Energy-Efficient Coding and Modulation Methods for Interference Suppression in Wireless Sensor Network Systems

An Energy-Efficient DS-CDMA communication and modulation technique for reducing multiple access interference (MAI) as well as for reducing power consumption in DSCDMA wireless sensor LAN systems is presented. Source symbols are represented by a special coding, termed Minimum Energy coding (ME coding), which exploits redundant bits for saving power when transmitted via RF links with On-Off-Keying. Since, in OOK, energy is consumed mostly when high bits are transmitted, the ME coding represents source symbols with the least number of high bits, given an extended word length. This ME coding is applied to DS-CDMA sensor network systems in order to reduce MAI as well as to reduce power consumption. When each channel uses the ME coding combined with a spreading code and OOK, fewer high bits are transmitted and thereby the probability of multiple channels sending signals at the same time is lowered. This implies that the MAI is reduced. It is shown that, with the new low MAI ME coding, signal-tonoise ratio is significantly increased and that the error probability is lowered as well. First, the architecture of this low MAI DS-CDMA communication system is described, followed by the principle of interference reduction. Second, a signal model is obtained for the new source coding and transmission process, and the SNR and error probability are analyzed. Third, guidelines for designing the optimal length of codeword, the number of communication channels, and power efficiency are obtained. Finally, a RF sensor network system is designed and simulation is conducted to verify the theoretical results and demonstrate the low MAI and low energy features of the sensor network. Thesis Supervisor: Haruhiko H. Asada Title: Professor of Mechanical Engineering