Non-Orthogonal Multiple Access
Minimizing Latency and Optimizing User Selection for 5G and Beyond Networks
May 13, 2019Edge-Caching tackling in 5G.
May 13, 2019
Non-Orthogonal Multiple Access (NOMA)
NOMA is a multiple access technique that has attracted a lot of interest from researchers and is considered a strong candidate for handling many of the requirements of 5G and beyond networks. Multiple access techniques allow multiple users to share the allotted resources in an effective manner. Recent studies categorize these techniques as Orthogonal Multiple Access (OMA) and Non-Orthogonal Multiple Access (NOMA).
In OMA, each user can exploit orthogonal communication resources within either a specific time slot (Time Division Multiple Access – TDMA), a frequency band (Frequency Division Multiple Access – FDMA), or a code (Code Division Multiple Access – CDMA) to avoid multiple access interference.
In contrast, NOMA allows multiple users to utilize the resources concurrently. Despite the fact that some degree of interference is introduced, the spectral efficiency is improved significantly. In addition, the number of active users that can be supported by the network increases significantly, hence the benefits of NOMA are of great interest. More specifically, the superiority of NOMA over OMA is determined as follows:
| Category | OMA | NOMA |
| Spectral efficiency and throughput | In OMA, a specific resource is assigned to each user regardless if it experiences a good or a bad channel condition; thus, the overall system suffers from low spectral efficiency and throughput. | In NOMA the same resource is assigned to multiple mobile users, with both good and bad channel conditions, at the same time. Hence, the resource assigned for the weak user is also used by the strong user, and the interference can be mitigated through SIC processes at the receiver. |
| User fairness, low latency, and massive connectivity | In OMA, the user with a good channel condition has a higher priority to be served while the user with a bad channel condition must wait for access, which leads to a fairness problem and high latency. This approach can not support massive connectivity. | On the other hand, NOMA can serve multiple users with different channel conditions simultaneously; therefore, it can provide user fairness, lower latency, and higher massive connectivity. |
In addition, it’s important to note that NOMA is compatible with the current and future communication systems and does not require significant modifications on the existing architecture. Also, a form of NOMA called multi-user superposition transmission (MUST), can be found in 3GPP Release 13.
Limitations:
However, NOMA is a recent technique and is still under development. There are also several aspects which can limit the which need to be resolved. The most important ones are listed below:
- The computational complexity is higher because each user has to decode not only its own signal, but the signal of other users too (to apply SIC), leading to a longer delay.
- The channel state information (CSI) must be known at the base station (BS), which could result in a feedback overhead.
- If any errors occur during the SIC process, then the error probability of successive decoding will be increased. To avoid such error propagation, the number of users utilizing the same resources must be limited.
- The users grouped together must satisfy considerable channel gain differences in order to improve the overall throughput, which is why NOMA is often applied in pairs.
Approach:
On of the main techniques of NOMA is through power domain, where different power levels are allocated for each user during downlink transmissions. More specifically, the higher the channel gain, the lower the power level. Assuming a pair of users, one strong user (User 1) and one weak user (User 2) where User 1 has better channel conditions than User 2 (h_1>h_2), then the power allocated to User 1 will be less than the power allocated to User 2 (p_1 < p_2). User 1 applies SIC to eliminate the interference caused by the weak user’s signal while User 2 treats User 1’s signal as noise, as shown in Fig. 1.
Figure 1: Power-domain NOMA process within in a dominant and a weak of users
In our work, we aim to resolve the limitations of NOMA by exploring the benefits of C-RAN architectures where the necessary information is collected and processed at a centralized BBU pool. More specifically, game theoretic approaches are used to allocate shared resources properly to allow users apply SIC without any errors or delays, select the most appropriate association of the users with the serving base stations and pair the NOMA users in a manner which will maximize throughput. Another aspect that we look into is ways to minimize the users’ energy consumption and the latency which is critical for new applications which are being envisioned such as the real-time tactile internet which promises to allow instant feedback from the network thus enabling virtual reality and mission-critical industrial automation.
