This paper presents an experimental evaluation of communication during human walking motion, using the medium access control (MAC) evaluation system for a prototype ultra-wideband (UWB) based wireless body area network for suitable MAC parameter settings for data transmission. Its physical layer and MAC specifications are based on the draft standard in IEEE802.15.6. This paper studies the effects of the number of retransmissions and the number of commands of GTS (guaranteed time slot) request packets in the CAP (contention access period) during human walking motion by varying the number of sensor nodes or the number of CFP (contention free period) slots in the superframe. The experiments were performed in an anechoic chamber. The number of packets received is decreased by packet loss caused by human walking motion in the case where 2 slots are set for CFP, regardless of the number of nodes, and this materially decreases the total number of packets received. The number of retransmissions and the GTS request commands increase according to increases in the number of nodes, largely reflecting the effects of the number of CFP slots in the case where 4 nodes are attached. In the cases where 2 or 3 nodes are attached and 4 slots are set for CFP, the packet transmission rate is more than 95%. In the case where 4 nodes are attached and 6 slots are set for CFP, the packet transmission rate is reduced to 88% at best.