LaNWiFi.NeTWorld's Wireless Technology Center // Wireless For Freedom.

• Tradeoff of power. Since one signal needs to be sent to at least two different nodes, more power is needed to send data from source to destination. On the other hand, with diversity, the transmit power at each node can be smaller than in the noncooperative mode. Thus, it is necessary to develop a power allocation algorithm such that minimum transmit power is used to maintain the user cooperative diversity.
• Tradeoff of transmission rate. In cooperative communication, a node needs to relay other nodes’ data and also transmit its own data. Thus, its transmission rate is reduced. However, due to diversity, channel coding rates can be higher, which increases the transmission rate. Thus, the actual transmission rate may not be reduced. In order to keep the transmission rate as high as possible, we need to consider tradeoff between diversity and reduced chance of transmission.
• Interference. With cooperative diversity, interference may also be increased, because the same signal is sent more than once in the network. However, diversity may reduce the transmit power level, which can compensate for the increased interference.
• Cooperation assignment scheme. This is concerned with finding other cooperative nodes for each node so that diversity can be achieved. In a one-hop infrastructure based network, cooperation assignment may be just a simple task. However, in a multihop distributed network such as WMNs, cooperation assignment is rather complicated, because it has to take into account many factors such as diversity gain, power, interference, and even fairness among nodes.
• New requirements on network nodes. Although diversity can be achieved through cooperative communications, algorithms are still needed to retrieve original data from multiplexed signals. This requires additional processing power on either transmitter or receiver. The functionality of cooperative communications may also need change of hardware in each node.

• Amplify and forward scheme. In this scheme, the partner node amplifies the received signal together with noise and forwards the amplified signal. The drawback of this scheme is that amplifying and forwarding analog signals is a nontrivial requirement for hardware. The advantage of this scheme is that no inter-user channel state information is needed, and conventional channel estimation can be used.
• Decode and forward scheme. In this scheme, a node decodes the signal and then retransmits it. An example of this scheme is presented in. The drawback of this scheme is that a node may not be able to correctly detect its partner’s signal. In this case, relaying signals will propagate errors. Another drawback of this scheme is that inter-user channel information is needed at the receiver.
• Coded cooperation. Each node’s data stream is encoded via a channel coding scheme into blocks. The codeword is partitioned into two segments with N1 and N2 bits, respectively.When a node transmits data, it first sends N1 bits in the codeword, and also decodes data from its partner node. If decoding is successful, the node will calculate the second segment of its partner node and then transmits the N2 bits. Otherwise, it will transmit its own second segment. This scheme does not need inter-user channel information. However, since cooperation is conditional, the receiver needs to know whether transmit nodes have cooperated or not. Other coded cooperation schemes are also discussed in.
• Selection relay. In order to avoid error propagation, a threshold test must be performed before relaying signals. When it is satisfied, the relay is done. Otherwise, a node needs to enter noncooperative mode. In this sense, coded cooperation is also a type of selection relay scheme.

Several schemes have proposed to detect original signals from multiplexed signals. The simplest scheme is to let relayed signals and original signals arrive at different time intervals. However, such a TDMA scheme may not be efficient, and also has difficulty in time slot allocation. We can also use CDMA to multiplex relayed and original signals. In addition, frequency division multiple access is possible. Whatever multiple access scheme is used, algorithms need to be developed to detect original signals with as good performance as possible. To date, user cooperative communications have been proved effective in wireless networks. However, the research is still in its preliminary stage, for three reasons. First, most of the existing schemes have only considered the problem of how to effectively forward data by cooperative nodes. How to assign cooperative nodes so as to optimize power, transmission rate, and interference requires new algorithms. Second, no practical solution has been proposed yet. For example, most existing algorithms assume central control is available, simple network topology has been considered, and only simple multiple access is considered in these algorithms. Such constraints render the existing algorithms unsuitable for WMNs. Third, many existing schemes rely on the knowledge of inter-user channel station information, which may not be available. New algorithms that do not or only partially rely on such information need to be developed. Nevertheless, user cooperative diversity is a promising technology, because: a) it potentially enables low-cost network nodes to have antenna diversity and thus increase overall network capacity; b) it can work together with multi-antenna systems to further increase network performance; c) it can help to resolve collision without using reservation based MAC protocol or collision avoidance procedures.