Other Projects

Performance Analysis of Inter-vehicle Communication (IVC) Networks

Inter-vehicle communication (IVC) has the potential to greatly improve road traffic safety and efficiency. The equipped vehicles in an IVC system form a special mobile ad hoc network (MANET), which is known as vehicular ad hoc networks (VANET). The characteristic of VANET is that the mobile communication nodes have special mobility patterns: the positions of nodes are limited to road network topology, and the density of vehicles may vary dramatically due to the driving behaviors. The network topology changes quickly due to the high mobility of vehicles and it is difficult to maintain a stable multi-hop communication path between two communication nodes. When the penetration rate of communication unites is low, it will lead to more challenges for the vehicles to set up a routing path. We thus want to evaluate the communication performance of IVC networks in different traffic scenarios: 1) two-lane bidirectional traffic streams, 2) multilane dynamic traffic streams and 3) a unidirectional dynamic traffic flow with shockwaveand rarefaction wave.

The study is performed by combining traffic flow theories (Lighthill-Whitham-Richards (LWR) model and Car Following model) and VANET communication Simulation in NS2. The GPSR protocol is implemented in NS2 in Ubuntu Linux. By changing the transmission range of communication units and penetration rate of equipped vehicles, we studied several performance measures, such as average throughput, end-to-end delay and number of hops. Monte Carlo method is adopted to overcome the randomness caused by low penetration rate of equipped vehicles. The simulation is conducted using TCL language and the results are analyzed using Bash and Gawk.

 Inter-Vehicle Communication 1Inter-Vehicle Communication 2Inter-Vehicle Communication 3

“Sunplus Cup” Electronics Design Competition: Mini Fire Engine Design

Our team were expected to design and make a mini fire engine that could locate “fire” (lighted candles which appear in random positions), choose and follow the best route to approach fire, evade barriers, extinguish the fire, and return to the garage.

We proposed the entire system scheme from scratch, designed the circuits and soldered components on the breadboards. We used respectively flame detector, infrared photodetector, and reflecting-infrared sensors to locate the fire, detect the barriers, and navigate the car (by detecting white lines on a black background). The overall control system is implemented with microchip SCM (single chip microcontroller) SPCE061A of SUNPLUS, which is programmed using C language in Keil C51. After receiving the signals from the sensors, the control system can deal with the data, calculate the best route, and proceed thorough controlling the left and right motors. The engine is able to compute and display the distance to the destination along the way. Once the task is completed, an alarm is provided using both sound and LED lights. For hardware, we cut and polished the vehicle’s base and wheels. The whole system is described in a report "Intelligent Electric Fire Engine" with 22,650 Chinese words. We won the Second Place Award in this competition.

Engine 1Engine 2Engine 3