An Enhanced Approach for Reliable Bulk Data Transmission Based on Erasure-Resilient Codes in Wireless Sensor Networks

We suggest an enhanced approach for reliable bulk data transmission of image files, multimedia video files, and successive log files for monitoring systems over wireless sensor networks. Recently, some approaches have been proposed that use multiple blocks in a frame. These approaches simply drop the block that contains the error and the error-free blocks are transmitted to other nodes. This paper proposes an enhanced FEC scheme, called M-FEC, which reduces the overhead and the total number of transmissions. M-FEC fragments bulk data into many small blocks and these blocks are encoded using rateless codes. Only the erroneous blocks that are detected in a frame are dropped and a relay node waits for other incoming blocks in order to construct a full-sized frame. This simple approach can improve the performance of the FEC scheme over WSNs. Our experiments show that M-FEC reduces the number of received frames at the sink node by almost half compared to the previous approach. We compared our results for different degrees of Wi-Fi interference by choosing particular IEEE 802.15.4 channels and running our experiments for three different types of Wi-Fi interference. 1. Introduction Wireless sensor nodes with real-time monitoring and multimedia capabilities can collect and transmit large images, videos, and successive log files from the environment to a sink [1–4]. Bulk data transmissions require reliable transmission from a source node to a sink node. The Automatic Repeat reQuest (ARQ) and Forward Error Correction (FEC) schemes are the most widely used error control techniques. The ARQ error recovery scheme requires retransmissions that increase cost and overhead even if improved retransmission algorithms are adapted. Two types of FEC (bit-level and packet-level FEC) have been proposed to guarantee reliability. While the bit-level FEC recovers from bit errors within a frame by including redundancy within each frame, the packet-level FEC recovers from packet loss through the use of redundancy packets. In the bit-level FEC method, a proper amount of redundancy has to be added because excessive redundancy wastes resources over a good quality channel and insufficient redundancy does not allow for recovery from errors over a low-quality channel. Owing to the unpredictable changes in wireless channel quality, it is very difficult to gauge the FEC encoding redundancy to match the current channel error rate. In the packet-level FEC method, the sender takes the original contents (i.e., the original symbols) as input, generates the encoded symbols, and then transmits