International Journal of Computer Applications (0975 – 888) Volume 47– No.19, June 2012 12 IFCCDC: A Fuzzy Control based Congestion Detection and Control in Wireless Sensor Networks Rekha Chakravarthi Research Scholar , Sathyabama University C.Gomathy PhD ,Prof & Head ,Dept of Electronics & Control Engineering , Sathyabama University ABSTRACT In Wireless Sensor Networks (WSN’s), congestion plays an important role in degrading the performance of the network. Under idle condition, the network load is very low whereas when an event is detected the network load becomes high which leads to congestion. Due to congestion the overall performance of the network degrades. Hence it is necessary to detect and control congestion. In this paper an efficient technique to detect and control congestion has been proposed. The congestion is detected by calculating a metric called Congestion Degree (Cd). It is the ratio between packet inter arrival time and packet inter service time. Once the congestion is detected, it is notified using Implicit Congestion Notification (ICN) signaling. On receiving the congestion notification signal, the transmission rate is controlled in order to reduce congestion. Further congestion control is implemented using Fuzzy Logic Controller. The performance of the network is measured for delivery ratio with different transmission rate and the PDR is compared with CODA. General Terms Wireless Sensor Network, Fuzzy Logic , Algorithm , Congestion Control. Keywords Wireless Sensor Networks (WSNs), Congestion, Congestion Degree (C d ), Fuzzy Logic Controller . 1. INTRODUCTION The Wireless Sensor Networks (WSNs) consists of several sensor nodes which will monitor the environmental conditions such as temperature, pressure, vibration, etc. Basically the sensor nodes sense the information and process the information. Then they transmit the information as data packets to the sink node. When an event occurs in the network, large number of sensor nodes will report their data packets to sink. When large number of sensor nodes are active in transmitting the data packets the load becomes heavy and traffic will also increases. This might leads to congestion in the network. The very important issue in the WSNs is congestion. There are various reasons for congestion. They are Buffer overflow, Channel contention, Packet collision, Many-to-one nature, etc. [11]. Buffer overflow occurs when the number of incoming packets is greater than the available buffer space. [12] Channel contention occurs between different flows and different packets of a flow. Packet collisions indicate lower level congestion and leads to packet drops. The many-to-one nature of event communication between multiple sources and sink causes bottleneck around sink. 2. RELATED WORK In [1], to detect congestion each node calculates its node rank based on the parameters buffer Size, hop count, channel busy ratio and MAC overhead. When the node rank crosses a threshold value T, the sensor node will set a congestion bit in every packet it forwards. If the congestion bit is set, the downstream node calculates the Rate Adjustment Feedback based on the rank and propagates this value upstream towards the source nodes. The source nodes will adjust their transmission rates dynamically based on this feedback. In Hop-by-Hop congestion control technique [2], the Effective Queue Length (EQL) is measured. EQL is used to determine whether the congestion is occurred or not. The technique called CONSEQ (CONtrol of Sensor Queues) is used to control the congestion. The CONSEQ aims to reduce congestion and, thereby, decrease the delay and energy consumption due to packet losses and retransmission in WSN. In [3], CODA, an energy efficient congestion control scheme for sensor networks was proposed. CODA (COngestion Detection and Avoidance) comprises three mechanisms: (i) receiver-based congestion detection; (ii) open-loop hop-by- hop backpressure; and (iii) closed-loop multi-source regulation. CODA detects congestion based on queue length as well as wireless channel load at intermediate nodes. Furthermore it uses explicit congestion notification approach and also an AIMD rate adjustment technique. In [4], the author proposes a Prioritized Heterogeneous Traffic-oriented Congestion Control Protocol (PHTCCP) to control congestion. It uses packet service ratio to detect congestion. Packet service ratio is defined as the ratio of average packet service rate and packet scheduling rate in each sensor node. PHTCCP uses hop-by-hop rate adjustment for controlling the congestion. The output rate of a node is controlled by adjusting the scheduling rate.