, health, language etc.…), making it dependent on nothing more than the movements of the user’s hands, thus allowing them to be free from all hardware while interfacing, and making the system surpass in its simplicity other systems that include held or touched interfaces. It utilizes the Kinect sensor’s image input parts (IR projector and IR camera) for computer vision along with an open source software called Processing for reception and analysis of input data to decision making on what actions and steps to perform, and even for communicating with the database. A MySQL database will be responsible for holding the information of the system and to handle them (adding, deleting, retrieving, etc.…). Even though the system will be designed for a library to display information about the books available in it, however the proposed system design will be broad enough to allow its implementation in many other domains, making people’s lives more efficient and innovative. Implementation steps are listed along with a detailed explanation of how the system works. The simulation results showing the system performance in different environment conditions and some screenshots are also displayed.

Doppler Effect in the Acoustic Ultra Low Frequency Band for Wireless Underwater Networks

It is well known that communication is affected by the change in frequency of a signal for an observer that is moving relative to the source. As a result of the motion of either the source or the observer, successive waves are emitted from a position that may get closer or further to the observer. This phenomenon, known as Doppler effect, also affects underwater acoustic waves used for communications between Autonomous Underwater Vehicles (AUVs), Underwater Sensors (USs) and remote operators. There have been few studies on the impact of Doppler shift in underwater communications. Assuming underwater communications using acoustic waves, in this paper we study the Doppler effect in relation to the half-power bandwidth and distance in the Ultra Low Frequency (ULF) band (i.e., from 0.3 to 3 kHz). We investigate two specific issues. Firstly, the maximum shift that can be expected on underwater links in the ULF band. Secondly, the maximum frequency drift, and associated patterns, that can happen during the reception of data frames. Numeric simulations are conducted. The analysis is based on scenarios that show the existence of significant Doppler effect. More specifically, we show that Doppler effect, under narrow half-power bandwidth, is significant with respect to the half-power bandwidth in the ULF band, for long distance communications. Furthermore, we show that Doppler effect patterns are not necessarily linear.

LISA: Listening-based immediate SCH access scheme in VANETs

Vehicle Ad-Hoc Networks was developed in order to improve the safety and comfort while driving a car. Thereby, IEEE 802.11p and IEEE 1609.4 are used for this purpose. However, these protocols showed some limitations, and this is showed through the low probability of successful transmission for safety messages and through the low throughput achieved for service messages. This paper proposes new mechanism, Listening-based Immediate SCH Access (LISA), which enhances the channel utilization, i.e. improving the probability of successful safety packets transmission and increasing the throughput for service packets. Using LISA scheme, the probability of successful transmission has increased by 14% when the probability of transmission is equal to 80%, and the throughput has increased up to five times compared to the original protocol.

ANC-based broadcasting technique in vehicular networks

In wireless networks, trying to avoid interferences by using allocation and scheduling techniques is not an effective and permanent solution. In a high density network where nodes can have a large communication range, the channel capacity can be insufficient to support all broadcast messages even if they are perfectly scheduled. This is the case of the control channel (CCH) in the IEEE 1609.4 standard for Wireless Access in Vehicular Environments. Moreover, in broadcast messages, there is neither acknowledgement nor retransmissions. Also, the expiry time restriction of safety messages pushes each vehicle to diffuse its message during one CCH interval. To deal with these problems, we propose a new approach which separates two collided messages on the signal level. Our method, which is technically based on the Analog Network Coding (ANC), allows to recover collided packets like the ZigZag decoding technique but without needing retransmissions. Simulations results show a high performance for our scheme in packet reception ratio and channel capacity.

3D sensor-based Moving Human Tracking Robot with Obstacle Avoidance

This paper, represents a design constituting of a robot that uses an Arduino to drive motors enabling it to move around the room in a direction determined by a Processing sketch that uses the Kinect to scan the room and determine where the robot should move. The design works by tracking the center of mass of a human having a skeleton, and avoiding obstacles that interfere it`s movement. The approach used in this system is attaching the computer and Kinect to the robot so that its vision system can move along with it as it explores the space. To accomplish this, a toy car was used as a robot in order to support the weight of a laptop or other small computer as well as the Kinect.