LTE uses a time-frequency-space grid to allocate resources to users in the DL and in the UL as shown below;
Users are allocated a certain number of resource blocks consisting of 12 consecutive subcarriers during a time slot of 0.5 ms. In the time domain, DL and UL transmissions are organized in radio frames of 10 ms, both for TDD and FDD. The same coding and modulation is applied to all groups of resource blocks belonging to the same L2 PDU (packets at MAC level) scheduled to one user within one TTI (subframe) and within a single stream. Therefore, the smallest granularity for assigning different modulations or profiles is 12 consecutive subcarriers during a 1 ms period.
In total, each resource block (PRB) consists of 84 resource elements (12 subcarriers during 7 OFDM symbols). However, some of the resource elements within a resource block will not be available as they are already occupied for reference signals or control information. Even though a downlink resource block is defined in terms of PRBs the downlink resource block assignment is carried out in terms of pairs of PRBs, where each pair consists of two, in the time domain, consecutive resource blocks within a subframe. Additionally, QPSK is used for all control information and data can be transmitted with different modulations like QPSK, 16-QAM, and 64-QAM. Thanks to the flexible resource allocation in time, frequency and space, and the scalable bandwidths supported by LTE, it is expected that at least 200 users per cell should be supported in the active state for a bandwidth of up to 5 MHz and at least 400 users for higher bandwidths. More users are supported in a passive (sleeping) state. In a 20 MHz bandwidth, which is the maximum allowed bandwidth, up to 100 Resource Blocks (RB) could be allocated to different users as indicated below.
A RB is the minimum frequency-time unit allocated to a single user and consists of 12 subcarriers. Assuming 400 active users per cell should be supported, each user should be allocated 1 RB every 4 consecutive sub-frames (1 ms TTI). The throughput per user in this worst-case scenario will be around 250 kbps. This throughput is for the downlink (DL) direction and will slightly decrease when considering the overhead of the control information. The cell-edge performance is a more challenging task since inter-cell interference can degrade the system capacity. A way to limit this interference is through intercell interference coordination, which can be achieved by restricting the transmission power in different parts of the spectrum in combination with a flexible frequency planning. A possible flexible frequency planning is a soft frequency reuse scheme which is characterized by a frequency reuse factor 1 in the central region of a cell, and a less efficient frequency reuse near the cell edge. When the mobile station (or UE) is near the base station (or eNB), the received power of the user signal is strong, and the interference from other cells is weak. So at the inner part of the cell, all the sub-carriers can be used to achieve high data rate communication as shown below;
In the above illustration, mobile stations 11 and 12 are connected with base station 1, while mobile stations 21 and 22 are connected with base station 2, and mobile stations 31 and 32 are connected with base station 3. It can be noticed that mobile stations 11, 21 and 31 are located at the intersection of 3 cells, while mobile stations 12, 22 and 32 are in the center of their respective cells. For the three mobile stations at the cell edge, different sub-carriers are allocated to avoid the co-channel interference. For the mobile stations near the base station, all the sub-carriers can be allocated and a lower output power can be used, to achieve a high frequency reuse. In this example, the frequency reuse factor is 3 for the cell edge and 1 for the inner part of the cell. For the inner part of the cell, through the limitation of the transmission power, some isolated islands are formed that do not interference each other. To conclude, the benefits of using this flexible frequency planning with soft frequency reuse scheme are:
• High bitrate at the cell center;
• Less interference at the cell edge, making easier the channel estimation, synchronization,
and cell selection;
• Overall improved spectral efficiency of the network.
To achieve the benefits of soft frequency reuse, a tight coupling between frequency allocation and output power is needed. This can be achieved by a thorough network planning carried out during the roll out phase of the network. However, with the trend towards smaller and smaller cells, a more distributed approach towards power and frequency planning is needed. In recent years, Self-management (Self-X) technologies that fully automate the tasks of managing (i.e. configuring, monitoring, and optimizing) a cellular network are emerging as an important tool in reducing costs for the service provider and will be a distinguishing feature of LTE networks. In this Self-X technology is used for the self-configuration of fractional frequency reuse for LTE that uses only local information. Their solution achieves a trade-off between locality of the information, optimality and stability of the solution.
