Search

This paper considers the resource allocation problem of Space Division Multiple Access (SDMA)/ Multiple Input Single Output (MISO)/Orthogonal Frequency Division Multiple Access (OFDMA) systems. Equal power allocation (EPA) is used among subcarriers to reduce the complexity of the problem. Furthermore, a novel algorithm called proportional rate greedy (PRG) is proposed to choose users and allocate powers among selected users for each subcarrier. Such EPA-PRG algorithm can be applied to different fairness issues. From simulation, it can both achieve high throughput and ensure the fairness among users without increasing the complexity.

Parallel RTS Processing MAC Protocol (PRP-MAC) can improve network throughput, but data packets are not always sent in a synchronized way. So PRP-MAC still produces the packet collisions in the network, which impact the performance of the network. In this paper, we proposed an improved PRP-MAC called IPRP-MAC protocol. This protocol can avoid collisions by using 3 types of CTS packets which are used to make sure the transmitters sent the data packets synchronously. Finally, the simulation results show that IPRP-MAC can enhance the network throughput compared to PRP-MAC.

It is well known that MIMO technology can provide additional space degrees of freedom and increase the channel capacity greatly. However, it is a challenge to design the MAC layer when using MIMO in ad hoc networks. In FT (Follow Traffic) scheme, RTS/CTSs are sent simultaneously and used to estimate the traffic load. The node allocates the channel estimation resources in order of decreasing received power of RTS. Though FT scheme greatly improves the throughput, the fairness among nodes is declined and the complexity of receivers is increased. To address the problem, this paper presents a scheme of MAC layer based on random number allocation strategy, in which the node allocates the channel estimation resources in an increasing order of random number in each RTS. The throughput performance of the proposed scheme is suboptimal, but it ensures the fairness and reduces the complexity of the receivers.