Wireless Body Sensor Networks (WBSNs) are deployed in health-related applications and their usage has rapidly increased in the past decade. Collecting various human vital signals such as heart rate or blood pressure has been made possible with the usage of WBSNs that can be implanted in or attached on the human body for monitoring and other purposes. WBSN technology can be beneficial for patients in terms of being monitored almost at all times, and signalling professional care-givers if required. Although WBSNs have significant advantages in both medical and non-medical fields, their characteristics create quite a number of issues, including mutual interference, inefficient frequency spectrum utilisation, limited resources, communication capabilities and storage capacity, and energy constraints. One of the most challenging issues is the mutual interference caused by neighbouring WBSNs. As the number of co-located IEEE 802.15.4-based WBSNs becomes larger in an operating frequency, their performance will be eventually influenced by the mutual interference. In fact, mutual interference eventually causes significant performance degradation due to inefficient channel utilisation. We have investigated the impact of mutual interference on WBSN performance and a number of effective solutions is proposed to enable IEEE 802.15.4-based WBSNs to share and better utilise time- and frequency resources. More specifically, the adaptive resource allocation schemes considered in this thesis aim to improve the packet transmission reliability and to reduce the overall energy consumption in the presence of mutual interference. To accomplish this, a number of distributed schemes are proposed that step-by-step improved the packet transmission reliability and reduced the overall energy consumption. Thereafter, the sensitivity of the performance of WBSNs is examined against the variation of several important system parameters for the considered schemes. This analysis has not only determined the percentage contribution of each considered system parameter to the overall performance of WBSNs, it also suggested that frequency adaptation needs to be augmented by adjusting the transmission timing of the involved WBSNs to achieve performance levels that come close to what can be achieved by an allocation computed by a centralised algorithm. An adaptive scheme is designed and introduced that is not only able to deal with the mutual interference, but which can also be implemented on real WBSNs without substantial modifications to the IEEE 802.15.4 standard. The results obtained from both simulation-based and experimental studies have shown that the combination of both adaptive frequency hopping and adaptive phase shifting strategies can significantly improve the overall performance of WBSNs under strong mutual interference.