Diversity-and-multiplexing Tradeoff For The Multiple Nodes Cooperative Network

The development of next generation wireless communication systems demand high transmission rate and need to guarantee the quality-of-service. However, because of the scattering envirment the signals received at the destination may fading which degrades the performance of wireless communications. MIMO system can introduce spatial diversity to combat fading. Additionally, taking advantage of the rich scattering environment, MIMO increases spatial multiplexing.In this paper, we present three models. The first one is a MARC model. The second one is a multiple-relays model. The third model is a wireless communication system with four nodes. We consider their achievable DMT upper bounds, because it represents the relationship between diversity gain and multiplexing gain. On one hand, it is a powerful tool to evaluate the performance of different multiple-antenna schemes at high SNR. On the other hand, it is easy to tackle and is strong enough to show insightful comparisons among different cooperative schemes. The main research work of this article is as follows:First, we build a new MARC model, and the difference between our model and the traditional model is the users in our model are cooperative uses. In this model, all the nodes are signal-antenna and the relay is full-duplex. We divide the model into two systems:nonclustered system and two clustered systems (including case one clustered system and case two clustered system). Then we study the DMT upper bounds of the two systems and the achievable performance of DF protocol. The results show that:case one clustered system has the best DMT performance of this model. Because of the limitation of multiplexing gain, cooperative system can simulate MIMO system only when the multiplexing gain is low. The DMT performance of our model is better than the traditional model, due to the cooperative between users. The signal-antenna makes the DF protocol achieve the best DMT upper bound in the nonclustered system and case two clustered system.Second, we show a new multiple-relays model. In this model, relays can receive and decode the others signals. In this model, all the nodes are signal-antenna and the relay is full-duplex. We divide the model into two systems:nonclustered system and clustered system. Then we investigate the DMT upper bounds of the two systems and the achievable performance of DF protocol. The results show that:the DF protocol can not achievable the DMT upper bound when the system is clustered, but when the system is nonclustered it can achieve. Due to signal-antenna nodes, through clustered system can simulate MIMO system, it can only simulate MIMO when the multiplexing gain is low.The last model is a wireless model with four nodes. In this model, there are two transmission nodes and two receive nodes, and all nodes are signal-antenna. Therefore, we divide it into MIMO system, Broadcast channel, MAC channel and cooperative system. We discuss the DMT upper bounds of these four systems and show the comparison figure. From the comparison figure we can see that:Broadcast channel and MAC channel have the same DMT upper bound, and their DMT performance is worse than others. MIMO system has the best DMT performance. And cooperative system can not completely simulate MIMO system.This paper can help us to solve how to calculate the achievable DMT when nodes in a model have many antennas and the relays take half-duplex.