Cooperation in clustered cellular systems

The performance limitations of conventional cellular wireless networks in terms of throughput and coverage are by now well recognized. This is due to extensive deployment of 2G and 3G systems, and has stimulated the search for new approaches to alleviate these drawbacks. A key technology that has been identified to fulfill this goal is cooperation, to be employed at either the base station (BS) or mobile station (MS) levels. As far as the BS level is concerned, multi-cell processing (MCP), sometimes referred to also as a distributed antenna system, prescribes joint encoding/decoding of the signals transmitted/received at the BSs through the exploitation of the high-capacity backbone connecting the BSs. Cooperation at the MS level in the context of cellular networks has been studied under different names such as mesh, hybrid or multi-hop cellular networks, and is based on specific forms of relaying by the MSs. Analysis of MCP (i.e., BS cooperation) has been so far mostly based on the assumption that all the BSs in the network are connected to a central processor via links of unlimited capacity. In this case, the set of BSs effectively acts as a multiantenna transmitter (downlink) or receiver (uplink) with the caveat that the anttenas are geographically distributed over a large area. Since the assumption of unlimited-capacity links to a central processor is quite unrealistic for large networks, more recently, there have been attempts to alleviate this condition by considering alternative models. In [1], [2], and [3] a model is studied where only a subset of neighboring cells is connected to a central joint processor. In [4] [5] (uplink) and [6] (downlink) a topological constraint is imposed where there exist unlimited capacity links only between adjacent cells, and message passing techniques are employed in order to perform joint decoding in the uplink. Reference [7] focuses on the uplink and assumes that the links between all the BSs and a central processor have finite capacity (finite-capacity backhaul). Finally, in [17] sequential DPC scheme of [8] is applied to local encoding with cluster codebook information (CI). BS cooperative technologies are enabled by the presence, respectively, of inter-BS (backbone) links that are not exploited by conventional cellular systems for the purpose of encoding. These links can be either wireless, orthogonal or not, thus possibly affecting the interference or bandwidth budgt of the network, or wired, hence requiring additional deployment efforts. Even with fully connected backhaul, sometimes we should not performance whole-network cooperation since deployment efforts may be very high when exchange information between all the BSs and cooperation gain may be trivial if including BSs which are far away from MSs. Hence, we expect Fig. 1. clustered multicell system